Indian Philosophy: A Reader [1st Edition] 0367147890, 9780367147891, 100072817X, 9781000728170, 0367816512, 9780367816513, 1000728102, 9781000728101, 100072803X, 9781000728033

Table of contents :
Introduction: some critical concepts of Indian philosophy by JONARDON GANERI
I: THE CONCEPT OF PHILOSOPHY
1 On the concept of philosophy in India by BIMAL KRISHNA MATILAL
2 Rationality in Indian philosophy by ARINDAM CHAKRABARTI
3 Intellectual India: reason, identity, dissent by JONARDON GANERI
II: PHILOSOPHY AS A SEARCH FOR THE SELF
4 The Upaniṣads by JOEL BRERETON
5 Hidden in the cave: the Upaniṣadic self by JONARDON GANERI
III: BUDDHIST PHILOSOPHY OF MIND
6 Indian theories of mind by GEORGES DREYFUS AND EVAN THOMPSON
7 From the five aggregates to phenomenal consciousness: towards a cross-cultural cognitive science by JAKE H. DAVIS AND EVAN THOMPSON
8 Subjectivity, selfhood and the use of the word ‘I’ by JONARDON GANERI
IV: SELF
9 The self as a dynamic constant: Rāmakaṇṭha’s middle ground between a Naiyāyika eternal self-substance and a Buddhist stream of consciousness-moments by ALEX WATSON
10 Arguing from synthesis to the self: Utpaladeva and Abhinavagupta respond to Buddhist No-selfism by ARINDAM CHAKRABARTI
11 ‘I am of the nature of seeing’: phenomenological reflections on the Indian notion of witness-consciousness by WOLFGANG FASCHING
V: METAPHYSICS
12 The Nyāya-Vaiśeṣika theory of universals by KISOR CHAKRABARTI
13 Objectivity and proof in a classical Indian theory of number by JONARDON GANERI
VI: EPISTEMOLOGY
14 A realist view of perception by BIMAL KRISHNA MATILAL
15 Nyāya perceptual theory: disjunctivism or anti-individualism? by ANAND JAYPRAKASH VAIDYA
VII: LANGUAGE
16 The context principle and some Indian controversies over meaning by B. K. MATILAL AND P. K. SEN
17 Bhartṛhari’s view of sphoṭa by BIMAL KRISHNA MATILAL
18 “Ākāśa” and other names by JONARDON GANERI
VIII: LOGIC
19 Semiotic conceptions in the Indian theory of argumentation by BIMAL KRISHNA MATILAL
20 Jaina logic and the philosophical basis of pluralism by JONARDON GANERI

Citation preview

INDIAN PHILOSOPHY The selection of essays in this volume aims to present Indian philosophy as an autonomous intellectual tradition, with its own internal dynamics, rhythms, techniques, problematics and approaches, and to show how the richness of this tradition has a vital role in a newly emerging global and international discipline of philosophy, one in which a diversity of traditions exchange ideas and grow through their interaction with one another. This new volume is an abridgement of the four-volume set, Indian Philosophy, published by Routledge in 2016. The selection of chapters was made in collaboration with the ­editors at Routledge. The purpose of this volume is to reintroduce the heritage of ­‘Indian Philosophy’ to a contemporary readership by acquainting the reader with some of the core themes of Indian philosophy, such as the concept of philosophy, philosophy as a search for the self, Buddhist philosophy of mind, metaphysics, epistemology, language and logic. Jonardon Ganeri is a philosopher, specializing in philosophy of mind and in South Asian and Buddhist philosophical traditions.

I NDIAN P HILOSOPHY A Reader

Edited by Jonardon Ganeri

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 selection and editorial matter, Jonardon Ganeri; individual chapters, the contributors The right of Jonardon Ganeri to be identified as the author of the editorial material, and of the authors for their individual chapters, has been asserted 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 has been requested for this book ISBN: 978-0-367-14789-1 (hbk) ISBN: 978-0-367-81651-3 (ebk) Typeset in Bembo by codeMantra

CONTENTS

Acknowledgements viii Introduction: some critical concepts of Indian philosophy

1

JO NA R D ON GA N ER I

I:  THE CONCEPT OF PHILOSOPHY 1 On the concept of philosophy in India 7 BI M A L K R I SH N A MATI L A L

2 Rationality in Indian philosophy 17 A R I NDA M CH A K R A B A RTI

3 Intellectual India: reason, identity, dissent 36 JO NA R D ON GA N ER I

II:  PHILOSOPHY AS A SEARCH FOR THE SELF 4 The Upaniṣads 53 JO E L BR ER ETON

5 Hidden in the cave: the Upaniṣadic self 65 JO NA R D ON GA N ER I

III:  BUDDHIST PHILOSOPHY OF MIND 6 Indian theories of mind 87 G E O R GES D R EYFUS A N D EVA N TH OMPSO N

v

viCONTENTS

7 From the five aggregates to phenomenal consciousness: towards a cross-cultural cognitive science 116 J A K E H . DAV I S A N D EVA N TH OMPSON

8 Subjectivity, selfhood and the use of the word ‘I’ 129 J O NA R D ON GA N ER I

IV:  SELF 9 The self as a dynamic constant: Rāmakaṇṭha’s middle ground between a Naiyāyika eternal self-substance and a Buddhist stream of consciousness-moments 145 A L E X WATSON

10 Arguing from synthesis to the self: Utpaladeva and Abhinavagupta respond to Buddhist No-selfism 166 A R I NDA M CH A K R A B A RTI

11 ‘I am of the nature of seeing’: phenomenological reflections on the Indian notion of witness-consciousness 182 WO L FGA N G FA SCH I N G

V:  METAPHYSICS 12 The Nyāya-Vaiśeṣika theory of universals 203 K I S O R CH A K R A B A RTI

13 Objectivity and proof in a classical Indian theory of number 218 J O NA R D ON GA N ER I

VI:  EPISTEMOLOGY 14 A realist view of perception 241 B I M A L K R I SH N A MATI L A L

15 Nyāya perceptual theory: disjunctivism or anti-individualism? 255 A NA N D JAYPR A K A SH VA I DYA

VII:  LANGUAGE 16 The context principle and some Indian controversies over meaning 281 B. K . MATI L A L A N D P. K . SEN

17 Bhart ṛhari’s view of sphoṭa B I M A L K R I SH N A MATI L A L

302

CONTENTS

vii

18 “Ākāśa” and other names 313 JO NA R D ON GA N ER I

VIII:  LOGIC 19 Semiotic conceptions in the Indian theory of argumentation 335 BI M A L K R I SH N A MATI L A L

20 Jaina logic and the philosophical basis of pluralism 344 JO NA R D ON GA N ER I

Index 359

ACKNOWLEDGEMENTS

The publishers would like to thank the following for permission to reprint their material: Johns Hopkins University Press for permission to reprint Jonardon Ganeri, ‘Intellectual India: Reason, Identity, Dissent’, New Literary History, 40, 2009, 247–263. Columbia University Press for permission to reprint Joel Brereton, ‘The Upanishads’, in W. T. de Bary and I. Bloom (eds), Eastern Canons: Approaches to the Asian Classics (New York: Columbia University Press, 1990), pp. 115–135. Oxford University Press for permission to reprint Jonardon Ganeri, ‘Hidden in the Cave: The Upaniṣadic Self ’, in The Concealed Art of the Soul (Oxford: Oxford University Press, 2007), pp. 13–38. DISCLAIMER

The publishers have made every effort to contact authors/copyright holders of works reprinted in Indian Philosophy: A Reader.This has not been possible in every case, however, and we would welcome correspondence from those individuals/companies whom we have been unable to trace.

viii

Introduction Some critical concepts of Indian philosophy Jonardon Ganeri

I

The selection of essays in this volume aims to present Indian philosophy as an autonomous intellectual tradition, with its own internal dynamics, rhythms, techniques, problematics and approaches, and to show how the resources of this tradition have a vital role in a newly emerging global and international discipline of philosophy, one in which a diversity of traditions exchange ideas and evolve through their interaction with one another. ‘Indian philosophy’ is the designation for a vast and still vastly under-­researched body of inquiry into the most fundamental topics ever to engage the human mind. Reflecting India’s north-­western borders with the Arabic and Persianate worlds, its north-­eastern boundaries with Tibet, Nepal, Ladakh and China, as well as the southern and eastern shores that afford maritime links with the lands of Theravāda Buddhism, Indian philosophy has been written in many languages, including Pali, Prakrit, Sanskrit, Malayalam, Urdu, Gujarati,Tamil,Telugu, Bengali, Marathi, Persian, Kannada, Punjabi, Hindi,Tibetan, Arabic and Assamese. From the time of the British colonial occupation, it has also been written in English. It spans philosophy of law, logic, politics, environment and society, but is most strongly associated with wide-­ranging discussions in the philosophy of mind and language, epistemology and metaphysics (how we know and what is there to be known), ethics, metaethics and aesthetics, and metaphilosophy. The reach of Indian ideas has been vast, both historically and geographically, and it has been and continues to be a major influence in world philosophy. In the breadth as well as the depth of its philosophical investigation, in the sheer bulk of surviving texts and in the diffusion of its ideas, the ­philosophical heritage of India easily stands comparison with that of China, Greece, the Latin West or the Islamic world.

1

2

JONARDON GANERI

It is striking how twentieth-­century Indian thinkers, including Gandhi, Nehru, ­ mbedkar and Tagore, and philosophers like Krishnachandra Bhattacharyya and A ­Anukulchandra Mukerji, who formulated the philosophical grounds of an intellectual decolonization, drew inspiration from the rich heritage of philosophy in India. Moreover, as the discipline of academic philosophy begins to address its legacy of exclusion and to evolve into a genuinely diverse and pluralistic field, India’s past represents an unquantifiably precious part of the human intellectual biosphere. For those who are interested in the way in which culture influences structures of thought, for those who want to study alternative histories of ideas, and for those who are merely curious to know what some of the world’s greatest thinkers have thought about some of the most intractable and central philosophical puzzles about human existence and experience, Indian philosophy is a domain of unparalleled richness and importance. Meanwhile, for those who draw upon India’s intellectual past to give sustenance to contemporary programmes of social and political reform, it is essential that the whole of this past, the fullness of its vast diversity and richness, is available and acknowledged. And in its potential for cross-­fertilization with ideas from other philosophical cultures – Greek, Chinese, European, African, Arabic and Anglophone – Indian philosophy has a richness that any creative thinker can and should draw upon. II

This volume is an abridgement of the four-­volume set Indian Philosophy (2016). The purpose of that set was to provide a one-­stop shop of secondary literature in Indian philosophy, a resource that any instructor or lecturer could use in offering a module on Indian philosophy within a philosophy curriculum.This new, abbreviated, edition aims to acquaint the reader with some of the core concepts of Indian philosophy. Chapters 1 to 3 concern the very conception of philosophy. Chapter 1 introduces readers to the idea of philosophy in India.There are significant issues around the question of the concept of philosophy in different intellectual cultures. Chapter 2 develops the idea that philosophy is a practice of the use of reason and looks at Indian conceptions of rationality. This theme is developed further in Chapter 3 and extended in a number of ways. Chapters 4 and 5 introduce readers to one of the earliest forms of philosophical speculation in India, the Upaniṣads. The Upaniṣads are beautiful but elusive texts, but one thing that is clear is that they place the search for the true self at the core of their philosophy. The topic of the nature of the self is continued in Chapters 6 and 7, which present ­Buddhist ideas about the nature of the mind and its subjective character. Yet B ­ uddhists hold no monopoly in the philosophy of mind, and Chapters 9, 10 and 11 explore fascinating ideas about the nature of the self in the non-­Buddhist philosophical schools of Kashmiri Śaivism and Advaita Vedānta. The volume then turns to topics in metaphysics and epistemology. Chapters 12 and 13 discuss the fascinating metaphysical theories of the school of Vaiśeṣika. Chapters 14 and 15 introduce readers to Indian epistemology, especially the Nyāya theory of perception as a source of knowledge.

INTRODUCTION

3

The final section of the book is devoted to issues in philosophy of language and philosophical logic. Chapters 16, 17 and 18 provide discussion of some of the most original and important Indian contributions to the study of language and meaning. Chapters 19 and 20 do the same with respect to Indian theories of argumentation. This volume thus covers a large range of topics, certainly enough to form the basis of a semester-­long course in Indian philosophy. We recommend that instructors combine the readings in this volume with audio content from the podcast series A History of ­Philosophy without any Gaps, the India section of which is jointly written by Peter ­Adamson and Jonardon Ganeri (freely available at https://historyofphilosophy.net and now also published as a book, Classical Indian Philosophy: A History of Philosophy without Any Gaps,Volume 5, Oxford University Press, 2020).

P A R T

I

THE CONCEPT OF PHILOSOPHY

CHAPTER

1

On the concept of philosophy in India Bimal Krishna Matilal

Source: Philosophical Essays: Anantalal Thakur Felicitation Volume (Calcutta: Sanskrit Pustak Bhandar, 1987), pp. 358–369. One often faces, while talking about ‘Indian philosophy’, the following question:1 ‘Is there anything called ‘philosophy’ in India, specifically in the way the term is understood in modern West?’ One common answer is to cite the Sanskrit term (darśana.) However, I propose to answer it differently, for ‘darśana’ in my view comes very close to the English word ‘view’, or maybe ‘philosophical viewpoint’. I shall start with a counter question; ‘Could we define “philosophy” even in its modern Western sense, a definition that will be unanimously accepted by all philosophers in the West?’ The Greek etymological sense (philein, to love—sophia, wisdom) is usually referred to for a quick answer to the question: what is philosophy? And at a popular level,‘philosophy’ still stands for private wisdom. At a technical level, however, the term is understood to mean the criticism and systematization of all knowledge (drawn from empirical science) as well as knowledge itself. Or, as a recent Western philosopher writes, it ‘is characteristically argumentative and essentially directed towards the determination of what logical relations do and do not obtain’ (A. Flew, A Dictionary of Philosophy, p. ix). This is not exactly ‘love of wisdom’, but not entirely unrelated to it. Paradoxically, Sanskrit has a better term for ‘philosophy’ in the modern sense. I believe it is better to start with the term ‘ānvīkṣikī’, a term used as early as Kauṭilya to refer to a discipline similar to what we call philosophy today. Kauṭilya, after enumerating the four main branches of learning, vidyāsthāna (‘philosophy’, religious education or Scriptures, trade-­and-­commerce, and the science of polity), defines ‘ānvīkṣikī’ as follows (this view is repeated also in many places in the later literature such as Manu):2

7

8

BIMAL KRISHNA MATILAL

Investigating, by means of reasoning what is good and what is evil in the context of scriptural/ religious education, material gain and loss in the context of learning trade-­and-­commerce (vārtā = economics), what is good and what is not so as a policy in the context, of learning polity and public behaviour—philosophy (ānvīkṣikī) confers benefit upon people, keeps mind steady in prosperity and adversity and imparts proficiency in thought, speech and action.

The same term is explained by Vātsyāyana at c. ad 300 (repeated by Jayanta in c. ad 850 and others) as ‘examination of what is obtained through perception (observation) and what is stated in the scriptures, by the pramāṇas (adequate evidence)’. In fact, it is the contention of Vātsyāyana that ānvīkṣikī is to be distinguished from other disciplines, specially from the ‘spiritual’ texts or the books of wisdom such as the Upaniṣads by virtue of its being characterized by a study of the pramāṇas, etc. For the Naiyāyikas, however, the term means the discipline that is called Nyāya-­prasthāna. But if we are not parochial, like the Naiyāyikas, we can take it to mean any school of ‘argumentative philosophy’, as Kauṭilya did. An incidental comment to guard against any misunderstanding here. The well-­known German indologist, H. Jacobi3 in 1911, interpreted the term ānvīkṣikī as philosophy and attempted to find in it a programme for an independent, theoretically-­oriented ‘science’. He was however severely criticized by P. Hacker,4 in 1958, who argued that Jacobi in his conclusion was wrongly influenced by his own ‘prejudice’, i.e., his fascination for the Western emancipation of philosophy from theology. In other words, Jacobi, according to Hacker, wrongly superimposed a Western situation on ancient India. I do not wish to enter into the terminological dispute here, for I believe it to be a non-­starter in this context. Nor am I searching for a Sanskrit term that would be exactly parallel to ‘philosophy’ in the West, for it is, by the very nature of the situation, bound to be an impossible task. In how many different ways has ‘philosophy’ been used in the West over a period of, say, a thousand years? Human beings decidedly are not so uniformly non-­different from each other that we may expect to find an exact copy of one historical contingency of the Western tradition in another, in this case Indian, tradition. On the other hand, the point to be made is that reflective self-­understanding and rational wisdom were present as much in the Occident as in classical India. And this gives us the rationale for talking meaningfully about ‘philosophy’ in classical India. Two further observations in this connection would be in order. First, Hacker was wrong in his critique of the point (apparently made by Jacobi) that ‘philosophy’ (ānvīkṣikī) in India constituted an autonomous domain. For there was a stream of critical and reflective tradition in India, which might not admittedly have developed into a completely open-­ended free enterprise of thought (as ‘philosophy’ in the West, particularly in the sense of the so-­called Cartesian freedom from prejudice, is understood to have). However, even this idea of ‘freedom from prejudice’ in the Cartesian conception of ‘philosophy’ is a construction of the Western historiographers. This idea has recently been called into question. I believe the Indian thinkers beginning from the Buddha down to Raghunātha perceived themselves as ‘free unprejudiced thinkers’ as much as their Western counterparts, in the early, medieval and even modern periods, perceived (and still do consider) themselves to be such thinkers. Such self-­perception, which is in part self-­deception, is an indispensable ingredient of even a ‘philosophical’ mind. However, our task is not to delve into the psychology of philosophers, whether Western or Indian.

PHILOSOPHY IN INDIA

9

Second, it seems undeniable that there was a conscious and sustained effort on the part of a number of authors (whom we tend to call philosophers today) of ancient and classical India to re-­define the scope of their śāstras ‘treatises’ and describe their method as rational, unbigotted and non-­authoritarian. In some cases, there has been explicit denunciation of the religious rituals and the fruits they are supposed to bring about (see Īśvarakṛṣṇa, below). However, there were certain presuppositions, which we today in the later part of the twentieth century do, and those ancients and medievals did not, think of as ‘prejudices’. However, in our modesty, we may admit that although we, as moderners, are free from many ‘prejudices’ which the premoderners and traditional Oriental societies had unconsciously nurtured, we can hardly claim that we have no prejudice whatsoever. If modesty allows this much, then it would be too finicky to demand an exact word in ancient Sanskrit, whose meaning would totally coincide with whatever meaning we at this time in history, care to attach to the word ‘philosophy’. What would be reasonable to enquire into and expect however is whether there was a tradition in India’s intellectual history (of the classical period) which primarily concerned itself with rational inquiry and consciously wanted to avoid religious dogmas. This is what I will try to show. In the opening lines of his celebrated Nyāyabhāṣya, Vātsyāyana Pakṣilasvāmin has remarked:5 Since there is success in activities based upon the knowledge of the object/goal (artha), the knowledge that is generated by pramāṇas (means or evidence for knowledge), (we can say that) the pramāṇas are connected (invariably) with the object/goal.

This rather important sentence has presented exegetical difficulties. There are five words in this sentence (in Sanskrit), of which two are compound words. There are about five terms which need explanatory notes because of their being slightly technical in nature. Of them, pramāṇa and artha are the two most difficult terms for the modern translators. In the above (translation), I have given the general sense of the sentence as far as practicable. It is a very condensed statement of an elaborate argument against the ‘sceptics’ (i.e. the Mādhyamikas). As I have argued elsewhere, the Mādhyamika fulfilled the role of the sceptics in the Indian tradition, although they were not sceptics themselves; they were Buddhists.6 Vātsyāyana gives the gist of his argument in a few sentences that follow, the opening sentence being a condensed version of it, almost ‘aphoristic’ (sūtra-­like) in its nature:7 There is no knowledge of the artha (object/goal) without a pramāṇa. (And) there is no success in activities without some knowledge of the artha. Having known the artha by pramāṇas, the subject (jñātṛ = knower) desires either to obtain or avoid that artha.The physical effort of that person (the subject) prompted (caused) by his desire to obtain ‘success’ of such activities means their connection with the results (fruits). Having desired either to obtain or to avoid, the subject makes ‘physical’ efforts and thereby either obtains or avoids the artha. The word ‘artha’ means both happiness as well as its causal factors and unhappiness as well as its causal factors. The arthas of pramāṇas are in fact innumerable, the living creatures (to whom they belong) being also innumerable.

10

BIMAL KRISHNA MATILAL

Saiṃīhā has been translated as ‘physical efforts’, but the word itself is ambiguous and may in fact mean all three kinds of efforts noted by the Sanskrit philosophers, physical, mental and verbal (by speech). Vātsyāyana, here, has himself explained the compound word ‘success-­in-­activities’ as well as the enigmatic ‘artha’. The concept of artha is based upon what we may call a tacit philosophical theory. The idea is that everything in this world, every object, every fact, can be the cause of the happiness or unhappiness of some creature or other in some way or other.Vātsyāyana also compares a pramāṇa with a lamp light (dīpa). Just as a lamp shows things are existent as well as establishes the non-­existence of those that are not shown, similarly pramāṇas reveal those that are existent, and by implication establish the non-­ existence of those that have no pramāṇa or evidence. Uddyotakara (c. ad 600), in the beginning of Nyāyavārttika, adds  a  few paragraphs where he deals with the question; what should a philosophical treatise concern itself with. In a clear and concise manner, he gives an outline of the nature of philosophy as a discipline, and shows admirably well how this theoretical discipline is connected with the practical life. He argues that a theory of knowledge and evidence (pramāṇa) is an integral part of philosophy as a discipline and that such a theory is also connected with a theory of action. I quote these paragraphs in full:8 This ‘treatise’ (śāstra) speaks about what is good for human beings. The special purpose of a ‘treatise’ is to explain the true nature of such things (artha) as are not known through ordinary means of knowledge, such as perception and inference. And only such students are entitled to study this ‘treatise’ as are endowed with the intellectual capacity to comprehend the nature of things which they have not learnt through ordinary means of knowledge, such as perception and inference. There are four kinds of people: those who have (fully) comprehended (the nature of things), those who have not comprehended at all, those who are in doubt, and those who have (completely) misunderstood. Among these four, one who has (fully) comprehended (i.e. one who knows) is in a position to teach; all the others have the need for knowledge, and hence are those who should be taught. The latter group of people learn through perception when they depend upon sense-­ object contact; they learn through inference when they depend upon perception of evidence and remembrance (of the invariance relation). Finally with regard to matters (not amenable to perception and inference, e.g., the concept of the highest good), they require verbal testimony; this is supplied by the ‘treatise’ (śāstra).

In the above I have followed the text in Dr A.Thakur’s edition.There is a variant preferred in two other editions,V.P. Dvivedin’s as well as the Calcutta Metropolitan’s, according to which the definition of ‘a treatise’ would be: The special purpose of a treatise is to explain the true nature of such things as are (already) known through ordinary means of knowledge, such as perception and inference.

Thus this text drops the negative particle and the two editors of the Calcutta Metropolitan edition added a footnote showing preference for the variant without the negative particle in spite of the contrary evidence from  a large number of manuscripts. Hence,

PHILOSOPHY IN INDIA

11

apparently in their opinion, the śāstra or the treatise deals with matters that are already known (presumably to the teachers) through perception and inference. Since the students (the three latter groups of people) lack the required knowledge, the teachers teach them through śāstra what they (the teachers) already know. However, let me continue with the remarks of Uddyotakara: [What is good?] The good consists of both happiness and avoidance of evil. In fact there are two types of good: those that are perceived and those unperceived. The perceived good is pleasure or well-­being. The avoidance of evil is unperceived. This avoidance again is of two kinds: the non-­ultimate and the ultimate. The non-­ultimate is accomplished through the removal of such temporary causes of evil (pain) as  a  thorn (or poison). The ultimate avoidance (of evil) is realised through the cessation of twenty-­one perpetual sources of evil (unhappiness). They comprise: (1) the body, (2–7) six faculties of sensing and cognising, (8–13) six types of (the given, the sensibles and the thinkable), (14–19) six types of awareness (five kinds of perceptual awareness and the non-­perceptual awareness), (20) happiness, and (21) unhappiness (evil). Of these twenty-­one varieties of evil, the body is regarded as evil because it is the abode of all unhappy experience; the faculties, the given objects and the awareness of them are so regarded because it is through their agency (or instrumentality) all unhappy experience (evil) arises. (Even) happiness is regarded as evil (a disvalue) because, it is invariably accompanied by some source of evil or other. And unhappiness is evil by its very nature. The cessation of all these is possible by the abandonment of their sources, viz., dharma and a-­dharma. The future dharma and a-­dharma(merit and demerit) can be abandoned by not engaging oneself in future (good and bad) action. The already acquired dharma and a-­dharma can be ‘abandoned’ by exhausting them through the consequential experience of happiness and unhappiness.

Here it is interesting and at the same time important to note the thematization of the twenty-­one items—all noted as duḥkha, the evil or the undesirable items which humans should try to avoid. Body etc. are instrumental for our mostly unhappy experience. Even the so-­called happy experiences are invariably intermixed with unhappy ones. Hence ‘philosophy’ or ‘treatise’ is supposed to teach us the way to the ultimate good (niḥśreyasa). The good life, according to this philosophical outlook, is one that turns us away from these duḥkhas, the undesirables, and propels us permanently into the direction of the ‘ultimate good’. It is useful to compare this notion of ungood or duḥkha with the three kinds of duḥkha in the Sāṃkhya school as well as the three-­fold duḥkha of the Buddhists. However, let me continue further with Uddyotakara. There are two kinds of people. Some are attached (to happiness etc.); the others are non-­ attached. Attachment means clinging to the objects (of enjoyment and suffering). The attached persons cling to such objects. Non-­attachment means lack of clinging to such objects. The unattached persons do not cling to such objects. Accordingly, there are two types of activities: … the non-­attached acts only in one way. He tries to avoid the ‘undesirable’ (what will cause bondage). He acts with the design ‘I will avoid the undesirable’. He does not have any particular attachment to anything. The ‘attached’ person, on the other hand, acts in two ways. He tries to obtain the object he desires and shuns that which is undesirable … He acts from both motivations. He proceeds to obtain with the design ‘I wish to obtain this desirable object’ and he avoids with the idea ‘I wish, to shun this undesirable object’. Again,

12

BIMAL KRISHNA MATILAL

the action of the attached has two modes. One mode of action flows from pramāṇa and hence is crowned with success, and the other flows pseudo-­pramāṇa (that which masquerades as a means of knowledge, but in fact a means of some erroneous judgement) … Both pramāṇa and pseudo-­pramāṇa yield certainty about the generic character of the object (in both cases we ascertain ‘it is an apple’). Both therefore lead to action. In the first case, it is successful (i.e. the desired object is obtained) in the second case, it is not.

All these statements are reasonably non-­parochial and it seems to me that a general theory of what ‘philosophy’ means or what we would like to call ‘philosophy’ in the Indian context is outlined here very clearly. In the Sāṃkhya, which was probably the earliest philosophical system in India, a similar notion of ‘philosophy’ or ‘treatise’ was present. For example, Īśvarakṛṣṇa (c. ad 400) begins his treatise by explicit reference to duḥkha as the necessarily given human condition and its possible amelioration.9 Being tormented by the three kinds of duḥkha (unhappiness: 1. pertaining to our body and mind, 2. caused by other beings, 3. caused by nature and the supernatural), there arises a desire to know (a question) about their causes. If these causes are due to the perceived (mundane) factors, we take care of them accordingly (by using antidotes). But no, the unhappiness does not vanish permanently (hence the desire to know the means of its cessation persists) (verse 1). Even the scriptural acts are on a par with the perceived factors. For they are impure, unfair, and their results are perishable. Hence the opposite (imperishable state as the result) is better, which arises due to knowledge of the manifest and unmanifest (verse 2).

It is argued here that for the permanent removal of the persistent duḥkha or unhappiness, one should turn to the study of the ‘treatise’ or philosophical treatise, for neither ordinary measures to alleviate suffering nor the religious rituals or acts to reach heaven in the next life are capable of guaranteeing the permanent cessation of unhappiness.We may note that the first verse signifies that for removing mundane sufferings we do not resort to ordinary means to destroy their causal factors, but the persistent recurrence of sufferings persuades some of us to look for a permanent remedy. The second verse shows that turning to religion (Vedic religious rituals etc. in this case) would be no better, for (a) some religious acts call for violence (in this case killing animals) and hence are morally impure, (b) some are unjust and iniquitous (for they supposedly make some people rich and powerful and others poor and weak in the next life), and (c) the projected outcome of all of them (limited time in heaven) is perishable. Hence the wise turn to śāstras (in this case, philosophical treatises) to find out a way to the ultimate good. Denunciation of the organized religion of the day is prominent here. I shall now introduce Jayanta, who is somewhat parochial, but in presenting the pūrvapakṣa (opponent’s position) he openly admits that: (a) ānvīkṣikī, which is nyāyavidyā or nyāyavistara, is one of the vidyāsthānas, disciplines or subjects to be learnt (Jayanta quotes from Yājña-­valkya smṛti), and (b) there were at least six different schools of ānvīkṣikī at the time, called ṣaṭ-­tarkī (they were Sāṃkhya, Bauddha, Jaina, Cārvāka or Lokāyata, Vaiśeṣika and Nyāya).10 It was significant that Haribhadra (who predates Jayanta) included all these

PHILOSOPHY IN INDIA

13

six in his Ṣaḍdar-­śanasamuccaya (a collection of six systems of philosophy of darśana), but also added Mīmāṃsā or the Jaiminīya mata. Towards the end he tackles the question of the number ‘six’ mentioned in the title as follows:11 Thus I have described briefly the philosophies of the āstikas.Verse 77 cd. Those who regard that there are only five systems of āstika philosophies, do not think that the Naiyāyika view differs from the Vaiśeṣika view.Verse 78. For them, the number ‘six’ in ‘six philosophies’ is fulfilled by adding the view of the Lokāyata. Hence I describe this view.Verse 79.

Thus I believe that at some point in history there was not much difference between the use of the term ‘ṣaṭ-­tarkī’ and ‘ṣaḍ-­darśana’, and hence the equation ānvīkṣikī = darśana (a rūḍha śabda) = philosophy was possible. The use of the word ‘darśana’ in this context is intriguing. It may be that from this time onwards the word came to be used in the sense of a philosophical view. This is a far cry from the way some moderners and Westerners would like to interpret the term ‘darśana’. Taking a cue from the etymology of the word, they would translate it as ‘vision’ or ‘intuition’ and thereby they ride their hobby horse by arguing that Indian philosophy in this way is very different from the analytical and rational approach of Western philosophers. To repeat Vātsyāyana, ānvīkṣikī was called ānvīkṣikī because of its special treatment of the theory of pramāṇas as evidence. A note on the word ‘āstika’ is in order here. It does not mean a theist or a follower of the Vedas. It means only those views which acknowledge the existence of an ‘after-­world’ (paraloka) as well as the moral principle that dispenses pāpa and puṇya (demerit and merit) according to good or bad actions. This is how the term has been explained in the commentary. The Lokāyatas were the only nāstikas for they argued that these entities, afterworld, demerit and merit, did not exist. In today’s terminology, not simply an atheist, but any moral or religious sceptic would be called nāstika. More important, however, is the omission of trade-­and-­commerce and polity from the list given by Yājñavalkya. Jayanta gives an answer to this puzzle. He says that this list of learnable subjects is compiled with the idea that they all should be teaching us about the four goals of human life (puruṣārtha), which include mokṣa as well as the doctrine of the unobservable fruits (demerit and merit) of our actions. The two, trade and polity, are squarely within the scope of observable results and purposes. Hence they have been left out in the list of Yājñavalkya. However, we should remember that Kauṭilya’s treatise was concerned mainly with the three goals, dharma, artha, and kāma and this is also corroborated by the Mahābhārata account of the origin of rājadharma, in Chapter 5 of the Śāntiparva (Mbh. 12.5929). Gaṅgeśa12 (c. ad 1350) located ānvīkṣikī among  a  list of eighteen subjects of learning (vidyāsthāna), which included all the fourteen of the Yājñavalkya list and added: Āyurveda, Dhanurveda, Gāndharvavidyā and Arthaśāstra (see Mathurānātha’s commentary). Gaṅgeśa asserted that ānvīkṣikī was the most honourable vidyā or subject of study (abhyar-­hitatama), which was only  a  resonance of Kauṭilya’s favoured and oft-­quoted pronouncement (see Rucidatta’s Prakāṣa-­ṭīkā). In Gaṅgeṣa’s list however the two, vārtā (trade and commerce) and polity, will be included under arthaśāstra, and thus it avoids the anomaly of the Yājñavalkya list.

14

BIMAL KRISHNA MATILAL

Gaṅgeśa wrote only on the four pramāṇas, his theology was only an appendix to the Anumāna-­khaṇḍa. And this was recognized as an epoch-­making book. Hence it would be interesting to see how he tried to connect his study of pramāṇas with the ultimate goal, mokṣa. In fact he says very little. The general idea was already expressed by Uddyotakara. The śāstra generates the right kind of knowledge of things (which includes knowledge of the self) and this in turn sequentially destroys the process of transmigration and the consequent sufferings or duḥkha. Mokṣa is thus automatically achieved in the way it has been described in Nyāya-­sūtra 1.1.2. Two points may be of some significance for us in the opening sentence of Gaṅgeśa. First, he mentions that the whole ‘jagat’ (world) is steeped in suffering and hence ‘philosophy’ (ānvīkṣikī) provides the road to their release.The commentators disagreed about interpretation of this word ‘jagat’. Raghunātha said that it meant the souls that are in the process of transmigration (Jagat = moving in). Mathurānātha13 argued that the rūḍha meaning of jagat is anything (vastu), any object not just souls. But certainly pots etc. are not suffering and hence would not be released. Thus, says Mathurānātha, the suffering state qualifies the things (vastu) and separates the suffering souls from those that are not suffering (pots etc. and free souls and God). And the release pertains only to the suffering souls, for whom the śāstra ‘treatise’ is composed. Second, Dharmarāja Rāmakṛṣṇa mentions an opponent’s view (pūrvapakṣa) which has a sociological significance. ‘Jagat’ means all sufferers and they must include the women and śūdras who have no right to study the Vedas. This view is clearly ascribable to Raghunātha, for he glosses ‘jagat’ as ‘saṃsāryātmajātam’, and this is quoted by Rāmakṛṣṇa. Hence according to Raghunātha’s cryptic statement, Gaṅgeśa was saying that ‘philosophy’ or ānvīkṣikī is open to all, not restrictive to the male members of the three varṇas. Unfortunately such informal social critique often goes unnoticed by us today (specially by modern historians and sociologists). It would be stupid to neglect the strong undercurrent of criticism of religious and social practices by the classical thinkers. In the old Sāṃkhya of Īśvarakṛṣṇa we have explicit rejection of these practices. Raghunātha’s tongue-­in-­cheek rejection of caste distinctions is also significant in this context. Rāmakṛṣṇa, however, being an orthodox vaidika, readily rejected Raghunātha’s view.14 Gaṅgeśa’s more significant statement (for our own purpose) comes at the end of the first paragraph. Referring to the sixteen categories (padārtha) enumerated in the ­Nyāyasūtra of Akṣapāda 1.1.1., he says:15 I am going to deal with the nature of pramāṇa here (the first one) among these (sixteen categories) because on pramāṇa depends the establishment of everything.

Thus a rationale was given for his principal concern for pramāṇa or the source of knowledge. Mathurānātha comments that this exclusive concern with the theory of knowledge is vindicated by the fact that this is well in line with what is called ānvīkṣikī ‘philosophy’, which again produces the fruit of the right kind of knowledge that is instrumental to the ultimate goal of mankind, mokṣa or cessation of suffering. The connection seems to us to be a bit tortuous, but not entirely fictitious. Obviously, there is a touch of parochialism here when it is claimed that knowledge of a particular system of metaphysics is the liberating knowledge. Even without taking notice of

PHILOSOPHY IN INDIA

15

parochialism or sectarianism one can say that knowledge of things as they are (tattva) is the liberator, for knowledge gives control, specially control over our actions and if a particular type of action creates ‘bondage’, another type of action (or non-­action) may lead to freedom or cessation of suffering. And if the liberating knowledge cannot be gained except through ānvīkṣikī or ‘philosophy’, then the position is proved. I am inclined to treat pramāṇa as evidence for knowledge but not in the sense (at least not entirely in the sense) of the justificatory ground for knowledge. This may amount to  a  misconstrual, according to one view. Pramāṇa is what ‘makes’ knowledge. Since knowledge is always an episode (an inner event in Indian philosophy, in fact, a sub-­category of mental occurrence) pramāṇa has also a causal role to play. It is the ‘most efficient’ cause of the knowledge-­episode. Knowledge yields de-­termination of an object x or a fact that p (artha-­pariccheda) as the result, and a pramāṇa is ‘instrumental’ in bringing about that result. This is the causal role of  a  pramāṇa. Similarly when a knowledge-­claim is made about x or about the fact that p, the opponent may demand a pramāṇa for the claim made. For example, the Buddhist will ask the Nyāya opponent ‘what is your pramāṇa for the claim that the soul exists?’ This question has been sometimes amplified as: is it from a perceptual pramāṇa or evidence (pratyakṣataḥ)? Or from an inferential pramāṇa (anumānataḥ)? In this context, pramāṇa is obviously fulfilling the evidential role. Hence a pramāṇa is both a cause and a ‘because’—has both causal and evidential role. In India, soteriology must be understood in an extended sense. It depends on one’s seeing the world properly and acting according to its value. This is the general characteristic shared by all the systems. Such action is conditioned by the seeing. And pramāṇas are instrumental (kāraka-­pradarśaka) in this seeing as well as ‘persuaders’ (pravartaka) in so seeing. In this second role, a pramāṇa is an evidence. In Buddhist philosophy too, Diṅnāga asserted in the beginning of his monumental work, Pramāṇasamuccaya, that in the person of the Buddha all the pramāṇas were realized and thus he connected soteriology with philosophy.16 What if this seeing or knowledge is called in question. If the knowledgehood of knowledge is doubted (as it could be in most cases), then, the Nyāya school says, we need an explicit evidence to establish the knowledgehood through an inference. It is in some of these cases that Vātsyāyana appealed to the test of ‘successful activity’. There may be other ways17 by which the knowledgehood may be inferentially established. Or, it may be self-­established or self-­evident, if we believe in the svataḥ theory. But in each case it seems that in knowing knowledge (i.e. prāmāṇya-­jñapti) pramāṇa has also an evidential role. According to the Nyāya position, knowledgehood is a contingent property of an awareness, and to know that I know I must know that this contingent property characterizes the awareness in question (and knowing simply the awareness itself would not do) and for this I need a pramāṇa, or an evidence. This is, however, not exactly foundationalism, nor the ‘justified true belief ’ situation. Still, this is conceivably a new but totally intelligible model for epistemology which needs further exploration. On the whole, I believe it has been established here that there was a rich and multi-­dimensional stream of thought in the Indian tradition, which was not entirely different from what has come to be called ‘philosophy’ in the West.

16

BIMAL KRISHNA MATILAL

NOTES AND REFERENCES 1. This happened, for example, in a conference during the Festivals of India in France, organized at the College International de Philosophie, Paris under the leadership of Jacques Derrida, during 23–26 October 1985. 2. Kauṭilya Arthaśāstra (ed. R.P. Kangle, Bombay), ch. 1. 3. H. Jacobi, ‘Zur Frühgeschichte der indischen Philosophie’, 1911, reprinted in B. Kolver (ed.), Herman Jacobi, Kleine Schriften, Wiesbaden, 1970. 4. P. Hacker, ‘Ānvīkṣikī’, WZKSO 2 (1958). 5. Vātsyāyana, Nyāyabhāṣya (ed. A. Thakur, Mithila, 1967), opening sentence. 6. Matilal, B.K., Perception, Clarendon, Oxford, 1986, pp. 46–65. 7. Vātsyāyana, op. cit., first paragraph. 8. Uddyotakara, Nyāya-­vānttika (ed. A. Thakur), opening paragraphs. The Metropolitan edn. is from Calcutta University, edited by T. Tarkatīrtha and A.M. Tarkatīrtha.V.P. Dvivedin’s edn. is from Chowkhamba,Varanasi, 1916. 9. Sāṃkhyakārikā (ed. R. Tripathi),Varanasi, 1970. 10. Jayanta, Nyāyamañjarī (ed. Gaurinath Sastri), Chaukhamba Sanskrit Series, Varanasi, 1982, pp. 9–10. See also Yājñavalkya Smṛti (1949, Bombay, ed. N. Acharya), ch. 1, verse 3. 11. Haribhadra, Ṣaḍdarśanasamuccaya (Chowkhamba, Benares, 1957), pp. 63–4. 12. Gaṅgeśa, Tattvacintāmaṇi, Pratyakṣa, with Māthurī (Delhi reprint, Motilal, 1973), pp. 114–16. 13. Gaṅgeśa, Māthurī, pp. 115–16. 14. Gaṅgeśa, Tattvacintāmani, Pratyakṣa, with Rucidatta and Rāmakṛṣṇa (Kendriya ­Sanskrtavidhyapitha, Tirupati, 1973), pp. 115–16. 15. Gaṅgeśa, Māthurī, p. 116. 16. See M. Hattori, Dignāga On Perception, Harvard University Press, Cambridge (Mass.), 1968. 17. See Matilal, Perception, ch. IV.

CHAPTER

2

Rationality in Indian philosophy Arindam Chakrabarti

Source: Eliot Deutsch and Ron Bontekoe (eds),  A  Companion to World Philosophies (­Oxford: Blackwell Publishers, 1997), pp. 259–278. INTRODUCTORY REMARKS

You cannot say “thank you” in Sanskrit. It would be ridiculous to deduce from this (as ­William Ward,  a  British Orientalist, did in 1822) that gratefulness as  a  sentiment was unknown to the ancient Indian people. It is no less ridiculous to argue that rationality as a concept is absent from or marginal to the entire panoply of classical Indian philosophical traditions on the basis of the fact that there is no exact Sanskrit equivalent of that word. For one thing, there are several words for the science or art of reasoning: for example, “ānvikṣikī,” “tarkaśāstra,” “nyāya.” (And one of these – namely, “ānvikṣikī,” and its role in the Indian metatheory of branches of learning or knowledge – will occupy us in a separate section of this article). There are also very ancient words for the institutions of rational debate and public problem-­solving contests (for example, a “brahmodya”), like the famous one reported in the Bṛhādaraṇyaka Upaniṣad (third book) in which Yājñavalkya steals the show in the court of the philosopher-­king Janaka. There are also words for the special form of reflecting by means of anticipated “pro” and “contra” arguments (“ūhāpoha,” “­manana,”“yukti-­vicāra”), which one is urged to cultivate as part of a contemplative culture. Second, even if there were no such closely cognate words, that would hardly license the conjecture that the concept is foreign to the Vedic people. Of course, identity-­criteria for concepts are hard to formulate. But, as the following discussion would demonstrate, articulated concepts of what makes a belief, an action, an interpretation, a preference, a choice of means or an end reasonable could be detected everywhere in classical Indian thought. Those concepts may not be easily recognizable as concepts of rationality, since unlike the 17

18

ARINDAM CHAKRABARTI

standard Western concepts of rationality, the typically Indian notions of rationality are, on an average, non-­hedonistic, non-­individualistic, non-­positivistic, and aim at surrendering the personal ego to an impersonal tradition or to some universal consciousness. In this article, I shall first try to diagnose four major worries that can make even unprejudiced surveyors of Indian thought wonder whether the concept of rationality, with its positive value-­overtone, is at all compatible with the general tenor of classical Indian philosophies. Although it is advisable to be suspicious of any talk of the “general tenor” of all Indian philosophies (except Buddhism, because I have avoided discussing it here), these deep worries are well grounded. By trying to face them, I think, we can get a better grip on the special contribution that Indian ways of thinking – in their inexhaustible variety of in-­house disagreements and their passion for intra-­traditional polemics – made to the multivalent concept of rationality. After responding to these four major worries, I take up specific aspects of the Indian theoretical engagement with logical, epistemic, hermeneutic, ethical, aesthetic, and soteriological rationality. Besides demonstrating the diversity and enormity of this field, I hope my account will suggest the way in which the tensions between scriptural testimony and reason, mysticism and logic, poetry and analysis, action and theory were celebrated, partially resolved and partially allowed to remain unresolved by the ancient and medieval Indian thinkers, and how some contemporary Indian interpreters are coming to terms with these inner tensions within their own pluralistic cultural legacy. HOW ARE HUMANS SPECIAL?

One important way in which the adjective “rational” has been understood in the West is as standing for the differentiating feature distinguishing human beings from other animals. It must have been part of popular thinking even in ancient India that cognitive and ratiocinative powers distinguish humans from beasts because a very important Hindu religious text (Durgā Saptaśatī, a part of Mārkanḍeya Purāṇa) warns against this idea and says in no uncertain terms: “True, humans are knowledgeable, but they are not the only ones; for even birds and beasts all have knowledge of some sort.” Then it goes on to give examples of apparently self-­sacrificing behavior on the part of bird-­mothers feeding their young knowing, as it were, that they are totally dependent at that stage. Because of the continuity across species underlying the reincarnation theory (which enabled the wise, rational, morally sensitive Buddha to be born in many sub-­human bodies or even Vishnu the god to assume the form of a fish or a pig), on the one hand, and the orthodox Vedic society positing caste divisions within human society as a natural given, on the other, the divide between humans and sub-­humans was never such a major theme among the traditional Indians. One can, however, point out three very striking ways in which the privilege of man has been thematized in the tradition. First, in the Aitareya Āraṇyaka (which is a part of the Vedic corpus) man (puruṣa, gender-­neutral) is said to be privileged because in him the self (ātman) is more manifest; he alone discerns what he perceives and can liberate his understanding and speech from the immediate needs of hunger and thirst. Saying that beasts and birds are unable to foresee and plan, this text comments beautifully: “Man knows tomorrow … and by the mortal aspires after the immortal.”

RATIONALITY IN INDIAN PHILOSOPHY

19

Second, Śabara in his commentary to the Mīmāṃsā Sūtra (VI, 1, 5) remarks on this typically human capacity to wait and (as the Ṛgveda X, 117 says, warning the short-­sighted ungiving amasser of wealth) “look down the longer path.” Rising above the proximate (āsanna) self-­interest, a human being can perform “sacrifices” – both in the literal and the ritualistic sense – for the sake of “unseen” results in a remote future. A sacrifice (yajña) is therefore taken by the Bhagavad Gītā (4.28–33) as the human rational activity par excellence, so much so that even “trying to ascertain the meaning of Vedic sentences by logical arguments” is called “knowledge-­sacrifice” (nyāyena vedārtha niścayaḥ jñānayajnaḥ). At the very end of Nirukta,Yāska (400 bce) personifies Reasoning as a wise seer (ṛṣi), one “tarka-­ṛṣi” who would guide the interpretation of obscure Vedic texts by humans. Thus while both animals and humans are conscious pleasure-­seekers, only humans are capable of dharma  –  considerations of piety and morality, right or wrong conduct. Hence the popular Sanskrit adage, found in some versions of the Hitopadeśa: “Without dharma, a man eating, sleeping, and engaging in sex is no different from a beast.” The link between the ethico-­r itualistic concept of dharma and rationality is very clear in texts like Yogavāśiṣṭha Rāmāyaṇa, which devotes an entire chapter (II, 14) to the nature and importance of vicāra (reflective analysis) without which human life is pointless. This is how the chapter starts; “With an intellect purified by understanding of the scriptures, the person who is aware of what causes what (kāraṇa-­jña), must constantly examine and analyze himself.” The chapter ends after 45 verses explaining the indispensability of reasoned reflection for moral, spiritual as well as worldly life with this echo of Socrates: “It is better to be born as a mud-­frog or as a slime-­worm or as a snake in a cave than to be an unreflective unexamining human being.” But all this praise for intellectual and moral superiority is qualified by a most interesting twist given by the Mahābhārata to its explanation of why “there is nothing nobler than humanity” in the story of a poor scholarly Brahmin (Mbh. XII, 174). Knocked over by a rich man’s carriage, he wants to die on the street out of utter frustration. Indra, the King of Gods takes the form of a jackal and tells him that he should make the best of his life as a human being because he is far better off than beasts who cannot even scratch themselves properly with their tails. For several verses this story goes on to explain why humans are great because they have a pair of hands with ten pliable fingers with which they can take out thorns, and make tools and shelters and clothing. Thus not so much as homo sapiens but as homo faber do humans rule over other creatures. What does this have to do with rationality? Foreshadowing what I say, in the next paragraph, about rationality and the use of hand gestures to communicate (for example, in the Indian but not exclusively Indian ways of greeting with folded hands, giving reassurance, and so on), let us compare this “manual” view of man with the following remark by Kant: The characterization of man as a rational animal is already present in the form and organization of the human hand, partly by the structure and partly by the sensitive feeling of the fingers and finger-­tips. (Anthropology, p. 323, italics Kant’s)

The point that the Mahābhārata makes about the hands making humans “free” to clean or protect themselves or to dominate or torture others is exactly the point that Kant also

20

ARINDAM CHAKRABARTI

makes in the remark: “By this, nature has made him fit for manipulating things not in one particular way but in any way whatsoever, and so for using reasons” (ibid.). A dimension of these wonderful possibilities that the hands open up for us which Kant did not speak of (and perhaps would not have considered part of rationality) is the endlessly expressive and creative use of the hands, palms and fingers in dance and drama that Bharata, in the fourth chapter of his magnum opus Nāṭya Śāstra, lists as 140 different “poses” forming the basic postural alphabet (karaṇas and angahāras). That would be a typically Indian extension of the “reason of the hands.” It is by systematic reflection and debate about every aspect of life that the classical Indian learned traditions self-­consciously tried to display the distinctiveness of human existence without making too much of a fuss about just the cognitive specialty of humans. An extreme example of the use of definition and dialectic in every sphere of rational thought – from etymology to erotica – can be given from the Kāmasūtra (VI, 2, 27–8) where Vātsyāyana considers the view of those who “argue that massage is a form of embrace because it is a tactile contact.” He observes that “there are three reasons why it is not so, because [in love-­practice] they occur at different times, because they serve different purposes, and unlike an embrace which is ideally reciprocal a massage is non-­mutual.” The commentator adds a little reductio to the effect that “if any tactile contact were an embrace, then even a kiss would be so.”When the zeal to theorize and support every theory with a reason was carried to this extreme in ancient Indian thought, one cannot but be surprised at the fact that “lack of theoretical orientation” is usually the first charge that is brought against the Indian mind. FOUR WORRIES DISSIPATED

Of the four features of Indian thought that could be pointed out as reasons for skepticism about the very idea of an Indian conception of Rationality – the first that draws our attention is the alleged practical or goal-­oriented character of Indian philosophy. Not only are pleasure (kāma), power/wealth (artha), piety/righteousness (dharma), and final liberation from suffering (mokṣa) distinguished as the four – and only four – alternative and actually pursued goals of life, but even branches of knowledge or subjects of study are divided accordingly. Thus we have kāmaśāstra, arthaśāstra, dharmaśāstra, and a specially prestigious mokṣaśāstra dealing with these four goals respectively. Philosophy is often identified with the last of these, so that there remains no room for the pure theories of logic, mathematics, knowledge, reality, or morality, undertaken simply for the sake of intellectual satisfaction. Along with this ancient overarching theory of the fourfold “ends of man” (puruṣārtha), philosophical treatises followed the general pattern of opening with a statement of “purpose” or “use” (prayojana) – for “why should an intelligent person undertake a study or an action until he is told what purpose it serves?” (Kumārila Bhaṭṭa, Śloka vārtika, 1.1.12) How can the out-­and-­out theoretical notion of reason and rational inquiry develop in an intellectual atmosphere so obsessed with practical concerns or with the goal of the end of suffering? There are several ways in which this worry can be allayed. As early as the fourth century bce, Kauṭilya, in his Arthaśāstra, divides disciplines (vidyā) into four: scripture (the three Vedas, trayī), agriculture and commerce (vārtā), politics and

RATIONALITY IN INDIAN PHILOSOPHY

21

public administration (daṇḍa-­nīti), and finally  –  “the light of all other disciplines, the methodology of all other practice, and the foundation of all moral virtues” –  ānvīkṣikī, the investigative reflective science which examines beliefs acquired through observation and testimony by the means of correct knowledge (pramāṇaih arthaparīkṣaṇam). The very recognition of a metascience (ānvīkṣikī) which would examine what is moral and what is immoral in the Vedas (dharmā dharmau trayyām), what is efficient and what is inefficient in the sciences of material acquisition, and good and bad policies in the science of government – weighing their strength of evidence by arguments – and the identification of this metascience with philosophy (examples given by Kauṭilya include sāṃkaya, yoga, and the lokāyata materialistic philosophy) unquestionably proves that even the recognition of the purposefulness of rational inquiry or action was part of a theoretical orientation of these ancient Indian thinkers. The practical purpose of a study itself became a theoretical topic of discussion. Indeed, Jayanta (c. 950 ce) begins his book Nyāyamanjarī (Logic Blossoms) by constructing a paradox of inquiry (rather like the Platonic “Meno’s Paradox”) based on this requirement that one know the use of every study that one begins. If you insist on first knowing its use and then being interested in a subject, then you are involved in a circularity. After mastering the subject alone you know exactly what purpose it can really serve; but you are not even interested to start learning the subject unless you first know what its use is. So the required knowledge of use cannot be a thorough critical knowledge on pain of this circularity.

One of the reasons why the tension between theory and practice never became important in classical Indian thought is that practice, as it was laid out in the arts (of dancing, building, medicine and poetry), never was blind, and the theories themselves claimed to be somehow livable in practice. The distinction, however, between action (karma) and knowledge (jñāna) has been recognized since the Upaniṣadic times. A healthy competition between the ritualists, on the one hand, who took sacred descriptive speech to be subservient to Vedic injunctions concerning what is to be done, and the followers of the path of pure knowledge, on the other, who took prescriptions to be subservient to metaphysical parts of sacred speech, continued for at least a thousand years. Both the ritualists and the metaphysicians, nevertheless, had their own ānvīkṣikīs or analytical methodologies. So, to quote J. N. Mohanty, “the Hindu mind was constantly engaged in theorizing about practice” and also committed to the idea that “a purely theoretical cognition will lead to the satisfaction of the highest practical interest” (Mohanty, 1995). The scheme of the four ends of life (puruṣārthas) to which we alluded above is itself a rational theorization about what humans live for. Recent (that is, late twentieth-­ century) Indian philosophers have taken up this fourfold scheme for thorough critical examination. Daya Krishna, for instance, has provocatively asked, “What is the puruṣārtha of intellect (buddhi),” given that the intellect just wants to raise questions and get the answers right, not desiring sensual pleasure or political or economic power, not trying to be virtuous, and not always aiming at freedom from suffering? We should not be too far from the mainstream of Indian thought if we answer this question by positing knowledge of the self or understanding of texts to be the puruṣārtha of intellect. Alternatively, we could develop K. C. Bhattacharya’s (Studies in Philosophy, p. 142) idea that rational reflection is

22

ARINDAM CHAKRABARTI

primarily reflection on pain and is “a freeing process” – thus linking it up with liberation. The link between an intellectual discipline, such as āvīkṣikī, and self-­knowledge or textual interpretation will become clear as we deal with the second and third “worries” concerning rationality in India. The second major worry is this: how could autonomous rationality have developed in Indian thought given that most philosophizing was done with an allegiance to the unquestioned authority of the Vedas or some other root-­text the truth of which was taken for granted? Where the forces of tradition and verbal testimony are so dominant, how could reason – in the Western sense of the term – flourish? Three lines of response, of varying degrees of power, and by no means mutually exclusive, can be adopted in the face of this second worry. First, while it is correct that mere or “dry” reasoning (śuṣka tarka) has been belittled by great Indian thinkers like Śaṃkara because it is groundless, unstable and conflict-­generating (Brahmasūtrabhāṣya, II.1.11), or because, as Bhartṛhari noted before him, “what expert reasoners have concluded with great logical acumen and effort is disproved by yet other more expert logicians,” not all Indian thought is blindly supportive of scriptural authority. Having abused the authors of the three Vedas as impostors and cheats, the materialist-­skeptical Cārvāka philosophers from very ancient times rejected all trust in religious texts as irrational. Of course, in their case, even inferences, and especially inductive generalizations, are epistemically unjustified, and insofar as testimony is reduced to a form of inference, our reliance on testimony too loses all rational respectability. Right from the Buddha’s own sermons up to the sophistication of Buddhist epistemology in the Yogācāra-­Sautrāntika school, the Buddhist mind shows opposition to unexamined “say so” as evidence. The Buddha urges his disciples not to believe his own words upon the basis of his personal authority, but to test them by reasoning and individual experience. Accordingly, only perception and inference are admitted as sources of knowledge in Buddhist epistemology, and testimony is either rejected or reduced away. Similarly, in what has been called the “Tradition of Rationalist Medicine” (Chattopadhyaya, 1980, pp. 85–115), appealing to religious or scriptural authorities in the context of clinical practice has been regarded by Caraka (from the very early Christian era) as committing the fallacy of irrelevance. We shall see, in a subsequent section on ancient Indian medical reasoning, how “medical integrity” was supposed to consist in reliance on empirical data, inductive probability, practical efficacy, and not on religious authority. Within the mainstream orthodox schools, Śaṃkhya – in spite of its lip-­service to the Vedas and the “word of the expert” as sources of knowledge – very clearly relied on its own variety of reflective reasoning as the sole means of attaining such knowledge as would lead to the pure and permanent cessation of suffering. In the two opening couplets of Sāṃkhya Kārikā, observed worldly means of removing pain and scripture-­prescribed ritualistic means of removing pain are both rejected as unsatisfactory, because even the heavenly pleasures (after death) promised as rewards for the performance of Vedic rituals are exhaustible, mixed with pain and surpassable in degree. The only method of attaining an inexhaustible, unmixed and unsurpassable state of freedom from pain is rational reflection on the distinction between the manifest (effects), unmanifest (cause) and consciousness (which is neither effect nor cause). Sāṃkhya, therefore, is at heart an out-­and-­out

RATIONALITY IN INDIAN PHILOSOPHY

23

reason-­based system of thought with its own basic presuppositions, such as the three fundamental guṇas and the doctrine of the pre-­existence of effect in the material cause, defended by a series of internally coherent arguments. Second, even Vedānta and Mīmāṃsā – the two pillars of Vedic orthodoxy – assign a crucial role to reasoning and critical argumentation in extracting the correct meaning from sacred sentences of the “heard” revelation (śruti). Far from being antagonistic, reason and scripture coexist peacefully together in coupling compounds strewn all over Vedānta literature (for example, s̄ruti-­yukti, tarkāgama, śāstranyāya). What is this assisting role that reason plays inVedic hermeneutics? In the Upaniṣads – the philosophical cream of theVedas – one finds statements like “you are that (Brahman)”. In order to make sense of such identity claims, the reader must first “distill” the meaning of “you,” which is coreferential with the reader’s (ideally, listener’s) use of “I.” Causal links are established by what in Vedānta is called the method of presence in presence and absence in absence (anvaya-­vyatireka). Now, signification or designation is taken as a special case of a causal link, because there is a lawlike connection between the utterance of a word and the consequent grasp, by the hearer, of a meaning. The Upaniṣads start from a proto-­materialistic conception of the self (the referent of “I”) as the food-­constituted body. Śaṃkara the commentator uses this method of presence and absence to reject, one after another, these “object”-­natured candidates for selfhood – the body, the life-­breath, the inner sense, the intellect – because the self seems to be present even in the absence or non-­functioning of these elements (in death, dreams, deep sleep, and so on). This method of elimination leaves only a pure non-­individuated subjective consciousness as the possible meaning of “I.” A similar isolation of relevant signification is performed on the word “that” (which directly stands for God and the totality of physical and mental entities of the universe). When the direct or primary referents of these two terms are seen to be in partial conflict, because the individual embodied “I” is not prima facie identical with “that” world or all that there is, the method of rational exegesis is employed. Secondary significance of words is generally derived from the literal sense by extension (for example, “crying over spilt milk” comes to include a whole lot of spilt other things) and elimination (for example, in “the opinion of the house,” “house” does not signify house at all). Thus through “retaining a part and rejecting a part” of their literal sense, the words “I” and “that” are taken in their secondary significance. That part of their distinct literal meaning where they intersect: namely, pure subjective consciousness, is taken as the emerging oblique meaning of the scriptural identity statement. What can never be spoken of – the Ātman-­Brahman – could thus be got at indirectly by its only testimony, the sentences of an authorless revelation. This is an oversimplified summary of the intricate interpretive technique through which reasoning is used to distill the indirect meaning of the “great sentences” of the Veda. Thus the text is trusted as the sole proof of the Ātman-­Brahman (the Self which is All) but it is subjected first to a tradition-­tested method of critical scrutiny. It is only relentless reasoning which can help us hold on to the distinction between the self (reality) and the not-­self (appearance), and without such reasoned discrimination, no blind parroting of the scriptures would get us anywhere. As far as the role of testimony is concerned, even scripture is a ladder to be kicked away after the saving knowledge of non-­duality dawns. Thus Śaṃkara’s faith in the “truth” of Vedic text is also ontologically provisional. Tarka (reasoning) may be baseless by itself (apratiṣṭha),

24

ARINDAM CHAKRABARTI

but even śruti (scripture) is, in the final analysis, ignorance that helps cure ignorance.While discussing the “instability” of autonomous reasoning, Śaṃkara, interestingly, considers and dismisses an objection which it is worthwhile to mention. “This alleged refutability and non-­decisiveness of reason,” the objection goes, “should be recognized as a good feature rather than a weakness, insofar as it keeps room for correction and improvement.” After all, if you have conflicting Vedic texts, reasoning is your only basis for adjudication! Third, we could question the very assumption that relying primarily on an impersonal unquestionable tradition is necessarily irrational. There are two ways in which not only the compatibility between reason and testimony but the essentiality of commitment to the tradition as a necessary condition for rationality can be brought out.The first way can be called the Nyāya–Dummett way and the second the Mīmāṃsā–Gadamer way. By emphasizing the irreducible role of knowledge from words (śabda pramāṇa) in the acquisition and use of language, Nyāya epistemology exposes  a  fundamental error of Lockean individualistic epistemology. As Michael Dummett remarks, It is not a rule of etiquette, or a device for saving time, that we should accept what others tell us: It is fundamental to the entire institution of language. (Motilal and Chakrabarti, p. 266)

There is no rationality without social interaction, because as Wittgenstein showed us, no one can be a private rule-­follower. But there is no social interaction without understanding of others’ speech. And, as Dummett and Davidson, in spite of other major differences, both insist  –  there is no understanding of others’ speech without  a  basic presumptive trust in their testimony. It follows, therefore, that there is no rationality without a basic trust in the veracity of competent speakers  –  “be that  a  sage,  a  lay Aryan or  a  mleccha foreigner” (to quote Vātsyāyana, the fifth-­century Nyāya commentator) – unless there is reason to suspect ignorance or deceit or lack of commitment. While this way of putting testimony back into the heart of rationality proceeds through the inescapable trustworthiness of fellow-­speakers of a language, the other way, adopted by Kumārila Bhaṭṭa and recently articulated by John Taber using insights from Gadamer, turns on an underlying distrust of individual speakers and treats a speaker-­less body of received tradition to be the only possible source of moral knowledge. Perception or empirically grounded inference never gives us any knowledge of what ought to be done. The verdict of our conscience or moral emotions is highly unreliable. The only ineluctable source of knowledge of right and wrong action, under the Mīmāṃsā view, which is firmly rooted in the epistemological doctrine of the intrinsic validity of all knowledge, is the impersonal objective (beginningless) prescriptive sentences handed down by one’s own cultural tradition. The only way meaningful speech could be unreliable is by being spoken by fallible individuals. If it is not spoken by anyone – as Mīmāṃsā takes the Vedas to be – then it is intrinsically knowledge-­yielding.The knowledge it yields is also unique, because from no other source of knowledge can you have any rational insight into morality. To quote Gadamer, “The real force of morals … is based on tradition. They are freely taken over but by no means created by  a  free insight grounded on reasons.” (Truth and Method, p.  281) Of course, neither Gadamer nor any modern person can swallow the orthodoxy of Kumārila that the Vedic tradition alone is the source of all moral knowledge. But it may be necessary to rectify the Enlightenment idea that a fully autonomous external critique of tradition is

RATIONALITY IN INDIAN PHILOSOPHY

25

possible or desirable purely on the basis of personal rationality. A creative but sympathetic understanding of the traditionalist ethics of Mīmāṃsā may enable us to appreciate why it is perfectly rational for a Veda-­rooted Indian to assert “I ought to feed the guest first because the Veda Says so” and then claim, like Wittgenstein, that he has hit the rock-­bottom of reason-­giving and that is where his spade turns. But by over-­emphasizing discursive knowledge – whether derived from reasoning or from the word of the reliable authority – am I not falsifying the very spirit of Indian intellectual traditions, which eventually aspire after non-­discursive, ineffable, direct mystical insight into a reality that transcends reason? How could any conception of rationality have developed in a philosophical milieu so deeply devoted to some kind of ecstatic vision where “the subject and the object are … eternally and absolutely lost in unity, and the din of phenomenal existence is forever hushed in the calm of sweet repose?” This is the third “worry” that I wish to address. Mahamahopadhyaya Gopinath Kaviraj, whom  I  have just quoted, compiled, in his classic 1924 paper “The Doctrine of Pratibhā in Indian Philosophy,” all the references to such supra-­rational supra-­sensuous intuitive awareness of “all things past, present and future in a simple flash” which was called “pratibhā.” Even from that pro-­mystical survey of Sanskrit sources, two things become clear: first, not all schools of philosophy believed in the possibility or centrality of such supramental experience; second, even the schools that relied heavily on pratibhā, like the Grammarians, did not understand by it anything which goes against or beyond reason. Before discussing the anti-­mystical position of Mīmāṃsā, let us take a quick look at the role of the so-­called “immediate spiritual experience” (aparokṣa anubhūti) in Advaita Vedānta. Contrary to the claims of Neo-­Hindu writers like Radhakrishnan, Śaṃkara nowhere claims that non-­dualism is based on direct mystical experience. To quote Halbfass: that experience which the Veda itself teaches as  a  transcendent soteriological goal, the sheer undisguised presence of Brahman, should not be confused with “personal experiences”. … Instead of being a documentation of subjective experience, the Veda is an objective structure which guides, controls and gives room to legitimate experience, as well as legitimate argumentation.  …  It is an objective, transpersonal epiphany, an authorless, yet didactically well-­organized body of soteriological instruction. (Halbfass, 1988, p. 388)

Rāmānuja, the most prominent rival-­Vedāntist, is even more unequivocal in asserting that no yogic trance can give knowledge of Brahman. Such Yogic visions, according to him, are, after all, results of intense imagination and vivid reproduction of previous sense-­ experiences. Such transient experiential states cannot be depended upon as a source of knowledge. The most orthodox traditionalists are the harshest critics of mystical intuition or even divine omniscience. Kumārila ridicules the possibility that “all things could be experienced directly at once in a synoptic vision” as on a par with the possibility that we could “hear colors.” Kaviraj, who is pained by this “bitter opposition,” finds the following rationale behind it. As we have already noted above, the impersonal and exclusive authority of the Veda is the corner-­stone of Mīmāṃsā. Personalities, however elevated or divine, even of an alleged Deity, are taken as sources of fallible and limited experiences. As Kaviraj

26

ARINDAM CHAKRABARTI

insightfully remarks, “The very fact of being a subject involves the inevitable relativity of consciousness fatal to omniscience.” So “omniscient person” is an oxymoron. Also, given the presence of the Vedic corpus as the ineluctable source of impersonal and hence intrinsically valid knowledge, any all-­knowing person – divine or human – is strictly redundant. Finally, if we look at the favorable accounts of intuitive knowledge in Patañjali’s Yoga or Bhartṛhari’s word-­non-­dualism (śabdādvaita), we come to realize that it is a deepened or sophisticated form of a reflective rationality which in Yoga takes the form of discrimination (viveka-­khyāti), and in the hands of the Grammarian takes the form of innate or instinctive capacity for synoptic grasp – for example, of the sentence-­meaning as a whole. Both of these contexts are, interestingly, deeply linguistic.  A  yoga-­practitioner’s prajñā (insight) becomes clearer and clearer, according to the Yoga-­sūtra (1.42–43), first by reflecting upon the distinctness of a word, the concept, and the object, all three of which are confused inextricably in our ordinary consciousness. As one recognizes the conventional character of the word–object relation and the memory-­mediated character of the concept associated with the word, the pure object in its individuality is supposed to shine more and more distinctly in isolation from its linguistic and cognitive cloaks.The last step of this stripping the pure object of words and imaginations through which the non-­discursive “dropping out of the mind” is achieved is indeed hard to explain. But Kaviraj makes an excellent attempt: It sounds absurd to say that the object alone remains without the citta or jñāna to take cognizance of it, but what is meant seems to be that the citta through extreme purity, becomes at this stage so tenuous as to be in fact a luminous void; it does not exist.

Notice that this elimination of thought and language is achieved through a very careful rational reflection on the distinction between the sound “cow,” the idea of a cow, and the particular animal cow in its uniqueness. Although Bhartṛhari emphasizes the indescribability of pratibhā, he also finds its presence in birds and beasts and describes it as “testified to by the inner experience of every man.” It is nothing esoteric. In a latent form it is innately present in every creature that can follow a rule. It is that instinctive flash of understanding through which, in practical situations, people come to decide what to do. Pratibhā is, indeed, a multifaceted concept. In its specially cultivated extra-­ordinary forms, it can amount to clairvoyance or yogic powers of remembering one’s own previous birth, and so on. But its central significance in Bhartṛhari’s philosophy is more that of tacit knowledge of the total sentence-­meaning which arises in a spontaneous way after individual word-­meanings and their syntactical connections are separately grasped, so that the final flash of understanding remains inexplicable even to the understander. Such meaning-­g rasping tacit knowledge lies at the heart of all interpretive rationality, rather than going beyond or against it. The fourth and final worry is of a different nature. Unlike the general anxiety regarding the alleged practical orientation, respect for authority and “mystical empiricism” of Indian philosophical systems, which were supposed to make the emergence of the idea of autonomous positivistic theoretical rationality unlikely, this has to do with a specific absence that India’s intellectual traditions suffer from. The concept of deductive or logical necessity which is paradigmatically illustrated by proofs in mathematics and formal

RATIONALITY IN INDIAN PHILOSOPHY

27

logic and the allied notion of a priori knowledge seem to be simply missing from the entire arena of Indian philosophical disciplines. Truths, even eternal truths, are recognized always as truths of fact, and never as truths of reason. Inferences have always been formulated as involving an inductive general premise (the vyāpti or rule of pervasion, about which more later) and never as demonstrative proofs. This absence of the notions of formal validity and logical necessity is linked at bottom with the failure to appreciate the idea of possible but non-­actual worlds and, in a roundabout way, also with the tendency to define knowledge as “true presentative awareness” without any further “justification” clause! These are serious allegations which have bothered contemporary Indian logicians and epistemologists like B. K. Matilal, J. N. Mohanty, and Sibajiban Bhattacharya. Broadly, two different sorts of response have been made to this crucial complaint. First, that in recognizing very clearly the law of non-­contradiction nyāya-­vaiśeṣika logicians do display adequate sensitivity to the concept of logical truths or necessity; and that in Buddhist logic deductive inferences find a place; and finally that the role of hypothetical reasoning (tarka) as buttressing evidence for universal generalizations shows that Indian logic has room for the modal notions of necessary connections between non-­actualized possibilities. This is the line adopted by Matilal (1982) in “Necessity and Indian Logic” (chapter 7 of Logical and Ethical Issues of Religious Belief). Second, that Nyāya, at least, is not even trying to do logic in the Frege–Russell sense, but is doing a phenomenological– epistemic logic in which eidetic rules are discovered not concerning compatibility and incompatibility between propositions but concerning compatibility and incompatibility of episodes of awareness (jñāna) in virtue of their contents.This is the line taken by Mohanty (1992) in his “The Nature of Indian Logic” (chapter 4 of Reason and Tradition in Indian Thought). The second approach is more challenging and opens up newer notions of logicality, which of course are not all that new in Asia. But let us discuss briefly Matilal’s approach first. Formulations of the law of non-­ contradiction are very easy to find in ancient and medieval Sanskrit writings. Udayana famously formulates the law of excluded middle as his  a  priori justification for  a  dichotomous division producing the sevenfold taxonomy of the Vaiśeṣika categories. Just as a classificatory rationality is operative in the division of all things first into presences and absences, and then the presences into non-­relations and relations, the non-­relations into independent and dependent, and so forth, so too purely analytic inferences are accommodated when definitions are taken as inference-­tickets, albeit for “inferences with merely differentiating marks” – for example, “Earth differs from all non-­earth because it has odor,” when having odor is taken as the defining mark of earth. Unlike other inferences, which require as supporting examples a series of cases where the mark (whatever property is mentioned in the “because”-­clause displaying the sign, probans or ground) coexists with the property to be inferred, “in [these] inference[s] we do not need certification from a positive example. … Naiyāyikas … would insist that this inference (or its conclusion) is necessarily true, for the opposite situation would be impossible or a logical contradiction” (Matilal 1982, p. 142). The Buddhist logicians talk about  a  universal concomitance which is not based on causal connection but issues simply from conceptually guaranteed class inclusions – for example, all elms are trees. From this they develop  a  theory of “inner connection” or “necessary pervasion” (antarvyāpti) which the fallibilist Nyāya school did not support.

28

ARINDAM CHAKRABARTI

The most important evidence for an Indian sensitivity to logical necessity comes from their use of “tarka,” which has been translated as hypothetical or subjunctive reasoning. Ancient Indian atheists would strengthen their position by arguing: If God were the maker, then he would possess a body, make strenuous efforts etc; But God cannot have a body or make efforts to overcome obstacles. Hence God is not the maker.

For the inherent modus tollens to work, one must have a clear conception of connection between two false sentences or two unmaterialized possibilities. There is a vast literature on the role of tarka starting with Bagchi’s (1953) pioneering Inductive Reasoning and proceeding to K. Chakrabarti’s (1992) Definition and Induction. In spite of the acceptance of its probative value for generalizations, it remains puzzling why a sentence like “If p then q,” in which both p and q are known to be false, is treated as expressing non-­knowledge (aprāmā). Notwithstanding extremely sophisticated distinctions made between self-­contradictions (such as the liar-­sentence) and pragmatic self-­refutations (like “I am not aware of this”), the Naiyāyikas use “barren woman’s son” and “horn of a rabbit” as empty terms without discrimination. Here is Matilal’s explanatory conjecture: Suppose my car is red. This fact, a contingent fact, has already defeated (excluded) the possibility of its being non-­red. Does this “excluded” possibility then join the group of impossibilities? No clear and explicit answers emerge from the Indian philosophers except in their discussion (Udayana) of citing a nonexistent entity as an example. In any case, consideration of the excluded possibilities have somehow been thought idle in the Indian context. (Matilal, 1982, pp. 150–1)

Mohanty, on the other hand, is not baffled by this apparent lack of interest in purely formal necessity/possibility on the part of the Indian logicians. Since Indian logic arose out of two different contexts – namely, as a science of adjudication of actually happening debates and disputations, and as part of a general epistemology in which inferential cognition as a form of veridical awareness had to be normatively thematized – it was natural that component sentences of a valid argument were usually confined to those which were materially true. Soundness was more important than mere validity. This epistemic and causal character of Indian logic becomes clear when we look at its definition of a fallacy or defect of a mark. (As a reminder, in the inference “a has f, because it has g,” g is the mark.) A defective mark or the defect of a mark is defined as “the object of such a true cognition as acts to prevent an inferential cognition.” The idea is this: consider the fallacy of locuslessness (āśsrayāsiddhi). “Sherlock Holmes must have been a psychopath because he took drugs.” The veridical awareness that Sherlock Holmes never existed would, according to Nyāya, prevent the serious inference that he actually was  a  psychopath, because the property of drug-­addiction as a putative mark would find no real locus. We surely have a different conception of logic (Mohanty calls it “Logic 2”) here, not only because the fallacy above would not count as a formal fallacy in Western logic but because the rejection or cancellation of a proposed inference that the discovery of the defect is supposed to lead to is conceptualized as “preventing.” In later Neo-­Nyāya schools, the relation of

RATIONALITY IN INDIAN PHILOSOPHY

29

preventor-­and-­prevented cognitions became a hot topic of discussion. My perception that the sun has risen prevents any immediately succeeding inference that it is still night-­time. But is this just a psychological generalization? Once we discover the fault in the mark, the inferential cognition is blocked or prevented. Is that just a causal law? Insofar as these rules are formulated in terms of the content-­structures of the cognition and not in terms of any other mentalistic features of cognition, they must be structurally grounded. Logicians in the Indian context were partly evolving an ethics of belief and partly discovering naturalistic laws of compatible and incompatible cognitive acts (in the same subject) in virtue of their intentional content (viṣayatā) which is further broken down into the roles of qualificand (viśeṣyatā), qualifier (prakāratā), relation (samsargatā), and so on. The logic of awareness gets fascinatingly complex and the philosophical logic involved in trying to decide the ontological status of these cognition-­conferred intentional roles of actual items which figure as objects, qualifiers and relational links becomes awfully subtle. But it is neither a formal logic that studies relations between sentence meanings or Fregean thoughts belonging to a non-­mental non-­material third realm, nor a psychologistic logic that studies merely how the mind functions with regard to cognitive acts. It is a logic of universal intentional structures interrelated in a normative way. Thus fallacies are called “apparent marks” (hetu-­ābhāsa) a correct knowledge of which prevents inference because, Mohanty remarks, “as rational beings we cannot make a fallacious inference: we only appear to be doing so” (Mohanty, 1992, p. 113). As far as the justification-­clause in the definition of knowledge is concerned, S­ ibajiban Bhattacharya has made a virtue out of the common complaint that the Indian concept of prāma never developed beyond the second rejected definition of Plato’s ­Theaetetus – namely, knowledge as true belief. By using simple principles of epistemic logic, he proves (see, “Epistemology of Testimony and Authority” in Knowing from Words) the following: “If knowledge is something more than mere true belief, then I should be able to know when I have knowledge and when I have mere true belief. Theorem 1 suggests, however, that no one is able to make that discrimination in his own case.” This is a pretty strong stand to take, because Western epistemology starts from the distinction between knowledge and true belief – the space between the two being precisely the space of reason. Perhaps all that Bhattacharya has proved is that that distinction has no role to play in self-­ascriptions or claims of knowledge. The rich skeptical literature in Sanskrit (see, Jayarāśi and Śrīharṣa) clearly demands a stricter notion of knowledge, because it is never tired of pointing out that accidentally slipping into a true belief by a route which could equally well have produced error is no knowledge (and that is all we can do!, hence the ­skeptical pessimism). But Bhattacharya’s attempt cautions us against taking the usual Western account of knowledge as justified true belief plus something else (which the Gettier-­ industry has been hard at work on) as gospel truth.  A  more realistic rationality may require that we be happy with true beliefs achieved through epistemically virtuous means. AXIOMATIC GRAMMAR

“To adhere to Indian thought,” remarks L. Renou, “means first of all to think like a ­grammarian.” The model of theoretical knowledge that appealed most to the Vedic mind

30

ARINDAM CHAKRABARTI

was, as we find in the hymn to knowledge (Ṛg Veda X.71), that of “the seers fashioning speech by their mind, sifting as with sieves corn-­flour is sifted.” Patañjali, who quotes this at the beginning of his “great commentary” on Pāṇini’s grammar, makes it clear that words are not “fashioned” by the grammarian like pots are made by the potter. Pāṇini was trying to come up with an adequate description of already existent ordinary and Vedic Sanskrit usage while remaining faithful to the criteria of simplicity and brevity.The method of this fourth century bce grammarian has been compared to that of Euclid insofar as Pāṇini’s rules or aphorisms (sūtras) are divided into three kinds: a the defining sūtras which introduce technical terms, b the theorems which normatively describe linguistic facts while demonstrating their legitimacy through formation and transformation rules, and c the metatheorems (paribhāṣā-­sūtra) explaining how rules have to be applied in particular cases (see, “Euclid and Pāṇini” in Staal 1988). There are obvious differences between Euclid’s subject matter, which demands a higher degree of generality and a priori validity of the axioms, and Pāṇini’s subject-­matter, namely, well formed words and sentences, to accommodate the contingent variety of which, rules have to be formed ad hoc. But Pāṇini’s rigorous standards of consistency, completeness, and the avoidance of redundancy (amounting to independence) among his basic rules make the comparison illuminating. This grammatical model of generating all correct speech-­ units by sifting them through the parsimonious “sieve” of a set of systematically arranged sūtras (memorizable strings of words) became the paradigm for philosophers in the early Common Era in India. Thus we have the Yoga-­sūtras, the Mīmāṃsā-­sūtras, the Vaiśeṣika-­ sūtras and, from the point of view of the study of epistemic and dialogical rationality, the all-­important Nyāya-­sūtras. RATIONALITY IN MEDICAL PRACTICE

Physicians in ancient India must have regularly held meetings for co-­operative as well as combative debates in the presence of some expert judges. The Nyāya list of “tricks of reasoning” and “sophistical rejoinders” is thought to have emerged out of this older tradition of the science and art of diagnostic and therapeutic debates. The enormous medical text Caraka Saṁhita, which reports on this tradition, divides the entire practice of medicine into four factors: (1) the physician; (2) the substances (drugs and diet); (3) the nurse; and (4) the patient.The four essential qualifications of the physician are: (1a) a clear grasp of the science learnt; (1b) a wide range of experience; (1c) general skillfulness; and (1d) cleanliness. The four key factors concerning drugs and diet are: (2a) abundance of supply; (2b) applicability; (2c) their many imaginable uses or multifacetedness, which is now called the “broad-­spectrum” nature; and (2d) richness. The four qualifications of the nurse are: (3a) a knowledge of attending techniques; (3b) skill; (3c) caring involvement with the patient; and (3d) cleanliness. The most interesting quartet of desiderata is that concerning the rational patient, who must have: (4a) a good memory (so as not to forget her own case-­history!); (4b) obedience to the doctor’s instructions; (4c) courage; and (4d) the verbal ability to describe the symptoms.

RATIONALITY IN INDIAN PHILOSOPHY

31

Interestingly enough, in spite of the presence of karma-­theory in popular as well as theoretical consciousness, the text does not list “accumulated good karma” as a condition of the curable patient. As long as the patient is cooperative and courageous and articulate, his chances of getting well are good! In the tenth chapter of Caraka Saṁhita we find the most fascinating argumentation about the efficacy of medical practice, where cases of cures without doctors, drugs or nursing as well as the death of a well attended and medically treated patient are discussed as counterexamples to the alleged necessity and sufficiency of medicine. We do not have space here to go into the details of this discussion. But what emerges out of this context is a clear sense of the need to defend the probabilistic workability/truth of a scientific practice/theory in the face of reasonable doubts arising from empirical data. (See ­Chattopadhyaya, 1980, pp. 107–13.) LOGICAL FORM AND CONFIRMATION OF UNIVERSAL CONCOMITANCE

The center-­piece of the Nyāya-­theory of reasoning has been the problem of formulating and justifying the relation of invariable co-­location or pervasion (vyāpti) between the mark (typically illustrated by smoke) and the property to be inferred (typically, fire). In this quick survey we shall first look at the way the universality of such a relation was captured, without resorting to extensional quantification theory, in terms of property, location and absence.Then we shall summarize the ways in which skeptical attacks against the possibility of the non-­circular empirical confirmation of such universal concomitance were resisted. Gaṅgeśa, the father of Neo-­Nyāya (thirteenth century ce) discusses 29 different formulations of the definition of pervasion, rejecting 21 of them as flawed. Pervasion has been defined as “non-­deviation,” “natural connection,” “relation of effect to its efficient cause,” “accompaniment of all cases of one term with the other term,” “unconditional relation,” and as “not being located in any place where the property to be inferred is absent,” to mention just a few of its definitions. The last of these – that is, pervasion’s definition in terms of non-­co-­location with the absence of the major term (loosely identifiable with the property to be inferred, for example, fire) – has been rejected because it fails to cover cases where the major term is an unnegatable property which is located everywhere – for example, existence or nameability. Interestingly, “This cup is nameable because it is knowable” is regarded as a sound inference even though the mark and the property to be inferred are both unnegatable. The final definition of pervasion, somewhat simplified, could be translated as follows: “Pervasion is the mark’s property of being co-­located with such an inferable property as is not the absentee of any absolute absence which is co-­located with the mark.” The subtleties of these definitions are brought out by at least five centuries (that is, until the eighteenth century ce) of scholastic creativity, which continued to invent counterexamples, adding further qualifications to avoid undercoverage, overcoverage and inapplicability. A very small part of this literature is available in English now in F. Staal’s (1988) Universals (“Means of Formalisation in Indian and Western Logic” and “Contraposition in Indian Logic”), in Matilal’s (1985) Logic, Language and Reality, and in C. Goekoop’s (1967) The Logic of Invariable Concomitance.

32

ARINDAM CHAKRABARTI

Coming to the issue of inductive justification of pervasion (vyāpti-­graha), the Nyāya philosophers conceded Hume’s point that there are no a priori necessities in nature. Against Sāṃkhya, which believes in inherent substantial identity between cause and effect, Nyāya believes that fire is one thing and smoke is another. But their account of contingent causal connection is realistic rather than psychologistic. Unlike Hume, Nyāya does not regard demonstrative a priori knowledge and irrational animal faith as the only options. Boldly affirming that it is perfectly rational to embrace fallible certitude after one has taken all possible empirical precautions to eliminate doubt (especially when asking for deductive necessity is irrational), Nyāya bases its knowledge claims of invariable concomitance on: (1) multiplication of instances; (2) subjunctive supporting arguments showing the material absurdity of the negation of the universal generalization to be established; (3) direct perceptual acquaintance with all cases of the inferable property through the experienced universal inherent in the observed cases (this is  a  controversial doctrine of seeing-­all-­ through-­seeing-­the-­class-­character-­in-­one); and (4) showing how skeptical doubt concerning induction lands one in practical paralysis insofar as one cannot help believing that food will always nourish, fire will always burn, and words will often be understood. “Doubt reaches its limits in pragmatic self-­refutation.” (The clearest account of this is to be found in K. Chakrabarti’s Definition and Induction.) DHARMA-­R ATIONALITY AND SELF–OTHER COMPARISON

I have said earlier that Vedic orthodoxy regarded tradition as the sole spring of moral knowledge. Such rigidity was naturally called into question by the tradition itself, and in the Mahābhārata and the Dharmaśāstras we see an obsessive attempt to arrive at contextualized as well as universal criteria for determining the right way to live. Some parts of practical rationality were explained in terms of means–end adaptations, so that a command like “You must perform fire-­r ituals since you wish to attain heaven” would be understood in terms of three pieces of belief: (1) that such performance is do-­able by the command-­receiver; (2) that it would promote a desirable result, in this case, heaven; and (3) that it does not involve undesirable results which are counterbalancing – for example, the sin of killing animals. But there were other parts of practical reason which were deontological, and the Mahābhārata tries to come up with criteria for such universal dharmas (duties). The briefest statement of such a criterion is: Do not inflict upon others what is intolerable to yourself. This, in short, is dharma and it is other than what one naturally desires. (Mbh. XIII, 113, 8)

Probing deeper into the source of this morality, the next verse says, In refusals, generosity, pleasure, pain, approval, and disapproval a human being finds a decisive source of knowledge (pramāṇa) by comparison with oneself.

This coheres perfectly with the symbolic narrative of  a  just businessman who teaches dharma even to higher-­caste brahmins because he has achieved “equal attitude towards all living beings.” As he puts it, “My weighing scale remains balanced equally for everybody” (Mbh. XII, 262, 10).

RATIONALITY IN INDIAN PHILOSOPHY

33

But what does one do when there is a conflict of duties? The Mahābhārata is full of such moral dilemmas. Traditions and texts cannot solve them because they themselves are often in conflict with one another. That is why one needs intelligence (buddhi) and learning (vidyā), a special training in the pramāṇas (means of knowledge), to purify the moral knowledge derived from handed-­down tradition. Dharma cannot afford to be intellectually blind or uncritical. Matilal quotes the following advice from Manu Saṃhita: What is to be done? If such a doubt arises with regard to a conflict of dharmas an assembly of not less than ten persons should deliberate and reach a decision – this assembly will be constituted by three Vedic scholars, one logician, one dialectician or debater, one expert in semantics, etymology, and three laymen from three different age-­groups – a student, a householder, and a retired person. (Matilal in Rationality in Question, p. 203)

One sees the beginnings of a discourse-­model of practical rationality quite clearly here! REASON IN EMOTIONS: THE RASA THEORY OF AESTHETIC RELISH

A similar process of universalization (sādhāraṇikaraṇa) or depersonalization, which has been called detached enjoyment through the “heart-­universals,” lies at the center of mainstream Indian aesthetic theory, which runs from Bharata’s treatise on drama through Anandavardhana and Abhinavagupta up to K. C. Bhattacharya’s early twentieth-­century reflections on aesthetic rapture. Bharata (first or second century ce) provides almost a functionalist logic of emotion in his classic sixth chapter, where he gives the cryptic formulation: “rasa (aesthetic rapture) comes from a combined functioning of the determinants (vibhāva) and the consequent expressions (anubhāva) and the passing states.” The basic idea, somewhat shrouded by a millennium’s worth of commentators’ controversies, seems to be this: a certain depersonalized enjoyment (for example, at the representation of fury) is a function of the stable latent emotional disposition (for example, anger) and is achieved through the input of some determinants (for example, insult, vengefulness, threat, jealousy, assault) and the output of some consequent outer expressions (for example, red eyes, frowning, the biting of lips, the grabbing of one hand with another, and so on) via the passing sentiments (for example, energy, restlessness, obstinacy, perspiration, trembling, and so on). Abhinava Gupta, who comes after a long tradition of commentators and thinkers on poetics and aesthetics, takes this obscure theory and develops an exceedingly complex epistemology of artistic enjoyment out of it. What is relevant for our purpose is to realize the importance of the process of the impersonalization of an emotional situation through which even the pain of actual or possible, real or imaginary others can become objects of aesthetic rapture by becoming, as it were, one’s own. The same mechanism of self–other comparison, in two radically different ways, yields both moral imagination as well as aesthetic enjoyment. Without the viewer’s loss of ego, emotional identification with a tragic hero should lead to pain rather than pleasure. K. C. Bhattacharya describes this beautifully in his 1930s essay “The Concept of Rasa.” “When I imagine, with delight, an old man affectionately watching his grandchild play with a toy, my sympathy with the ­grandfather’s sympathetic feeling for the child goes through the grandfather’s heart-­universal. The beauty of a child at play appears to me through a kind of knowledge-­by-­identification.

34

ARINDAM CHAKRABARTI

My personality is as it were dissolved and yet  I  am not caught in the object like the child. I freely become impersonal.” K. C. Bhattacharya’s theory of aesthetic appreciation as “sympathy with sympathy” becomes rather obscure and contentious and he admits that this idea of the “heart-­universal” is “semi-­mythological.”Yet there is a certain right-­ mindedness in his interpretation of the phenomenon of de-­individuation through which even the humdrum or the ugly or the ludicrous become beautiful, and because of which, in spite of total empathy with the tortured heroine on the stage, one does not feel the need to go and attack the actor playing the villain, because while one enjoys “suffering” with the felt-­heroine-­in-­general one cannot hit a villain-­in-­general. CONCLUSION: THE LOGICAL WAY TO LIBERATION

We started this survey of the ideas of rationality by mentioning the term “ānvīkṣikī,” which is usually taken as synonymous with logic, Nyāya or the philosophical systems in general. What is profoundly interesting is that this metascience or logic of all other disciplines is said to be identical with the liberating science of the spirit (ātmavidyā).Through medicine and healthy diet, ethically correct behavior, intelligent and legally constrained utility-­ maximization and aesthetic enjoyment, we constantly engage in the most basic reasonable activity of all – that is, the avoidance of suffering. But as long as we are unclear about the nature of our selves, we continue to make errors which lead to recurring pains. Different philosophical systems, in consonance with, but not solely depending upon, one’s own received tradition, try to give us knowledge of the self.This is as much a rational project as it is a practical, moral and spiritual one. As far as the major classical Indian philosophies such as Nyāya, Sāṃkhya, Vedānta, and the Kaśmir Śaiva schools are concerned, rationality is theoretically studied and practically used so that the thinking agent can ultimately lose her individual ego, the object-­directed outward mind or intellect which pretends to be the self, the pleasure-­seeking wish-­generating cognitions which make one other-­dependent, and hence unfree.The practice of Dharma which forces you to compare yourself to others, the performance of duties without desire for reward, the distillation of emotions through aesthetic universalization, a healthy dose of reasoned skepticism about reason’s ability to yield stable knowledge  –  all of these lead to that saving self-­knowledge which brings freedom from suffering. According to Akṣapāda Gautama (first century ce), in order to achieve such liberation you must have accurate knowledge of the means of knowledge, the objects of knowledge, doubt, purpose, example, tenet, the components of a syllogism, hypothetical reasoning, the determination of a conclusion, truth-­finding discourse, defensive debate, polemics, fallacies, tricks, retorts, and the conditions of defeat. Thus in Indian thought spirituality and rationality merge into one discipline called āvīkṣikī. BIBLIOGRAPHY Bagchi, Sitansusekhar 1953: Inductive Reasoning: a Study of “Tarka” and its Role in Indian Logic (Calcutta). Bhattacharya, K. C. 1983: Studies in Philosophy (Delhi: Motilal Banarassidas). Biderman, Shlomo and Scharfstein, Ben-­Ami (eds) 1989: Rationality in Question (Leiden: E. J. Brill). Chakrabarti, Kisor K. 1995: Definition and Induction (Honolulu: University of Hawaii Press). Chattopadhyaya, Debiprasad 1980: “Tradition of Rationalist Medicine in Ancient India,” in Philosophy: Theory and Action (Poona). Goekoop, C. 1967: The Logic of Invariable Concomitance (Dordrecht: Kluwer).

RATIONALITY IN INDIAN PHILOSOPHY

35

Halbfass, Wilhelm 1988: India and Europe (Albany: State University of New York Press). ——— 1991: Tradition and Reflection (Albany: State University of New York Press). Kant, Immanuel 1974: Anthropology from  a  Pragmatic Point of View, trans. Mary J. Gregor (Holland: ­Martinus Nijhoff). Kaviraj, Gopinath 1990: Selected Writings of M. M. Gopinath Kaviraj (Varanasi: Centenary Celebration Committee). Masson, J. L. and Patwardhan, M.V. 1970: Aesthetic Rapture (Poona: Deccan College). Matilal, B. K. 1977: Logical Illumination of Indian Mysticism (Oxford: Oxford University Press). ——— 1982: Logical and Ethical Issues of Religious Belief (Calcutta: University of Calcutta). ——— 1985: Logic, Language, and Reality (Delhi: Motilal Banarsidass). Matilal, B. K. and Chakrabarti, Arindam (eds) 1994: Knowing from Words (Dordrecht: Kluwer). Mohanty, J. N. 1992: Reason and Tradition in Indian Thought (Oxford: Clarendon Press). ——— 1995: “Theory and Practice in Indian Philosophy,” Australasian Journal of Philosophy, March. Saha, Sukharanjan 1987: Perspectives on Nyaya Logic and Epistemology (Calcutta: K. P. Bagchi). Solomon, Esther A. 1978: Indian Dialectics,Vols 1 and 2 (Ahmedabad). Staal, Frits 1988: Universals: Studies in Indian Logic and Linguistics (Chicago: University of Chicago Press).

CHAPTER

3

Intellectual India Reason, identity, dissent Jonardon Ganeri

Source: New Literary History, 2009, 40, 247–263. AMARTYA SEN AND THE REACH OF REASON

India has a long and multifaceted history of argumentation and public reasoning. In his magnificent book The Argumentative Indian: Writings on Indian Culture, History and Identity,1 Amartya Sen provides this history with a global context. Public reasoning is fundamental to both democratic politics and secular constitutional arrangements, and it is no accident that India, with its extensive traditions of tolerance and the admission of dissenting voices in public discourse, should have deep democratic and secular instincts. This is something, Sen suggests, which more narrowly sectarian understandings of India have lost sight of, and he recommends that we keep in mind figures such as the Indian emperors Ashoka and Akbar, both of whom strongly encouraged public debate and respect for heterodox voices, and also Indian skeptics like the atheist Jāvāli, so vividly depicted in the Rāmāyaṇa. Sen insists that such individuals are as much the precursors of a modern Indian identity as any other figure drawn from Indian history. Thus, “public reasoning is central to democracy … parts of the global roots of democracy can indeed be traced to the tradition of public discussion that received much encouragement in both India and China (and also Japan, Korea and elsewhere), from the dialogic commitment of Buddhist organization” (182), and it is therefore a mistake to think of democracy and secularism as “Western values” which India has embraced.The demonstrably global origin of traditions of public reasoning undermines any thought that the West has a distinctive claim upon liberal values, but also, just as importantly, it undercuts arguments that there are things called “Asian values” which are antithetical to ideas of democracy, secularism, and human rights (134–37). There are no “cultural boundaries” in the reach of reason or in the availability of values like tolerance and liberty (280). 36

INTELLECTUAL INDIA

37

Another central idea for Sen is that reason is “before identity,” that each of us is free to reason about what is of value and significance to us in whatever situations we find ourselves, that neither religion nor community nor tradition imposes upon us an identity fixed in advance. There is a relationship between agency and freedom, and a contrast between agency and well-­being, a full sense of agency involving not only “control over decisions” but also “the freedom to question established values and traditional priorities” (240), including the freedom to decide that religious or communitarian affiliations are of less significance than one’s literary, political, or intellectual commitments:“We have choice over what significance to attach to our different identities.There is no escape from reasoning just because the notion of identity has been invoked” (352). So reasoning has center stage both in shaping individual identities and in deliberating about public goods. And in seeking to give structure and substance to those deliberations, the whole of India’s past is available, a past that has been deeply international and profoundly interreligious. Sen finds in Rabindranath Tagore’s assertion that the “idea of India ‘itself militates’ against the intense consciousness of the separateness of one’s own people from others” two thoughts in tandem, one that opposes the idea of India as a mere federation of separate and alienated religious communities, the other opposing an isolationist conception of India in the world (349). In the global circulation of ideas, India has always been a major player, and the combination of “internal pluralism” and “external receptivity” has fashioned for India a “spacious and assimilative Indian identity” (346). Though rightly emphasizing India’s argumentative traditions and historical accommodation of dissenting voices, the lack of detail in Sen’s description of those traditions is striking. There is in Sen’s work little mention of any actual analyses of public reasoning in India, no reference, for example, to seminal works on dialectic and argumentation such as the Kathāvatthu, the Nyāyasūtra, or the Vādavinoda. Still less is there any significant description of the resources of practical reason in India’s intellectual past, or of the ways identities are understood as fashioned and not found. Sen’s understanding of the “reach” of reason, of the utility of critical public discussion, and of rationality in human psychology, hardly makes reference to India’s long tradition of thought about these topics. This is puzzling. It is as if the mere fact of this tradition of argumentation is sufficient for Sen, that the substance of that tradition is irrelevant. Or else it is as if Sen presumes a priori that the substance must coincide in all important respects with the concept as it features in contemporary work on political and social theory. In short, while Sen speaks freely about exemplary political figures like Ashoka and Akbar, he is largely silent on the intellectual figures who have provided India with its theoretical resources and self-­understandings. Sen observes with decisive clarity how a false contrast between the intellectual traditions of India and the West is brought about by the biases in the respective histories that are told, that “[i]n comparing Western thoughts and creations with those in India, the appropriate counterpoints of Aristotelian or Stoic or Euclidian analyses are not the traditional beliefs of the Indian rural masses or of the local wise men but the comparably analytical writings of, say, Kauṭilya or Nāgārjuna or Āryabhaṭa. ‘Socrates meets the Indian peasant’ is not a good way to contrast the respective intellectual traditions” (158). What I find astonishing, though, is that this is the only time Sen mentions Nāgārjuna in the whole book, and of the many great Indian intellectuals who have thought so long and so hard about reason, theoretical, practical, and public, Nāgārjuna is lucky in getting a mention at all (no Nyāya

38

JONARDON GANERI

philosophy is mentioned, for example, and the term “Nyāya” is not even in the index). Profoundly aware of the pitfalls, Sen seems nevertheless to fall into them. Misleading impressions aside, the danger here is that liberal secularism is made to win too easily. The mere suggestion that one should reason more and better might seem to fail to engage at anything other than an abstract admonitory level; it might sound more like enthusiasm than practical advice. This is the deeper worry that motivates attacks on the Enlightenment’s appeal to reason, a more serious worry than simply that reason has so often been abused for totalitarian ends. What a proper response to that worry requires is a detailed engagement with the resources of reason in India as they are actually in play or can be brought into play. The cause of Sen’s silence is, I think, that his primary interest is in provisions of the secular state itself and not in what it is to be a participant within it. A secular state tries to work out how to structure its policies and institutions in such a way that there is symmetry in the state’s dealings with any particular group, religion, class, or individual. The appropriate model of reason here is the one that Rawls seeks to capture with his use of the term “public reason.” Public reason is the mode of deliberation that brings people of diverse philosophical, religious, and moral convictions into a state of rational accord with respect to a matter of mutual concern and common interest. In a pluralistic community tolerant of difference, it is essential that the resources, reach, and requirements of public reason be properly understood, and no such understanding is acceptable that gives discriminate advantage to the particular view of any one of the parties in the deliberation. For a straightforward if overly simple example of the workings of public reason, consider the following account of the process leading to the preparation of a common communiqué at a recent meeting of the G8 industrialized countries: “Acting under the policy parameters set by their political masters … the job [of the negotiating officials or sherpas] is to move words and phrases in and out of square brackets. If a phrase stays in square brackets, it is not agreed. If it comes out of brackets, consensus is reached.”2 But the view from the side of the institutions and the state is blind to the question that is most pressing for the participants themselves. That is the question of how a specific individual, whose access to resources of reason has a particular shape and character, finds within those resources the materials to engage in public and practical acts of reason. How, for example, does such an individual find their way to a conception of deliberative thinking about the public good? How do they form a conception of thinking about other participants in the public space for whom the resources of reason are different? It is important, certainly, to point out, as Sen does, that Hinduism contains within itself many dissenting voices and heterodox opinions; but the difficult question is to understand how those dissenters, no less than the mainstream, made sense of their dissent. One needs to show how Hinduism has within itself models of rational deliberation that make possible the dissenting voices and internal critiques and how those models also make available to Hindus a conception of what it is to reason about the public good. An analogous exercise needs to be repeated for each of the participants in a secular state; only in this way can each reach an understanding of what Rawls has appropriately described as the “overlapping consensus” in which each group, for its own reasons and on the basis of its own resources of reason, makes sense of and agrees to a common position or policy. The exercise

INTELLECTUAL INDIA

39

can be seen happening within Islam through a contemporary revival of the practice of ijtihad (reasoning and interpretation based on the sacred texts).3 Space for such an approach depends in part on the identification of a neutral secularism, that is, a secularism which demands that politics and the affairs of state are unbiased or symmetric in respect of different religions (313), in contrast with an understanding of secularism that sees it as requiring the prohibition of any religious association in public or state activities (19). Prohibitory secularism requires the resources of reason to be wholly disenchanted in public, even those upon which individuals draw. Neutral secularism imposes this requirement only on state invocations of public reason; its requirement on individuals is that their appeal to private resources or reason does not bias them in favor of their own resources or lead to an asymmetry in their reasoned dealings with others. (A Buddhist analysis of debate from the time of Ashoka, the Kathāvatthu, provides an astonishingly subtle theorization of this idea of biaslessness in public reasoning.) The background worry is, of course, that in developing the resources of reason within Hinduism, Islam, or Buddhism, we are in some way making reason subordinate to tradition and religious command. Sen reads Akbar as resisting that threat with a strong insistence on the autonomy of reason (274). My argument is that we can respect the need for autonomy without restricting reason’s resources to those merely of allegedly value-­free disciplines such as rational choice theory. I am saying that the appeal to India’s traditions of argumentation and public reasoning is hollow if it does not engage with the detail of those traditions. For only in this way does the full panoply that a well-­informed “argumentative Indian” has available to himself or herself come to the fore, in contrast with the restricted vision of a sectarian approach. Sen’s wonderful discussion of the plurality of India’s calendrical systems, and the significance of that plurality, only serves to emphasize the absence of any corresponding presentation of the range and richness of Indian conceptions of reason (a range that would include, to use the Sanskrit terminology, at the minimum, tarka, nyāya, ūha, and yukti; that is to say, reason based on hypothesis, on examples and general rules, on extrapolation from paradigms, and on empirical generalization). Likewise, in speaking of the resources for fashioning Indian identities, the wealth of material about self, agency, and identity, Hindu, Buddhist, Muslim, and Jaina, orthodox and heterodox, needs in some measure to be gainsaid. My point is simply that these are the intellectual resources that an Indian entering an argument can and should bring to the table. In the case of figures such as Tagore and Gandhi, this is clearly exactly what happened, Gandhi’s reintroduction of the idea of ahiṃsa as a defence against inhumanity being a good case in point. Only in this way can the claims of more sectarian thinkers, that theirs is the sole true inheritance of India, effectively be silenced. Let me give an example. There is, in the Nayasūtra, an elegant discussion of the various ways in which an opinion or principle might count as “settled” (siddhānta). An opinion might be “settled” because there is a general consensus, which is further analyzed as a situation in which the opinion is accepted by some parties, including one’s own, and outright rejected by none.4 Alternatively, an opinion might be “settled” in the sense of being accepted by some parties, including one’s own, but rejected by others (1.1.29). An opinion is considered “settled” in a third sense if it is used as a premise in further derivations (1.1.30), and in a fourth if it is entertained provisionally for the purpose of considering its merits (1.1.31).The idea of consensus implicit in the first conception of a settled

40

JONARDON GANERI

opinion is a valuable and useful one, for it reveals a way to see how achieving consensus might require something weaker than universal endorsement by all parties. Indeed, if the officials at the G8 meeting had been cognizant of this theory, they might have wanted to refine their bracketing method so as to distinguish between the case where a phrase is accepted by some but rejected by others and the case where it is accepted by some and rejected by none. In so far as it allows for progress toward a consensus that might not have otherwise been achievable, this would represent a substantive contribution to the actual machinery of public reason. My point is that this analysis of settled opinion is a resource of reason that a well-­informed “argumentative Indian” has at his or her disposal, something that can shape the nature of his or her participation in public debate. It is by acquainting ourselves with such detail that we get a true sense of the “India large” of which Sen so admirably speaks. And Sen is surely right in his diagnosis of the cause of what he calls the “extraordinary neglect of Indian works on reasoning, science, mathematics and other so-­called ‘Western spheres of success’” (80) in the “comprehensive denial of Indian intellectual originality” that one sees in the appalling colonial writings of James Mill and others (more precisely, in the twin effects of the “magisterial” and the “exoticist” approaches to India; 154–55, 160).The effects of this systematic deprecation are still to be seen in the reluctance of many scholars today to take seriously the intellectual substance of the Indian texts. Sen is right too when he speaks of the need for a “corrective regarding Indian traditions in public reasoning and tolerant communication, and more generally what can be called the precursors of democratic practice” (80). The recovery of bits of theory such as the one I have briefly sketched (and it was, of course, developed to a much greater deal of sophistication than I have revealed) is the way to make good that necessity. Fragments of theory like this one, though presented in the texts as abstracted from any concrete context, were most certainly the product of engagement with the day-­to-­day business of reasoning publicly about matters of common interest with others who did not share one’s views. I have said that intellectuals like Gandhi and Tagore were most certainly aware of the “India large,” full of resources for reason. Sen observes how Tagore was resistant to anything that seemed to smack of the application of mechanical formula, that “[t]he question he persistently asks is whether we have reason enough to want what is being proposed, taking everything into account” (119). At least two resources from India’s intellectual past are available to support such a question. One is the prolonged debate in the Mahābhārata over the rights and wrongs of a lie, uttered by Yudhiṣṭhira in a moment of crisis.There are anticipations in this discussion of the problem that was to vex Kant, whether it is right to lie to the malicious pursuers of an innocent person. The other, perhaps even more interesting, is the strong vein of particularist moral reasoning found in the highly intellectual Mīmāṃsā school. The Mīmāṃsā theoreticians develop an account of practical reasoning that is situational and adaptive, driven by particular cases, and extremely versatile. This, again, is a resource of great value for any “argumentative Indian.” Indeed, it is through the imaginative exploitation of such models that dissenters and skeptics find the resources of reason with which to develop their critique. Pointing to the brute existence of skeptical voices like that of Jāvāli is only the beginning of the story. What we really need to know is how a skeptic like Jāvāli adapted and

INTELLECTUAL INDIA

41

manipulated the tools of justification and argument at his disposal so as to make possible his dissent. If nothing else, that would be a step toward understanding how heterodox voices might similarly empower themselves in global public discourse today. In Sen’s observation of the way one sees “the colonial metropolis supplying ideas and ammunition to post-­colonial intellectuals to attack the influence of the colonial metropolis” (133), we see an expression of the typical skeptic’s maneuver. Again, the general pattern of such a move would be familiar to an Indian intellectual with access to the materials for reasoned thought that an expansive conception of India makes available. For precisely such strategies have been put into practice by skeptical critics of orthodox Hinduism, voices that include the Buddhist Nāgārjuna, the freethinking critic Jayarāśi, and the dissenting Vedāntin Śrīharṣa (all of whom, it must be said, achieve a far greater degree of sophistication and methodological self-­awareness than does Jāvāli). The influence of their criticism derives precisely from its using the very resources of reason used by its target. For while “reasoned humanity” should certainly be open to sound criticism from any quarter, it is a fact of human nature that it is much harder to be receptive to criticism formulated by outsiders in outside terms than to criticism made from within one’s own terms. The reason for this is that rational criticism is effective and not merely enthusiastic when it has the potential to become self-­criticism. This averts what would otherwise be a distant call upon the “sovereignty of reason.” For reason to be effective, it must be engaged. And it must be admitted that charged or difficult situations create impediments to the engagement of reason. Acknowledging human frailty ought not, however, to lead us to think that it is better simply to fall back on our moral instincts, such as our instincts to respond to others with respect and sympathy. Difficult as it is, reason needs to be engaged in at least the following areas: i) in publicly scrutinizing policies whose unintended but injurious effects are all too easy for policymakers to ignore (such as famine); ii) in reflecting on our values and priorities in the course of fashioning an identity; iii) in questioning the appropriateness of our emotional reactions; iv) in order to “transcend ideology and blind belief ”; v) to cultivate our moral sentiments themselves and revise our first perceptions (275–81). That leaves us to address the issue introduced by the rhetoric of colonialism, made salient in the work of cultural anthropologists; and kept in circulation by the politics of a “clash of civilizations”: “The question that has to be faced here is whether such exercises of reasoning may require values that are not available in some cultures” (282), in particular the values of liberalism—tolerance, mutual respect, the dignity of humanity, rights, justice. Sen’s answer is simply that there are no cultural barriers to the availability of such values as these, that the idea that there are such cultural barriers is an ideology of the West, reinforced by its self-­serving deprecation of the intellectual traditions of countries like India. Sen is right to see in the figure of Akbar an excellent illustration of the point. Ruling India in the sixteenth century, Akbar practiced the “path of reason” (rahi aql), which led him to reject traditionalism, respect all religions, and develop a “tolerant multiculturalism” (288). These ideas did not disappear with Akbar himself, and one can see in the India of the seventeenth century a commitment to the values of liberalism, among both the political classes and the intellectuals, as significant as in seventeenth-­century Europe (with which India already had many cultural and mercantile relations).

42

JONARDON GANERI

Voltaire’s friend,Vauvenargues, said that “a truly new and truly original book would be one which made people love old truths.”5 Sen’s book is truly new and truly original, and will make us love India’s “old truths” about reason, identity, and dissent. I have argued that such a rediscovery of India’s long tradition of argumentation and public reasoning must be substantive and not merely gestural. Nor should we think of this as a merely archaeological affair, best left to historians; for as it has also been very well said: “No philosopher understands his predecessors until he has re-­thought their thought in his own contemporary terms; and it is characteristic of the very greatest philosophers, like Kant and Aristotle that they, more than any others, repay this effort or re-­thinking.”6 We must re-­think the arguments of the Indians. Reason itself will be enriched if we do so. REASON ENRICHED

I have argued that a modern Indian identity should avail itself in a substantive way of resources of reason from the whole of India’s past. In what follows, I will discuss examples across the spectrum: from the work of Buddhist intellectuals at the time of Ashoka to post-­Independence fiction, from ancient Vedic ritual theory to early modern Islamic and Jaina syncretism. Reasoning, it has been suggested, is implicated in “choice over what significance to attach to our different identities” (352). The Stoics put forward a theory according to which there are four “personae,” each of which has a role in practical decision making. Their personae are: one’s rational being, the position one is born into, the choices one makes, and what fortune deals. The first of these personae is a part of our common humanity, something we share; the other three are individual. Insofar as it is possible to deliberate about one’s identity, therefore, the important point is that two forces will always be at work, one which takes our common humanity as its starting point, the other making choices based on our individual needs, natures, inheritances, and situations. The dialectical interplay between these two forces has been a running theme in Indian discussions of selfhood, the quest for a universal self having been put firmly on the intellectual map by the Upanishads two millennia or more ago. It is not easy to decide, in a Kantian fashion, what to do on the basis of the principles of pure reason and yet no easier to exercise real choice in matters that shape the sort of person one is to be. So we tend to move back and forth between the two, in a sort of perpetual oscillation. Something of this is caught brilliantly by a Tamil short story writer of post-­Independence India, whose pen name was “Mauni,” the silent one. He uses the narrative device of the double to investigate the to-­and-­fro between our cosmopolitan and our individual selves. Mauni explores the nature of the self, and the role of choice in fashioning identity, in four brilliant stories written between 1954 and 1971: “From Death, Creation,” “Beyond Perception,” “Error,” and “Wasteland.” His most famous line sums up the tension I am speaking about: “Of what, then, are we the moving shadows?”7 One response to the oscillation is to attempt to cultivate a more fixed relationship with our rational being, our cosmopolitan self. We might think of this as involving something like a “return” to a natural state of being, an idea that is expressed by the Sanskrit term svāsthya, “coinciding with oneself,” and we might think of the relevant way to cultivate such a return as a reining in of individual emotional attachments (which texts like the Mahabharata describe as a “taming” of the self). To put that thought another way, what

INTELLECTUAL INDIA

43

we do is cultivate the ability to choose not to attach value or significance on the basis of our “private passions.” Indian intellectuals in the Nyāya tradition have given a decidedly epistemological account of this accomplishment, according to which it is knowledge that renders such choice possible. Possessing knowledge, therefore, is necessary for agency and freedom. Knowledge enables the cultivation of a rational, cosmopolitan self. Cultivating an ability to make choices is a way of giving orientation to reason, and it can be helpful to draw further on nautical imagery and to contrast orientation by means of the Pole Star with orientation by means of a compass. The Pole Star is a fixed point in the distance, upon which we set our sights. In the present context, what this corresponds to is what Kant called a regulative ideal. Nirvāṇa and mokṣa are regulative ideals; so too perhaps is the ideal life of a Buddha or a sage. The important point is that, although we take aim at them as a way of giving direction to our choices, it is not necessary that we should expect ever to arrive there. Orientation by means of a compass is quite different. What it corresponds to in the domain of practical reason is the use of maxims and principles in making decisions. Just as one can use a compass even in deep fog, when one’s destination is not visible, so one can let moral rules or religious commands be one’s guide in the course of deciding what to do. As we have mentioned above, Indians like Tagore have been skeptical about so orienting reason; others, though, have brought reason to its defence. A skeptical voice is heard in the following passage of the Mahābhārata: There was an ascetic by the name of Kauśika, lacking very much knowledge of the scriptures. He lived in a spot far from the village, at the confluence of several rivers. He made a vow, saying, “I must always speak the truth.” He became celebrated as a speaker of the truth. At that time some people, running in fear from a band of robbers, entered the wood where Kauśika lived. The robbers, filled with rage, searched carefully for them there. Then, approaching Kauśika, the speaker of truth, they questioned him, saying “o holy one, by which path have a group of men gone a little while ago? Answer us this question, which we ask in the name of Truth. If you have seen them, tell us.” Thus adjured, Kauśika told them the truth, saying, “Those men have entered this wood crowded with many trees and creepers and plants.” Exactly so did Kauśika give them the information. Then those cruel men, it is reported, locating the people they sought, killed all of them. As a consequence of that great sin, which consisted in the words he spoke, Kauśika, ignorant of the subtleties of morality, descended into hell.8

The scriptures of which Kauśika is unaware are, presumably, those Dharmaśāstra passages which explain the circumstances under which untruthfulness is permissible.The clear implication is that Kauśika is to be held responsible for the consequences of his thoughtless commitment to truth telling, that he is culpably ignorant of the analysis of moral reasoning available in the literature. It is important to understand how the resources of reason can make internal dissent possible. This is especially the case with respect to the broad family of culturally similar traditions that is Hinduism, for Hinduism has often been regarded by its opponents as intolerant of dissent and by its proponents as speaking with a single voice. Of many skeptical voices within Hinduism, let me mention one that challenges the moral authority of the Vedas on rational grounds. The argument appeals to broad principles of rational interpretation: the Vedas, it is said, are verifiably mistaken, internally inconsistent, and pointlessly

44

JONARDON GANERI

repetitious (Nyayasūtra 2.1.57). As speech acts, the argument continues, they resemble the delusional ramblings of a drunkard; they carry no epistemological authority. An uncharitable view of religious tolerance might lead one to expect this skeptical argument to be met with censure and condemnation, but in fact it is joined in argument. Other principles of rational interpretation are advanced that resolve the inconsistencies and explain the repetitions, and a justification of the assent worthiness of the Vedic pronouncements is sought in a general epistemology of testimony. I have suggested that there are as many concepts of reason in India as there are calendars. An important contrast within orthodox Hinduism is reflected in the use of the terms hetu, “evidence-­based rationality,” and tarka, “hypothesis-­based rationality.” Manu, the author of the most influential of the law books, is disappointingly unequivocal in his criticism of the unconstrained use of evidence-­based reason, but he is considerably more willing to allow hypothesis-­based rationality (Manu).9 A careful examination of the resources of such rationality reveals that there is an underlying model of considerable flexibility and power. This model of rationality is based on two sorts of principles: i) principles for the selection of paradigmatic cases or exemplars, and ii) principles for the mapping of truths about the paradigms onto truths about other cases, based on rules of adaptation and substitution. One might imagine how one reasons when one is trying to change the battery of a new car, a process that involves remembering the procedure that worked on the old car and adapting it to fit the different layout and design of the new one. Clearly this “blueprint + adaptation” model is situational and particularist. I believe that it came to serve as the basis of a general theory of moral reasoning, leaving behind its origins in Vedic hermeneutics. And, as many texts make clear, it makes possible the existence of dissent and disagreement, for different decisions about what counts as an appropriate adaptation, and also what counts as a relevant paradigm, can always be advanced and defended. As a resource to be drawn upon in reasoning about one’s choices, the model is a highly versatile one. The details of this theory are found in works on Dharmaśāstra and Mīmāṃsā, especially in commentaries on the Manusṃrti and in Kumārila’s Tantravārttika. Public reasoning under a secular constitution demands a framework which is symmetrical or unbiased in its accommodation of a plurality of standpoints. Jaina and Buddhist dialecticians have done important theoretical work here. The tolerance of diversity of the third century bce Buddhist emperor Ashoka is well known, and the council he convened with the purpose of settling doctrine disputes between different Buddhist groups was run according to a codified theory of impartial debate. Debate is so structured as to give each party a fair and equal opportunity to rehearse their arguments and for counterarguments to be presented. Guiding the entire debate is an endeavor, not to find a winner and a loser, but to tease out the hidden assumptions that may lie in the background of some given position, so that there can be a clarification of what is at stake and what each party is committed to. The policy of making debates have the clarification of commitments as their function, rather than confrontation and victory or defeat, has important advantages. One is that it is easier to concede, if one comes to see that one’s position rests on hidden commitments that one would not endorse. There can be progress in public reasoning without there having to be winners and losers. The text which Ashoka’s dialectians produced is called the Kathāvatthu.10

INTELLECTUAL INDIA

45

If neutrality is one requirement of public reasoning, another is that there be common ground. Indian theoreticians describe this shared ground as the dṛṣtānta. or udāharaṇa; I will call it the “anchor” in a debate. Anchors are what ensure that acts of public reasoning engage with the participants, that what is disputed and what is implied are tied to their frames of reference. At one level, this is simply a reflection of the “blueprint + adaptation” model of decision making, shifted from the domain of individual practical reasoning to the domain of public consensus-­seeking reasoning (from svārtha to parārtha, in the terminology of the Indian theorists). Any given case that is to serve as the starting point in a public discussion must be such that its relevant features are agreed upon by all participants to the dialogue; otherwise, the act of public reasoning will not even get off the ground. So the existence of anchors is a requirement in an act of public reasoning. It is possible to develop an account of anchoring from ideas described in the ancient Ritual Sūtras as well as in early Nȳaya sources. I have been arguing that the full range of resources from intellectual India is available in the fashioning of modern Indian identities. J. L. Mehta has said about one such resource, the Rig Veda, that: “We in India still stand within that Wirkungsgeschichte and what we make of that text and how we understand it today will itself be a happening within that history.”11 I think we should generalize what Mehta says to include all the texts and traditions of India and to broker Indian identities in the global diaspora as well as in India itself. INDIA AND THE WORLD: GLOBALIZATION IN THE SEVENTEENTH CENTURY

India in the seventeenth century, the century after Akbar, was a place of great intellectual excitement. Muslim, Hindu, and Jaina intellectuals produced work of tremendous interest, ideas circulated around India, through the Persianate and Arabic worlds, and also between India and Europe. If ever there were a concrete embodiment of the open and spacious “idea of India” to which Tagore gave voice, it was here. Let me illustrate my point by focussing on a single year: 1656. In India, this was the year in which a long running process of religious isomorphism, pioneered by Akbar’s minister Abu’l Fazl and orchestrated around Ibn al Arabi’s idea of “Unity in Being” (waḥdat al-­wujūd), reached fulfilment in Dārā Shukoh’s grand project to translate fifty-­two Upanishads into Persian, a project for the sake of which he assembled in Varanasi a large team of bilingual scholars. Dārā believed that he could establish that the differences between Hinduism and Islam were largely terminological, and even that the Upanishads could be read as a sort of commentary upon the Qur’an. 1656 would also be the year in which the French philosopher and physician François Bernier would leave behind him the France of les libertins érudits, on a journey that would bring him soon to Mughal India. The leading disciple of the great empiricist Pierre Gassendi, it was Bernier who would eventually, on his return to France, make Gassendi’s work available to French and British audiences. Before doing so, however, he was to spend years as the court physician first of Dārā Shukoh and then of Aurangzeb, spending his spare time translating the works of Gassendi and Descartes into Persian, and discussing philosophical ideas at his home with the by now redundant members of Dārā Shukoh’s team of pandits, as well as with his scientifically minded patron Dānishmand Khān.To any one of them, Gassendi’s empiricist atomism would have resounded a familiar

46

JONARDON GANERI

Vaiśeska tone. Coincidentally, no doubt, this was to be the time when three of Varanasi’s leading Navya Nȳaya philosophers would compose brilliant and original Vaiśesika works: Mādhavadeva’s Nyāyasāra, Raghudeva’s Dravyasārasaṃgraha, and most particularly Jayarāma Nyāyapañcānana’s Padārthamālā, a work completed in 1659. All three were also signatories to a 1657 “letter of public declaration” (nirṇayapatra) by a large group of Varanasi Sanskrit scholars, an historically important act of public reason in its own right. If Europe had not yet been awoken to Gassendi’s empiricist alternative to Cartesian rationalism, 1656 was nevertheless to be a significant year there too, for it was the year in which Spinoza received his cherem in Amsterdam, an excommunication from the Portuguese Jewish community as a result of some anticipation of the heretical views he would later systematize; the pantheistic/atheistic doctrine that “God or Nature” is the only substance, the denial of miracles, of prophecy, and of scriptural revelation. While China was perceived by Pierre Bayle and Nicolas Malebranche to be a land inhabited by Spinozists, it would fall to Abraham Hyacinthe Anquetil-­Duperron, while introducing his translation of Dārā Shukoh’s Persian Upanishads into Latin, to declare that the Upanishads indeed represented a creed of “true Spinozism.”12 Bernier himself reported on the influence of the “Unity of Being” idea that had underpinned Dārā Shikoh’s project, which, he says, “has latterly made great noise in Hindoustan, inasmuch as certain Pendets or Gentile Doctors had instilled it into the minds of Dara and Sultan Sujah, the elder sons of Chah-­Jehan.”13 Claiming that this same doctrine can be found in Plato and in Aristotle, and that it is almost universally held by the “Gentile Pendets” of India, he adds that “this was also the opinion of [Robert] Flud[d,] whom our great Gassendy has so ably refuted.”14 (Fludd drew on Neoplatonist resources in a search for hidden connections between a purely intelligible realm and the realm of sensation.) Bernier describes the idea in question as being that God … has not only produced life from his own substance, but also generally everything material or corporeal in the universe, and that this production is not formed simply after the manner of efficient causes, but as a spider which produces a web from its own navel, and withdraws it at pleasure.The Creation then, say these visionary doctors, is nothing more than an extraction or extension of the individual substance of God, of those filaments which He draws from his own bowels; and, in like manner, destruction is merely the recalling of that divine substance and filaments into Himself.15

Bernier concludes that this idea has led him to take as his motto the slogan that “There are no opinions too extravagant and ridiculous to find reception in the mind of man.”16 No reader of David Hume’s Dialogues Concerning Natural Religion will fail to identify either the image of the spider or the slogan with which Bernier concludes. It is as likely that Hume read Bernier directly as that he received the simile from Bernier’s correspondent Pierre Bayle, whose Historical and Critical Dictionary was published in French in 1697 and includes information from ancient Greek sources about the Indian Brahmins as well as from contemporary travellers’ accounts.While it was certainly rhetorically convenient and effective for Hume to ridicule the notion in its Indian formulation, his attack on the rationalistic explanation of the unity of the world brought the career of that widely admired cosmopolitan idea decisively to an end.

INTELLECTUAL INDIA

47

To illustrate the importance of these relationships, it is perhaps worth stressing just how well connected Bernier was with the circles that fashioned the early Enlightenment in Europe, the libertins érudits and the nouveaux pyrrhoniens, and how quickly ideas circulated between India and Europe. Among his correspondents, for example, was the prominent new skeptic, François de La Mothe Le Vayer; and if John Dryden had already penned his play Aureng-­Zebe in 1675, this was only because Bernier’s Travels had by then been translated into English by the first Secretary of the Royal Society, Henry Oldenburg. With Gassendi’s work translated into Persian even before it was available in French, and the monistic pantheism of the Upanishads and Dārā Shukoh already in France and England years before Spinoza’s Ethics were published, what more dramatic evidence could there be of the rapid global circulation of ideas in the 1660s and 70s? Some years before, Dārā Shukoh sat next to Shāh Jahān as they received a deputation by the Varanasi poet Kavīndrācārya Sarasvati, who had come to argue that a tax on Hindu pilgrimage to Varanasi and other pilgrimage sites be repealed. He engaged them in philosophical conversation and indeed described to them the contents of Śaṅkara’s ­Brahmasūtrabhāṣya. When he returned triumphant, the speeches in his honor were led by the Navya Nȳaya scholar Jayarāma Nyāyapañcānana, a senior public intellectual of the period and someone whose students received the support and encouragement of Kavindra. Himself the student of the great Navadvīpa scholar Rāmabhadra Sārvabhauma, Jayarāma had impeccable credentials in Navya Nȳaya. He and his school of philosophy might be said to represent to the intellectual world of early modern India what Gassendi and Mersenne did to early Enlightenment Europe. Indeed, just as Gassendi sought to reintroduce the empiricist atomism of Epicurus through a colossal Latin commentary on the Epicurean texts, the Syntagma Philosophicum, so Jayarāma would be the first person in many centuries to write a commentary upon the early Vaiśeṣik sources, his Siddhāntamālā. In his discussion and defence of atomism, Jayarāma refers intriguingly to a group he calls “the new skeptics” (navyanāstikāḥ); I have yet to determine exactly who he had in mind. This is the point at which to say something about Yaśovijaya Gani, by far the most brilliant and productive Jaina intellectual of the seventeenth century.The tolerant pluralism of the Jainas had attracted Akbar, but Yaśovijaya’s story holds a particular interest because he provides a link between the two traditions of thinking I have been describing, the theistic monism of the Upanishads, Dārā Shikoh and Ibn al’Arabi on the one hand, and the empiricist atomism of the new Vaiśeṣikas on the other.Yaśovijaya spent an extended apprenticeship studying Navya Nȳaya in Varanasi, his stay coming to an end shortly before Dārā’s translational project began. Yaśovijaya himself, after a period writing Jaina philosophical treatises using the techniques and methods of Navya Nȳaya, then switched to writing works of a more spiritual orientation, several dealing with concepts of the self. Both Dārā and Yaśovijaya would appeal to the venerable metaphor of the ocean and its waves to explain the relationship between apparent multiplicity and underlying unity they sought to articulate in their common quest for an interdoctrinal rapprochement. In Yaśovijaya’s case, however, the unifying conception is more like that of a shattered mirror, in which each fragment reflects some part or aspect of the reflected object, a representation of which is to be built up by reconstructing all the partial images; while for Dārā it is as if multiple images of some one thing, some clearer and some indistinct, are reflected in the surfaces

48

JONARDON GANERI

of parallel facing mirrors. In the later work of Yaśovijaya, there is, I would suggest, a clear expression of a range of values associated nowadays with political liberalism. It is clear that there was a good deal of movement between Varanasi and Navadvīpa (in Bengal) at this time; indeed Raghudeva’s scribe Mahādeva Puntāṃkara tells us that he made a number of trips himself in order to study with the Navya Nȳaya pandits there. One person to make at least part of that journey was the Englishman John Marshall. Marshall was a friend of the Cambridge Platonist Henry More, to whom indeed he entrusted his literary remains. He found work with the East India Company and travelled between Hugli and Patna in the early 1670s, where he died in 1677. What makes him remarkable is his interest in Indian ideas. Having learned Persian and Arabic, he prepared English translations of two works, the Sāma Veda from Madhusūdana’s Persian, and the Bhāgavatapurāṇa. Two centuries later, E. B. Cowell would declare that if Marshall’s researches had been published “in 1680, they would have inaugurated an era in European knowledge of India, being in advance of anything which appeared before 1800.”17 In Marshall’s diaries we find him drawing attention to the existence of the story of the Flood in the Hindu scriptures,18 presumably as it is found in the Bhāgavatapurāṇa itself, this over a century before William Jones made comparative mythology into a science. Marshall also knew of the Hindu conception of the world as a cobweb emanating from God,19 already mentioned by Bernier. Marshall’s descriptions are especially remarkable in being free of the condescending tone of most later European writing from India. CONCLUSION

The “idea of India” is indeed an open, assimilative, and spacious one, sustaining a plurality of voices, orthodox and dissenting, of many ages, regions, and affiliations. Modern Indian identities in the global diaspora, as much as in India itself, can call upon all these voices and traditions, re-­think them, adapt and modify them, use the resources of reason they make available in deliberation about who to be, how to behave, and on what to agree. That is a fundamental freedom, one which ought not to be surrendered in binding oneself to narrower, constricted understandings of what India is. NOTES 1. Amartya Sen, The Argumentative Indian: Writings on Indian History, Culture and Identity (New York: Picador, 2006) (hereafter cited in text). 2. “Sherpas Call Tune for Political Masters,” The Guardian, June 30, 2005. 3. David Smock, “Itjihad: Reinterpreting Islamic Principles for the Twenty-­First Century,” United States Institute of Peace, Special Report 125 (Aug. 2004), available online at http:// www.usip.org/pubs/specialreports/sr125.html. 4. Gautama, Nȳayasūtra, ed. Anantalal Thakur (Delhi: Indian Council of Philosophical Research, 1997) (hereafter cited in text by book, chapter, and verse). 5. Vauvenargues, Réflections et maximes, §400; quoted from The Reflections and Maxims of Luc de Clapiers, Marquis of Vauvenargues, trans. F.G. Stevens (London: Humphrey Milford, 1927). 6. Peter Strawson, Individuals: An Essay in Descriptive Metaphysics (London: Methuen, 1959), 11.

INTELLECTUAL INDIA

49

7. Mauni, Short Stories, translated by Lakshmi Holmström (New Delhi: Katha, 1997), 48. See my “‘What you are you do not see, what you see is your shadow’: The Philosophical Double in Mauni’s Fiction,” in The Poetics of Shadow:The Double in Philosophy and Literature, ed. Andrew Ng Hock Soon (Hanover: Ibidem-­Verlag, forthcoming). 8. Mahabharata of Krishna-­Dqaipayana Vyasa, trans. K. M. Ganguli, 3rd ed. (New Delhi: M ­ unshiram Manoharlal Publishers, 1976), 8.69.40–46. 9. Manu’s Code of Law: A Critical Edition and Translation of the Mānava-­Dharmaśāstra, ed. Patrick Olivelle (New York: Oxford Univ. Press, 2005). For Manu on evidence-­based reason see 2.10–11; for hypothesis-­based rationality see 12.106. 10. Points of Controversy, or, Subjects of Discourse: Being a Translation of the Kathāvatthu from the Abhidhammapitaka, ed. S. Z. Aung and C.A.F. Rhys Davids. Pali Text Society, Translation Series no. 5 (London: Luzac & Co., 1960). 11. J. L. Mehta, “The Rgveda: Text and Interpretation,” in Philosophy and Religion: Essays in Interpretation (Delhi: ICPR, 1990), 278. 12. M. Anquetil-­Duperron, trans., Oupnek’hat, id est, Secretum tegendum, translation of the Upanishads, vol. 2 (Paris: Levrault, 1801–2), 659, 665. 13. François Bernier, Travels in the Mogul Empire, A.D. 1656–1668, trans. Archibald Constable (1891; London: Humphrey Milford, Oxford Univ. Press, 1916), 345. 14. Bernier, Travels in the Mogul Empire, 346–47. 15. Bernier, Travels in the Mogul Empire, 347. 16. Bernier, Travels in the Mogul Empire, 349. 17. E. B. Cowell qtd. in John Marshall’s entry in Biographical Register of Christ’s College 1505– 1905, and of the Earlier Foundation, God’s House, 1448–1505, ed. J. A.Venn (Cambridge: Cambridge Univ. Press, 1910–1913), 592. 18. John Marshall in India: Notes and Observations in Bengal, 1668–1672 (London: Oxford Univ. Press, 1927), 181. 19. John Marshall in India, 186.

P A R T

II

PHILOSOPHY AS A SEARCH FOR THE SELF

CHAPTER

4

The Upanis. ads Joel Brereton

Source: William T. de Bary and Irene Bloom (eds), Approaches to the Asian Classics (New York: Columbia University Press, 1990), pp. 115–135. The Upanishads are early texts of the Hindu tradition which set forth the foundations of the world and the true nature of the self. They are formally quite diverse, for they include narratives, dialogues, verses, and the teachings of ancient sages. The principal Upanishads, which were composed probably between 600 and 300 b.c.e., constitute the concluding portion of the Veda, the most ancient and conventionally the most fundamental scripture of Hinduism. According to most reckonings, there are fourteen Vedic ­Upanishads, and these can be assigned a relative chronology on the basis of their literary form and language. The oldest are in prose. Among them are the Bṛhad Āraṇyaka, Chāndogya, Kauṣītakī, Taittirīya, and Aitareya Upaniṣads. A second, generally later group of Upanishads were written in verse. These include the Kaṭha, Īśā, Muṇḍaka, and Śvetāśvatara Upaniṣads. Finally, the youngest are also in prose, but in a style which is closer to classical Sanskrit than to the more archaic language of the oldest Upanishads. These include the Maitrī and Māṇḍūkya Upaniṣads. This division is only approximate, for individual U ­ panishads may contain material from different periods. It is not quite accurate to say that the Upanishadic period ended with these texts. The term “Upanishad” became the designation of a genre rather than of specific texts, for works called Upanishads continued to be written through the Middle Ages and even into the early modern period. Tradition holds that there are 108 Upanishads—108 is a sacred number—but even more texts claim the status of Upanishads. These later Upanishads fall into different types which reflect the major developments in Hindu religious thought and practice. Some discuss topics and ideas introduced in the earlier Upanishads, others teach yoga and the renunciation of the world, and still others reflect sectarian worship of the classical Hindu deities. These Upanishads are significant, but none of them has had 53

54

JOEL BRERETON

the enduring influence on Indic thought that the Vedic Upanishads have had. This essay, therefore, will be concerned entirely with the latter. The Vedic Upanishads play a critical role in the history of Indic religion. Literarily, they are the last sections of the Brāhmanas, which are commentaries on the great public rites of the Vedic tradition. This literary position also reflects their historical place, for they are later than most of the Brāhmanas, and their thought developed out of the Brāhmanas. Historically, their period was one of transition, when the foundations of classical Hindu religion were established and archaic forms of Vedic religion were superseded. Therefore, many basic elements of Hindu religion were first clearly articulated in the Vedic Upanishads. These include the ideas of karma and rebirth; instructions concerning yoga, meditation, and asceticism; the concept of a self beyond the individual self; and the view that there is a single reality hidden by the multiple forms of the world. The Upanishads are significant also because they continued to exercise a decisive role in Hindu religious history long after their composition. In fact, unlike the rest of the Veda, they still remain a major source of inspiration and authority within Hinduism. The most influential schools of Hindu religious thought are the schools of the Vedānta, a term that means “the end of the Veda.” They have that name because they understand the Upanishads, which are the last part of the Veda, to be the essence of Vedic teaching and the ultimate authority regarding the true nature of things. Of these schools, one has dominated both Indic and Western interpretation of the Upanishads. This school is the Advaita Vedānta, the “Nondualist Vedānta,” which was established by a teacher named Śankara, who lived around 700 c.e. According to Śankara, the Upanishads teach that ultimately there exists one, and only one, indivisible reality. Hence the world and all its distinctions are less than fully real, and whatever reality they do possess derives from that one reality. Śankara must occasionally struggle hard to ground this view in the Upanishads, for although much of Upanishadic thought does stress the coherence and final unity of all things, it is not easily reduced to his or any other simple formula. Other schools of the Vedānta differed profoundly with Śankara and with one another in their interpretations of the Upanishads. And these differences point to an important characteristic of the Upanishads. Like Western scriptures, they are not catechisms of direct answers to religious questions, which obviate the need for any further reflection. Rather, they stimulate thought and challenge interpretation. They have this character for several reasons. First, even individual Upanishads are far from systematic. They were composed by different people, living at different times and in different areas of northern India. Moreover, those who put the Upanishads into their final shape combined and even interwove various teachings. As a result, an argument may move in different directions even within one section of one Upanishad. Second, especially the older Upanishads used the language of symbols and concrete images rather than that of abstraction. Some symbols were traditional, but the Upanishadic sages also created new images and refashioned old ones, especially the symbols of the Vedic ritual. Such language gives Upanishadic diction the appeal of concrete narrative and the resonance of images, but it also opens them to very different interpretations. Thus, though the Upanishads are often understood philosophically, they call for not so much a systemization and specification of meaning as a reading which permits the associations and expansions of poetic discourse.

55

THE UPANIṢADS

While the complexity of Upanishadic language and literary history precludes any easy summary of their teaching, there is a broad theme that encompasses much of their thought. In general, each Upanishadic teaching creates an integrative vision, a view of the whole which draws together the separate elements of the world and of human experience and compresses them into a single form. To one who has this larger vision of things, the world is not a set of diverse and disorganized objects and living beings, but rather forms a totality with a distinct shape and character. The Upanishads often create such an integrative vision by identifying a single, comprehensive and fundamental principle which shapes the world. One term by which the Upanishads designate that fundamental principle is “brahman.” For later followers of the Vedānta, the brahman has a particular definition and a specific character, but for the Upanishads, the brahman remains an open concept. It is simply the designation given to whatever principle or power a sage believes to lie behind the world and to make the world explicable. It is the reality sought by the householder who asks a sage: “Through knowing what, sir, does this whole world become known?” (Muṇḍaka Upanisṣad 1.13). Such an aspiration for a total vision of things is not unique to the Upanishads or to India. One distinctive turn which the Upanishads give to this vision is their understanding that the fundamental principle of everything is also the core of each individual. The typical designation for that core is the ātman, the “self.” Thus, in Upanishadic terms, the brahman is discovered within the ātman, or conversely, the secret of one’s self lies in the root of all existence. Within this common approach, however, the Upanishads differ among themselves in the shape they give to that vision of totality and the means by which they create it. To clarify this vision of the Upanishads and to show their diversity, I will present five paradigms that these teachings follow. Each paradigm is a method or pattern through which the Upanishads construct a totality out of the multiplicity of the world. These five Upanishadic paradigms are: (1) the correlation of different aspects of reality to one another; (2) the emergence of the world from a single reality and its resolution back into it; (3) a hierarchy which leads ultimately to the foundation of all things; (4) a paradoxical coincidence of things which are ordinarily understood to exclude or oppose one another; and (5) a cycle which encompasses the processes of life and the world. These paradigms do not exhaust the variety of Upanishadic teachings, but collectively they suggest their range. CORRELATION

One means of demonstrating unity behind apparent diversity is by displaying correspondences among things belonging to different domains. This was not a new technique in Indic thought, since already the Brāhmanas had made extensive use of the same paradigm. But while the Brāhmanas sought such correlations within the domain of the ritual and between the domains of the ritual and the outside world, the Upanishads search primarily for those that exist within and among the human and natural domains. The Upanishads have several ways of identifying these correspondences. One is to link various parts of the world to a single, structured entity. That is, they take a symbol or set of symbols which is comprehensible and ordered, and then relate items from different domains to it. In this way, the symbol becomes a map. Just as one understands an unfamiliar

56

JOEL BRERETON

territory by comparing various places and objects to areas on a map, so one comprehends the world by associating its parts to a known object. In the Upanishads, the ordering object or domain is normally something concrete. Thus in the opening of the Bṛhad Āraṇyaka Upaniṣad, the sacrificial horse is the image of the world. The passage begins: “Now, the head of the sacrificial horse is dawn; his eye, the sun; his breath, the wind; his open mouth, the fire which is common to all” (Bṛhad Āraṇyaka Upaniṣad 1.1.1). The passage thus identifies the head of the horse with the symbols of the main divisions of space. The sun represents the heaven; wind, the midspace between earth and heaven; and fire, the earth. It then goes on to equate other parts of the land and air with the body of the horse, the seas with the sacrificial vessels which stand on either side of the horse, and various living beings, from gods to humans, with the different names by which the horse is addressed in the ritual. The passage thus reduces the whole world to the form of the horse, and by doing so, it makes the world a single, comprehensible object. Therefore, those who reflect on the horse understand and embrace everything in their contemplation. The use of a ritual object like the sacrificial horse in such a system of correlations is more typical of Brāhmanas than of Upanishads. For the latter, the most frequent correlation is between the macrocosm and the human body. That is, they equate the parts of the body to the constituents of the visible world, so that the whole world becomes the image of the human form. In this way, instead of appearing external and alien, the world becomes a familiar place and a place in which humans occupy a central position. The Aitareya Upaniṣad, for example, opens with a narrative which tells how the world and the body became the twin images of one another. Although this narrative has the form of a creation story, it is better read not as describing the actual process of creation, but as establishing the connections that now exist within the world. In the beginning, there was only the self, the ātman. This self resolves to create. It first produces the basic form of the world, and then from the waters, it draws forth a being that the text calls the “person.” This person is egg-like, for it is without faculties of sense or action.Therefore the self next creates these faculties by bringing them forth from the person one by one.Then from these faculties, it extracts the characteristic product or activity of each one, and from these in turn, it brings forth a corresponding aspect of the world. Thus, from the mouth of the person comes speech and then fire; from its nostrils, breath and wind; from its eyes, sight and sun; from its ears, hearing and the four directions; from its skin, hair and plants; from its heart, thought and the moon; from its navel, inhalation and death; and from its penis, semen and water. The associations which provide the basis for most of these correlations are not difficult. The one exception is the odd correlation of the navel, inhalation and death. On the one hand, the inhalation is drawn toward the navel, which connotes birth and life. But on the other, inhalation is linked with eating, which implies human physicality and therefore human mortality. Hence, inhalation also signifies death. The creation of the human being follows in the next part of the narrative. The macrocosmic realities just created fall into the sea once again, back to their place of origins.This descent into the sea figuratively expresses the disorganization of these powers. They need an order, and to find it, they enter into the human form once again. Fire becomes speech and enters the mouth; wind becomes breath and enters the nostrils; the sun becomes sight and enters the eyes, and so on. The sequence exactly reverses the order of their creation.

57

THE UPANIṢADS

Thus, the elements of the world are created from the cosmic person, and then they return again to the human form in which all people share. Hence, the world mirrors the human body, and its parts correspond to human faculties. Finally, at the end of the process, the self itself enters into the newly created human form. In this way, the self, which is the origin of all, becomes the self of each human being. Thus, as the body corresponds to the visible world, the self of the world coincides with the self of the individual. Both physically and spiritually, therefore, the human being is a perfect microcosm. The correspondence between the microcosm and the macrocosm illustrates how the Upanishads fashion a vision of totality through correlation. This correspondence unifies the world in the form of the person, and therefore makes the world comprehensible as a whole. Furthermore, it also implies that the world and the power that controls it are not outside, bearing down upon and threatening the individual. Rather, because the parts of the world are equivalent to the parts of a person, humans include everything within themselves. EMERGENCE AND RESOLUTION

The Upanishads also create a sense of the unity of all things through viewing creation as an emergence from a single reality and destruction as a return to that reality.The Muṇḍaka Upaniṣad provides some striking images of this process: “As a spider spins and gathers (its web), as plants grow upon the earth, as head and body hair (grow) from a living person, so everything here arises from the imperishable” (Muṇḍaka Upaniṣad 1.1.7).The image of the spider and its web is especially strong. The spider stands alone at the beginning, spins its web out of itself, and finally draws it back into itself. In the same way, the “imperishable” brahman emits the world from itself and ultimately reabsorbs it back into itself. Thus, the world is only the outward projection of the brahman. One of the most influential realizations of this paradigm occurs in Chāndogya Upaniṣad 6. The chapter takes the form of a dialogue between a sage named Uddālaka and his arrogant son Śvetaketu. After studying the Vedas for twelve years, Śvetaketu comes home very proud of his achievement. His father then asks him if during his education, his teachers taught him that “by which what has been unheard becomes heard, what has been unthought becomes thought, what has been unknown becomes known” (Chāndogya Upaniṣad 6.1.3). To understand clay or copper, for example, means that one also knows the character of all clay and copper utensils, because their qualities will reflect the substance of their composition. In the same way, he asks, has Śvetaketu been taught the basic reality which comprises all things and through which the character of all things is known? Śvetaketu admits that he knows nothing of this, and Uddālaka begins his explanation of that reality. In the beginning there was only being. From being then follows an evolution toward increasing materiality. Being, which is imperceptible, first gives rise to heat, which can be felt. Then heat gives rise to water, which can be felt and seen, and finally water gives rise to food, which can be felt, seen, and tasted. Food connotes full materiality. Uddālaka then confirms these relations by seeing their reflection in natural processes. Heat does produce water in humans, for when they are hot, they sweat, and rainwater does produce

58

JOEL BRERETON

food. Similarly, human experience corroborates the reverse process, the resolution of each factor into the previous one. According to Uddālaka, people become thirsty when heat absorbs water and hungry when water absorbs food. Once he has established the fundamental evolution from being, Uddālaka then shows how everything, and especially human beings, are the products of the intermingling and segmentation of the three evolutes. In the body, for example, each of the evolutes divides into three: food becomes mind, flesh, and excrement; water becomes breath, blood, and urine; heat becomes speech, marrow, and bone. Thus, the human being is nothing but a composite of the three basic factors. Likewise, the continuous process of the emergence and resolution of the three evolutes is reflected in the process of dying. When people die, they first lose consciousness, but they still breathe. This is the reabsorption of food by water, for mind and consciousness derive from food, but breath comes from water.Then, when water returns to heat, the breath ceases, for water corresponds to breath. Even without the breath, however, the body remains warm, for heat still remains. Finally, as the heat and the life of the person dissipate into being, the body becomes cold and death becomes complete. In Uddālaka’s vision, therefore, being is the ultimate, pervasive source of mind, body, and the world, and life continues through a constant process of movement out of and into being. In the second half of his teaching, Uddālaka drives home its significance for Śvetaketu’s self-understanding. In one section, he asks his son to bring him a fruit from the nyagrodha tree, one of the sacred fig trees of India. He tells him to cut it open to find the seed, and then to cut the seed open.When he does so, Śvetaketu finds the seeds in the fruit, but then he cannot find anything in the seed.Yet within that seed is an essence, says Uddālaka, and that invisible essence is the source of the life of the huge nyagrodha tree. “Believe me, my child,” he concludes, “that which is this finest essence—this whole world has that as its self.That is the real.That is the self.Thus are you, Śvetaketu” (Chāndogya Upaniṣad 6.12.3). Being is that essence, for being is the imperceptible source of all creation. This being is therefore the self of the nyagrodha, the self of the world, and the self of Śvetaketu. Here again, as in the Aitareya Upaniṣad, the true self is not the individual self, but rather the identity that one shares with everything else.There is no true distinction among living beings, for they all emerge from being and retreat to it. All things, both animate and inanimate, are unified in being, because they are all the transformations of being.To understand the nature of being, therefore, is to have the knowledge that Uddālaka promised Śvetaketu, the knowledge that accounts for everything, the knowledge of the totality of things. HIERARCHY

A third method of organizing experience is through constructing a hierarchy. That is to say, the Upanishadic sages set up a system of levels that shows which powers include other powers or which are dependent on which others. Ultimately, by moving toward progressively deeper levels, the sage identifies the fundamental principle on which everything else is established. In one sense, this is the most characteristic technique of Upanishads, for it is from it that the Upanishads have their name. The word “upaniṣad,” though usually translated “secret teaching” or the like, originally meant the subordination of one thing to another. The purpose of arranging things in such a progression is finally to identify the dominant

THE UPANIṢADS

59

reality behind an object. In Chāndogya Upaniṣad 8.7–12, for example, the god Indra and the anti-god Virocana approach the creator deity, Prajāpati. They ask him the true nature of the self. Rather than giving them the real answer straightaway, Prajāpati first tells them that the true self is nothing but the self that one sees in a reflection. This answer satisfies Virocana; “So then, with tranquil heart Virocana went to the anti-gods. To them he declared this upaniṣad: ‘Oneself (ātman) is to be satisfied here. Oneself is to be served. He who satisfies his own self here, who serves himself—he gains both worlds, this one here and yonder (heavenly world)’” (Chāndogya Upaniṣad 8.8.4).Virocana’s upaniṣad holds that the physical self is subordinate to no other self but is the true and fundamental self. However, to revere the physical self as ultimate is a very dangerous upaniṣad. As Indra and ­Virocana depart with this teaching, Prajāpati observes that whoever “will follow this upaniṣad, whether they be gods or anti-gods, will pass away” (Chāndogya Upaniṣad 8.8.4). They will perish, for they do not identify themselves with an immortal self but with a self that dies. This truth is finally recognized by Indra, who returns to Prajāpati for further instruction. Gradually, Prajāpati leads Indra into more profound definitions of the self until at last, after Indra has studied with him for 101 years, Prajāpati reveals to him the reality behind the physical self and the true foundation of the self. The definition of the self which Indra and Virocana seek is one of the dominant concerns of Upanishads. And of all the Upanishadic teachers, the one most closely associated with the quest for the self is a sage named Yājnavalkya. Yājnavalkya appears in Bṛhad Āraṇyaka Upaniṣad 3–4, first in a verbal contest with other brahmins and then in dialogue with King Janaka of Videha. According to Yājnavalkya, the foundation of the self is the subject upon which all consciousness depends and all action is established. Because that true self is always the perceiving subject, it itself can never become an object of thought or perception: “You could not see the seer of seeing. You could not hear the hearer of hearing.You could not think the thinker of thinking.You could not know the knower of knowing” (Bṛhad Āraṇyaka Upaniṣad 3.4.2). That same self, however, is also the self of the world.Yājnavalkya calls it the “inner controller,” which sustains both the body and the world, but which neither the person nor the world knows (Bṛhad Āraṇyaka Upaniṣad 3.7). It is the “imperishable,” at whose command “the sun and the moon,” “heaven and earth,” and “moments, hours, days and nights, fortnights, months, seasons and years remain distinct,” and at whose command, “some rivers flow eastward, others westward” (Bṛhad Āraṇyaka Upaniṣad 3.8.9). The only way to understand this self is to strip it of any positive content; “This is described as ‘not this, not that’ (neti, neti). It is ungraspable, for it is never grasped. It is indestructible, for it is never destroyed. Without attachment, for it is not attached. Unbound, (yet) it is never unstable, never injured” (Bṛhad Āraṇyaka Upaniṣad 4.4.22). The end of this passage suggests the personal significance of this view of the self. To identify oneself with a self which stands above any recognizable object is to become invulnerable. Mind, body, emotions—all these are susceptible to pain and suffering. But all these can be made objects of knowing, and therefore there must be a knowing subject that is deeper and more fundamental than they. It is that unknowable, knowing subject which is the true self.Why then fear sickness, suffering, or death? These affect what the self perceives; they do not affect what the self is. Even if the self cannot become an object of thought, it can still be experienced. To investigate the experience of the self,Yājnavalkya examines dreamless sleep, which for him

60

JOEL BRERETON

is a state in which everything other than the self is forgotten. In this state, the self emerges alone and as it truly is. This state of the self in itself is one of complete fulfillment: “Now, this is that form of his which is beyond pleasure, in which evil is removed, which is free of fear” (Bṛhad Āraṇyaka Upaniṣad 4.3.21). For one who experiences the true self, there is nothing wanted and nothing lacking, and therefore neither yearning nor grief can trouble that person. In this state, the self never loses consciousness, “for there is no complete loss of knowing for the knower, because (the knower) is indestructible” (Bṛhad Āraṇyaka Upaniṣad 4.3.30). However, because there is nothing to be perceived, that consciousness has no object. Yājnavalkya captures this aspect of the experience of the self through an analogy: “Just as a man, when embraced by his dear wife, knows nothing within or without, even so this embodied self, when embraced by the conscious self, knows nothing within or without” (Bṛhad Āraṇyaka Upaniṣad 4.3.21). As the moment of sexual climax attenuates the awareness of any object and the distinction between oneself and the other, so the self knows nothing outside of itself. One’s own self (the “embodied self ”) experiences only itself as the “conscious self,” the ultimate subject of all knowing. However there is a problem in all of this. Yājnavalkya’s analogies of deep sleep and sexual intercourse apparently contradict one another, since people normally experience deep sleep as a state of unconsciousness, not one of objectless consciousness. For this reason, other Upanishadic teachers followed Yājnavalkya in identifying the true self as the ultimate subject, but rejected his identification of the state of the self and the state of deep sleep. Recall that in the dialogue between Prajāpati and Indra (Chāndogya Upaniṣad 8), Indra rejected Prajāpati’s first explanation of the true self as the bodily self. After further definitions of the self and after 96 years of study, Prajāpati tells Indra that the self is experienced in deep sleep. “Then with tranquil heart, (Indra) went forth. But even before reaching the gods, he saw this danger: ‘Obviously, now this (self) does not know itself, (does not know that) “This am I,” neither does it know (other) beings here. It becomes one gone to destruction. I see nothing useful in this’” (Chāndogya Upaniṣad 8.11.1). Indra finds the teaching inadequate because it suggests that when everything is removed from the self, only unconsciousness remains. So once more, Indra returns to Prajāpati, and after another five years, Prajāpati reveals to him the true self. Similarly, the Māṇḍūkya Upaniṣad teaches that the self is experienced in a fourth state beyond waking, dreaming, and dreamless sleep. According to this text, that fourth state is not unconsciousness, but neither is it consciousness, at least not consciousness as it is normally understood. The state of the imperceptible self is unique and not identifiable with any other state. These discussions of the self and of the experience of it are a reminder that the Upanishadic search for a definition of totality is not only an intellectual one. To be sure, the Upanishads seek the satisfaction of a comprehensive understanding of the world and the psychologically fulfilling identification with a deathless self beyond the individual self. But in addition, the Upanishadic views of self and the world also shape experiences in which the unknown self is directly realized. Some Upanishads make clear that the self is not something that can be taught, and that successful realization of self is not in the conscious control of the one seeking it: “This self cannot be attained by instruction, nor by intellect, nor by much learning. It can be attained only by the one whom it chooses. To him that self reveals its own form” (Kaṭha Upaniṣad 2.23, Muṇḍaka Upaniṣad 3.2.3). According to

61

THE UPANIṢADS

these texts, the knowledge of the self comes as if it were a revelation that breaks in upon the mind, and not as something intellectually achieved. Yet within the Upanishads, there are attempts to devise methods of reaching the self. The Śvetāśvatara Upaniṣad, one of the latest Upanishads, comes close to classical yogic methods of finding the self. According to Śvetāśvatara Upaniṣad 2.8ff., a yogi should sit with the body erect and steady, checking bodily movements and restraining the breath. Thus quieting the senses and the mind, the yogi begins to move inward toward the core of the self and the world: “Fog, smoke, sun, wind, fire, fireflies, lightning, crystal, the moon— these are the preliminary appearances” (Śvetāśvatara Upaniṣad 2.11). Deprived of external distraction, the mind experiences these forms of light, which emerge from the inner light of the self. Eventually, the yogi moves through the layers of the self to its innermost core: “Just as a mirror stained by dust shines brilliantly when it is well cleansed, so the embodied (self), seeing the nature of the self, becomes single, its goal attained, free from sorrow” (Śvetāśvatara Upaniṣad 2.14).The true self is the end of the journey, for it is the foundation on which are built all the other states of mind and self. Comparing the approach of Yājnavalkya and his successors and that of Uddālaka shows both the range and the common direction of the Upanishads. The two approaches begin with different paradigms for defining totality. Uddālaka’s method is to describe the principle from which the world evolves and into which it devolves. In contrast, Prajāpati, Yājnavalkya, and the yoga teacher of the Śvetāśvatara Upaniṣad distinguish different levels of existence and experience and then move through these levels until they uncover the final, fundamental level. Thus, according to Yājnavalkya, to shift from the world experienced while awake to that of dreamless sleep is to advance to the deepest level of the self. Similarly, in the Śvetāśvatara Upaniṣad, different meditative states mark the progress toward the true self. Their approach is based on a hierarchical ordering of reality. Not only do Yājnavalkya and Uddālaka operate with different paradigms, but also they begin from opposite directions. Yājnavalkya’s is essentially an internal approach, which seeks the fundamental principle by turning back the layers of the self. Uddālaka, on the other hand, begins externally, with the world outside the self.Yājnavalkya investigates the self psychologically by observing its different states of consciousness, while Uddālaka analyzes the physical world and reduces its constituents and processes to a simple system. But both try to confirm their teachings by observation, either of internal states or of external realities. And both, whether they move from inside to outside or outside to inside, finally locate a principle which is simultaneously at the core of both the self and the world. For Yājnavalkya, the self that is the unknowable subject of all knowing is also the self of the whole world. For Uddālaka, being gives rise to the world, but it is also the self to which all living things return at death and from which they come forth at birth.Thus both teachers create visions of totality, even though they construct them differently. PARADOX

Oddly, and perhaps even paradoxically, paradoxes can also create a unified vision. In a paradox, normally distinct objects are unexpectedly related or even equated to one another. If we encountered a very poor but happy couple, we could say that “paradoxically, they are rich in their poverty.” Normally, wealth and poverty exclude one another, but in these

62

JOEL BRERETON

people, material want and emotional richness coincide. In the same way, the Upanishads can use paradox to bring together things that appear to be separate in order to create a larger whole. The best known Upanishadic paradox is the teaching of the sage Śāṇḍilya. Actually, this teaching occurs for the first time in a late portion of the Śatapatha Brāhmana, but it is repeated almost verbatim in Chāndogya Upaniṣad 3.14. Śāṇḍilya’s teaching fuses the extremes of reality in a paradox. It concludes: “that self of mine in the heart is smaller than a grain of rice or of barley or a mustard seed or a grain of millet or the kernel of a grain of millet; that self of mine in the heart is greater than the earth, greater than the midspace, greater than the sky, greater than these worlds” (Chāndogya Upaniṣad 3.14.3). The self is the most intimate part of a person, the very center of one’s being, and therefore it is the smallest of the small. Yet, at the same time, it surpasses everything. The paradox thus undercuts any exclusion or any separation of an individual from the rest of the world, for there is nothing beyond the self. If the self is the very smallest and the very largest, what is not encompassed by it? For one Upanishad, the Īśā Upaniṣad, paradox is a central strategy.This entire poem consists of sets of paradoxes and antinomies. In V. 4, for example, the poet says that the One, the fundamental principle of things, is “unmoving” and yet it is swift, “swifter than the mind.” It is unmoving because it is eternal; it is swifter than the mind because it is inconceivable. Therefore the paradox, like that of the couple’s wealth and poverty, does not express a real contradiction, only a rhetorical one. But this rhetorical paradox lays the basis for a real paradox in the next verse. There the upaniṣad says that the One “moves and does not move. It is far away and it is near. It is within everything and outside of everthing” (Īśā Upaniṣad 5). Here the running-standing model from the previous verse is restated as a moving-not moving opposition. The rest of the verse then develops this opposition. The consequence of moving is that the One is far; the consequence of not moving is that it remains near.The limit of distance is to be outside everything; the limit of nearness is to be inside. The paradoxical claim is that the One is both. In some way, the One is beyond time and space, and yet it is also within the world as the source of everything.Timelessness and time, perfection and movement—these only appear to be opposites, for in actuality, the One is all of them. Both the teaching of Śāṇḍilya and that of the Īśā Upaniṣad, therefore, connect a single principle to opposite and apparently exclusive extremes. By so linking the extremes, they imply that this principle comprehends everything else as well. There is nothing that the principle does not include, nothing that remains separate from it and from everything else within it. In that way, the self of Śāṇḍilya and the One of the Īśā Upaniṣad become symbols of the totality of things. CYCLES

A final strategy for creating an integrative vision is to represent world processes as a cycle. Of the various constructions of a vision of totality, this has the clearest foundation in the earlier Veda.The Vedic rituals followed the repeating sequences of natural events: the alternation of day and night, the phases of the moon, the seasons of rain and no rain, and the succession of years. Through the rituals, therefore, people understood and regulated their lives according to the course of nature. Even more explicitly than this ritual tradition, the Upanishads consolidate life and death, the succession of natural events and the divisions of time into recurring cycles.

63

THE UPANIṢADS

One such cycle follows the movement of water, which represents the essence of life. This pattern appears in the parallel texts, Chāndogya Upaniṣad 5.3.7ff. and Bṛhad Āraṇyaka Upaniṣad 6.2.9ff. According to them, the gods fill the moon with soma, which is the holiest of all ritual offerings and which represents the elixir of life. As the moon wanes, soma is poured out as rain, which falls to earth where it nourishes the plants and becomes their sap and juice. Men ingest it with their food and pass that life essence to women through their semen. Within a woman, semen gives birth to a new human being. When that person dies and the body is cremated, the life essence rises once again with the smoke. For some, the life essence then returns to the moon and the cycle begins again. This vision thus integrates the birth and death of humans into the natural movement of water down to earth and up again to the sky. But that cycle does not continue for everyone. For those who understand this process and do not allow themselves to be entangled in the world controlled by this pattern, the cycle is broken at death. They go on to realms from which there is no further birth. These passages, therefore, contain two concepts which were to become central to the Indic tradition: the idea that time moves in cycles and the idea of rebirth. The earliest Upanishads present rebirth as a concept which few know. In Bṛhad Āraṇyaka Upaniṣad 3.2, for example, a brahmin named Ārtabhāga quizzes Yājnavalkya again and again about a person’s destiny after death. Finally, Yājnavalkya says, “‘Ārtabhāga, my friend, take my hand. We two alone will know of this. This matter of ours is not (to be discussed) in public.’ So the two went away and deliberated. What they spoke of was action (karma). What they proclaimed was action. Now, by good action one becomes good, by bad action bad” (Bṛhad Āraṇyaka Upaniṣad 3.2.13). This passage points toward the classical concept of karma, which comes to mean not only action but also the effect of action on one’s own destiny. According to the developed tradition, by the accumulation of good karma, one secures a good rebirth, by bad karma, a bad one. Later Upanishads take the reality of rebirth and karma for granted, and frame their goals in terms of them. The constant cycle of rebirths comes to express entrapment by death and suffering, and therefore release from that cycle becomes the critical achievement. Thus, the Kaṭha Upaniṣad starkly opposes the two possible fates: “On the one hand, he who is without understanding, who is unmindful and ever impure, does not attain the goal and goes into the cycle of rebirths. But he who understands, who is mindful and ever pure, attains the goal from which he is not born again” (Kaṭha Upaniṣad 3.7f.). This negative turn gives the cyclical paradigm an equivocal status. In the Brāhmanas and early Upanishads, the natural cycles provided models for making sense of the world and human life, and even in later texts, they never completely lost that significance. However, these cycles could also be seen not just as embracing life but also as imprisoning it. For this reason, in order to break out of the cycle of constant death and rebirth, this paradigm had to be rejected as the final truth about human life. THE INTEGRATIVE VISION

An integrative vision of things was not the only concern of the Upanishads, but it was a central one. This survey suggests some of the reasons for its significance. First and fundamental was the aesthetic and intellectual satisfaction in the ability to see things as a whole. The vision comprehends the world, and by it, people know who they are and where

64

JOEL BRERETON

they are. People understand that they are a part of everything, in fact, that they are at the very center of everything, and they know that everything is a part of them. Second, this vision was a powerful knowledge. From early in the Vedic tradition, to know the truth of something was to have control over it. To know the world is therefore also to master the world and to direct one’s fate. Especially the later Upanishads insist that insight into the true nature of things effects the highest attainment of all, the attainment of a final release from all temporal and spatial limitation. Third, the vision of totality was a transforming vision. Above all, it required a reevaluation of what one truly is and therefore what is truly consequential. In their various ways, the Upanishads argue that people are really not what they appear to be. They seem to be individuals, vulnerable to suffering and death, subject to their private destinies. That individual self, however, is not the true self. Death cannot affect the true self, nor can anything else, for the self precedes and embraces everything. The person who truly sees the self in this way, therefore, should have neither desire nor fear, for that person knows that no harm can come to the self. Fourth, this vision was a compelling experience. The Upanishads are not first-person records of religious or mystical experiences but rather the intellectual forms which molded and reflected such experiences. As a result, they do not give direct access to the experience of knowing the ātman or the brahman. Nevertheless, they do suggest the drama of that experience. The vision is only fully known when one becomes the true self or when one directly perceives the world in its singleness. Why did this vision develop at this time and in this place? One reason was that its basis was firmly established in the Brāhmanas, which anticipate the Upanishadic aspiration to see the world as a totality. Moreover, the search for an integrative vision and the ability to experience this vision are not unique to India. Other cultures have also created paradigms for consolidating the diverse elements of experience, and other societies have given rise to mystical movements. The reason for the crystallization of Upanishadic thought may lie partly in the social and political changes occurring at the end of the Vedic period. This was not only a time of intellectual change, but one which saw the creation of cities and the consolidation of tribes into states.The stress on personal more than corporate identity, which can accompany these developments, may have provoked a nostalgia for a sense of the whole and for a definition of the self which did not isolate the individual. At the same time, the Upanishads are a result of the very individuality they seem to compromise. They are the compositions of creative individuals, and they address themselves to individuals. Some Upanishads imply that the realization of the truth can occur only by a supremely individualistic act: breaking away from society and retreating to the forest for a life of study and meditation. Thus, the Upanishads may be the product of a contradiction, but if so, it is one of the creative contradictions which have driven the development of Indic culture and its complexity. SUGGESTED TRANSLATIONS Patrick Olivelle, The Early Upaniṣads (Oxford: Oxford University Press, 1998).

CHAPTER

5

Hidden in the cave The Upanis. adic self Jonardon Ganeri

Source: The Concealed Art of the Soul (Oxford: Oxford University Press, 2007), pp. 13–38. The I, the I is what is deeply mysterious. Wittgenstein, Notebooks, 1914–16

1.1  THE RELUCTANCE OF THE SAGE

The recalcitrant sages of the Upaniṣads1 are coy but not covert.They do not, in general, conceal their true beliefs with false words; they are not insincere. Tardiness not trickery is their leading trait, an immense reluctance to ‘spill the beans’, a coyness rooted not so much in a self-serving secretiveness or a disinclination to share the knowledge that gives them power,2 but which exists rather as a response to a deep respect for the power of that knowledge, and a recognition of the need not to be frivolous either with the knowledge itself or with its potential recipients. Yājñavalkya, certainly, has about him a degree of furtiveness when he replies to Āryabhāga’s demand to be told what happens to a man after he dies: ‘My friend, we cannot talk about this in public; let’s go and discuss it in private’ (BU 3.3.2); or when in response to Janaka’s persistent questioning, he panics, thinking, ‘The king is really sharp! He has flushed me out of every cover’ (BU 4.3.33); or when, having tried to palm off Uḍasta Cākrāyaṇā with a cryptic account of the self, Uṣasta exclaims: ‘That’s a fine explanation! It’s like saying “This is a cow and that is a horse!” Come on, give me a real explanation … of the self that is within all’ (BU 3.4.2). But even he displays the more typical mood when he cracks under Gārgī’s prolonged interrogation: ‘Don’t ask too many questions, Gārgī, or your head will shatter apart! You are asking too many questions about a deity about whom one should not ask too many questions. So, Gārgī, do not ask too many questions!’ (BU 3.6.1). The sage’s reluctance is born not of subterfuge but frustration and, perhaps, fear. 65

66

JONARDON GANERI

Good use is made of the trope of the reluctant sage by the Upaniṣadic storyteller: to engender in the reader a sense of respect for the profundity of the wisdom about to be imparted, as well as to convey the idea that a gift is being given and a very precious one at that. More particularly, though, the reluctant sage is made to speak as one who is having his arm twisted, and the whole tone is ‘Look, I didn’t want to tell you this, but you have forced me to.’ The reluctant sage seems with his reluctance to want to divest himself of responsibility, implying that the pupil alone is responsible for whatever consequences may follow the disclosure. Often indeed it is the virtuosity of the pupil that elicits this disclosure, an exceptional feat of endurance or asceticism, an uncanny disinterest in the pleasures of life.Virtuosity, but not necessarily virtuousness—it is not so much the moral worthiness of the pupil that turns the trick, but their apparent willingness to sacrifice all, their readiness to accept the consequences without complaint. For then the sage can truly think ‘You took this upon yourself—don’t blame me!’ Reluctance and fear are clearly intertwined in the story of Dadhyañc Ātharvaṇa. We learn from Ṛgveda 1.116.12 of a sage Dadhyañc who has discovered the secret truth called the ‘honey-doctrine’ and who is sworn to secrecy by Indra, with the threat that if he were to disclose this doctrine Indra would cut off his head. The Aśvins want the secret but also know that Dadhyañc will not risk his neck to divulge it. They hit upon the following brilliant contrivance: they themselves cut off Dadhyañc’s head and replace it with the head of a horse. Now, when Dadhyañc reveals the secret and Indra severs his horse’s head, the Aśvins can restore to him his original form—so that he will have revealed the secret and still evaded Indra’s wrath! This is the story alluded to in the following verse: As thunder discloses the rain, O Heroes, I disclose that wonderful skill you displayed for gain; When Dadhyañc Ātharvaṇa revealed the honey, Through the horse’s head to you. You fixed a horse’s head, O Aśvins, On Dadhyañc Ātharvaṇa; True to his word, O mighty ones, He revealed to you Tvaḍṭṛ’s honey, That remains a secret with you. (BU 2.5.16–17)

With the head substitution the Aśvins displayed a ‘wonderful skill’ indeed, an ingenuity that diffused Indra’s threat and coaxed the reluctant Dadhyañc into revealing to them the honey-truth, a doctrine about the self that is the honey of all beings. Not that he had any objection in principle to telling them, nor did they trick him into an indiscretion—it was fear alone that held Dadhyañc’s tongue. Let’s next take up the story of Naciketas, the retelling of which in the Kaṭha Upaniṣad develops earlier renditions in Ṛgveda 10.135 and Taittirīya Brāhmaṇa 3.11.8.1–6.3 The ­kernel of the story is the lecture Naciketas receives from Yama on the topic of the self. The framing narrative tells how Yama came, reluctantly, to be giving that lecture. The pious boy Naciketas watches as his father performs the sarvamedha or ‘sacrifice of all’, a ritual in which one gives away all one’s most important things as sacrificial gifts. N ­ aciketas notices that the cows being given are not, in fact, the best; indeed ‘they have been milked

THE UPANIṢADIC SELF

67

dry, they are totally barren—“joyless” are called the worlds to which a man goes who gives them as gifts’ (KaU 1.3). Perhaps it is with the intention of exposing his father’s hypocrisy that ­Naciketas then asks to whom he, Naciketas, will be given, and he twice repeats the question until his father yells back ‘I’ll give you to Death!’ (KaU 1.4). With equanimity Naciketas accepts his fate and goes to the abode of Yama, the god of death, where he remains without food for three days waiting for Yama to return. Yama is conscious that he has shown a Brahmin great disrespect in allowing him to stay at his home without hospitality and therefore grants Naciketas three wishes of his own choosing. In the Taittirīya rendition, but not in the Kaṭha, there is a suggestion that Naciketas tricked Yama, deliberately choosing to arrive when he knew Yama would be away, thereby manufacturing the obligation, and indeed that he did this on the advice of his repentant father. Be that as it may, the three gifts Naciketas asked for were to be allowed to return home to his now pacified father, to be told how to perform effectively the fire-altar ritual and render permanent its results, and finally to be told what happens to a man after he dies: There is this doubt about a man who is dead. ‘He exists,’ say some; others, ‘He exists not.’ I want to know this, so please teach me. This is the third of my three wishes. (KaU 1.20)

Of the three wishes, Yama executes the first without hesitation, assuring Naciketas that he will return home to a father who will be affable in the future. Yama also delivers the second wish without fuss, explaining the method of constructing the fire-altar and adding that this fire-altar will henceforth bear the name ‘Naciketas’. When it comes to the third wish, however,Yama is suddenly reluctant: As to this even the gods of old had doubts, for it’s hard to understand, it’s a subtle doctrine. Make, Naciketas, another wish. Do not press me! Release me from this. (KaU 1.21) Choose sons and grandsons who’d live a hundred years! Plenty of livestock and elephants, horses and gold! Choose as your domain a wide expanse of earth! And you yourself live as many autumns as you wish! (KaU 1.23) And if you would think this is an equal wish: You may choose wealth together with a long life; Achieve prominence, Naciketas, in this wide world; And I will make you enjoy your desires at will. (KaU 1.24) You may ask freely for all those desires, Hard to obtain in this mortal world;

68

JONARDON GANERI

Look at these lovely girls, with chariots and lutes, Girls of this sort are unobtainable by men— I’ll give them to you; you’ll have them wait on you; But about death please don’t ask me, Naciketas. (KaU 1.25)

Yama’s extreme reluctance to speak of death has a beseeching, desperate note, but of what can he be so afraid? Is it that Yama, the god of death, is being forced to speak about himself, to tell the secret of his own name? For Naciketas is certainly right that if even the gods of old had doubts about the matter then no one but Death himself is fit to explain it (KaU 1.22).Yama bribes, cajoles, implores Naciketas but in the end cannot refuse him. The culminative effect of the frame narrative is to box Yama into a corner where he has no option but to give his speech in spite of all his trepidation.Whether or not he has been deceived, his unwillingness to break the laws of good custom and indeed his own word, finally outweighs his immense reluctance to speak. Yama compliments Naciketas on his wisdom and steadfastness (KaU 2.1–11) and begins his discourse on the self. A third example of Upaniṣadic reluctance is met with in the famous story about Indra and Virocana, in Chāndogya 8.7.1–8.12.6.The gods and demons both have heard tell that Prajāpati speaks of a self ‘by discovering which one obtains all the worlds, and all one’s desires are fulfilled’ (CU 8.7.2). Like two polar explorers, the god Indra and the demon Virocana set out carrying firewood and live as celibates in Prajāpati’s presence.Thirty-two years later Prajāpati finally asks them what they want, and being told of their quest to discover the self he fobs them off with an answer he knows to be false: ‘Look at yourselves in a pan of water. And let me know if there is anything you do not perceive about yourselves.’ So they looked into a pan of water. Prajāpati asked them: ‘What do you see?’ And they replied: ‘Sir, we see here our entire body (ātman), a perfect likeness down to the very hairs of the body, down to the very nails.’ Prajāpati told them then: ‘Adorn yourself beautifully, dress well, and spruce yourself up, and then look into a pan of water.’ So they adorned themselves beautifully, dressed well, and spruced themselves up, and then looked into a pan of water. Prajāpati asked them: ‘What do you see?’ And they replied: ‘Sir, as the two of us here are beautifully adorned, well dressed, and all spruced up, in exactly the same way are these, sir, beautifully adorned, well dressed, and all spruced up.’ ‘That is the self; that is the immortal; that is the one free from fear; that is brahman,’ ­Prajāpati told them. And the two of them left with contented hearts. Seeing the two depart, Prajāpati observed: ‘There they go, without learning about the self, without discovering the self!’ (CU 8.8.1–4).

To his credit, Indra soon realizes that the mirrored self cannot be the self for which he seeks (CU 8.9.1—he realizes that if the external appearance can be made beautiful, so too can it become lame and crippled), and he returns for further instruction. Prajāpati makes him wait another thirty-two years and then tries to fob him off again, this time with ‘The one who goes happily about in a dream—that is the self; that is the immortal; that is the one free from fear; that is brahman’ (CU 8.10.1). Indra falls for it a second time but again returns unconvinced. He is again told to wait thirty-two years and is

THE UPANIṢADIC SELF

69

then once more fobbed off, with ‘When one is fast asleep, totally collected and serene, and sees no dreams—that is the self; that is the immortal, that is the one free from fear; that is brahman’ (CU 8.11.1). Indra goes away happy, again returning when he realizes his error. Prajāpati now says ‘I will explain it to you further, but only under the following condition—stay here for another five years’ (CU 8.11.3). Indra waits and his patience is finally rewarded. But what are we to make of Prajāpati’s astonishing reluctance, of his willingness to deceive not just the demon Virocana but Indra too, of his insistence on long periods of harsh living as a condition for receiving his instruction? And why did Indra believe Prajāpati’s final answer, when he might well by now have wondered whether he was not simply being fobbed off one more time? Indra is not doing anything that might better ready him for the new revelation during the long years; indeed the only times he seems to improve his understanding are when he goes away content and then begins to question; in short, when he is not at Prajāpati’s home. There is, to be sure, a graded teaching about the nature of the self in this Upaniṣadic story, and it is very successful as pedagogic narrative. Indeed, careful attention to narrative form here gives much insight into the nature of the ‘self-knowledge’ being sought. What is remarkable is the literary device used here, in which progressively more sophisticated accounts of the self are presented as the 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 the 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.4 Indra could not begin even to appreciate the virtues of the less obvious doctrine that the true self is the self one encounters in a dream had he not already understood as wrong the more obvious idea about the true self being the body’s reflected image or one’s appearance to others. And now Prajāpati’s reluctance might be seen as following from a further idea, that for one doctrine about the self to be a preparatory condition of another, it must not merely be understood that it is false, but this understanding must be the result of one’s own personal investigation and ­discovery.5 If that is so, then Prajāpati could not have told Indra the final doctrine straight away, nor could he have told him that any of the preceding doctrines was false. He could only ‘feed’ Indra a preparatory doctrine and wait to see if Indra discovers it to be false. Prajāpati’s reluctance, then, is of a different order than the fearful silence of Dadhyañc or the cautious reserve of Yama; it is a reluctance born of didactic necessity when what is being taught is the nature of the self. If we compare the Upaniṣadic representation of the sage with Plato’s representation of Socrates, we see that the sage’s reluctance is no ordinary reluctance. Rather, it is a feigned (perhaps even ironic) reluctance, designed to encourage and motivate the hearer or reader in the direction of a quest for the truth, a concealed truth about a hidden self. To borrow a term from Kierkegaard, the persona of the reluctant sage is a literary device for the ‘indirect communication’ of a truth about the soul.6 Later in this book, I will compare Prajāpati’s approach to teaching about the self with that of the Buddha (see chapter 4).

70

JONARDON GANERI

1.2  METAPHORS OF THE CAVE

And what did reluctant Yama, the lord of death, tell Naciketas? He told him of the existence of a hidden self, a self the discovery of which will free a man from grief and sorrow. He told him of a self that lies ‘hidden in the cave’ (nihito guhāyām): Finer than the finest, larger than the largest, is the self that lies here hidden in the cave of a living being. Without desires and free from sorrow, a man perceives by the creator’s grace the grandeur of the self. (KaU 2.20)

The metaphor of the cave as the hiding place of the soul is to be repeated several times in the Kaṭha (2.12, 3.1), and other Upaniṣads appeal to the same formula: Truth and knowledge, The infinite and brahman: A man who knows them as hidden in the cave, hidden in the highest heaven; Attains all his desires, together with the wise brahman. (TU 2.1.1) It is large, heavenly, of inconceivable form; yet it appears more minute than the minute. It is farther than the farthest, yet it is here at hand; It is right here within those who see, hidden within the cave. (MuU 3.1.7)

What is this cave wherein the self hides? The metaphor is absent from the two earliest Upaniṣads—Bṛhadāraṇyaka and Chāndogya—which speak instead of a ‘space within the heart’ (hṛdayaākāśaḥ), and that is not a metaphor but an important element of early Upaniṣadic physiology and psychology. Patrick Olivelle (p. 23) has summarized well the older view: The heart has a cavity at the centre and is surrounded by the pericardium. Channels or veins run from the heart to the pericardium and to other parts of the body.The cavity of the heart is the seat of the vital powers and the self and plays a central role in the explanations of the three states of awareness—waking, dreaming, and dreamless sleep—as well as of death. In sleep, the cognitive powers distributed throughout the body during the waking hours are gathered together in the cavity of the heart. The space of this cavity is homologized with cosmic space (see CU 3.12.7–9), and in the dream state the person travels about this space seeing and enjoying the same type of things that he experienced while awake. During deep and dreamless sleep, the self slips out of that cardiac space and enters the veins going from

71

THE UPANIṢADIC SELF

the heart to the pericardium; there it remains oblivious to everything (see BU 2.1; 4.3–4). At death the self, together with the vital powers, departs from the heart along a channel and exits through either the crown of the head (TU 1.6) or the eye (BU 4.4.2).

Many commentators identify later references to a ‘cave’ with this space within the heart (Olivelle’s translation often amplifies the phrase ‘hidden in the cave’ as ‘hidden in the cave of the heart’). Perhaps, however, the allusion to a cave is not to be taken quite so literally, but rather simply as a metaphor for the idea that the self is hidden.7 The noun guhā ‘cave’, deriving from a verbal root guh-, whose meaning is ‘to cover, hide, conceal, keep secret’, carries the senses of ‘a cave, cavern, hiding place; hiding, concealing’, just as the participial form guhya denotes ‘to be concealed, covered or kept secret; mysterious; a secret, ­mystery’.8 That is the sense it has when the same image surfaces in the Mahābhārata as part of a description of the elusiveness of dharma, where an identification with the space within the heart would clearly be out of place: The scriptures are many and divided; the dharmaśāstras are many and different. Nobody is called a sage until and unless he holds a different view [from anyone else]. The truth of dharma lies concealed in the dark cave. (MB 1.191.25)

In this it resonates with those other metaphors of concealment that set the tone of the Upaniṣadic text: The face of truth is covered with a golden dish. Open it, O Pūḍan, for me, a man faithful to the truth. Open it, O Pūḍan, for me to see.

(BU 5.15.1; IU 1.15)

Hidden in all the beings, this self is not visibly displayed. Yet, people of keen vision see him, with eminent and sharp minds. (KaU. 3.12)

That the soul is concealed—this is the force of the phrase ‘hidden in the cave’ as it occurs in the Kaṭha,Taittirīya and Muṇḍaka.The self within is not transparent; it stands in need of discovery. Discovery by whom? By oneself, presumably. But what can hide the self from itself? And how, if concealed, is it to be disclosed? The Upaniṣadic author buries the self in order to make possible a project of self-discovery, but must not bury the self too deeply or the viability of that very project will be undermined.9 1.3  WHAT CONCEALS THE SELF?

To a modern mind the problem of the self ’s concealment may seem obscure. Am I not condemned forever to be in my own company? The more immediate problem seems to

72

JONARDON GANERI

be not how to find the self but how to escape it, in the all-too-brief moments when one ‘loses oneself ’ in a reverie or film or good novel. One straightforward idea is that the self is just so small that it is very hard to spot. In the Bṛhadāraṇyaka, the self is compared with a grain of rice or barley (BU 5.6.1). In the Chāndogya, it is ‘smaller than a grain of rice or barley, smaller than a mustard seed, smaller even than a millet grain or a millet kernel’ (CU 3.14.3); but it is also the width of a span (CU 5.18.1). In the Kaṭha, again, the self is the size of a thumb (KaU 2.12, 6.17). Then again we are told several times that the self pervades and saturates the body that hides it: This very breath, which is the self consisting of intelligence, penetrates this bodily self up to the very hairs of the body, up to the very nails. Just as a razor within a case or a termite within a termite hill, so this self consisting of intelligence penetrates this bodily self up to the very hairs of the body, up to the very nails. (KsU 4.20) Like oil in sesame seeds and butter in curds, like water in the riverbed and fire in the firedrills, so, when one seeks it with truth and austerity, one grasps that self (ātman) in the body (ātman)—that all-pervading self, which is contained [in the body] like butter in milk. (SU 2.15–16)

That is the theme too in a very famous passage which speaks of the self as having five ‘sheaths’: Different from and lying within this man formed from the essence of food is the self consisting of lifebreath, which suffuses that man completely … Different from and lying within this self consisting of breath is the self consisting of mind, which suffuses this other self completely … Different from and lying within this self consisting of mind is the self consisting of perception, which suffuses this other self completely … Different from and lying within this self consisting of perception is the self consisting of bliss, which suffuses this other self completely. (TU 2.3.2–5)

The view that the soul has the same extension as the body has been maintained by the Jainas, one proprioceptive argument being that the soul must always be in contact with the senses and the senses are located on the body’s surface.10 Even if we are reluctant to take literally the notion that the soul has spatial boundaries, the argument points in the direction of a second way to make sense of the idea that the self is hidden by the body, namely that it does not enter the ‘field’ of perception of any of the sense faculties. For, as the Kaṭha says: The Self-existent One pierced the apertures outward, therefore, one looks out, and not into oneself. A certain wise man in search of immortality, turned his sight inward and saw the self within. (KaU 4.1)

THE UPANIṢADIC SELF

73

The self, then, is mis-placed; and it is not only because the self is on the ‘wrong side’ of the senses that it cannot be sensed but also because it lacks the sensible qualities: It has no sound or touch, no appearance, taste, or smell. (KaU 3.15ab) Not by speech, not by the mind, not by sight can he be grasped. How else can that be perceived, other than by saying ‘He is!’ (KaU. 6.12)

If the self fails to be an object for the senses, however, how is it to become an object of desire, for desire seems to track the sensory (on seeing something, one wishes either to obtain or avoid it)? This question leads us to an important passage at the beginning of Chāndogya chapter 8, a preparatory passage for the story of Indra and Virocana to come (see above). Here the discovery of the self is related to the discovery of true (satya) desires masked by false (anṛta) ones. The passage begins in a by now familiar way: Now, here in this fort of brahman there is a small lotus, a dwelling place, and within it, a small place. In that space there is something—and that’s what you should try to discover, that’s what you should seek to perceive. (CU 8.1.1)

What is the something? It is the self: That is the self free from evils—free from old age and death, free from sorrow, free from hunger and thirst; the self whose desires and intentions become true …. So, those here in this world who depart without having discovered the self and these true desires do not obtain complete freedom of movement in any of the worlds, whereas those here in this world who depart after discovering the self and these true desires obtain complete freedom of movement in all the worlds. (CU 8.1.5–6)

The true desires—and with them the true self—are, however, hidden by other, false desires: Now, these true desires are masked by the false. Although they are true, they have the false for a mask, for when someone close to him departs from this world, he doesn’t get to see him here. On the other hand, people who are close to him, whether they are alive or dead, as well as anything else that he desires but does not get—all that he finds by going there, for these true desires of his masked by the false are located there. Take, for example, a hidden treasure of gold. People who do not know the terrain, even when they pass right over it time and again, would not discover it. In exactly the same way, all these creatures, even though they go there every day, do not discover this world of brahman, for they are led astray by the false. (CU 8.3.1–2)

74

JONARDON GANERI

The self, we are now told, is just too close to be seen. Or rather, being present whenever we see, we do not see it. Are the false desires the ‘outward desires’, mentioned in Kaṭha 4.2, pursued by fools who seek the stable in the unstable? The senses, those outward apertures, are forever drawing us away from what is indeed close at hand. As for the true desire, is it a desire for that which is stable (and indeed immortal), a desire not for the thing seen but for the reason there is seeing, for the principle behind sight, a desire not for the thinking but for the reason there is thought? If so, then no wonder that without knowing the terrain, we pass over it time and time again. This idea receives support from a passage in the Bṛhadāraṇyaka, where Yājñavalka also distinguishes two sorts of desire, and with them two kinds of self.When a person acts in accordance with his ordinary desires he is made good by good action and made bad by bad action: What a man turns out to be depends on how he acts and on how he conducts himself. If his actions are good, he will turn into something good. If his actions are bad, he will turn into something bad. A man turns into something good by good action and into something bad by bad action. And so people say: ‘A person here consists simply of desire.’ A man resolves in accordance with his desire, acts in accordance with his resolve, and turns out to be in accordance with his action. (BU 4.4.5)

This passage echoes Yājñavalka’s furtive advice to Āryabhāga in BU 3.3.2, where he likewise taught that a man is made by what he does—the earliest statement of the doctrine of karma (see also Appendix B). But Yājñavalka goes on now to speak of another sort of man, a man whose only desire is the self: Now, a man who does not desire—who is without desires, who is freed from desires, whose desires are fulfilled, whose only desire is the self (ātmakāma)—his vital functions do not depart. Brahman he is, and to brahman he goes. On this point there is the following verse: When they are all banished, those desires lurking in one’s heart; Then a mortal becomes immortal, and attains brahman in this world. (BU 4.4.6)

The true desire of such a person is not a desire for anything sensible, but a desire for the self, a desire for the principle by which there is sensing. This is the self of which the Kauḍītaki says (KsU 3.8),‘it does not become more by good actions or in any way less by bad actions’. This is the true self hidden in the cave, the cave of the body and its embodied desires. Action unmotivated by desire is a transparent expression of the self. Desire conceals the self. These Upaniṣadic ideas about concealment by desire find their fullest statement in the Bhagavadgītā, where action without desire is held to constitute the soul, while actions produced in response to desire destroy it: In action alone is your proper interest, never in [its] fruits. Let not your motive be the fruit of action, nor your attachment to inaction. (BG 2.47)

THE UPANIṢADIC SELF

75

As fire is enveloped in smoke, and a mirror [obsured] by dust, as an embryo is concealed by the womb, so is this [self ] covered by that [desire]. Knowledge is concealed by this perpetual enemy of the knower, this insatiable fire in the form of desire, o son-of-Kuntī. (BG 3.38–9) 1.4  THE SELF IS NOT AN OBJECT OF CONSCIOUSNESS

The metaphor of the self as hidden in the cave, I will now argue, is best interpreted as giving voice to the doctrine that the self is not a possible object of consciousness. Our search for the self among the objects of consciousness is a forlorn one, and indeed is ill-conceived. If the self is not within the purview of the senses and mind, that is not because it has nothing to do with sensing and thinking; in fact, just the opposite—being what makes sensing and thinking possible, it is ‘too close’ to be seen. Not being an object among others in the world of comprehended objects, why shouldn’t it best be described through the use of paradoxical conceptions?11 That the self the Upaniṣadic sages were in search of is not itself an object of consciousness follows from Yājñavalka’s clever explanation of the matter to his wife, Maitreyī. The account occurs twice in the Bṛhadāraṇyaka: at 2.4.14, and again at 4.5.15, where a short but important elucidatory passage is inserted into the text. Here is 2.4.14: When there is a duality of some kind, then the one can smell the other, the one can see the other, the one can hear the other, the one can greet the other, the one can think of the other, and the one can perceive the other. When, however, the Whole has become one’s very self (ātman), then who is there for one to smell and by what means? Who is there for one to see and by what means? Who is there for one to hear and by what means? Who is there for one to greet and by what means? Who is there for one to think and by what means? Who is there for one to perceive and by what means? By what means can one perceive him by means of whom one perceives this whole world? Look—by what means can one perceive the perceiver? (BU 2.4.14)

The text of 4.5.15 inserts between the last and the penultimate sentences: About this self (ātman), one can only say ‘not—, not—.’ He is ungraspable, for he cannot be grasped. He is undecaying, for he is not subject to decay. He has nothing sticking to him, for he does not stick to anything. He is not bound; yet he neither trembles in fear nor suffers injury.

The insertion tells us what it is that the argument is meant to show—that the self is not an object of thought, nor even of reference, or as Yājñavalka even more bluntly explains to Uḍasta Cākrāyaṇa: You can’t see the seer who does the seeing; you can’t hear the hearer who does the hearing; you can’t think of the thinker who does the thinking; and you can’t perceive the perceiver who does the perceiving. The self within all is this self of yours. (BU 3.5.2)

Yājñavalkya’s clever argument explains the transition from ‘the perceiver does not perceive himself ’ to ‘the perceiver is not perceived’. It is taken for granted that perceiving is

76

JONARDON GANERI

an anti-reflexive relation, but it does not yet follow that the perceiver is not and cannot be perceived, that the relation is antisymmetric (i.e. that if one is a perceiver, then one is not a thing perceived). The possibility to be excluded is that A perceives B and B perceives A, or some wider circle of perception.Yājñavalkya argues by universalization of the relata, supposing first that the perceiver is the All and then that the perceived is the whole world. A perceiver who is All, clearly, leaves nothing to perceive and is therefore not itself perceived by something other than itself; similarly, if the All is perceived, and perceiving is anti-reflexive, then the perceiver cannot be among the world of things perceived. The Upaniṣadic self is not, then, an object of consciousness. To say that it is instead the subject of consciousness will not advance the discussion very far, for this has been granted all along; what we have now done is to reject the idea that the subject of consciousness can be thought of as an object of consciousness. Indeed, we are now better able to make sense of Prajāpati’s graded instruction to Indra. Of the three bogus accounts of the self he offered, the first two are attempts to represent the self as a possible object of consciousness. Perhaps it is an object of sensory awareness, the reflected image in a pool of water? Indra is not fooled. Perhaps, then, if not a sensory object, it is an object of dream consciousness, ‘the one who goes happily about in a dream’ (CU 8.10.1)? Indra takes his time, but sees through the ruse. So let us give up on the idea of the self as an object of consciousness, and say instead that it is the residue that is there in dreamless, contentless sleep—but this ‘self ’, Indra clearly saw, is not a self at all: ‘this self as just explained, you see, does not perceive itself fully as, “I am this”; it does not even know any of these beings here. It has become completely annihilated’ (CU 8.11.1). Finally, Prajāpati provides an account of that self by discovering which one obtains all the worlds and all one’s desires are fulfilled: Now, when this sight here gazes into space, that is the seeing person, the faculty of sight enables one to see. The one who is aware: ‘Let me smell this’—that is the self; the faculty of smell enables him to smell. The one who is aware: ‘Let me say this’—that is the self; the faculty of speech enables him to speak. The one who is aware: ‘Let me listen to this’—that is the self; the faculty of hearing enables him to hear. The one who is aware: ‘Let me think about this’—that is the self; the mind is his divine faculty of sight. This very self rejoices as it perceives with his mind, with that divine sight, those objects of desire found in the world of brahman. It is this self that the gods venerate, as a result of which they have obtained all the worlds, and have had all their desires fulfilled. Likewise, when someone discovers this self and comes to perceive it, he will obtain all the worlds and have all his desires fulfilled. (CU 8.12.4–6)12

The conception of self described here is distinct from and an improvement over two ­others: the conception of self as agent, and the conception of self as controller. The agentive self is the one who sees, the one who thinks, the one: By whom impelled, by whom compelled, does the mind soar forth? By whom enjoined does the breath, march on as the first? By whom is this speech impelled, with which people speak?

77

THE UPANIṢADIC SELF

And who is the god that joins the sight and hearing? (KaU 1.1)

The controller self is the self who rides the body-chariot: Know the self as a rider in a chariot, and the body, as simply the chariot. Know the intellect as the charioteer, and the mind, as simply the reins. (KaU 3.3) The senses, they say, are the horses, and sense objects are the paths around them; He who is linked to the body, senses, and mind, the wise proclaim as the one who enjoys. (KaU. 3.4)

The self described by Prajāpati is no mere agent, no mere controller. It is the one who stands still, observing itself as it watches, as it hears and thinks. It is not merely the one who sees, nor the one who decides to look, but the one who is aware: ‘let me see’. It is not a detached or impassive self, disengaged from its own desires and actions, but it is aware of itself even as it pursues them—it is that in virtue of which the subject of consciousness is self-conscious, something that is, necessarily, not itself an object of consciousness.13 1.5  THE HIDDEN CONNECTION BETWEEN ONE SELF AND ALL

There is a hidden truth about this hidden self, the self that looks out from the body’s nine portals but is not itself a possible object of gaze.The Upaniṣad—the ‘hidden connection’ or ‘secret teaching’14—is that the self that gazes out from within my body is the same as the self that gazes out from within yours. The principle (brahman) behind thinking is the same for each and every thinking self (ātman). Indeed it is this principle that the self truly consists in, this ‘by knowing which a man comes to know this whole world’ (MuU 1.3), the ‘highest object of the teachings on hidden connections, an object rooted in austerity and the knowledge of the self ’ (SU 1.16).This is what Yama, the lord of death, taught Naciketas: As the single wind, entering living beings, adapts its appearance to match that of each; So the single self within each being, adapts its appearance to match that of each, yet remains quite distinct. As the sun, the eye of the whole world, is not stained by visual faults external to it; So the single self within every being, is not stained by the suffering of the world, being quite distinct from it. (KaU 5.10–1)

78

JONARDON GANERI

The concealed Upaniṣadic self is an impersonal self, the same for all. Knowing this, how could the various happenings that befall one as one threads a path through life matter so much? How could they be invested with personal significance to one who knows that one is not more this person than that? With a robust sense of self, one has to be content with one’s own pleasures and pains; with only an impersonal sense of self, the pleasures (and also pains) of everyone are one’s own: ‘You who know this self here, the one common to all men, as somehow distinct—you eat food. But when someone venerates this self here, the one common to all men, as measuring the size of a span and as beyond all measure, he eats food within all the worlds, all the beings, and all the selves’ (CU 5.18.1). That the true self is the same within each of us follows from the rule of substitution (ādeśa) which the sage Āruṇi taught to his son Śvetaketu (CU 6.1.1–7). In grammar and ritual, a rule of substitution tells us when one word or object can stand in for another, it being generally understood that the substitute resembles and behaves like the original.15 Olivelle comments that ‘such rules within the Upaniṣadic tradition are said to be “secret”, thus approximating the meaning of Upaniṣad’ (p. 501). Here is what Āruṇi told his recently educated and overly proud son: ‘Śvetaketu, here you are my son, swell-headed, thinking yourself to be learned, and arrogant; so you must have surely asked about that rule of substitution (ādeśa) by which one hears what has not been heard before, thinks of what has not been thought of before, and perceives what has not been perceived before?’ ‘How indeed does that rule of substitution work, sir?’ ‘It is like this, son. By means of just one lump of clay one would perceive everything made of clay—the transformation (vikāra) is a verbal handle, a name—while the reality is just this: “It’s clay.” It is like this, son. By means of just one copper trinket one would perceive everything made of copper—the transformation is a verbal handle, a name—while the reality is just this: “It’s copper.” It is like this, son. By means of just one nail-cutter one would perceive everything made of iron—the transformation is a verbal handle, a name—while the reality is just this: “It’s iron.” That son, is how this rule of substitution works.’ (CU 6.1.3–6)

Thereupon Āruṇi teaches Śvetaketu how to look for the hidden core of things, the sap pervading the tree, the seed in a banyan fruit, the salt in saltwater, and repeatedly he says, ‘The finest essence here—that constitutes the self of this whole world; that is the truth; that is the self. And that[’s how] you are (tat tvam asi), Śvetaketu’ (CU 6.8.7).16 Knowing the ‘archetype’ or ‘model’, and the rule that tells us how to derive the new using the old as a pattern,17 puts us in a position to gain knowledge of the new. Here the term ‘minute essence’ (aṇiman) is used in the ‘vanilla essence’ sense, implying an extract fine and minute. Elsewhere, the metaphor of the salt in salt-water carries more the idea of ‘self ’ as a mass term, referring to a single but distributed spread of cognition: When a chunk of salt is thrown in water, it dissolves into that very water, and it cannot be picked up in any way.Yet, from whichever place one may take a sip, the salt is there! In the same way this Immense Being has no limit or boundary and is a compact mass (ghana) of perception (vijñāna). (BU 2.4.12)

THE UPANIṢADIC SELF

79

This passage, which immediately precedes Yājñavalkya’s clever explanation of why the perceiver cannot be perceived, hints at a further reason why the self is not a possible object of consciousness—lacking a proper boundary (limit, surface), it cannot be pinned down. All we succeed in doing, when we look for the self within, is to sample cognitions, much as we sample the salt-water or the chunk of salt. Knowing this to be the ‘essence’ of the self, and knowing the rule by which one self is substituted for another (what has to change, what remains constant), we derive knowledge of all. Knowing oneself is a matter of knowing the quintessence of thinking, and this is common to all. Given indeed that the self is not an object, what other story could one tell? For it follows that no criterion for the individuation of objects could be used to individuate distinct selves. The ‘thin’ self we discover in the phenomenology of thinking itself is not the ‘thick’ self supported by differences between individuals with respect to the contents of their thoughts (see further the end of chapter 7). Indeed, and ironically, it is nevertheless a self behind which one can hide. Italo Calvino has made the point well: ‘Just as the author, since he has no intention of telling about himself, decided to call the character “I” as if to conceal him, not having to name him or describe him more than this stark pronoun’.18 The Upaniṣadic self is a mask behind which individuals can hide their individuality, even as that individuality masks the self. 1.6  FINDING THE SELF IN THE PERIPHERY OF THINKING

If the self had simply been lost, we might have hoped to find it again. As matters stand, we are in worse shape: it is not that we have simply mislaid the self, but that the self is hidden, in principle, from view. Our search for the self as one object among others in the world is forlorn. So in what can self-discovery consist, given what we now know, that it does not consist in the discovery of a self ‘out there’? For the Upaniṣadic sage, self-discovery is the supreme quest and the source of moral and spiritual fulfilment; but they have hidden the self so well as to make their own quest seem hopeless. To the question ‘in what does the disclosure of the self consist?’ there seem now to be two possible answers. One is to identify a non-objectual mode or aspect of experience, the enjoyment of which is constitutive of self-knowledge.This is the way of the mystic, but need not necessarily lead to mysticism.The other way is the way of the quietist, who will seek to explore the moral and experiential consequences of the discovery that the self is not a possible object of consciousness, without trying to find a substitute for the knowledge that cannot be had (an epistemic humility).Yājñavalkya, again in his conversation with Maitreyī, makes what appears to be a constructive proposal of the first sort: It is like this. When a drum is being beaten, you cannot catch the external sounds; you catch them only by getting hold of the drum or the man beating that drum. Or when a conch is being blown, you cannot catch the external sounds; you catch them only by getting hold of the conch or the man blowing that conch. Or when a lute is being played, you cannot catch the external sounds; you catch them only by getting hold of the lute or the man playing that lute. (BU 2.4.7–9; 4.5.8–10)

The context has made it amply clear that, when it comes to the human subject, what cannot be caught is the self. So the proposal is that, when a sensation is being sensed,

80

JONARDON GANERI

you catch the sensing self only by getting hold of the sensing; when a cognition is being cognized, you catch the thinker only by getting hold of the thinking. Here is a way to reach the self—not by grasping it as an object, but catching it in its activity of sensing and thinking. Just as it is hopeless to catch a sound in the air, so it is impossible to catch the self as if it were a thing. If we cannot catch the sound once it has been released, we can catch it at the moment of its production—catch the producing of the sound. If we cannot catch the self as an object among others in the world, we can catch it in the very act of thinking. This tallies nicely with the Chāndogya description mentioned before (CU 8.12.4–5). My proposal, in other words, is that we read the passage as saying that the self is caught in the phenomenological quality of thinking, in the flavour of the experience of ‘what it is like’ to think. There is something that it feels like, from within, to be thinking, and in focusing upon this one is participating in a non-objectual awareness of the self. This is something that cannot directly be ‘taught’, and it is the reason for the oblique literary form assumed by the Upaniṣadic narrative. The very narrative form that the story of Indra assumes, as an allegory of concealment, functions protreptically, to turn the mind of the reader, to redirect their search in a new direction. I will say more about this in chapter 4. Consider the following passage, again from the second report of Yājñavalkya’s conversation with Maitreyī: As a mass of salt has no distinctive core and surface; the whole thing is a single mass of ­flavour—so indeed, my dear, this self has no distinctive core and surface; the whole thing is a compact mass (ghana) of cognition (prajñāna). (BU 4.5.13)

In this passage, which parallels BU 2.4.12 quoted above, the use of the metaphor of salt is given a new meaning. The search for the self is not a search for the cognitive core of the psyche, but for the quintessence that pervades it and can be extracted (pra-jñāna). The description of the self as being without a distinctive core and surface tells strongly against the correctness of attributing to the Upaniṣadic thinkers a ‘substratum’ theory of self. How exactly does it feel to experience the self ? It is like this [says Yājñavalkya, this time to Janaka]. As a man embraced by a woman he loves is oblivious to everything within or without, so this person embraced by the self consisting of knowledge is oblivious to everything within or without. Clearly, this is the aspect of his where all desires are fulfilled, where the self is the only desire, and which is free from desires and far from sorrows. (BU 4.3.21) Where a man sees, hears, or discerns no other thing—that is plenitude.Where one sees, hears, or discerns some other thing—that is scarcity …. Plenitude, indeed, is below; plenitude is above; plenitude is in the west; plenitude is in the east; plenitude is in the south; and plenitude is in the north. Indeed, plenitude extends over this whole world. Now the substitution of the word ‘I’—I am, indeed, below; I am above; I am in the west; I am in the east; I am in the south; I am in the north. Indeed, I extend over this whole world. Next, the substitution of self—The self, indeed, is below; the self is above; the self is in the west; the self is in the east; the self is in the south; and the self is in the north. Indeed, the self extends over this

THE UPANIṢADIC SELF

81

whole world. A man who sees it this way, thinks about it this way, and perceives it this way; a man who finds pleasure in the self, who dallies with the self, who mates with the self, and who attains bliss in the self—he becomes completely his own master; he obtains complete freedom of movement in all the worlds. (CU 7.5.1–2)

What I am calling an awareness of the ‘what it feels like to be thinking’ is, the first of these passages says, a non-objectual experience of pure rapture. In the second passage, it is described as an experience that does not represent the subject as having a spatial location. Contrast that with ordinary perceptual experience, whose egocentric frame-of-reference places me in the centre of a network of spatial relationships with the objects perceived. The Upaniṣadic self isn’t a self shown but not stated, as it was for Wittgenstein; rather, it is the self that hovers in experience’s phenomenal shadow. It is a sense of being present everywhere; or better, a sense of being stripped of a sense of being somewhere. Even though it is present to each of us, the phenomenal character of thinking itself is barely noticeable, hidden as it is behind the ‘false desires’ of worldly cognitive involvement. In the cultivated sensibility of the Upaniṣadic sage, however, it is strong and loud, completely satisfying: Just as a disk smeared with clay, once it is cleaned well, shines brightly, so also an embodied person, once he has perceived the true nature of the self, becomes solitary, his goal attained and free from sorrow. (SU 2.14)

The self is discovered, then, in the shadowy edges of experience. It is a further step, and one taken only in the more theistic late Upaniṣads, to identify this experience with a mystical awareness of the divine. This is the true nature of self, distinct from the bits and pieces of cognition, which are mere designations of the ‘knowledge’ (prajñāna) in which the self consists: Which of these is the self? Is it that by which one sees? Or hears? Or smells odours? Or utters speech? Or distinguishes between what is tasty and what is not? Is it the heart and the mind? Is it awareness? Perception? Discernment? Cognition? Wisdom? Insight? Steadfastness? Thought? Reflection? Drive? Memory? Intention? Purpose? Will? Love? But these are various designations of cognition (prajñānasya nāmadheyāni) …. Knowledge (prajñāna) is the eye of all that, and on knowledge it is founded. Knowledge is the eye of the world, and knowledge, the foundation. Brahman is knowing. It is with this self consisting in knowledge that he went up from this world and, having obtained all his desire in the heavenly world up there, became immortal. (AU 3.1–4)

I will conclude with a final and necessarily tentative conjecture. The discovery of self has a phenomenology, and that phenomenology is the esoteric phenomenology of disclosure itself. The anxiety that has been instilled by a constant reiteration of the message that something of great value has been lost, something that makes the difference between immortal freedom and generations in hell, the sheer relief to be had on being told that

82

JONARDON GANERI

what was lost is now found, is itself an ecstatic rapture.What is the catalyst for that ecstatic release? Perhaps not even an actual discovery. In her study in the ethics of secrets, Sissela Bok says: [a]wareness of the allure and the dangers of secrecy … is central to human experience of what is hidden and set apart. Rooted in encounters with the powerful, the sacred, and the forbidden, this experience goes far deeper than the partaking of any one secret. Efforts to guard secrets, probe them, or share them often aim for this deeper and more pervasive experience. If we do not take this into account … then we shall but skim the surface; and the secrets, once revealed, will seem paltry and out of proportion to all that went into guarding them.19

Exactly so here. No actual secret knowledge about the self, no matter how insightful, could by itself explain the enormous reluctance of the Upaniṣadic sage to speak. That reluctance has another function—to generate a fearsome sense of secrecy, the release from which constitutes an experience of euphoric bliss, and so directs the mind to where it should, all along, have been looking.20 NOTES 1. All references are to the edition and translation of Patrick Olivelle: The Early Upaniṣads: An Annotated Text and Translation (New York: Oxford University Press, 1998). Another recent translation, Valerie J. Roebuck’s The Upanisads (London: Penguin Books, 2003), is also excellent. On the chronology, Olivelle says that, ‘any dating of these documents that attempts a precision closer than a few centuries is as stable as a pack of cards’ (p.  12). The current consensus is that the prose Bṛhadāraṇyaka [BU] and Chāndogya [CU] are the oldest, and pre-Buddhist; they are also edited texts with different chronological strata, but roughly ­seventh to sixth centuries BCE. The other three prose Upaniṣads, Taittirīya [TU], Aitareya [AU] and Kauṣītaki [KsU], are probably pre-Buddhist too, and sixth to fifth centuries BCE. Of the more theistic verse Upaniṣads, the order is probably Kena [KaU], Kaṭha [KaU], īśā [IU], Śvettāśatara [SU] and Muṇḍaka [MuU], and the date somewhere in the last few centuries BCE. Recent work on the date of the Buddha’s death has recommended between 375 and 355 BCE in place of the usually cited 486 BCE. 2. The prominent exception is SU 6.22: ‘This supreme secret was proclaimed during a former age in the Vedānta. One should never disclose it to a person who is not of a tranquil disposition, or who is not one’s own son or pupil.’ 3. See H. D.Velankar, ‘The Ṛgvedic origin of the story of Naciketas’, in Mèlanges d’Indianisme a la Mèmoire de Louis Renou (Paris: Èditions e. de Boccard, 1968), pp. 763–72. 4. A preparatory condition is thus the inverse of a presupposition, something that must be true for the sentence presupposing it to be evaluable as either true of false. 5. It is said more than once that ‘This self cannot be grasped by teachings or by intelligence or even by great learning’ (KaU 2.23; MuU 3.2.3). But compare KaU 2.8, which seems to assert precisely the opposite: ‘Yet one cannot gain access to it, unless someone else teaches it. For it is smaller than the size of an atom, a thing beyond the realm of reason.’ 6. Søren Kierkegaard, Concluding Unscientific Postscript (Princeton: Princeton University Press, 1992). 7. Compare Democritus’ famous statement: ‘Of truth we know nothing, for truth is in a well’ (fr. 117; Diog. L. ix, 72). Democritus (fl. 430 BCE) is reported to have travelled widely, and

THE UPANIṢADIC SELF











83

Diogenes Laertius informs us that ‘Some say that he associated with the “naked philosophers” in India’ (Diog. L. ix, 35). 8. Apte, sv. guh-, guhya. J. L. Mehta,‘The Ṛgveda: text and interpretation’, in his Philosophy and R ­ eligion: Essays in Interpretation (Delhi: Indian Council for Philosophical Research, 1990), p. 283:‘Another myth with a metaphor at its core is that of the demon Vala, the Encloser, the cavity that shuts in. Here again we have a common noun turned into a proper name, what appears to us today the mythologization of an abstract idea, in this case, that of enclosing or encaving.’ 9. As Śaṅkara was well-aware, a paradox threatens: for inquiry into the self to be possible, the self must be neither already known nor altogether unknown. The Brahmasūtra Śāṅkara Bhāṣya (ed.), with commentaries N. A. K. Shastri and V. L. Shastri Pansikar (Bombay: Nirnaya Sagar Press, 1917), pp. 78–83 (under Brahmasūtra 1.1.1). On this version of the paradox of inquiry and its later elaborations, see Amber Carpenter and Jonardon Ganeri, ‘Meno goes to India: The new paradox of inquiry’ (forthcoming). 10. In the Vaiśeṣikasūtra we find an argument that the soul is conscious only of that with which the body is in contact (VS 3.2.1, 5.2.15–16). VS 7.1.28–29, however, state that the soul is ‘large’ (mahat), which the commentators take to imply omnipresence (parama-mahattva). For discussion: Keiichi Miyamoto, The Metaphysics and Epistemology of the Early Vaiśeṣikas (Pune: Bhandarkar Oriental Research Institute, 1996), chapter 2. 11. For example, CU 3.14.3. On the Upaniṣadic use of paradox: Joel Brereton, ‘The Upaniṣads’, in W. T. de Bary and I. Bloom (eds), Approaches to the Asian Classics (New York: Columbia University Press, 1990), pp. 115–35. 12. I hear a similar explanation in Proclus, in his commentary on Euclid’s Elements: ‘And indeed prior to both of these [cognitions and desires], there is something unitary in the soul, which often says “I am perceiving”, “I am reasoning”, “I have appetite”, and “I will”. It is conscious of all these activities and cooperates with them.’ Trans. Richard Sorabji, in The Philosophy of the Commentators 200–600 AD (London: Duckworth, 2004), vol. 1, p. 151. Again, Roderick Chisholm: ‘There are many references in literature to a discovery … which seems to be of the first importance but which can be put only in some such sentence as “I am me”. What is discovered in such cases? … It is the discovery one makes when one is first aware of the unity of consciousness; it is thus a discovery about those things one has been directly attributing to oneself. One suddenly becomes aware of the fact that they are all being attributed to the same thing …. And how does one come to see this? It would be correct to say: “One has only to consider it to see that it is true.” But it is, apparently, something that many people never happen to consider.’ Roderick M. Chisholm, The First Person: An Essay on Reference and Intentionality (Brighton: The Harvester Press, 1981), p. 90. 13. The idea can be traced through Schopenhauer to Wittgenstein, and from him to more recent writers, such as Sidney Shoemaker. See Appendix A. 14. For analyses of the etymology, see the bibliography cited by Olivelle, The Early Upaniṣads, p. 24, n. 29. 15. Eivind Kahrs, Indian Semantic Analysis:The ‘nirvacana’Tradition (Cambridge: Cambridge University Press, 1998), chapter 5. See also my ‘The ritual roots of moral reason,’ in Kevin Schilbrack (ed.), Thinking Through Rituals: Philosophical Perspectives (London: Routledge, 2004), pp. 207–33. 16. Olivelle’s translation of the famous equation tat tvam asi is controversial; it is more normally rendered ‘You are that’, the neuter pronoun tat construed anaphorically to back-refer to brahman. Olivelle argues, following Brereton, that the rules of Vedic syntax do not permit a neuter pronoun to stand in opposition to a masculine noun (tvam, ‘you’). He therefore renders tat adverbially. Joel Brereton, ‘Tat tvam asi in context’, Zeitschrift der Deutschen ­Morgenländischen Gesellschaft¨136 (1986): 98–109; Olivelle, The Early Upaniṣads, pp. 560–1. At least one critic has objected, however, that in interpreting texts, principles of philosophical

84



JONARDON GANERI

charity sometimes override syntax; see Stephen H. Phillips, ‘Engagement with Sankrit philosophical texts’, in Rita Sherma and Arvind Sharma (eds), Hinduism and Hermeneutics:Towards a Pluralistic Discourse in the Study of Religion (New York: Oxford University Press, 2007). Valerie Roebuck is similarly sceptical: ‘[T]here are numerous places in the Upaniṣads where the authors have departed from the strict rules of grammatical gender to make a teaching point.’Valerie J. Roebuck, The Upaniṣads (London: Penguin Books, 2003), p. 423, n. 12. She translates CU 6.8.7 as, ‘This subtle part is what all this has as self. It is truth: it is the self. You are that, Śvetaketu.’ There is room to argue that in fact no departure from grammar is involved, that nominal sentences do not always demand agreement in gender. 17. Compare the use in ritual theory of the notion of tantra; cf. my ‘The ritual roots of moral reason’ (ibid.). 18. Italo Calvino, If on a Winter’s Night a Traveller (London:Vintage, 1998), p. 15. 19. Sissela Bok, Secrets: On the Ethics of Concealment and Revelation (New York: Pantheon Books, 1982), p. 5. 20. A modern example of this literary use of the trope of secrecy and concealment is found in Henry James’s story, ‘The figure in the carpet’. James describes the secret as the ‘bait on a hook’ and as the ‘cheese in a mousetrap’, enthralling yet frustrating the reader, who is thereby forced to reflect on the nature of their quest. Henry James, The Figure in the Carpet and Other Stories (London: Penguin Books, 1986). REFERENCES

Bok, Sissela. Secrets: On the Ethics of Concealment and Revelation (New York: Pantheon Books, 1982). Brereton, Joel. ‘Tat tvam asi in context’, Zeitschrift der Deutschen Morgenländischen Gesellschaft 136 (1986): 98–109. Brereton, Joel. ‘The Upaniṣads’, in W. T. de Bary and I. Bloom (eds), Approaches to the Asian Classics (New York: Columbia University Press, 1990), pp. 115–135. Calvino, Italo. If on a Winter’s Night a Traveller (London:Vintage, 1998). Carpenter, Amber, and Ganeri, Jonardon. ‘Can you seek the answer to this question?’, Australasian ­Journal of Philosophy 88 (2010): 571–94. Chisholm, Roderick M. The First Person: An Essay on Reference and Intentionality (Brighton: The ­Harvester Press, 1981). Ganeri, Jonardon. ‘The ritual roots of moral reason’, in Kevin Schilbrack (ed.), Thinking Through Rituals: Philosophical Perspectives (London: Routledge, 2004), pp. 207–33. James, Henry. The Figure in the Carpet and Other Stories (London: Penguin Books, 1986). Kahrs, Eivind. Indian Semantic Analysis: The ‘nirvacana’ Tradition (Cambridge: Cambridge University Press, 1998). Kierkegaard, Soren. Concluding Unscientific Postscript, ed. and trans. Howard V. Hong and Edna H. Hong (Princeton: Princeton University Press, 1992). Mehta, J.L. ‘The Rgveda: text and interpretation’, in his Philosophy and Religion: Essays in Interpretation (Delhi: Indian Council for Philosophical Research, 1990). pp. 272–292. Miyamoto, Keiichi. The Metaphysics and Epistemology of the Early Vaisesikas (Pune: Bhandarkar Oriental Research Institute, 1996). Phillips, Stephen H.‘Engagement with Sankrit philosophical texts’, in Rita Sherma and Arvind Sharma (eds). Hinduism and Hermeneutics: Towards a Pluralistic Discourse in the Study of Religion (New York: ­Oxford University Press, 2007). Sorabji, Richard. The Philosophy of the Commentators 200–600 AD (London: Duckworth, 2004), 4 vols. Velankar, H. D. ‘The Rgvedic origin of the story of Naciketas’, in Mélanges d’ Indianisme a la Memoire de Louis Renou (Paris: Editions e. de Boccard, 1968), pp. 763–72. Wittgenstein, Ludwig. Notebooks 1914–16, ed. G. H. von Wright and G. E. M. Anscombe (Oxford: ­Blackwell, 1961).

P A R T

III BUDDHIST PHILOSOPHY OF MIND

CHAPTER

6

Indian theories of mind Georges Dreyfus and Evan Thompson

Source: Philip David Zelazo, Morris Moscovitch and Evan Thompson (eds), The Cambridge Handbook of Consciousness (Cambridge: Cambridge University Press, 2007), 89–116. INTRODUCTION

In discussing Asian views of mind and consciousness, we must start from the realization that this topic presents insurmountable challenges. The diversity of Asian cultures from China to India to Iran is so great that it is impossible to find coherent ways to discuss the mental concepts of these cultures over and above listing these conceptions and noting their differences. Hence, rather than chart a territory that hopelessly extends our capacities, we have chosen to examine Indian views of the mind, with a special focus on the Indian Buddhist tradition, which can be traced back to the first centuries after the life of Siddhartha Gautama, the Buddha (566–483 bce), and which continued to develop in India through the 7th and 8th centuries ce. This approach allows us to present a more grounded and coherent view of the mind as conceived in the Indian philosophical tradition and to indicate some areas of interest that this tradition offers to cognitive scientists and philosophers of mind. In talking about the mind, it is important to define the term, for it is far from unambiguous. In most Indian traditions, the mind is neither a brain structure nor a mechanism for treating information. Rather, mind is conceived as a complex cognitive process consisting of a succession of related mental states. These states are at least in principle phenomenologically available; that is, they can be observed by attending to the way in which we experience feeling, perceiving, thinking, remembering, and so on. Indian thinkers describe these mental states as cognizing (jñā) or being aware (buddh) of their object.

87

88

GEORGES DREYFUS AND EVAN THOMPSON

Thus, the mind is broadly conceived by traditional Indian thinkers as constituted by a series of mental states that cognize their objects. This general agreement breaks down quickly, however, when we turn to a more detailed analysis of the nature and structure of the mind, a topic on which various schools entertain vastly different views. Some of these disagreements relate to the ontological status of mental states and the way they relate to other phenomena, particularly physical ones. Such disagreements are related to well-known ideas in the Western tradition, particularly the mind-body dualism that has concerned Western philosophy since Descartes. But many of the views entertained by Indian thinkers are not easily mapped in Western terms, as we see in this chapter. Most Indian thinkers do not consider the ontological status of mental states to be a particularly difficult question, for most of them accept that there is an extra-physical reality. Among all the schools, only the Materialist, the Cārvāka, reduces the mental to physical events. For its proponents, mental states do not have any autonomous ontological status and can be completely reduced to physical processes.They are just properties of the body, much like the inebriating property of beer is a property of beer. Most other thinkers reject this view forcefully and argue that the mind can neither be eliminated nor reduced to the material. Their endorsement of an extra-physical reality does not, however, necessarily amount to a classical mind-body dualism (of the sort found in Descartes’ Meditations or Plato’s Phaedo). Moreover, although they agree in rejecting the materialist view, they strongly disagree in their presentations of the mind. In this chapter, we focus mostly on the Buddhist tradition, exploring some of its views of the mind. One of the most salient features of this tradition is that its accounts of the mind and consciousness do not posit the existence of a self. According to this tradition, there is no self, and mental activity cannot be understood properly as long as one believes in a self. The Hindu tradition, by contrast, maintains that mental life does involve a permanent self. Thus, to contextualize Buddhist views of the mind, we begin with a brief presentation of some of the most important Hindu views.We then present the Buddhist Abhidharma and its analysis of the mind in terms of awareness and mental factors. Traditionally, the Abhidharma makes up one of the ‘three baskets’ into which Buddhists divide their scriptures – Sutra or sayings of the Buddha, Vinaya or monastic discipline, and Abhidharma, which systematizes Buddhist teachings in the form of detailed analyses of experience. In examining the Abhidharma, we examine the ways in which this tradition analyzes the different functions of the mind without positing the existence of a self. These analyses are in certain ways reminiscent of those in cognitive science that aim to account for cognitive processing without invoking a homunculus or ‘little man’ inside the head who oversees the workings of the mind (or merely passively witnesses the results; see Varela, Thompson, & Rosch, 1991, for further discussion of this parallel). The Abhidharma, however, is phenomenological; its concern is to discern how the mind works as evidenced by experience (but especially by mentally disciplined and refined contemplative experience). Although thus it is also epistemological, the Abhidharma does not present any developed epistemological analysis of the structure of mental states and the way they relate to their objects so as to produce knowledge. To cover this topic we turn to Dharmakīrti (c. 600 ce), one of the main Buddhist epistemologists, who offers a comprehensive view of the types of cognition and their relation to their objects.

INDIAN THEORIES OF MIND

89

The phenomenological analyses contained in the Abhidharma and the epistemological analyses of Dharmakīrti offer significant resources for cognitive scientists and philosophers of mind in their efforts to gain a better understanding of consciousness. These analyses also constitute the theoretical framework for the ways in which the Buddhist tradition conceives of meditation and mental training, both with regard to the phenomenology of contemplative mental states and the epistemology of the types of knowledge that these states are said to provide. Given the increasing scientific interest in the physiological correlates and effects of meditation and their relation to consciousness (see Chapter 19), it is important for the scientific community to appreciate the phenomenological and philosophical precision with which these states are conceptualized in the Buddhist tradition. ¯M SELF AND MENTAL STATES: A SA . KHYA VIEW

One of the most important views of the mind in the Hindu tradition is found in the Sāṃkhya school. Traditionally this school is said to have been founded by the philosopher Kapila, a legendary figure who may have lived as early as the 7th century bce, but the earliest Sāṃkhya text we possess dates from the 3rd century ce. The Sāṃkhya tradition is one of the six classical schools of Hindu philosophy (Nyaya, Vaisesika, Sāṃkhya, Yoga, Purva Mimamsa, and Vedānta). Its influence extends to the other schools, particularly the Vedānta school, which later became especially important in the development of Hindu thought.The Sāṃkhya was in fact less a school proper than a way of thinking based on the categorization of reality. It was crucial in the formation of Indian philosophical thinking before and after the start of the Common Era, and hence it is unsurprising that its view of the mind has been largely adopted in the Hindu tradition and beyond.1 The Sāṃkhya approach rests on a dualistic metaphysics built on the opposition between material primordial nature (pradhāna) or materiality (prakṛti) and a spiritual self (ātman) or person (puruṣa).2 Nature is the universal material substratum out of which all phenomena other than the self emerge and evolve. These phenomena, which make up the world of diversity, are physical transformations of the three qualities (guṇa) that compose primordial nature. These three qualities are sattva (transparency, buoyancy), rajas (energy, activity), and tamas (inertia, obstruction). They are principles or forces, rather than building blocks. All material phenomena, including the intellect and organs of perception, are understood to be made up of a combination of these three principles.The one principle not included in this constant process of transformation is the self, which is permanent, non-material, and conscious or aware. The self is also described as the conscious presence that witnesses the transformations of nature, but does not participate in them. As such it is passive, though it witnesses the experiences deriving from the transformations of the world of diversity.3 Although the Sāṃkhya analysis of mind is dualistic, it does not fit within classical mind-body dualism. For the Sāṃkhya, the mind involves a non-material spiritual element, namely the self.The self, however, is not the same as the mind. Rather, the self is the mere presence to or pure witnessing of the mental activities involved in the ordinary awareness of objects. This pure witnessing, untainted by the diversity of the material world, is not sufficient for mental activities, for mental activities are representational or semantic and require more than passive mirroring. Mental activity is the apprehension of an object, and this activity requires active engagement with objects and the formation of ideas and

90

GEORGES DREYFUS AND EVAN THOMPSON

concepts necessary for purposeful action in the world. The self cannot account for such activity, however, because it is changeless and hence passive. To account for our cognitive activities, we therefore need other elements that participate in the world of diversity. Because any element that participates in the world of change must emerge out of primordial materiality and hence be material, it follows that the analysis of mental states cannot be limited to their spiritual dimension (the self), but must also involve material elements. Hence, for the Sāṃkhya, mental activity requires the cooperation of the two fundamental types of substance that make up the universe, passive consciousness and material nature. Having described the Sāṃkhya metaphysics, we can now sketch its influential analysis of mental activity.4 This analysis starts with buddhi, which is usually translated as ‘the intellect’ and is the ability to distinguish and experience objects. This ability provides the prereflective and presubjective ground out of which determined mental states and their objects arise; it is also the locus of all the fundamental predispositions that lead to these experiences.The intellect emerges out of primordial matter and therefore is active, unlike the non-material and passive self. The self is described metaphorically as a light, for it passively illuminates objects, making it possible for the intellect to distinguish them. The intellect operates in a representational way by taking on the form of what is known. This representational ability works in two directions – toward the conscious and uninvolved self and toward the objects. The intellect, thanks to its quality of clarity and transluscence (sattva), takes on the form of the self by reflecting it. As a result, it seems as if the self experiences the diversity of objects, when it is actually the intellect that undergoes these experiences, the self being the mere witness of them. This ability of the intellect to usurp the function of consciousness helps the intellect in its apprehension of objects, for by itself the intellect is active but unconscious. Awareness of objects arises only when the intellect takes on the light of the self and reflects it on objects, much like pictures are created when light is projected onto a film. In this way, the intellect becomes able to take on the form of the object and thus to discern it. The intellect’s reflecting the self and taking on the form of an object are not, however, sufficient to fully determine experience. To become fully cognitive, experience requires the formation of subjective and objective poles. Experience needs to be the experience of a particular individual apprehending a particular object. The formation of the subjective pole is the function of the ‘ego-sense’ (ahaṃkāra), the sense of individual subjectivity or selfhood tied to embodiment. This sense colors most of our experiences, which involve a sense of being a subject opposed to an object. The determination of the objective pole, on the other hand, is the function of ‘mentation’ (manas), which oversees the senses and whose special function is discrimination. This function allows mentation to serve as an intermediary between the intellect and the senses. Mentation organizes sensory impressions and objects and integrates them into a temporal framework created by memories and expectations. In this way, our experience of objects in the world is created. Although the dualistic metaphysics associated with this view was rejected in the history of Indian philosophy, the Sāṃkhya model of the mind was taken over by other Hindu schools. It serves as a foundation of the philosopher Patañjali’s (c. 2nd century bce) Yoga view of mind, which is similar to the Sāṃkhya.5 The Yoga view also rests on the opposition between passive self and active mental activities (citta), a rubric under which intellect, ego-sense, and mentation are grouped. Similarly, Śaṃkara (788–820 ce), who savaged the

INDIAN THEORIES OF MIND

91

dualism of the Sāṃkhya, took over its model of the mind in his Advaita Vedānta, emphasizing the contrast between the transcendence of the self and the mental activities of the ‘inner sense’ (antaḥkarava) belonging to the person.6 Hence, the Sāṃkhya view can be taken as representative of the Hindu view of the mind, especially in its emphasis on the difference between a passive witnessing consciousness and mental activity. According to this view, as we have seen, mental events come about through the conjunction of two heterogeneous factors – a transcendent self and a diversity of mental activities. It is a basic presupposition of the Hindu tradition that mental life involves a permanent self. Yet because mental life also undeniably involves change, it cannot be reduced to this single, motionless factor of the self; hence the need for the complicated analysis briefly summarized here. This tension in accounts of the mind and consciousness between identity and change, unity and diversity, is of course also prevalent throughout Western philosophy and persists in cognitive science. We turn now to the Buddhist tradition, which presents a different perspective on this issue. THE ABHIDHARMA TRADITION AND ITS VIEW OF THE MIND

The Buddhist tradition is based on the opposite view of no-self (anātman). For the Buddhists, there is no self, and hence mental activity is not in the service of such an entity, but rather functions on its own. In short, for the Buddhists there is no self that is aware of the experiences one undergoes or the thoughts one has. Rather the thoughts themselves are the thinker, and the experiences the experiencer. How, then, do Buddhists explain the complexities of the mind? How do they explain mental regularities if there is no central controller to oversee the whole process? For an answer, we turn to the Abhidharma, one of the oldest Buddhist traditions, which can be traced back to the first centuries after the Buddha (566–483 bce). First elaborated as lists,7 the Abhidharma contains the earlier texts in which Buddhist concepts were developed and hence is the source of most philosophical developments in Indian Buddhism. But the Abhidharma is not limited to this role as a source of Buddhist philosophical development. It remained a vital focus of Buddhist thought and kept evolving, at least until the 7th or 8th century ce. In this chapter, we focus on two Indian thinkers from the 4th or 5th century ce, Asahga and Vasubandhu, and ignore the diversity of opinions and debates that has animated this tradition. The object of the Abhidharma is to analyze both the realm of sentient experience and the world given in such experience into its components in language that avoids the postulation of a unified subject. This analysis concerns the whole range of phenomena, from material phenomena to nirvāṇa (the state of enlightenment, understood as the direct realization of the nature of reality, including especially the lack of any essential self and the consequent liberation from suffering). For example, there are elaborate discussions of the four primary and four secondary elements that make up matter (see de la Vallée Poussin, 1971, l: 22). There are also lengthy treatments of the nature, scope, and types of soteriological practices prescribed by the Buddhist tradition, a central focus of the Abhidharma. But a large part of the Abhidharmic discourse focuses on the analysis of mental phenomena and their various components. It is this part of the Abhidharma that we examine in this chapter.

92

GEORGES DREYFUS AND EVAN THOMPSON

In considering experience, the Abhidharma proceeds in a rather characteristic way that may be disconcerting for newcomers, but reflects its historical origin as mnemonic lists of elements abstracted from the Buddha’s discourses. For each type of phenomenon considered, the Abhidharma analyzes it into its basic elements (dharma), lists these elements, and groups them into the appropriate categories (examples are given below).The study of the Abhidharma thus often revolves around the consideration of series of extended lists. In elaborating such lists of components of experience and the world given in experience, the Abhidharma follows the central tenets of Buddhist philosophy, in particular the twin ideas of non-substantiality and dependent origination. According to this philosophy, the phenomena given in experience are not unitary and stable substances, but complex and fleeting formations of basic elements that arise in dependence on complex causal nexuses. Such non-substantiality is particularly true of the person, who is not a substantial self, but a changing construct dependent on complex configurations of mental and material components. This analysis, which is diametrically opposed to the Sāṃkhya view, is not just limited to the person, but is applied to other objects. All composite things are thus analyzed as being constituted of more basic elements. Moreover, and this point is crucial, these basic elements should not be thought of as reified or stable entities, but as dynamically related momentary events instantaneously coming into and going out of existence. Thus, when the Abhidharma analyzes matter as being made up of basic components, it thinks of those components not as stable particles or little grains of matter, but rather as fleeting material events, coming into and going out of existence depending on causes and conditions. Similarly, the mind is analyzed into its basic components; namely, the basic types of events that make up the complex phenomenon we call ‘mind’. This Abhidharmic analysis is not just philosophical but it also has practical import. Its aim is to support the soteriological practices that the Buddhist tradition recommends. The lists of material and mental events are used by practitioners to inform and enhance their practices. For example, the list of mental factors we examine shortly is a precious aid to various types of meditation, providing a clear idea of which factors need to be developed and which are to be eliminated. In this way, the Abhidharma functions not just as the source of Buddhist philosophy but also informs and supports the practices central to this tradition. In the Abhidharma the mind is conceived as a complex cognitive process consisting of a succession of related momentary mental states. These states are phenomenologically available, at least in principle: They can be observed by turning inwardly and attending to the way we feel, perceive, think, remember, and so on. When we do so, we notice a variety of states of awareness, and we also notice that these states change rapidly. It is these mental states arising in quick succession that the Abhidharma identifies as being the basic elements of the mind. It should be clear from this preliminary characterization that in elaborating a theory of the mind the Abhidharma relies primarily on what we would call a first-person approach. It is by looking directly at experience that we gain an understanding of mind, not by studying it as an object and attending to its external manifestations. This approach of the Ahbhidharma is not unlike that of such Western thinkers as James, Brentano, and Husserl, who all agree that the study of the mind must be based on attention to experience (see Chapter 4). This approach is well captured by James’s famous claim that in the study of

INDIAN THEORIES OF MIND

93

the mind, “Introspective Observation is what we have to rely on first and foremost and always” (James, 1981, p. 185). As James himself recognizes, however, first-person observation of the mind, although it might seem a straightforward enterprise, is not a simple affair and raises numerous questions. What does it mean to observe the mind? Who observes? What is being observed? Is the observation direct or mediated? In addition to these difficult epistemological issues (some of which we take up in the next section), there are also questions about the reliability of observation.We are all able to certain degrees to observe our own minds, but it is clear that our capacities to do so differ. Whose observations are to be considered reliable? This question is significant for the Abhidharmists, who may include in their data not only ordinary observations but also the observations of trained meditators. This inclusion of observation based on contemplative mental training and meditative experience marks an important difference between the Abhidharma and James, as well as other Western phenomenologists. Nevertheless, the degree to which meditative experience is relevant to Buddhist theories of the mind is not a straightforward matter, as we see shortly. The comparison between the Abhidharma and James goes further, however, than their reliance on an introspective method. They also share some substantive similarities, the most important of which is perhaps the idea of the stream of consciousness. For the Abhidharma, mental states do not arise in isolation from each other. Rather, each state arises in dependence on preceding moments and gives rise to further moments, thus forming a mental stream or continuum (santāna, rgyud), much like James’s ‘stream of thought’. This metaphor is also found in the Buddhist tradition in which the Buddha is portrayed as saying, “The river never stops: there is no moment, no minute, no hour when the river stops: in the same way, the flux of thought” (de la Vallée Poussin, 1991, p. 69, translation from the French by Dreyfus). Unsurprisingly, there are also significant differences between James and the Abhidharma. One difference of interest to contemporary research is the issue of whether mental states arise in continuity or not (see Varela, Thompson, & Rosch, 1991, pp. 72–79). James’s view is well known: “Consciousness does not appear to itself chopped up in bits” (James, 1981, p. 233). Although the content of consciousness changes, we experience these changes as smooth and continuous, without any apparent break. The Abhidharma disagrees, arguing that although the mind is rapidly changing, its transformation is discontinuous. It is only to the untrained observer that the mind appears to flow continuously. According to the Abhidharma, a deeper observation reveals that the stream of consciousness is made up of moments of awareness, moments that can be introspectively individuated and described. Several Abhidharma texts even offer measurements of this moment, measurements one would expect to be based on empirical observation. Yet such claims are problematic, for different Abhidharma traditions make claims that at times are strikingly at odds with one another. For example, the Mahavibhāṣā, an important text from the first centuries of the Common Era, states that there are 120 basic moments in an instant.The text further illustrates the duration of an instant by equating it to the time needed by an average spinner to grab a thread. Not at all, argues another text: This measurement is too coarse. A moment is the 64th part of the time necessary to click one’s fingers or blink an eye (see de la Vallée Poussin, 1991, pp. 70–71). Although these measurements differ, one could argue that given the imprecision of premodern measurement, there is a rough agreement between these

94

GEORGES DREYFUS AND EVAN THOMPSON

accounts, which present a moment of awareness as lasting for about 1/100th of a second. This is already significantly faster than pyschophysical and electrophysiological estimates of the duration of a moment of awareness as being on the order of 250 milliseconds or a quarter of a second (see Pöppel, 1988; Varela, Thompson, & Rosch, 1991, pp. 72–79). But consider the claim made by a Theravada Abhidharma text that “in the time it takes for lightning to flash or the eyes to blink, billions of mind-moments can elapse” (Bodhi, 1993, p. 156). The time scale in this account, which is standard in the Theravada tradition, is faster by many orders of magnitude. This dramatic discrepancy alerts us to some of the difficulties of accounts based on observation. For whom are we to believe? On which tradition should we rely? Moreover, we cannot but wonder about the sources of these differences. Do they derive from the observations of meditators, or are they the results of theoretical elaborations? It is hard to come to a definitive conclusion, but it seems reasonable to believe that these accounts are not simply empirical observations, but largely theoretical discussions, perhaps supplemented by observation reports. Hence one must be cautious and not assume that these texts reflect empirical findings. Although some may, they are mostly theoretical elaborations, which cannot be taken at face value, but require critical interpretation. Finally, another Abhidharma text seems to muddy the waters further by claiming that the measure of a moment is beyond the understanding of ordinary beings. Only enlightened beings can measure the duration of a moment (de la Vallée Poussin, 1991, p. 73). Thus it is not surprising that we are left wondering! According to the Abhidharma, the mental episodes that compose a stream of consciousness take as their objects either real or fictional entities. This object-directed character of mind has been called ‘intentionality’ by Western philosophers, such as ­Brentano and Husserl. Brentano claimed that intentionality is an essential feature of consciousness and proposed it as a criterion of the mental. All acts of awareness are directed toward or refer to an object, regardless of whether this object is existent or not. We cannot think, wish, or dread unless our mind is directed toward something thought about, wished for, or dreaded, which thus appears to the mind. Therefore, to be aware is for something to appear to the mind.The Abhidharma seems to share this view, holding that every moment of cognition relates to particular objects, and hence it assumes that intentionality and consciousness are inseparable.8 The Abhidharma also holds that this stream of consciousness is not material. It is associated with the body during this lifetime, but will come to exist in dependence on other bodies after the death of this body. It is crucial to recognize, however, that the immaterial stream of consciousness is not a soul in the Platonic or Cartesian sense, but an impersonal series of mental events. Buddhist philosophers do not believe in an ontology of substances – that reality comprises the existence of independent entities that are the subjects of attributes or properties. Rather, they argue that reality is made up of events consisting of a succession of moments. Thus, mind and matter are not substances, but evanescent events, and mental and material events interact in a constantly ongoing and fluctuating process. Moreover, Buddhist philosophers partake of the general Indian reluctance to separate the mental and the material. Hence they do not hold that the divide between the material and mental spheres is absolute. Nevertheless, for the Buddhists, in contrast to the Sāṃkhya, there is a sharp divide between the mental, which is intentional and conscious, and other

INDIAN THEORIES OF MIND

95

elements. In this respect, Buddhists are perhaps the closest among Indian philosophers to a classical mind-body dualism. The Abhidharma, however, does not stop at a view of the mind as a succession of mental states, but goes much further in its analysis, breaking down each mental state into its components. According to the Abhidharma schema, which is to our knowledge unique, each mental state is analyzed as having two aspects: (i) the primary factor of awareness (citta), whose function is to be aware of the object, and (ii) mental factors (caitesika), whose function is to qualify this awareness by determining its qualitative nature as pleasant or unpleasant, focused or unfocused, calm or agitated, positive or negative, and so on. The philosopher Vasubandhu (c. 4th or 5th century ce), one of the great Abhidharmists, explains this distinction between awareness and mental factors as follows:9 Cognition or awareness apprehends the thing itself, and just that; mental factors or dharmas associated with cognition such as sensation, etc., apprehend special characteristics, special conditions. (de la Vallée Poussin, 1971, l: 30)

The basic insight is that mental states have two types of cognitive functions – (1) awareness and (2) cognitive and affective engagement and characterization. The mental state is aware of an object. For example, the sense of smell is aware of a sweet object. But mental states are not just states of awareness. They are not passive mirrors in which objects are reflected. Rather, they actively engage their objects, apprehending them as pleasant or unpleasant, approaching them with particular intentions, and so forth. For example, a gustatory cognition of a sweet object is not just aware of the sweet taste but also apprehends the object as pleasant, distinguishes certain qualities such as its texture, and so on. It also categorizes the object as being (say) one’s favorite Swiss chocolate. Such characterization of the object is the function of the mental factors. We now describe this distinction between the primary factor of awareness and mental factors in more detail.

The primary factor of awareness

The primary factor of awareness (citta) is also described as vijñāna, a term often translated as consciousness or cognitive awareness. It is the aspect of the mental state that is aware of the object. It is the very activity of cognizing the object, not an instrument in the service of an agent or self (which, as we have seen, the Buddhist philosophers argue is nonexistent). This awareness merely discerns the object, as in the above example where one apprehends the taste of what turns out to be one’s favorite Swiss chocolate. Thus Vasubandhu speaks of awareness as the “bare apprehension of each object” (de la Vallée Poussin, 1971, I: 30). In most Abhidharma systems, there are six types of awareness: five born from the five physical senses (sight, hearing, smell, taste, and touch) and mental cognition. Each type of sensory cognition is produced in dependence on a sensory basis (one of the five physical senses) and an object. This awareness arises momentarily and ceases immediately, to be

96

GEORGES DREYFUS AND EVAN THOMPSON

replaced by another moment of awareness, and so on.The sixth type of awareness is mental. It is considered a sense by the Abhidharma, like the five physical senses, though there are disagreements about its basis (see Guenther, 1976, pp. 20–30). Some Abhidharma texts, such as Asaṇga’s (Rahula, 1980), argue that these six types of consciousness do not exhaust all the possible forms of awareness. To this list Asaṇga adds two types of awareness: the store-consciousness (ālaya-vijñāna, kun gzhi rnam shes) and afflictive mentation (kliṣ†a-manas, nyon yid; Rahula, 1980, p. 17).10 The idea of a store-consciousness is based on a distinction between the six types of awareness, which are all described as manifest cognitive awareness (pravṛtti-vijñāna, ’jug shes), and a more continuous and less manifest form of awareness, which is the store-consciousness. This awareness is invoked to answer the following objection: If there is no self and the mind is just a succession of mental states, then how can there be any continuity in our mental life? Asaṇga’s answer is that there is a more continuous form of consciousness, which is still momentary, but exists at all times. Because it is subliminal, we usually do not notice it. It is only in special circumstances, such as fainting, that its presence can be noticed or at least inferred. This consciousness contains all the basic habits, tendencies, and propensities (including those that persist from one life to the next) accumulated by the individual. It thus provides a greater degree of continuity than manifest cognitive awareness on its own. The store-consciousness is mistaken by the afflictive mentation as being a self. In this way one’s core inborn sense of self is formed. From a Buddhist point of view, however, this sense of self is fundamentally mistaken. It is a mental imposition of unity where there is in fact only the arising of a multiplicity of interrelated physical and mental events. The sense of control belonging to one’s sense of self is thus largely illusory. There is really nobody in charge of the physical and mental processes, which arise according to their own causes and conditions, not our whims.The mind is not ruled by a central unit, but by competing factors whose strength varies according to circumstances. Thus Asaṇga, allegedly Vasubandhu’s half-brother, posits as many as eight types of consciousness, a doctrine usually associated with a particular Buddhist school, the Yogācāra. This school contains many interesting insights, without which there is no complete understanding of the depth of Buddhist views of the mind, but there is not space to discuss these insights here. Let us simply point out that there are some interesting similarities between the Yogācāra and the Sāṃkhya views. The store-consciousness, in acting as the holder of all the potentialities accumulated by an individual, is not unlike the ­intellect (buddhi), whereas the afflictive mentation seems similar to the ego-sense (ahamkāra). ­Furthermore, mental cognition does not seem too different from mentation (manas). These similarities indicate the reach of the Sāṃkhya model, even in a tradition whose basic outlook is radically different. Mental factors

Mental states are not just states of awareness; they also actively engage their objects, qualifying them as pleasant or unpleasant, approaching them with a particular attitude, and so on. Mental factors, which are aspects of the mental state that characterize the object of awareness, account for this engagement. In other words, whereas consciousness makes

INDIAN THEORIES OF MIND

97

known the mere presence of the object, mental factors make known the particulars of the content of awareness, defining the characteristics and special conditions of its object. They qualify the apprehension of the object as being pleasant or unpleasant, attentive or distracted, peaceful or agitated, and so forth. The translation of these elements of the mind (caitesika) as factors is meant to capture the range of meanings that the Abhidharma associates with this term. The relation between cognitive awareness and mental factors is complex. At times the Abhidharma construes this relation diachronically as being causal and functional. Factors cause the mind to apprehend objects in particular ways. At other times, the Abhidharma seems to emphasize a synchronic perspective in which cognitive awareness and mental factors coexist and cooperate in the same cognitive task.11 In accordance with its procedure, the Abhidharma studies mental factors by listing them, establishing the ways in which they arise and cease, and grouping them in the appropriate categories. Each Abhidharma tradition has a slightly different list. Here we follow a list of 51 mental factors distributed in 6 groups.12 The mental typology presented in this list has a number of interesting features in relation to more familiar Western philosophical and scientific typologies: • Five omnipresent factors: feeling, discernment, intention, attention, and contact • Five determining factors: aspiration, appreciation, mindfulness, concentration, and intelligence • Four variable factors: sleep, regret, investigation, and analysis • Eleven virtuous factors: confidence/faith, self-regarding shame, other-regarding shame, joyful effort, pliability, conscientiousness, detachment, non-hatred (lovingkindness), wisdom, equanimity, and non-harmfulness (compassion). • Six root-afflictions: attachment, anger, ignorance, pride, negative doubt, and mistaken view. • Twenty branch-afflictions: belligerence, vengefulness, concealment, spite, jealousy, avarice, pretense, dissimulation, self-satisfaction, cruelty, self-regarding shamelessness, other-regarding shamelessness, inconsideration, mental dullness, distraction, excitement, lack of confidence/faith, laziness, lack of conscientiousness, and forgetfulness. The nature of this complex typology becomes clearer when one realizes that these six groups can be further reduced to three. The first three groups contain all the neutral factors. They are the factors that can be present in any mental state, whether positive or negative. Hence these factors are neither positive nor negative in and of themselves. The next three groups are different. These factors are ethically determined. The eleven virtuous factors are positive in that they do not compel us toward attitudes that lead to suffering. They leave us undisturbed, open to encounter reality with a more relaxed and freer outlook. The twenty-six afflictive factors, on the other hand, disturb the mind, creating frustration and restlessness. They are the main obstacles to the life of the good as understood by the Buddhist tradition.The very presence of these factors marks the mental state as virtuous or afflictive. Thus it is clear that the Abhidharma typology is explicitly ethical. This presentation also offers interesting insights concerning the cognitive functions of the mind. In particular, the analysis of the five omnipresent factors – feeling, discernment,

98

GEORGES DREYFUS AND EVAN THOMPSON

intention, attention, and contact – shows some of the complexities of Abhidharmic thinking.These five are described as omnipresent because they are present in every mental state. Even in a subliminal state such as the store-consciousness these five factors are present. The other factors are not necessary for the performance of the most minimal cognitive task (the apprehension of an object, however dimly and indistinctly). Hence they are not present in all mental states, but only in some. One striking feature of this list is the pre-eminent place of feeling (vedanā, tshor ba) as the first of the factors. This emphasis reflects the fundamental outlook of the tradition, which views humans as being first and foremost sentient. But it also reflects a distinctive view of the cognitive realm that emphasizes the role of spontaneous value attribution. For the Abhidharma, a mental state is not only aware of an object but at the same time it also evaluates this object. This evaluation is the function of the feeling tone that accompanies the awareness and experiences of the object as either pleasant, unpleasant, or neutral. This factor is central in determining our reactions to the events we encounter, because, for the most part, we do not perceive an object and then feel good or bad about it out of considerate judgments. Rather, evaluation is already built into our experiences. We may use reflection to come to more objective judgments, but those mostly operate as correctives to our spontaneous evaluations. Feeling is not the only important factor, however. A mental state involves not only awareness and feeling but also discernment (saṃjñā, ’du shes also often translated as perception or recognition). This factor involves the mind’s ability to identify the object by distinguishing it from other objects.This concept of discernment presents some difficulties, however. In its most elaborate form, discernment is based on our semiotic ability to make distinctions, mostly through linguistic signs. But for the Abhidharma, the mind’s ability to identify objects is not limited to linguistic distinctions, however important they may be. Infants and non-human animals are understood to have the ability to make distinctions, although they do not use symbolic thinking. Are these prelinguistic cognitions nevertheless semiotic? Do they involve non-linguistic signs, or do they make distinctions without the use of signs? It seems plausible to argue that some of these states involve non-linguistic signs, as in the case of visual cognitions that distinguish objects on the basis of visual clues. For the Abhidharma, however, this question strikes deeper, because several meditative states in the Buddhist tradition are described as signless (animitta, mthan med).13 Can the mind in these states identify its object without making distinctions? Or is it the case that even in the case of signless states the mind still makes distinctions, although they are not linguistic or even conceptual? In a short chapter such as this one, we cannot delve into this issue, despite its relevance to the dialogue between Buddhism and the sciences of mind. Other factors are also significant. Intention (cetanā, sems pa) is a central and omnipresent factor, which determines the moral (not ethical) character of the mental state. Every mental state approaches its object with an intention, a motivation that may be evident to the person or not. This intention determines the moral nature of the mental state, whether it is virtuous, non-virtuous, or neutral. This factor is associated with the accomplishment of a goal and hence is also thought of as a focus of organization for the other factors. Also important is the role of attention (manasikāra, yid la byed pa), another one of the five omnipresent factors. It is the ability of the mind to be directed to an object. A contemporary commentator explains attention this way: “Attention is the mental factor

INDIAN THEORIES OF MIND

99

responsible for the mind’s advertence to the object, by virtue of which the object is made present to consciousness. Its characteristic is the conducting of the associated mental states [i.e., factors] to the object. Its function is to yoke the associated mental states [i.e., factors] to the object” (Bodhi, 1993, p. 81). Every mental state has at least a minimal amount of focus on its object; hence attention is an omnipresent factor. Attention needs to be distinguished from two other related factors. The first is concentration (samādhi, ting nge ’dzin), the ability of the mind to dwell on its object singlepointedly. The second is mindfulness (smṛti, dran pa, also translated as recollection), which is the mind’s ability to keep the object in focus without forgetting, being distracted, wobbling, or floating away from the object. Both abilities are not present in every mental state. Concentration differs from attention in that it involves the ability of the mind not just to attend to an object but also to sustain this attention over a period of time. Similarly, mindfulness is more than the simple attending to the object. It involves the capacity of the mind to hold the object in its focus, preventing it from slipping away in forgetfulness. Hence both factors, which are vital to the practice of Buddhist meditation (see ­Chapter 19), are included among the determining factors.They are present only when the object is apprehended with some degree of clarity and sustained focus. The factors discussed so far are mainly cognitive, but the Abhidharma list also includes mental factors we would describe as emotions. Consider the ethically determined factors, starting with the eleven virtuous ones: confidence/faith, self-regarding shame, otherregarding shame, joyful effort, pliability, conscientiousness, detachment, non-hatred (lovingkindness), wisdom, equanimity, and non-harmfulness (compassion). We would describe several of these factors, such as lovingkindness and compassion, as emotions. These two factors belong to what we would characterize as the affective domain, although here they are understood not with regard to their affectivity, but rather in relation to their ethical character.14 Hence they are grouped with other factors, such as wisdom and conscientiousness, that are more cognitive than affective. For the ­Abhidharma all these factors are grouped together. They are all positive in that they promote well-being and freedom from the inner compulsions that lead to suffering. The afflictive factors, on the other hand, are precisely those that lead to suffering.They are by far the most numerous group and are clearly a major focus of this typology: • Six root-afflictions: attachment, anger, ignorance, pride, negative doubt, and mistaken view. • Twenty branch-afflictions: belligerence, vengefulness, concealment, spite, jealousy, avarice, pretense, dissimulation, self-satisfaction, cruelty, self-regarding shamelessness, other-regarding shamelessness, inconsideration, mental dullness, distraction, excitement, lack of confidence/faith, laziness, lack of conscientiousness, and forgetfulness. Here again we notice that this list contains factors that look quite different. Some factors such as ignorance are clearly cognitive, whereas others such as anger and jealousy are more affective. They are grouped together because they are afflictive: They trouble the mind, making it restless and agitated.They also compel and bind the mind, preventing one from developing more positive attitudes. This afflictive character may be obvious in the case of attachment and jealousy, which directly lead to dissatisfaction, frustration, and restlessness.

100

GEORGES DREYFUS AND EVAN THOMPSON

Ignorance – that is, our innate and mistaken sense of self – is less obviously afflictive, but its role is nonetheless central here, because it brings about the other more obviously afflictive factors. Although there are many elements in the typology of mental factors that we can identify as emotions (anger, pride, jealously, lovingkindness, and compassion), there is no category that maps onto our notion of emotion. Most of the positive factors are not what we would call emotions, and although most of the negative factors are affective, not all are. Hence it is clear that the Abhidharma does not recognize the notion of emotion as a distinct category of a mental typology. There is no Abhidharma category that can be used to translate our concept of emotion, and similarly our concept of emotion is difficult to use to translate the Abhidharma terminology. Rather than opposing rational and irrational elements of the psyche, or cognitive and emotive systems of the mind (or brain), the Abhidharma emphasizes the distinction between virtuous and afflictive mental factors.Thus, our familiar Western distinction between cognition and emotion simply does not map onto the Abhidharma typology. Although the cognition/emotion distinction has recently been called into question by some scientists (see Chapter 29 and ­Damasio, 1995), it remains central to most of contemporary cognitive science and philosophy of mind. The Abhidharma typology offers a different approach, one in which mental factors are categorized according to their ethical character. This typology could prove fruitful for psychologists and social and affective neuroscientists interested in studying the biobehavioral components of human well-being (see Goleman, 2003). The analyses of mental factors we have reviewed indicate the complexity, sophistication, and uniqueness of the Abhidharma mental typology. For this reason, the Abhidharma is often called, somewhat misleadingly, ‘Buddhist psychology’.15 Yet the Abhidharma analysis does not answer all the questions raised by the Buddhist view of the mind as lacking a real self. In particular, it leaves out the issue of the cognitive or epistemic structure of the mental states that make up the stream of consciousness.To examine this issue, we turn to another Indian Buddhist tradition, the logico-epistemological tradition of Dignāga and Dharmakīrti (see Dreyfus, 1997; Dunne, 2004). BUDDHIST EPISTEMOLOGY

This tradition was started by Dignāga around 500 ce and was expanded significantly more than a century later by Dharmakīrti, the focus of our analysis. Its contribution was the explicit formulation of a complete Buddhist logical and epistemological system. The importance of this system in India can be seen in the continuous references to it by later Buddhist thinkers and the numerous attacks it received from orthodox Hindu thinkers. It gradually came to dominate the Indian Buddhist tradition, even eclipsing the Abhidharma as the prime focus of intellectual creativity. The concern of this tradition is the nature of knowledge. In the Indian context, this issue is formulated as this question: What is the nature of valid cognition (pramāṇa) and what are its types? Hindu thinkers tend to present a realist theory, which liberally allows a diversity of instruments of valid cognition. For example, the Sāṃkhya asserts that there are three types of valid sources of knowledge: perception (pratyakṣa), inference (anumāna), and verbal testimony (śabda). The Nyāya, perhaps the most important Hindu

INDIAN THEORIES OF MIND

101

logico-epistemological tradition, added a fourth type of valid cognition, analogy (upamāna). This fourfold typology provided the most authoritative epistemological typology in India. Buddhist epistemology, however, rejects these typologies and offers a more restrictive view, limiting knowledge to inference and perception. It is in its examination of inference as a source of knowledge that the Buddhist tradition analyzes reasoning, in particular the conditions necessary for the formation of sound reasons and all their possible types. Hence this tradition is often described, also somewhat misleadingly, as ‘Buddhist logic’.16 The interpretation of the word pramāṇa is itself a topic of debate among Buddhist and Hindu thinkers. For the latter, this word, in accordance with its grammatical form, refers to ‘means of valid cognition’. This understanding also accords with the basic view of this school that knowledge is owned by a subject, the self, to whom knowledge is ultimately conveyed. For example, the Nyāya asserts that knowledge is a quality of the self. It is only when I become conscious of something that I can be said to know it. This view is energetically rejected by Dharmakīrti, who follows the classical Buddhist line that there is no knowing self, only knowledge. Hence, pramāṇa should not be taken in an instrumental sense, but as referring to the knowledge-event, the word itself being then interpreted as meaning valid cognition. This type of cognition is in turn defined as that cognition that is non-deceptive (avisamvādi-jñāna): Valid cognition is that cognition [that is] non-deceptive (avisamvādi). Non-deceptiveness [consists] in the readiness [for the object] to perform a function. (Dharmakīrti, Commentary on Valid Cognition II: 1, translated by Dreyfus, in Miyasaka, 1971–2)

This statement emphasizes that pramāṇa is not the instrument that a knowing self uses to know things. There is no separate knowing subject, but just knowledge, which is pramāṇa. According to this account, a cognition is valid if, and only if, it is non-deceptive. Dharmakīrti in turn interprets non-deceptiveness as consisting of an object’s readiness to perform a function that relates to the way it is cognized. For example, the non-deceptiveness of a fire is its disposition to burn, and the non-deceptiveness of its perception is its apprehension as burning. This perception is non-deceptive because it practically corresponds to the object’s own causal dispositions, contrary to the apprehension of the fire as cold. The scope of the discussion of pramāṇa, however, is not limited to the analysis of knowledge, but constitutes a veritable philosophical method used in investigating other philosophical and even metaphysical topics. All pronouncements about the world and our ways of knowing it must rest on some attested forms of knowledge, such as perception and inference, if they are to be taken seriously. No one can simply claim truth, but must be able to establish statements by pinning down their epistemic supports. The advantage of this method is that it provides intertraditional standards of validation and the development of a relatively neutral framework within which philosophical and metaphysical claims can be assessed, without regard to religious or ideological backgrounds. This procedure is different from the Abhidharmic approach, which presupposes Buddhist ideas and vocabulary. In analyzing the mind, Dharmakīrti starts from the same view of mind as the ­Abhidharma. Mind is made up of momentary mental states that arise in quick succession. Each moment of consciousness comes to be and disappears instantaneously, making a place for other

102

GEORGES DREYFUS AND EVAN THOMPSON

moments of awareness. Moreover, each moment apprehends the object that appears to it and in the process reveals the object that is apprehended. In this way, each mental state cognizes its object. But as an epistemologist, Dharmakīrti investigates issues left out by the ­Abhidharma, tackling questions that are central to any philosophical exploration of the mind. In this chapter, we examine some of these questions. First, we consider Dharmakīrti’s analysis of the nature of cognitive events.We examine his view of the mind as apprehending representations of external objects, rather than the objects themselves, and the consequences that this view has for the issue of whether the mind is inherently reflexive (self-revealing and self-aware).We also examine Dharmakīrti’s theory of perception, as well as some of his views on the nature of conceptuality and its relation to language. Finally, we revisit the issue of intentionality, showing the complexity of this notion and attempting to disentangle its several possible meanings within the context of a Buddhist account of the mental. The reflexive nature of mental events

We commonly assume that we have unproblematic access to our environment through our senses. Even casual first-person investigation shows, however, that such access may well not be the case. There are cases of perceptual illusions, and even when we are not deceived, the perceptions of individuals vary greatly. Hence philosophy cannot take for granted the common-sense view of perceptual knowledge. Many Western philosophers have argued that our perceptual knowledge goes well beyond the sensible experiences that give rise to it. Although this claim is debatable, we cannot assume without examination that we understand the way in which cognition apprehends its objects. In thinking about the nature of cognition, Dharmakīrti relies crucially on the concept of aspect (ākāra), a notion that goes back to the Sāṃkhya, but has been accepted by several other schools. The idea behind this position, which is called in Indian philosophy sākāravāda (‘assertion of aspect’), is that cognition does not apprehend its object nakedly, but rather through an aspect, which is the reflection or imprint left by the object on the mind. For example, a visual sense consciousness does not directly perceive a blue color, but captures the likeness of blue as imprinted on cognition.Thus, to be aware of an object does not mean apprehending this object directly, but having a mental state that has the form of this object and being cognizant of this form. The aspect is the cognitive form or epistemic factor that allows us to distinguish mental episodes and differentiate among our experiences. Without aspects, we could not distinguish, for instance, a perception of blue from a perception of yellow, for we do not perceive yellow directly. The role of the aspect is thus crucial in Dharmakīrti’s system, for it explains a key feature of consciousness: Consciousness is not the bare seeing that direct realism and common sense suppose, but rather the apprehension of an aspect that represents this object in the field of consciousness.The aspect is not external to consciousness. It is not only the form under which an external object presents itself to consciousness but also the form that consciousness assumes when it perceives its object.Thus an aspect is a representation of objects in consciousness, as well as the consciousness that sees this representation. The implication of this analysis is that perception is inherently reflexive. Awareness takes on the form of an object and reveals that form by assuming it. Thus, in the process of

INDIAN THEORIES OF MIND

103

revealing external things, cognition reveals itself. This view of cognition as ‘self-luminous’ (svayam prakāśa) and self-presencing is not unique to Dignāga, its first Buddhist propounder, or to Dharmakīrti, his follower. It is also accepted by other thinkers, particularly the Hindu Vedāntins, who identify consciousness as the self and describe it as being ‘only known to itself ’ (svayavedya) and ‘self-effulgent’ (svayamprabha; see Gupta 1998, 2003; Mayeda, 1979/1992, pp. 22, 44). For Dignāga and Dharmakīrti, however, the inherently reflexive character of consciousness is not a consequence of its transcendent and pure nature, but of its consisting of the beholding of an internal representation. From one side, consciousness has an externally oriented feature, called the objective aspect (grāhyākāra). This feature is the form that a mental state assumes under the influence of an external object. The second side is the internal knowledge of our own mental states. It is called the subjective aspect (grāhakākāra), the feature that ensures that we are aware of the objective aspect, the representation of the object.These two parts do not exist separately. Rather, each mental state consists of both and hence is necessarily reflexive (aware of itself in being aware of its object). The necessary reflexivity of consciousness is understood by Dharmakīrti and his followers as a particular type of perception called self-cognition (svasaṃvedana). Self-cognition can be compared to what Western philosophers call apperception; namely, the knowledge that we have of our own mental states. It is important to keep in mind, however, that apperception does not imply a second and separate cognition directed toward a given mental state of which one is thereby aware. For Dharmakīrti, apperception is not introspective or reflective, for it does not take inner mental states as its objects. It is instead the selfcognizing factor inherent in every mental episode, which provides us with a non-thematic awareness of our mental states. For Dharmakīrti, reflexivity is a necessary consequence of his analysis of perception, according to which a subjective aspect beholds an objective aspect that represents the external object within the field of consciousness. Self-cognition is nothing over and above this beholding. Self-cognition is the intuitive presence that we feel we have toward our own mental episodes. We may not be fully aware of all the aspects and implications of our experiences, but we do seem to keep track of them. Tibetan scholars express this idea by saying that there is no person whose mental states are completely hidden to him- or herself.This limited self-presence is not due to a metaphysical self, but to self-cognition. Because apperception does not rely on reasoning, it is taken to be a form of perception. Apperception does not constitute, however, a separate reflective or introspective cognition. Otherwise, the charge that the notion of apperception opens an infinite regress would be hard to avoid. Dharmakīrti’s ideas are not unlike those Western philosophers who have argued that consciousness implies self-consciousness (see Chapters 3 and 4). Such philosophers include (despite their otherwise vast differences) Aristotle, Descartes, Locke, Kant, Husserl, and Sartre (see Wider, 1997, pp. 7–39). According to Locke, a person is conscious of his or her own mental states. He defines consciousness as “the perception of what passes in a man’s mind” (Essay Concerning Human Understanding ll: ii, 19). Leibniz, in his New Essays Concerning Human Understanding (ll: i, 19), criticizes Locke, pointing out that this view leads to an infinite regress, for if every cognitive act implies self-awareness, self-knowledge must also be accompanied by another awareness, and so on ad infinitum. This regress arises, however, only if knowledge of one’s mental states is assumed to be distinct from knowledge of external objects. This assumption is precisely what Dharmakīrti denies.

104

GEORGES DREYFUS AND EVAN THOMPSON

A consciousness is aware of itself in a non-dual way that does not involve the presence of a separate awareness of consciousness. The cognizing person simply knows that he or she cognizes without the intervention of a separate perception of the cognition. This knowledge is the function of apperception, which thus provides an element of certainty with respect to our mental states. Apperception does not necessarily validate these states, however. For example, one can take oneself to be seeing water without knowing whether that seeing is veridical. In this case, one knows that one has an experience, but one does not know that one knows.The determination of the validity of a cognition is not internal or intrinsic to that cognition, but is to be established by practical investigation. Several arguments are presented by Dharmakīrti to establish the reflexive nature of consciousness.17 One of his main arguments concerns the nature of suffering and happiness as it reveals the deeper nature of mental states. For Dharmakīrti, as for the Abhidharma, suffering and happiness are not external to consciousness, but integral to our awareness of external objects. Our perceptions arise with a certain feeling-tone, be it pleasant, unpleasant, or neutral; this feeling-tone is a function of the presence of the mental factor of feeling as described by the Abhidharma.This feeling needs to be noticed, however; otherwise we would not be aware of how the apprehension of the object feels. Because this noticing cannot be the function of another mental state without incurring the problem of an infinite regress, it must be the mental state apprehending the external object that becomes aware at the same time of the feeling.This conclusion indicates, for Dharmakīrti, the dual nature of mental states. In a single mental state, two aspects can be distinguished: (1) the objective aspect, the representation of the external object in consciousness, and (2) the subjective aspect, the apprehension of this appearance or self-cognition. For Dharmakīrti, a mental state thus has two functions. It apprehends an external object (ālambana) and beholds itself. The apprehension of an external object is not direct, but results from the causal influence of the object, which induces cognition to experience (anubhava) the object’s representation. Hence, mind does not experience an external object, but beholds an internal representation that stands for an external object. Cognition cannot be reduced to a process of direct observation, but involves a holding of an inner representation. This beholding is not, however, an apprehension in the usual sense of the word, for the two aspects of a single mental episode are not separate. It is an ‘intimate’ contact, a direct experiencing of the mental state by itself through which we become aware of our mental states at the same time as we perceive things. Theory of perception

This view of cognition as bearing only indirectly on external objects has obvious consequences for the theory of perception. The theory of perception is an important element of Dharmakīrti’s epistemology, for we have access to external reality first and foremost through perception, the primary valid cognition. But this access is not as unproblematic as one might think. Although it might seem commonsensical that perception results from our encounter with the world, in reality consciousness does not directly cognize the object, but only indirectly cognizes it. For Dharmakīrti, as we have seen, the mind has direct access only to the representational aspect caused by the object; the object itself remains

INDIAN THEORIES OF MIND

105

inaccessible to consciousness. The similarity between object and aspect – and hence between object and consciousnesss, the aspect being the cognitive form of the object that stands for the object in the field of consciousness – is the crucial element in this causal theory of perception. This similarity ensures that perception is not locked up in its own appearances, as conceptions are. Consciousness is not in direct contact with the external world, but only with an internal impression caused by the external object. Hence the external object remains hidden, though not completely. When pressed by these problems, Dharmakīrti sometimes shifts between the views of two different Buddhist philosophical schools, using one perspective to bypass problems that arise in the other. These two views are the Sautrāntika theory of perception, which is representationalist in the ways just described, and the Yogācāra theory, which is idealist and denies that there is anything outside of consciousness. Following Dignāga’s example and his strategy of ascending scales of philosophical analysis, Dharmakīrti holds that the Yogācāra theory is truer and hence higher on the scale of analysis. This theory denies that there are any external objects over and above the direct objects of perception. Thus its view of perception is phenomenalist: It reduces external objects to interpreted mental data, but such data are no longer taken to stand for external objects (because it is now held that nothing exists outside of consciousness). This theory, however, is counter-intuitive, and so Dharmakīrti refers to it only occasionally, preferring to argue on the basis of the commonsensical assumption that external objects exist. His theory of perception thus has a peculiar two-tiered structure, in which he presupposes the existence of external objects, which he then ultimately rejects to propound a form of idealism. Among these two tiers, the one Dharmakīrti most often refers to is the Sautrāntika representationalist theory of perception. According to this view, consciousness does not have direct access to external objects, but grasps objects via the intermediary of an aspect caused by and similar to an external object. He sometimes replaces this view by a Yogācāra view, which holds that internal impressions are not produced by external objects, but by internal tendencies. This shift into full-blown idealism allows Dharmakīrti to bypass the difficulties involved in explaining the relation between internal perceptions and external objects. Because there are no external objects, the problem of the relation between internal impressions and external objects does not arise. At this level, his philosophy of perception can be described as phenomenalist, for it holds that there is no external object outside of aspects. Another major feature of Dharmakīrti’s account is his sharp separation between perception and conception, a separation enshrined in his definition of perception as the cognition that is unmistaken (abhrānta) and free from conceptions (kalpanāpoḍha) (Commentary on Valid Cognition, III: 300 cd). Because perception is unmistaken and conception is mistaken, perception must be free from conception. This analysis of perception differs sharply from the dominant account in India, the epistemological realism of the Nyāya school and its assertion of the existence of a determinate (savikalpaka) form of perception. For the Nyāya, perception does not stop with the simple taking in of sensory stimuli, but also involves the ability to categorize this input. Although we may start with a first moment of indeterminate perception, in which we merely take in external reality, we do not stop there but go on to formulate perceptual judgments. Moreover, and this is the crux of the question, these judgments are for the Nyāya fully perceptual. They are not mistaken conceptual overlays, but true reflections of reality.

106

GEORGES DREYFUS AND EVAN THOMPSON

This commonsensical view of perception is not acceptable to Dharmakīrti, for it leads to an unenviable choice: either accept the reality of the abstract entities necessary for the articulation of the content of perception or reject the possibility of an unmistaken cognition. Because neither possibility is acceptable for Dharmakīrti, he holds that perception can only be non-conceptual. There is no determinate perception, for the judgments induced by perception are not perceptual, but are just conceptual superimpositions. They do not reflect the individual reality of phenomena, but instead address their general characteristics. Because those are only constructs, the cognitions that conceive them cannot be true reflections of reality. Hence for perception to be undistorted in a universe of particulars, it must be totally free from conceptual elaborations.This position implies a radical separation between perception, which merely holds the object as it is in the perceptual ken, and interpretation of this object, which introduces conceptual constructs into the cognitive process. This requirement that perception be non-conceptual is the cornerstone of the Buddhist theory of perception. But it creates problems for Dharmakīrti. It would seem that given his privileging of perception he should hold an empiricist view, according to which perception boils down to a bare encounter with reality and knowledge is given to the senses. Dharmakīrti should hold the view that the aspects through which we come to perceive reality are fully representational like Locke’s ideas, that they stand for external objects, and that their apprehension is in and of itself cognitive. Dharmakīrti’s view of perception, however, is more complex, for he shares with Sellars (1956) the recognition that knowledge, even at the perceptual level, does not boil down to an encounter with reality, but requires active categorization.We do not know things by sensing them, for perception does not deliver articulated objects, but only impressions, which by themselves are not forms of knowledge but become so only when they are integrated within our categorical schemes. For example, when we are hit on the head, we first have an impression. We just have a sensation of pain, which is not by itself cognitive.This sensation becomes cognitive when it becomes integrated into a conceptual scheme, in which it is explained as being an impact on a certain part of our body due to certain causes. It is only then that the impression of being hit becomes fully intentional. Prior to this cognitive integration, the impression, or to speak Dharmakīrti’s language, the aspect, does not yet represent anything in the full sense of the word. It only becomes so when interpreted conceptually. This view of perception agrees with Dharmakīrti’s analysis of the validity of cognitions, which consists in their being ‘non-deceptive’, a term interpreted in practical terms. Cognitions are valid if, and only if, they have the ability to lead us toward successful practical actions. In the case of perception, however, practical validity is not as straightforward as one might think. Achieving practical purposes depends on correctly describing the objects we encounter. It is not enough to see an object that is blue; we must also see it as being blue. To be non-deceptive, a cognition depends on the appropriate identification of the object as being this or that. Perceptions, however, do not identify their objects, for they are not conceptual.They cannot categorize their objects, but only hold them without any determination. Categorization requires conceptual thought under the form of a judgment. Such a judgment subsumes its object under an appropriate universal, thereby making it part of the practical world where we deal with long-lasting entities that we conceive of as parts of a determined order of things. For example, we sense a blue object that we

INDIAN THEORIES OF MIND

107

categorize as blue.The perceptual aspect (the blue aspect) is not yet a representation in the full sense of the word, because its apprehension, the perception of blue, is not yet cognitive. It is only when it is interpreted by a conception that the aspect becomes a full-fledged intentional object standing for an external object. Hence, Dharmakīrti’s account of perception leads us to realize the importance of categorical interpretation in the formation of perceptual knowledge, a position that is not without problems for his system, given his emphasis on the primacy and non-conceptuality of perception. Nevertheless, the merit of this analysis is that it disentangles the processes through which we come to know the world, explaining the role of perception as a way to contact the world while emphasizing the role of conceptual categorization in the formation of practical knowledge.

Thought and language

In examining thought (kalpanā), Dharmakīrti postulates a close association with language. In fact, the two can be considered equivalent from an epistemological point of view. Language signifies through conceptual mediation in the same way that thought conceives of things. The relation between the two also goes the other way: We do not first understand things independently of linguistic signs and then communicate this understanding to others. Dharmakīrti recognizes a cognitive import to language; through language we identify the particular things we encounter, and in this way we integrate the object into the meaningful world we have constructed. The cognitive import of language is particularly obvious in the acquisition of more complex concepts. In these cases, it is clear that there is nothing in experience that could possibly give rise to these concepts without language. Without linguistic signs thought cannot keep track of things to any degree of complexity. Dharmakīrti also notes that we usually remember things by recollecting the words associated with those things. Thus concepts and words mutually depend on each other. This close connection between thought and language, inherited from Dignāga, differentiates Dharmakīrti from classical empiricists, such as Locke and modern sense-data theorists, who believe in what Sellars (1956) describes as the ‘myth of the given’. Locke, for example, holds that concepts and words are linked through association. The word ‘tree’ acquires its meaning by becoming connected with the idea tree, which is the mental image of a tree. Hence for Locke the representation of the tree is not formed through language, but is given to sensation (Dharmakīrti’s perception). We understand a tree as a tree through mere acquaintance with its representation without recourse to concepts. Dharmakīrti’s philosophy is quite different, for it emphasizes the constitutive and constructive nature of language. This conception of language is well captured by one of Dharmakīrti’s definitions of thought: Conceptual cognition is that consciousness in which representation (literally, appearance) is fit to be associated which words. (Ascertainment of Valid Cognition 40: 6–7, in Vetter, 1966)

Thought identifies its object by associating the representation of the object with a word. When we conceive of an object we do not apprehend it directly, but through the

108

GEORGES DREYFUS AND EVAN THOMPSON

mediation of its aspect. Mediation through an aspect also occurs with perception, but here the process of mediation is different. In the case of perception there is a direct causal connection between the object and its representation, but no such link exists for thought. There is no direct causal link between the object and thought, but rather an extended process of mediation in which linguistic signs figure prominently. For Dharmakīrti, the starting point of this process is our encounter with a variety of objects that we experience as being similar or different. We construct concepts in association with linguistic signs to capture this sense of experienced similarity and difference. This linguistic association creates a more precise concept in which the representations are made to stand for a commonality that the objects are assumed to possess. For example, we see a variety of trees and apprehend a similarity between these objects. At this level, our mental representations have yet to yield a concept of tree. The concept of tree is formed when we connect our representations with a socially formed and communicated sign and assume that they stand for a treeness that we take individual trees to share. In this way experiences give rise to mental representations, which are transformed into concepts by association with a linguistic sign.The formation of a concept consists of the assumption that mental representations stand for an agreed-upon imagined commonality. Thus concepts come to be through the conjunction of the experience of real objects and the social process of language acquisition. Concept formation is connected to reality, albeit in a mediated and highly indirect way. But concept formation is also mistaken, according to this view. A concept is based on the association of a mental representation with a term that enables the representation to stand for a property assumed to be shared by various individuals. In Dharmakīrti’s nominalist world of individuals, however, things do not share a common property; rather, the property is projected onto them. The property is manufactured when a representation is made to stand for an assumed commonality, which a variety of individuals are mistakenly taken to instantiate. Hence this property is not real; it is merely a pseudo-entity superimposed (adhyāropa) on individual realities. This property is also not reducible to a general term. In other words, the commonality that we project onto things does not reside in using the same term to designate discrete individuals. Upon analyzing the notion of sameness of terms, we realize that identifying individual terms as being the same presupposes the concept of sameness of meaning, in relation to which the individual terms can be identified. Thus commonality is not due simply to a term, but requires the formation of concepts on the basis of the mistaken imputation of commonality onto discrete individuals. What does it mean, however, for a concept to be based on an assumed commonality? Here Dharmakīrti’s theory must be placed within its proper context, the apoha or exclusion theory of language, which was created by Dignāga. This complex topic is beyond the scope of this chapter. Suffice it to say that the apoha theory is a way to explain how language signifies in a world of individuals. Linguistic meaning poses a particularly acute problem for Dignāga and Dharmakīrti, for they are committed to a connotationist view of language, in which sense has primacy over reference. Such a view, however, is difficult to hold in a nominalist ontology that disallows abstract entities, such as meaning.18 The apoha theory tries to solve this conundrum by arguing that language does not describe reality positively through universals, but negatively by exclusion. Language is primarily meaningful, but this does not mean that there are real senses. Rather, we posit agreed-upon fictions that we construct for the sake of categorizing the world according

INDIAN THEORIES OF MIND

109

to our purposes. Thus ‘cow’ does not describe Bessie through the mediation of a real universal (cowness), but by excluding a particular (Bessie) from the class of non-cow. Matilal describes Dignāga’s view this way: Each name, as Dignāga understands, dichotomizes the universe into two: those to which the name can be applied and those to which it cannot be applied. The function of a name is to exclude the object from the class of those objects to which it cannot be applied. One might say that the function of a name is to locate the object outside of the class of those to which it cannot be applied. (Matilal, 1971, p. 45)

Although linguistic form suggests that we subsume an individual under a property, analysis reveals that words merely exclude objects from being included in a class to which they do not belong. The function of a name is to locate negatively an object within a conceptual sphere. The impression that words positively capture the nature of objects is misleading. This theory was immediately attacked by Hindu thinkers, such as Kumārila and Uddyotakara, who raised strong objections. One of them was that this theory is counter-intuitive, because we do not perceive ourselves to eliminate non-cows when we conceive of cows. Dharmakīrti’s theory of concept formation is in many ways an attempt to answer these attacks. It argues that the apoha theory is not psychological, but epistemological. In conceiving of objects we do not directly eliminate other objects, but instead rely on a representation that is made to stand in for an assumed commonality shared by several particulars. It is this fictional commonality that is the result of an exclusion. There is nothing over and above particulars, which are categorized on the basis of their being excluded from what they are not. The concept that has been formed in an essentially negative way is projected onto real things. In the process of making judgments such as ‘this is a tree’, the real differences that exist between the different trees come to be ignored and the similarities are reified into a common universal property, which is nothing but a socially agreed-upon fiction. The eliminative nature of thought and language is psychologically revealed when we examine the learning process. The word ‘cow’, for instance, is not learned only through a definition, but by a process of elimination. We can give a definition of ‘cow’, but the definition works only if its elements are known already. For example, we can define cows as animals having dewlaps, horns, and so on (the traditional definition of ‘cow’ in Indian philosophy). But how do we know what counts as a dewlap? Not just by pointing to the neck of a cow, but by eliminating the cases that do not fit. In this way, we establish a dichotomy between those animals that fit, and other animals or things that do not, and on the basis of this negative dichotomy we construct a fictive property, cowness. This construction is not groundless, however, but proceeds through an indirect causal connection with reality. Concepts are not formed a priori, but elaborated as a result of experiences. Dharmakīrti’s solution to the problem of thought and meaning is thus to argue that in a world bereft of real abstract entities (properties), there are only constructed intensional (linguistic) pseudo-entities, but that this construction is based on experience; that is, perception. This grounding in perception ensures that, although conception is mistaken in the way reviewed above, it is neither baseless nor random and hence can lead to the formation of concepts that will be attuned to the causal capacities of particulars.

110

GEORGES DREYFUS AND EVAN THOMPSON

Dharmakīrti and Abhidharma: intentionality revisited

Dharmakīrti’s analysis has in certain respects a great deal of continuity with the ­Abhidharma. Both view the mind as constituted by a succession of mental states in accordance with their ontological commitments, which privilege the particular over the general. Reality is made up of a plurality of elements (here moments of awareness), and generality, when it is not a figment of our imagination, is at best the result of aggregation. This emphasis on the particular derives from the central tenets of the Buddhist tradition; namely, non-substantiality and dependent origination. In Dharmakīrti’s epistemological approach, this emphasis expresses itself in valuing perception over conception, and in the problematic but necessary cooperation between the two forms of cognition. We do not come to know things by merely coming across them, but by integrating them into our conceptual schemes on the basis of our experiences. One question raised by this analysis concerns intentionality. The Abhidharma tradition had assumed all along that cognitions were intentional, but did not provide a systematic analysis of intentionality. Dharmakīrti fills this gap, analyzing the way in which various types of cognition bear on their objects. But because he makes a sharp distinction between perception and conception, his analysis does not yield a single concept of intentionality, but on the contrary leads us to realize that this central notion may have to be understood in multiple ways. The cognitive process starts with our encounter with the world through perceptions, but this encounter is not enough to bring about knowledge. Only when we are able to integrate the objects delivered through the senses into our categorical schemes can we be said to know them in the full sense of the word. Hence, if we understand intentionality as cognitive – that is, as pertaining to knowledge – we may well have to agree with Dharmakīrti that perception is not in and of itself fully intentional. Only when perception is coordinated with conception does it become intentional; hence it can be said to be intentional only in a derived sense of the word. Perception is not in and of itself cognitive, but only inasmuch as it has the ability to induce conceptual interpretations of its objects. This does not mean, however, that perception is completely blank or purely passive. It has an intentional function, that of delivering impressions that we take in and organize through our conceptual schemes. Hence, perception can be said to have a phenomenal intentionality, which may be revealed in certain forms of meditative experiences. Dharmakīrti alludes to such experiences when he describes a form of meditation, in which we empty our mind without closing it completely to the external world (Commentary on Valid Cognition III: 123–5, in Miyasaka 1971–2). In this state of liminal awareness, things appear to us but we do not identify them. We merely let them be. When we come out of this stage, the usual conceptual flow returns, and with it the conceptualization that allows us to identify things as being this or that. This experience shows, Dharmakīrti argues, that identification is not perceptual, but is due to conceptualization. In such a state, perception takes place but not conceptualization. Hence, perception is a non-conceptual sensing onto which interpretations are added. Due to the speed of the mental process, the untrained person cannot differentiate conceptual from non-conceptual cognitions. It is only on special occasions, such as in some form of meditation, that a clear differentiation can be made. There, the flow of thought gradually subsides, and we reach a state in which there is a bare sensing of things.

111

INDIAN THEORIES OF MIND

In this state, what we call shapes and colors are seen barely (i.e., as they are delivered to our senses without the adjunctions of conceptual interpretations). When one gradually emerges from such a non-conceptual state, the flow of thoughts gradually reappears, and we are able to make judgments about what we saw during our meditation. One is then also able to make a clear differentiation between the products of thoughts and the bare delivery of the senses and to distinguish cognitive from phenomenal intentionality. The analysis of intentionality, however, may have to go even further to account for all the forms of cognition known to Buddhist traditions. We alluded above to the Abhidharmic idea of a store-consciousness, a subliminal form of cognition that supports all the propensities, habits, and tendencies of a person. Although such a store-consciousness is usually asserted by the Yogācāra to support their idealist view, it is known to other traditions under other names and hence has to be taken seriously within a Buddhist account of the mind, regardless of the particular views that are associated with it. But given the particularities of this form of consciousness, its integration within a Buddhist view of the mind is not without problems. The difficulties come from the fact that the store-consciousness does not seem to have cognitive or even phenomenal intentionality. Because it does not capture any feature, it cannot be said to know its object, like conceptions. Because it is subliminal, it is difficult to attribute to it a phenomenal content able to induce categorization, like perceptions. How then can it be intentional? To respond to this question would necessitate an analysis that goes well beyond the purview of this chapter. Several avenues are open to us. We could argue that the storeconsciousness is not intentional and hence that intentionality is not the defining characteristic of the mental, but only of certain forms of cognitions.We would then be faced with the task of explaining the nature of the mental in a way that does not presuppose intentionality. Or we could extend the concept of intentionality, arguing that the store-consciousness is not intentional in the usual cognitive or phenomenal senses of the word, but rather that its intentionality consists in its having a dispositional ability to generate more explicit cognitive states. Some Western phenomenologists, notably Husserl and Merleau-Ponty, distinguish ‘object directed intentionality’ from ‘operative intentionality’ (see Chapter 4). Whereas the former is what we usually mean by intentionality, the latter is a non-reflective tacit sensibility, a spontaneous and involuntary level that makes us ready to respond cognitively and affectively to the world, though it is not by itself explicitly cognitive. This most basic form of intentionality is important in explaining our openness to the world. It also seems an interesting avenue for exploring the cognitive nature of the store-consciousness. CONCLUSION

We can now see the richness and the complexities of the Indian Buddhist analyses of the nature of the mind and consciousness. The Abhidharma provides the basis of these analyses, with its view of the mind as a stream of moments of consciousness and its distinction between the primary factor of awareness and mental factors. This tradition also emphasizes the intentional nature of consciousness, the ability of consciousness to be about something else. As we have seen, however, this concept is far from self-evident and needs further philosophical clarification.This clarification is one of the important tasks of Dharmakīrti’s philosophy. In accomplishing this task, Dharmakīrti critically explores the

112

GEORGES DREYFUS AND EVAN THOMPSON

variety of human cognitions, distinguishing the conceptual from the perceptual modes of cognition and emphasizing the constructed nature of the former and its close connection with language.Yet, as we have also seen, this philosophy is not always able to account for all the insights of the Abhidharma, particularly those concerning the deeper layers of consciousness. When we look at the Indian Buddhist tradition, we should not look for a unified and seamless view of the mind. Like any other significant tradition, Indian Buddhist philosophy of mind is plural and animated by debates, questions, and tensions.This rich tradition has a great deal to offer contemporary mind science and philosophy,including rich phenomenological investigations of various aspects of human cognition and exploration of various levels and types of meditative consciousness. This tradition also shows, however, that it would be naïve to take these investigations of consciousness as being objectively given or established. Rather, they are accounts of experience that are often intertwined with doctrinal formulations and hence are open to critique, revision, and challenge, like any other human interpretation. Indeed, these formulations need to be taken seriously and examined with the kind of critical spirit and rigorous philosophical thinking exhibited by Dharmakīrti. Only then, can we do justice to the insights of this tradition. GLOSSARY Sa ¯m . khya

Pradhāna: primordial nature or prakṛti, materiality. The primordial substance out of which the diversity of phenomena arise. It is composed of three qualities (guṇa): sattva (transparency, buoyancy), rajas (energy, activity), and tamas (inertia, obstruction). They are the principles or forces whose combination produces mental and material phenomena. Atman: spiritual self or puruṣa, person. The non-material spiritual element that merely witnesses the mental activities involved in the ordinary awareness of objects. Buddhi: usually translated as ‘the intellect’. It has the ability to distinguish and experience objects.This ability provides the prereflective and presubjective ground out of which determined mental states and their objects arise. It is also the locus of all the fundamental predispositions that lead to these experiences. Ahaṃkāra: egoity or ego-sense. This is the sense of individual subjectivity or selfhood tied to embodiment, which gives rise to the subjective pole of cognition. Manas: mentation. It oversees the senses and discriminates between objects. By serving as an intermediary between the intellect and the senses, mentation organizes sensory impressions and objects and integrates them into a temporal framework created by memories and expectations. Citta: mental activities or antahkarana, internal organ. This is the grouping of buddhi, ahaṃkāra, and manas. Pramāṇa: instrument of valid cognition of the self. The Sāṃkhya recognizes three such instruments: perception, inference, and testimony. The Nyāya adds a fourth one, analogy.

113

INDIAN THEORIES OF MIND

Buddhist

Citta: primary factor of awareness or vijñāna, consciousness. It is the aspect of the mental state that is aware of the object, or the bare apprehension of the object. It is the awareness that merely discerns the object, the activity of cognizing the object. Caitesika: mental factor. Mental factors are aspects of the mental state that characterize the object of awareness and account for its engagement. In other words, whereas consciousness makes known the mere presence of the object, mental factors make known the particulars of the content of awareness, defining the characteristics and special conditions of its object. Alaya-vijñāna: store-consciousness. This continuously present subliminal consciousness is posited by some of the Yogācāra thinkers to provide a sense of continuity in the person over time. It is the repository of all the basic habits, tendencies, and propensities (including those that persist from one life to the next) accumulated by the individual. Bhavaṇga citta: life-constituent consciousness. Although this consciousness is not said to be always present and arises only during the moments where there is no manifest mental activity, it also provides a sense of continuity for the Theravada school, which asserts its existence. Kliṣta-manas: afflictive mentation. This is the inborn sense of self that arises from the apprehension of the store-consciousness as being a self. From a Buddhist point of view, however, this sense of self is fundamentally mistaken. It is a mental imposition of unity where there is in fact only the arising of a multiplicity of interrelated physical and mental events. Pramāṇa: valid cognition. Not the instrument of a self but the knowledge-event itself. There are only two types of valid cognition admissible in Buddhist epistemology, pratyakṣa, perception, and anumāna, inference. Svasaṃvedana: self-cognition. This is the limited but intuitive presence that we feel we have toward our own mental episodes, which is due not to the presence of a metaphysical self but to the non-thematic reflexive knowledge that we have of our own mental states. Because self-cognition does not rely on reasoning, it is taken to be a form of perception. It does not constitute, however, a separate reflective or introspective cognition. Otherwise, the charge that the notion of apperception opens an infinite regress would be hard to avoid. NOTES 1. Presenting the Sāṃkhya view in a few lines is problematic given its evolution over a long period of time, an evolution shaped by the addition of numerous refinements and new analyses in response to the critiques of Buddhists and Vedāntins. For a quick summary, see Mahalingam (1977). For a more detailed examination, see Larson and Bhattacharya (1987). 2. Contrary to Vedānta, the Sāṃkhya holds that there are many individual selves rather than a universal ground of being such as Brahman. 3. The notion of a pure and passive ‘witness consciousness’ is a central element of many Hindu views about consciousness (see Gupta, 1998, 2003). 4. For a thoughtful discussion of this view of the mind, see Schweizer (1993).

114

GEORGES DREYFUS AND EVAN THOMPSON

5. Numerous translations of Patañjali’s Yoga Sutras are available in English. 6. For discussion of the Advaita Vedānta view of consciousness, see Gupta (2003, Chapter 5). For a philosophical overview of Advaita Vedānta, see Deutsch (1969). 7. For a glimpse of the origins of the Abhidharma, see Gethin (1992). 8. For Husserl, by contrast, not all consciousness is intentional in the sense of being objectdirected. See Chapter 4 and the final section of this chapter. 9. All quotations from this work are translated from the French by G. Dreyfus. 10. See Rahula (1980, p. 17). Although the Theravada Abhidharma does not recognize a distinct store-consciousness, its concept of bhavaçga citta, the life-constituent consciousness, is similar. For a view of the complexities of the bhavaçga, see Waldron (2003, pp. 81–87). 11. They are then said to be conjoined (sampayutta, mtshungs ldan), in that they are simultaneous and have the same sensory basis, the same object, the same aspect or way of apprehending this object, and the same substance (the fact that there can be only one representative of a type of consciousness and mental factor at the same time). See Waldron (2003, p. 205). 12. This list, which is standard in the Tibetan tradition, is a compilation based on Asaçga’s Abhidharma-samuccaya. It is not, however, Asaçga’s own list, which contains 52 items (Rahula 1980, p. 7). For further discussion, see Napper (1980) and Rabten (1978/1992). For the lists of some of the other traditions, see Bodhi (1993, pp. 76–79) and de la Vallée Poussin (1971, II: 150–178). 13. Although some of these states may be soteriologically significant and involve the ability to transcend duality, not all need be. The practice of concentration can involve signless meditative states, and so too does the practice of some of the so-called formless meditative states. 14. For a discussion of whether compassion and lovingkindness, seen from a Buddhist point of view, are emotions, see Dreyfus (2002). 15. For a brief but thoughtful discussion of the idea of Buddhism as a psychology, see Gomez (2004). 16. For discussion of the characteristics of Indian logic, see Matilal (1985) and Barlingay (1975). On Buddhist logic, see Kajiyama (1966). For an analysis of Dharmakīrti’s philosophy, see Dreyfus (1997) and Dunne (2004). 17. For a detailed treatment of Dharmakīrti’s arguments and their further elaboration in the Tibetan tradition, see Dreyfus (1997, pp. 338–341, 400–415). 18. For more on this difficult topic, see Dreyfus (1997) and Dunne (2004). REFERENCES Barlingay, S. S. (1975). A modern introduction to Indian logic. Delhi: National. Bodhi, B. (Ed.). (1993). A comprehensive manual of Abhidharma. Seattle, WA: Buddhist Publication Society. Damasio, A. (1995). Descartes’ error: Emotion, reason and the human brain. New York: Harper Perennial. de la Vallée Poussin, L. (1971). L’Abhidharmakoṡa de Vasubandhu. Bruxelles: Institut Belge des Hautes Etudes Chinoises. de la Vallée Poussin, L. (1991). Notes sur le moment ou ksana des bouddhistes. In Prasad, H. S. (Ed.), Essays on time. Delhi: Sri Satguru. Deutsch, E. (1969). Advaita Vedanta: A philosophical reconstruction. Honolulu: University Press of Hawaii. Dreyfus, G. (1997). Recognizing reality: Dharmakīrti’s philosophy and its Tibetan interpretations. Albany, NY: State University of New York Press. Dreyfus, G. (2002). Is compassion an emotion? A cross-cultural exploration of mental typologies. In Davidson, R. & Harrington, A. (Eds.), Visions of compassion: Western scientists and Tibetan Buddhists examine human nature (pp. 31–45). Oxford: Oxford University Press. Dunne, J. D. (2004). Foundations of Dharmakjrīti’s philosophy. Boston: Wisdom.

INDIAN THEORIES OF MIND

115

Gethin, R. (1992).The Mātrikās: Memorization, mindfuIness and the list. In Gyatso, J. (Ed.), In the mirror of memory (pp. 149–172). Albany, NY: State University of New York Press. Goleman, D. (2003). Destructive emotions.A scientific dialogue with the Dalai Lama. NewYork: Bantam. Gomez, L. (2004). Psychology. In Buswell, R. (Ed.), Encyclopedia of Buddhism (pp. 678–692). New York: MacMillan. Guenther, H. (1976). Philosophy and psychology in the Abhidharma. Berkeley, CA: Shambala Press, 1976. Gupta, B. (1998). The disinterested witness. A fragment of Advaita Vedanta phenomenology. Evanston, IL: Northwestern University Press. Gupta, B. (2003). Cit. consciousness. New Delhi: Oxford University Press. James, W. (1981). Principles of psychology. Cambridge, MA: Harvard University Press. Kajiyama,Y. (1966). Introduction to Buddhist logic. Kyoto: Kyoto University. Larson, J., & Bhattacharya, R. S. (1987). Encyclopedia of Indian philosophies: Sāṃkhya, A dualist tradition in Indian philosophy. Delhi: Motilal. Mahalingam, I. (1977). Sāṃkhya-Yoga. In Carr, B. & Mahalingam, I. (Eds.), Companion encyclopedia of Asian philosophy. London: Routledge Press. Matlilal, B. K. (1971). Epistemology, logic, and grammar in Indian philosophical analysis. The Hague: Mouton. Matilal, B. K. (1985). Logic, language, and reality. Delhi: Matilal Banarsidas. Mayeda, S. (1992). A thousand teaching: The Upadeŝashasrī. Albany, NY: State University of New York Press. Original work published 1979. Miyasaka,Y. (Ed.) (1971–2). Pramanavarttikakarika. Acta Indologica 2. Napper, E. (1980). Mind in Tibetan Buddhism. Ithaca, NY: Snow Lion. Pöppel, E. (1988). Mindworks:Time and conscious experience. Boston: Harcourt Brace Jovanovich. Rabten, G. (1992). The mind and its functions. Mt. Pélerin: Rabten Choeling. Original work published 1978. Rahula,W. (1980). Le Compendium de la Super-Doctrine de Asagga. Paris: Ecole Française d’Extrême-Orient. Schweizer, P. (1993). Mind/consciousness Dualism in Saṃkhya-Yoga philosophy. Philosophy and Phenomenological Research, 53, 845–859. Sellars, W. (1956). Empiricism and the philosophy of mind. In Feigl, H. & Scriven, M. (Eds.), Minnesota studies in the philosophy of science.Vol. 1: The foundations of science and the concepts of psychology and psychoanalysis (pp. 253–329). Minneapolis, MN: University of Minnesota Press. Varela, F. J., Thompson, E., & Rosch, E. (1991). The embodied mind: Cognitive science and human experience. Cambridge, MA: MIT Press. Vetter, T. (1966). Dharmakīrti’s Pramanaviniscayah 1. Kapitel: Pratyaksam. Vienna: Österreichische Akademie der Wissenschaften. Waldron, W. (2003). The Buddhist unconscious. London: Routledge Press. Wider, K. (1997). The bodily nature of consciousness: Sartre and contemporary philosophy of mind. Ithaca, NY: Cornell University Press.

CHAPTER

7

From the five aggregates to phenomenal consciousness Towards a cross-cultural cognitive science Jake H. Davis and Evan Thompson

Source: Steven M. Emmanuel (ed.), A Companion to Buddhist Philosophy (Oxford: John Wiley and Sons, 2013) pp. 585–597. INTRODUCTION

Buddhism originated and developed in an Indian cultural context that featured many first-person practices for producing and exploring states of consciousness through the systematic training of attention. In contrast, the dominant methods of investigating the mind in Western cognitive science have emphasized third-person observation of the brain and behavior. In this chapter, we explore how these two different projects might prove mutually beneficial.We lay the groundwork for a cross-cultural cognitive science by using one traditional Buddhist model of the mind – that of the five aggregates – as a lens for examining contemporary cognitive science conceptions of consciousness. The model of consciousness and meditative transformations of consciousness that we offer in this chapter is inspired by the accounts found in the Pāli Nikāyas. For this reason and for the sake of simplicity, we make reference especially to Pāli textual sources and terminology. Nevertheless, it is important to note at the outset that these texts admit of multiple possible readings. Our reconstruction differs in certain respects from the traditional interpretation of the five aggregates in the Theravāda Buddhist commentaries on the Pāli Nikāyas. Our aim, however, is not to give an historical account of what these concepts meant at any point in the development of Buddhist thought; and we make no claim that anyone in the Buddhist tradition, early or late, actually understood this model in the way we suggest. The model of attention, consciousness, and mindfulness that we draw from the Nikāya account of the five aggregates is of interest to us because it suggests promising new directions for scientific investigations of the mind. Put another way, 116

TOWARDS A CROSS-CULTURAL COGNITIVE SCIENCE

117

whatever value our model has lies not in any claim to historical authenticity but, rather, in its claim to being empirically accurate and productive of further research. Situating Buddhist views within recent scientific debates about consciousness allows us to see how these views might be tested experimentally and thereby opens up new understandings of what these ancient teachings mean for us today. At the same time, understanding the conceptual frameworks of the Buddhist teachings can help scientists to refine the theoretical frameworks they bring to research on meditation and consciousness. This opportunity is lost if we simply apply existing scientific frameworks to interpret data from experiments on Buddhist meditation practices. The burgeoning scientific literature on “mindfulness” meditation offers a case in point. This form of meditation can be broadly characterized by the aim to cultivate a lucid awareness of one’s own moment-to-moment bodily, emotional, perceptual, and cognitive processes. Seeing the potential for this technique in medical settings, Jon Kabat-Zinn pioneered in the 1980s the Mindfulness-Based Stress Reduction program (MBSR). MBSR is now offered in the secular context of hospitals and clinics around the world and has become the subject of a burgeoning scientific literature. Kabat-Zinn’s approach was influenced by Korean Zen Buddhist teachings as well as by Advaita Vedanta, and the particular technique he incorporated into MBSR was directly inspired by Theravāda Buddhist teachers drawing on texts from the Pāli discourses such as the Mahāsatipaṭṭhāna Sutta, or “Longer Discourse on Mindfulness” (DN.II.290–315). Yet, attempts in the scientific literature to formulate what mindfulness is have often proceeded in almost total independence from theoretical formulations of mindfulness practice contained in Buddhist textual traditions. In the absence of references to such traditional canonical sources, there has been an inordinate focus on one particular phrase Kabat-Zinn used in his seminal introductory guide for practitioners to describe mindfulness – namely, “paying attention in a particular way: on purpose, in the present moment, and nonjudgmentally” (Kabat-Zinn 2004, 4). When specific references occur in the scientific literature to the Buddhist textual sources, these references often consist in noting that the term “mindfulness” is a translation of the Pāli term sati. In Buddhist theory, however, the term sati carries connotations of memory and remembrance, making attempts to understand mindfulness as a present-centered, non-elaborative, and non-judgmental attention appear inaccurate and confused (see Bodhi 2011; Dreyfus 2011). Indeed, the term “mindfulness” seems to have been chosen by early translators of the Pāli texts because they saw parallels not with a notion of non-judgmental present-centered attention but, rather, between the Christian ethical notion of conscience and the textual usage of sati in the context of holding in mind and being inspired by certain truths, for the sake of improvement of one’s ethical character (Gethin 2011). The broad usage of the term sati is perhaps best captured by the colloquial English notion of “minding.” The Pāli texts employ sati in reference to everything from “minding” one’s livestock (MN.I.117) to “minding” one’s meditation object in practices such as lovingkindness (Sn.26), in addition to using sati specifically in the context of mindfulness meditation or, more literally, in the establishment of sati (sati-upaṭṭhāna).1 In this general sense, sati clearly can involve elaborative and evaluative cognitive processes. In the role sati plays in the context of mindfulness meditation, however, the involvement of memory may be of a more limited and specific kind.

118

JAKE H. DAVIS AND EVAN THOMPSON

In order to investigate properly a given type of meditation practice, scientists must take account of the traditional theoretical frameworks used to conceptualize and teach that practice (Lutz et al. 2007). We outline here how the traditional theoretical context of mindfulness practice can offer important suggestions for scientific research. In particular, the five aggregates model draws distinctions that are not always clearly formulated in contemporary cognitive science, but that are crucial for a scientific understanding of the function of mindfulness meditation. We suggest below how empirical hypotheses about the role of memory and its relation to attention and consciousness in mindfulness meditation can be refined in light of distinctions suggested in the Buddhist five aggregates model. A BUDDHIST MODEL OF THE MIND

The Buddhist five aggregates model parallels a number of distinctions drawn in cognitive science and therefore serves as a useful theoretical resource for developing a cross-cultural cognitive science of consciousness (Varela et al. 1991). In the Pāli texts the five aggregates (khandhas) are listed as rūpa, vedanā, saññā, saṅkhāra, and viññāṇa. Deciding what each of these words means, however, is not straightforward. Indeed, as we will see, interpreting the khandhas raises philosophical issues that directly connect with contemporary debates about consciousness. The first aggregate, rūpa, is often understood as referring simply to the physical matter of the body. In the Pāli dialogues, however, this term connotes not only the body’s solidity and extension but also its mobility, temperature regulation, fluid, and digestive systems, as well as its processes of decay. For this reason, some textual scholars suggest that rūpa is better understood as referring to the “lived body rather than simply its flesh” (Hamilton 2000, 29). On this reading, the conceptual framework of the five khandhas anticipates contemporary cognitive scientific and phenomenological accounts of the bodily basis of cognition, emotion, and consciousness (see Thompson 2007 for an overview). Bodily changes such as the contraction of the gut and the flush of blood in anger have long been recognized as central to emotion.William James (1884) proposed that emotions essentially are such bodily reactions, an idea that still plays an important role in emotion theory today; for example, the neuroscientist Antonio Damasio (2000) and the philosopher Jesse Prinz (2004) have both argued that emotions are constituted in part by bodily reactions. But emotion theorists also recognize a second aspect of emotion, one that takes us from rūpa construed as the living body to vedanā, the second of the five khandhas. This second aspect is the specific feeling tone belonging to a given emotion. Some emotions feel pleasant and others feel unpleasant. When we consciously feel joyful, the experience is pleasant, and when we feel fearful, the experience is unpleasant. Psychologists call this aspect of emotion its affective valence or hedonic tone (see Colombetti 2005 for the complicated history behind this concept of “valence”). The notion of affect valence provides a close analogue to the Buddhist notion of vedanā. In the Khajjaniya Sutta (SN.III.86–7), vedanā is defined as feeling pleasure, feeling pain, or feeling neither-pleasure-nor-pain. In the case of both concepts, valence and vedanā, the feeling tone of pleasant versus unpleasant is closely related to action tendencies of approach versus avoidance. From the modern neuroscience perspective, the bodily responses constitutive of an emotion, including an emotion’s valence and action tendency,

TOWARDS A CROSS-CULTURAL COGNITIVE SCIENCE

119

can be activated even when we do not report consciously feeling the emotion (LeDoux 2000). For example, we may exhibit bodily responses associated with fear, even though we do not report seeing anything fearsome or feeling fearful. Thus, like vedanā, valence motivates us at implicit as well as explicit levels. Moreover, recent work has shown that such implicit affect valence is not limited to emotional episodes and influences decisionmaking on everything from consumer choices to moral judgment (Loewenstein and Lerner 2003). This understanding of the pervasive role of affect valence in human psychology finds a parallel in the Buddhist suggestion that vedanā is present with every mental state, not just those Western psychology includes under the emotions. In understanding the function of meditative training in bringing about personal transformation, the habits of mind that dispose an individual to perceive and react to the world in certain distinctive ways are of obvious importance. These habits of mind fall under the fourth of the five aggregates, saṅkhāra. This category can be understood as comprising all volitional activities. These include volitions that lead to outward action or what we normally think of as the will. But they also include more internal processes, such as attention, manasikāra – literally, “making-in-the-mind.” Thus we can understand saṅkhāra as referring to implicit and habitual processing routines that shape how we perceive and behave and that typically escape explicit, cognitive awareness. Importantly, these habits of mind not only shape our inner and outer actions but are themselves formed through the repetition of certain kinds of inner and outer volitional activities.Thus, in addition to conditioning the other four aggregates, the saṅkhāras involve dynamic self-reference and self-conditioning: habits are formed and conditioned by habits (SN.III.87).2 This conception parallels recent models of cognitive events as self-forming processes arising from non-linear interactions between components at neural and motor levels (Cosmelli et  al. 2007). Complex (non-linear) dynamical systems have a feature known as sensitive dependence on initial conditions: a minute change in conditions at one point in time can greatly shift the trajectory of the system down the line. Similarly, the dynamic self-formation of the saṅkhāras allows for the possibility of radical transformation of one’s personality traits. In the particular case of mindfulness meditation, the suggestion is that, by intentionally attending to present experience instead of dwelling in reactivity to the remembered past or the imagined future, we can radically transform the habits of attention that surface at moments of feeling threatened or tempted, and thereby transform the way we react outwardly to such situations. Within this category of habits of mind, the role of attention is of particular interest for us here. In the Mahāhatthipadopama Sutta of the Majjhima Nikāya, for example, we find the following claim: If the internal eye-organ is intact, but an external form does not come into its range … If the internal eye-organ is intact, and an external form does come into its range, but there is not the bringing together born from that (tajja samannāhāra), there is not the appearance of a degree of consciousness born from that (tajja viññāṇabhāga). But when the internal eye-organ is intact, and an external form does come into its range, and there is the bringing together born from that, there is the appearance of a degree of consciousness born from that. (MN.I.190)

120

JAKE H. DAVIS AND EVAN THOMPSON

Despite other Pāli texts that omit the factor of “bringing together,” samannāhāra, in the account of perceptual processes, this factor is clearly crucial in the above formulation: an external form coming into the range of an intact eye is said to result in a share or degree of consciousness only with the addition of this factor of bringing together. The traditional Pāli commentary glosses samannāhāra as here meaning manasikāra (attention).3 As the above formulation suggests, manasikāra is understood in this theoretical framework as a universal kind of attention necessary for any moment of consciousness. It may therefore correspond in a rough way to the basic kind of alertness required for the basal, core-level consciousness that Parvizi and Damasio (2001) hypothesize to be dependent on subcortical structures such as the thalamus and brainstem, and which occurs independently of the direction of this consciousness to particular objects through selective attention. This core level of consciousness, which we discuss briefly below, stands in contrast to the more cognitive functions that allow one to identify, recall, and report what one experiences. These cognitive processes are the function of the third aggregate, saññā. In the Khajjaniya Sutta (SN.III.87), saññā is defined as cognizing (sañjānāti) that there is blue, that there is red, yellow, or white. The term saññā is often glossed as “perception,” but this interpretation is inadequate. As the Pāli scholar Peter Harvey explains, saññā is only one part of the perceptual process and … one can have a saññā of a mental object but cannot, in English, be said to “perceive” such an object … the word “saññā” and its verbal form “sañ-jānāti” clearly refer to some kind of knowledge or knowing which is done in an associative, connective, linking (sa-) way. (Harvey 1995, 141)

The Pāli texts contain some intriguing statements that suggest saññā may be akin to what the philosopher Ned Block (2007, 2008) calls “cognitive access,” defined as the ability to recall, report, and deliberate on a perceptual event.The Nibbedhika Sutta (AN. III.413), for instance, defines saññā as that which results in spoken communication (vohāra): “As one identifies (sañjānāti) it, so one says ‘I saw thus.’” Saññā is differentiated in the Buddhist model of the mind from viññāṇa, the fifth aggregate, often glossed as “consciousness.” It is tempting to relate this notion to what Block calls “phenomenal consciousness” (Block 2007, 2008). Whereas phenomenal consciousness consists in “what it is like” for a subject to have or to undergo an experience, cognitive access consists in having the content of an experience enter working memory so that one can identify and report on this content. Given this distinction, viññāṇa, defined as a moment of visual, auditory, tactile, olfactory, gustatory, or mental awareness, would be analogous to phenomenal consciousness, whereas saññā, defined as a recognitional ability, would be analogous to cognitive access. Yet this tentative analogy between Pāli Buddhist and cognitive science conceptions of consciousness needs refinement. Block conceives of phenomenal consciousness as a state of experiencing in a rich and vivid way certain objects or properties – for instance, a state of seeing red. Without such a notion of phenomenally conscious states as essentially including modality-specific content, it would make little sense to suggest, as Block does, that visual phenomenal consciousness might be realized by certain patterns of recurrent neural activity in visual areas of the brain (Block 2005). In contrast, Parvizi and Damasio

TOWARDS A CROSS-CULTURAL COGNITIVE SCIENCE

121

(2001) suggest that there is a basic, core level of consciousness, dependent on the thalamus and brainstem, that occurs independently of selective attentional processes in higher cortical areas. This core or ground-floor level of consciousness depends on a basic kind of alerting function distinct from the higher-level mechanisms of selective attention that come into play in determining what one is conscious of. On this view, the fact that there is a phenomenal feel – the fact that there is something it is like for a subject – depends on the basic alerting function. In contrast, the content of phenomenal consciousness – what it is like for a subject – depends also on how this consciousness is directed to particular objects and properties through selective attention. Put another way, the particular contents of phenomenal consciousness can be seen as modifications or modulations of a basal level of awareness dependent on the alerting function (see also Searle 2000). We suggested above that the Pāli Buddhist concept of manasikāra may be analogous to this alerting function, rather than to selective attention. Correspondingly, viññāṇa may be best understood from this cognitive science perspective as analogous to a basal level of awareness common to all phenomenally conscious states. We need to be cautious, however, in drawing any of the foregoing parallels between the third and fifth aggregates and cognitive science conceptions of cognitive access and consciousness. Currently there is no consensus in cognitive science about whether phenomenal consciousness and cognitive access are two different phenomena, or whether phenomenal consciousness depends constitutively on cognitive access.4 On the one hand, it seems odd to say that you can have a conscious experience that you do not know you are having. And if knowing that you are having a certain experience, such as a visual experience of the color red or a tactile experience of hardness, requires the cognitive functions of identifying the object or properties being experienced, then it seems problematic to postulate a type of experience that occurs independent of cognitive access. Furthermore, given that the principal scientific criterion for the presence of consciousness is behavioral report, and behavioral report requires cognitive access, how could such a subjective experience ever be investigated? On the other hand, it seems unsatisfactory to assume, in advance of the evidence, that having a conscious experience consists wholly in various cognitive operations, such as identifying its content or identifying oneself as having experienced that content. Proponents of drawing a distinction between phenomenal consciousness and cognitive access need only posit that some instances of phenomenal consciousness happen not to be cognitively accessed; they need not posit that there are subjective experiences that the subject cannot access or know about. Indeed, one function of phenomenal consciousness may be to make its content accessible for encoding in working memory, for the purposes of identification, recall, deliberation, and report (Prinz 2005, 2011; see also Block 2011, 567). Certain experiences may be too fleeting and rapid to stabilize in working memory, as various kinds of evidence have sometimes been taken to suggest (see Block 2011; Kouider et al. 2010). Nevertheless, such experiences may not be inaccessible in principle; for instance, it may be possible to gain greater cognitive access to them through the kind of mental training central to mindfulness meditation. We believe this last point indicates a major shortcoming in the current cognitive science discussions. These discussions have proceeded without significant consideration being given to the possibility that specific forms of mental training might be able to

122

JAKE H. DAVIS AND EVAN THOMPSON

produce new data about attention and consciousness. Mental training and its relevance for understanding consciousness are areas where Buddhist theory and meditation practice have much to contribute, as we discuss in the next section. VARIETIES OF ATTENTION TRAINING

Many Buddhist traditions distinguish between meditation practices aimed primarily at concentrating the mind and meditation practices aimed primarily at developing wisdom. In a Buddhist context, concentration practices range from cultivating states such as lovingkindness or, literally, friendliness (mettā) to practices aimed simply at cultivating a settled and unified state of mind (samādhi) through concentration on a meditative object, such as the sensations of the breath or a visualized image of a colored disk or a light. In these forms of meditation, practitioners counteract mind-wandering by repeatedly bringing the mind back to the subject of meditation. Concentration practices may have important contributions to make to our understanding of the processes responsible for stabilizing particular contents in consciousness. We can use studies of the perceptual phenomenon known as binocular rivalry to illustrate this point. In normal vision, the brain receives visual images from each eye that present slightly differing views on the same scene. In the experimental paradigm known as binocular rivalry, however, each eye is presented with a different image at the same time. For example, one eye may receive the image of a house while the other eye receives the image of a face. Subjects generally report seeing one image at a time but also that their perception switches unpredictably between the two images. Thus, although the stimulus remains constant, visual consciousness changes as the two stimuli compete for perceptual dominance. On the one hand, the visual image that is not consciously seen provokes significant neural responses selective to its particular features. For example, the image of a fearful face has been found to activate the amygdala, an area of the brain associated with perceiving emotionally salient stimuli (Williams et al. 2004). On the other hand, voluntary shifts in attention have been shown to affect which image becomes consciously seen (Ooi and He 1999). For this reason, binocular rivalry paradigms have provided an important source of evidence for debates over consciousness and its relation to attention. In an intriguing study of meditation and binocular rivalry, Olivia Carter and her colleagues found that long-term Tibetan Buddhist practitioners of concentration meditation were able to change the perceptual switching rate when they viewed the images while practicing this type of meditation (with eyes open focused on the display as the meditative object) (Carter et al. 2005). A large number of the practitioners reported that the amount of time one image remained perceptually dominant increased considerably while practicing concentration meditation as well as immediately after meditation. Three individuals reported that the image remained completely stable, with no switching, for an entire 5-minute period of concentration meditation. In some cases, one of the two images was completely dominant; in other cases, the non-dominant image remained faintly or partially visible behind the dominant one, so that the conscious perception was of two superimposed images. As Carter and her colleagues observe, “These results contrast sharply with the reported observations of over 1000 meditation-naïve individuals tested

TOWARDS A CROSS-CULTURAL COGNITIVE SCIENCE

123

previously” (ibid., 412). Thus, it may be that meditative training of voluntary attention enables long-term practitioners of concentration meditation to stabilize consciousness of one or the other image, or even to maintain conscious awareness of the non-dominant image, in a way that normal subjects are unable to do. If so, investigations of brain activity in meditators with expertise in concentration meditation may help shed light on the processes that make particular contents phenomenally conscious. The use of various methods of attention training for developing altered states of consciousness through strong concentration was widespread at the time of the Buddha. Buddhist texts relate how, before his enlightenment, the Buddha studied techniques for concentrating the mind under teachers such as Ālāra Kālāma and Uddaka Rāmaputta (MN.I.237–51). Yet these early Buddhist texts also emphasize that the method of mind training that the Buddha went on to discover for himself was novel, with results that differ importantly from those that were being taught by his contemporaries. In a modern context, we can take this claim to be an empirical one, subject to experimental test, and hence one that may be best approached through a cross-cultural cognitive science based on both Buddhist and cognitive scientific models of attention and consciousness. In addition to what cognitive scientists describe as the endogenous orienting network, which voluntarily allocates selective attention to a chosen object (Corbetta and Shulman 2002), concentration or “focused attention” styles of meditation involve a “monitoring” function necessary to detect when attention has wandered away from the chosen object (Lutz et al. 2008). Lutz and colleagues distinguish such “focused attention” practices from “open monitoring” practices, which may involve focused attention training at early stages of practice but use the development of the monitoring skill to be able eventually to drop any intentional selection or deselection within the field of present experience. Instead, meditators aim to remain attentive to whatever arises in moment-to-moment experience, without becoming lost in mind-wandering. Open monitoring styles of meditation include certain Tibetan Buddhist and Chan/Zen practices, as well as Theravāda mindfulness practices. Theravāda mindfulness meditation, or, more literally, the establishment of sati (satiupaṭṭhāna), involves returning the mind again and again to present-moment experience (for a discussion of this term, see Anālayo 2004, 29–30; Bodhi 2011). This practice thus includes an element of concentration, though different teachers emphasize the concentrative aspect to differing degrees. In other concentrative practices, one might return the attention again and again to a particular feeling of friendliness, or a particular mental image of color or light, thereby cultivating the continuity and stability of a particular object in the mind. In contrast, Theravāda mindfulness practice aims to develop a settled type of attention on objects that are constantly changing. Present experiences of heat or cool in the body, of anger or of joy, of concentration or of distractedness, constantly arise and pass away again. Theravāda Buddhist teachings claim that experiencing for oneself in this direct and lucid way the impermanent and unstable nature of all aspects of experience brings about a profound change in how one relates to oneself and others (e.g., Mahasi Sayadaw 1994). In our discussion of the five aggregates model above, we made a distinction between the particular sensory and mental contents of phenomenal consciousness and a basal level of consciousness dependent on an alerting function. Theravāda mindfulness meditation may hold particular promise for investigating this basal level of consciousness because this type

124

JAKE H. DAVIS AND EVAN THOMPSON

of meditation is said to enhance the clear awareness of whatever arises but without using focused attentional selection. One way such enhanced phenomenal consciousness may be achieved is by a reduction in elaborative cognitive processes – the proliferation of evaluative thoughts about moment-to-moment stimuli – combined with increased alertness. Recent experimental studies of Theravāda mindfulness meditation are consistent with this idea. Consider first a study of the effects of Theravāda mindfulness meditation on the socalled attentional blink. In this experimental paradigm, subjects have to identify two visual targets presented within 200 to 500 milliseconds of each other in a rapid sequence of other distracting visual stimuli. Subjects often notice the first target but fail to notice the second one, as if their attention had blinked.The standard explanation is that detecting the first target uses up the available attentional resources, so the second target is missed and not reported. A recent study showed that the ability to detect the second target was greatly improved after a three-month intensive Theravāda mindfulness meditation retreat, and that this improvement correlated with EEG measures showing more efficient neural responses to the first target (Slagter et al. 2007). Importantly, the participants were instructed not to meditate during the task, so the improved performance indicates that mindfulness meditation has lasting effects on attention outside of the context of meditation practice. The authors of this study suggest that mindfulness meditation may lead to less elaborative cognitive processing of the first visual target – less “mental stickiness” to it – and that this reduction facilitates the ability to identify and report the second rapidly occurring target. The idea that enhanced phenomenal consciousness is linked to a reduction in elaborative cognitive processing as a result of mindfulness practice is also supported by recent work on mind-wandering and its association with the brain’s so-called default mode network. The default mode network comprises a set of brain regions active in the resting state but whose activity decreases during externally directed and attention-demanding perceptual tasks (Buckner et  al. 2008); these regions have also been shown to be active during mind-wandering (Mason et al. 2007; Christoff et al. 2009), including mindwandering during focused attention meditation conditions (Hasenkamp et  al. 2012). Mindfulness meditation practice is associated with decreases in default mode network activation (Brewer et  al. 2011; Berkovich-Ohana et  al. 2012) and, importantly, with corresponding increased activation in visceral and somatic areas associated with interception (Farb et al. 2007, 2010). According to traditional descriptions, mindfulness becomes effortless at advanced stages of practice. As the Burmese meditation master Mahasi Sayadaw (1994) puts it, “in the act of noticing, effort is no longer required to keep formations before the mind or to understand them.” We noted above that, in the Pāli Buddhist framework, a basic and universal kind of attention, manasikāra, is held to be necessary for consciousness. The scholar-practitioner Anālayo suggests that sati “can be understood as a further development and temporal extension of this type of attention [manasikāra], thereby adding clarity and depth to the usually too short fraction of time occupied by bare attention in the perceptual process” (Anālayo 2004, 59). Whereas the focusing of attention in concentration practices involves activation of voluntary orienting networks, mindfulness practice may consist in enhancing the processes involved in sustaining alert consciousness more generally. If this were the case, then we should expect that long-term trait increases in one’s consciousness of subtle stimuli (as opposed to transitory state increases) would be evident even in resting states.

TOWARDS A CROSS-CULTURAL COGNITIVE SCIENCE

125

CONCLUSION

To study the effects of therapeutic interventions on the brain and the rest of the body, scientists need to employ conceptual constructs of the phenomenon under investigation that guide where and how they look. Thus, in studying the health benefits and psycho-physiological processes underlying mindfulness meditation, scientists have had to ask what precisely mindfulness is (Davidson 2010).Yet attempts in the scientific literature to define mindfulness have often proceeded in almost total independence from theoretical formulations of mindfulness practice contained in Buddhist textual traditions. Fortunately, a new conversation between Buddhist textual scholars and cognitive scientists about the construct of mindfulness is gaining momentum (see the collection of articles in the June 2011 issue of Contemporary Buddhism). Our goal in this chapter has been to provide some useful tools for this new conversation. In particular, building bridges between the five aggregates model and contemporary cognitive science can offer a way to understand more precisely the roles of attention, consciousness, and memory in Theravāda mindfulness meditation. Like other concentration practices, many forms of mindfulness meditation begin by employing working memory in directing selective attention – for instance, to the sensations of breathing. As we suggested above, however, the reduction of elaborative cognitive processing in mindfulness meditation may play a central role in advanced mindfulness practice, in particular by allowing for an increase in phenomenal consciousness of current stimuli. This mental transformation in turn has implications for what psychologists call “episodic memory” (the memory of particular experienced events), because increased phenomenal consciousness can facilitate accurate identification of what is experienced, as well as later recall and report. Drawing on the relation between the concepts of manasikāra and sati in the Pāli Nikāyas, we have further speculated that mindfulness meditation may function by enhancing the alerting function crucial for phenomenal consciousness. As we noted at the outset of this chapter, however, these texts allow multiple interpretations, and the conception of manasikāra that we employ may not line up neatly with traditional interpretations in the Theravāda Buddhist commentaries. We suggest that the proposed relation between manasikāra and sati be treated as a testable hypothesis.Whatever value our model may have lies in its ability to suggest fruitful directions for future work in the cross-cultural cognitive science of consciousness. We conceive of the discussion that we have undertaken here as one tentative step in a larger project of developing a cross-cultural cognitive science of Buddhist therapeutic interventions. One way to build on our discussion would be to develop a cognitive science perspective on the Buddhist claim that mindfulness counteracts not knowing, by increasing awareness of presently arising stimuli, and also counteracts knowing wrongly, by attenuating emotional distortions of attention, perception, and memory. Having taken that step from cognitive to emotional functions, a further project would be to examine critically, in the light of empirical work on attention, emotion, and moral psychology, the central Buddhist claim that certain emotional motivations are unskillful (akusala) and to be abandoned (pahātabbaṃ); that other qualities are skillful (kusala) and to be cultivated (bhavitabbaṃ); and that we can discern the difference for ourselves.

126

JAKE H. DAVIS AND EVAN THOMPSON

NOTES 1. The term satipaṭṭhāna has commonly been rendered as a (plural) noun, the (four) “foundations of mindfulness.” But the primary sense of the term is verbal and refers to the active practice of establishing mindfulness, as noted recently by prominent translators such as Bhikkhu Bodhi (2011, 25) and Thanissaro Bhikkhu (2011). For a critique of the more standard gloss of satipaṭṭhāna as “foundations of mindfulness” and the commentarial derivation of the term from paṭṭhāna on which this gloss is based, see Anālayo (2004, 29–30). 2. SN.III.87, “saṅkhāre saṅkhārattāya saṅkhatam abhisaṅkharonti.” 3. MN-a.II.229 (commentary to MN.I.190). In support of this interpretation, Harvey (1995, 129–30) notes that these terms are used as synonyms in the suttas, as at MN.I.445. 4. For a sampling of the debate, see Block (2005, 2011); Cohen and Dennett (2011); Kouider et al. (2010); and Lamme (2003). REFERENCES Anālayo (2004). Satipaṭṭhāna:The Direct Path to Realization. Cambridge: Windhorse. Berkovich-Ohana, Aviva, Glicksohn, Joseph, and Goldstein, Abraham (2012). Mindfulness-Induced Changes in Gamma Band Activity – Implications for the Default Mode Network, Self-Reference and Attention. In Clinical Neurophysiology 123, 700–10. doi: 10.1016/j.clinph.2011.07.048. Block, N. (2005). Two Neural Correlates of Consciousness. In Trends in Cognitive Sciences 9(2), 46–52. doi: 10.1016/j.tics.2004.12.006. Block, N. (2007). Consciousness, Accessibility, and the Mesh between Psychology and Neuroscience. In Behavioral and Brain Sciences 30, 481–548. doi: 10.1017/ S0140525X07002786. Block, N. (2008). Consciousness and Cognitive Access. In Proceedings of the Aristotelian Society 108(3), 289–317. doi: 10.1111/j.1467–9264.2008.00247.x. Block, N. (2011). Perceptual Consciousness Overflows Cognitive Access. In Trends in Cognitive Sciences 15, 567–75. doi: 10.1016/j.tics.2011.11.001. Bodhi, Bhikkhu (2011). What Does Mindfulness Really Mean? A Canonical Perspective. In Contemporary Buddhism 12 (1), 19–39. doi: 10.1080/14639947.2011.564813. Brewer, Judson A., Worhunsky, Patrick D., Gray, Jeremy R., Tang, Yi-Yuan, Weber, Jochen, and Kober, Hedy (2011). Meditation Experience is Associated with Differences in Default Mode Network Activity and Connectivity. In Proceedings of the National Academy of Sciences 108, 20254–9. doi: 10.1073/ pnas.1112029108. Buckner, Randy L., Andrews-Hanna, Jessica R., and Schacter, Daniel L. (2008). The Brain’s Default Network. In Annals of the New York Academy of Sciences 1124 (1), 1–38. doi: 10.1196/annals.1440.011. Carter, O. L., Presti, D. E., Callistemon, C., Ungerer,Y., Liu, G. B., and Pettigrew, J. D. (2005). Meditation Alters Perceptual Rivalry in Tibetan Buddhist Monks. In Current Biology 15, R412–R413. Christoff, K., Gordon, A. M., Smallwood, J., Smith, R., and Schooler, J. W. (2009). Experience Sampling During fMRI Reveals Default Network and Executive System Contributions to Mind Wandering. In Proceedings of the National Academy of Sciences 106, 8719. Cohen, Michael A., and Dennett, Daniel C. (2011). Consciousness Cannot Be Separated from Function. In Trends in Cognitive Sciences 15, 358–64. doi: 10.1016/j.tics.2011.06.008. Colombetti, G. (2005). Appraising Valence. In Journal of Consciousness Studies 12(8–10), 103–26. Corbetta, M., and Shulman, G. L. (2002). Control of Goal-Directed and Stimulus-Driven Attention in the Brain. In Nature Reviews Neuroscience 3, 201–15. Cosmelli, D., Lachaux, J. P., and Thompson, E. (2007). Neurodynamical Approaches to Consciousness. In The Cambridge Handbook of Consciousness. Cambridge: Cambridge University Press. Damasio, A. R. (2000). The Feeling of What Happens: Body and Emotion in the Making of Consciousness. London: Heinemann.

TOWARDS A CROSS-CULTURAL COGNITIVE SCIENCE

127

Davidson, R. J. (2010). Empirical Explorations of Mindfulness: Conceptual and Methodological Conundrums. In Emotion 10(1), 8–11. Dreyfus, Georges (2011). Is Mindfulness Present-Centred and Non-Judgmental? A Discussion of the Cognitive Dimensions of Mindfulness. In Contemporary Buddhism 12(1), 41–54. doi: 10.1080/14639947.2011.564815. Farb, N. A. S., Anderson, A. K., Mayberg, H., Bean, J., McKeon, D., and Segal, Z. V. (2010). Minding One’s Emotions: Mindfulness Training Alters the Neural Expression of Sadness. Emotion 10(1), 25–33. Farb, N. A. S., Segal, Z. V., Mayberg, H., Bean, J., McKeon, D., Fatima, Z., and Anderson, A. K. (2007). Attending to the Present: Mindfulness Meditation Reveals Distinct Neural Modes of Self-Reference. In Social Cognitive and Affective Neuroscience 2, 313. Gethin, Rupert (2011). On Some Definitions of Mindfulness. In Contemporary Buddhism 12(1), 263–79. doi: 10.1080/14639947.2011.564843. Hamilton, Sue (2000). Early Buddhism: A New Approach: The I of the Beholder. Richmond, Surrey: Curzon Press. Harvey, Peter (1995). The Selfless Mind: Personality, Consciousness and Nirvana in Early Buddhism. London: Routledge. Hasenkamp,Wendy,Wilson-Mendenhall, Christine D., Duncan, Erica, and Barsalou, Lawrence W. (2012). Mind Wandering and Attention during Focused Meditation: A Fine-Grained Temporal Analysis of Fluctuating Cognitive States. In NeuroImage 59, 750–60. doi: 10.1016/ j.neuroimage.2011.07.008. James, William (1884). What Is an Emotion? In Mind 9, 188–205. Kabat-Zinn, Jon (2004). Wherever You Go,There You Are. New York: Hyperion. Kouider, Sid, Gardelle,Vincent de, Sackur, Jérôme, and Dupoux, Emmanuel (2010). How Rich Is Consciousness? The Partial Awareness Hypothesis. In Trends in Cognitive Sciences 14, 301–7. doi: 10.1016/ j.tics.2010.04.006. Lamme, Victor A. F. (2003). Why Visual Attention and Awareness Are Different. In Trends in Cognitive Sciences 7(1), 12–18. doi: 10.1016/S1364–6613(02)00013-X. LeDoux, J. E. (2000). Emotion Circuits in the Brain. In Annual Review of Neuroscience 23, 155–84. Loewenstein, G., and Lerner, J. S. (2003). The Role of Affect in Decision Making. In Handbook of ­Affective Sciences. Ed. R. J. Davidson, K. R. Scherer, and H. H. Goldsmith. Oxford: Oxford University Press. Lutz, Antoine, Dunne, John D., and Davidson, Richard J. (2007). Meditation and the Neuroscience of Consciousness. In The Cambridge Handbook of Consciousness. Ed. P. D. Zelazo, Morris Moscovitch, and Evan Thompson. Cambridge: Cambridge University Press. Lutz, Antoine, Slagter, Heleen A., Dunne, John D., and Davidson, Richard J. (2008). Attention Regulation and Monitoring in Meditation. In Trends in Cognitive Sciences 12, 163–9. doi: 10.1016/ j.tics.2008.01.005. Mahasi Sayadaw (1994). The Progress of Insight: (Visuddhiñana-katha). Trans. Nyanaponika Thera. At www.accesstoinsight.org/lib/authors/mahasi/progress.html#ch6.11. Mason, M. F., Norton, M. I., Van Horn, J. D., Wegner, D. M., Grafton, S. T., and Macrae, C. N. (2007). Wandering Minds: The Default Network and Stimulus-Independent Thought. In Science 315, 393–5. doi: 10.1126/science.1131295. Ooi, T. L., and He, Z. J. (1999). Binocular Rivalry and Visual Awareness: The Role of Attention. In ­Perception 28, 551–74. Parvizi, J., and Damasio, A. (2001). Consciousness and the Brainstem. In Cognition 79, 135–60. Prinz, Jesse J. (2004). Gut Reactions: A Perceptual Theory of Emotion. New York: Oxford University Press. Prinz, Jesse J. (2005). A Neurofunctional Theory of Consciousness. In Cognition and the Brain:The Philosophy and Neuroscience Movement. Cambridge: Cambridge University Press, 381–96. Prinz, Jesse J. (2011). Is Attention Necessary and Sufficient for Consciousness? In Attention: Philosophical and Psychological Essays. Oxford: Oxford University Press, 174. Searle, John R. (2000). Consciousness. In Annual Review of Neuroscience 23, 557–78. doi: 10.1146/ annurev.neuro.23.1.557.

128

JAKE H. DAVIS AND EVAN THOMPSON

Slagter, Heleen A, Lutz, Antoine, Greischar, Lawrence L., Francis, Andrew D., Nieuwenhuis, Sander, Davis, James M., and Davidson, Richard J. (2007). Mental Training Affects Distribution of Limited Brain Resources. In PLOS Biology 5 (6), e138. doi: 10.1371/ journal.pbio.0050138. Thanissaro Bhikkhu (2011). Translator’s introduction to Maha-satipatthana Sutta: The Great Frames of Reference. Access to Insight. Retrieved December 16, 2011, from http://www.accesstoinsight.org/ tipitaka/dn/dn.22.0.than.html Thompson, E. (2007). Mind in Life: Biology, Phenomenology, and the Sciences of Mind. Cambridge, MA: Belknap Press. Varela, F. J., Thompson, E., and Rosch, E. (1991). The Embodied Mind: Cognitive Science and Human Experience. Cambridge, MA: MIT Press. Williams, Mark A., Morris, Adam P., McGlone, Francis, Abbott, David F., and Mattingley, Jason B. (2004). Amygdala Responses to Fearful and Happy Facial Expressions under Conditions of Binocular ­Suppression. In Journal of Neuroscience 24, 2898–904. doi: 10.1523/JNEUROSCI.4977-03.2004.

CHAPTER

8

Subjectivity, selfhood and the use of the word ‘I’ Jonardon Ganeri

Source: Mark Siderits, Evan Thompson and Dan Zahavi (eds), Self, No Self? (Oxford: ­Oxford University Press, 2010), pp. 176–192. . 1. ASANGA AND VASUBANDHU ON SELF-CONSCIOUSNESS AND CONSCIOUS ATTENTION TO ONESELF

A well-known Buddhist philosophy of mind has it that conscious experience is a s­ynthesis of five forms of activity: the processes of registering, appraising, stereotyping, readying, and consciously attending.1 These philosophers say that the last of these, conscious a­ ttention, is relative to a perceptual modality, so that, for example, consciously attending to what is being visually registered is different in kind from consciously attending to what is being registered in touch.2 They say as well that one can also consciously attend to what is going on in one’s mind.This is a mode of conscious reflecting (mano-vijñāna); it is a way of being self-aware, of consciously attending to one’s own psychological state. To these six varieties of conscious attention Asaṇga adds a seventh, which he calls ­simply manas ‘mind’/‘consciousness’, or else ‘defiled mind’ (klista-manas).3 It is, he seems to think, a distinct and more basic mode of being self-aware.4 Asaṇga argues that there must exist a non-perceptual modality of self-consciousness, which is distinctively ­associated with what is ‘mine’, as well as being the support of conscious reflection, and something that contributes to the persistence of one’s sense of self. Itself ethically neutral, it is n ­ evertheless responsible for the four vices to do with the self: [Question:] How does one know that manas in the sense of ‘defiled mind’ (klista-manas) exists? [Answer:] Without it, there could be no uncompounded ignorance, i.e. a basic ignorance not yet associated with all the diverse defilements but standing as their base. Besides, conscious 129

130

JONARDON GANERI

reflecting (manovijñāna) must also have a simultaneous support, as do the sensory consciousnesses which have such supports in their material organs. Such a simultaneous support can only be the ‘defiled mind’. Also, the very etymology of manas has to do with ‘mine’, which can be explained only by the ‘defiled mind’. Also, without it there would be no difference between the non-identifying trance and the cessation trance, for only the latter is free of defiled mind. Also, the sense of an existence of self is always existent in nonsaintly states: there must be some special consciousness to account for the persistence of this sense. The defiled mind is always defiled by the false view of self, pride of self, love of self, and ignorance (about self); but is itself ethically neutral.5

Asaṇga’s brother Vasubandhu claims that this mode of being self-aware undergoes a ‘transformation’ into what gets described as a self.6 He says that manas—‘consciousness’— is a way of being aware, associating it with the activity of ‘thinking’ (manana). It takes the store-consciousness (alaya-vijñana) as its foundation. It undergoes a transformation into something that we metaphorically call a self, but this transformation is the work of ­cognitive fabrication, and there is in fact no such thing: The metaphors of ‘self ’ and ‘items’ which develop in so many ways take place in the ­transformation of consciousness. Dependent on [the store-consciousness] there develops a consciousness called manas, ­having that as its basis, and having the nature of ‘thinking’. This transformation of consciousness is a cognitive fabrication, and as it is cognitively fabricated it does not exist.7

How are we to make sense of what is going on here? The import of the use of the terms ‘conceptual fabrication’ (vikalpa) and ‘metaphorical designation’ (upacāra), in connection with the self, is that the end-result of the transformation of pre-attentive self-­consciousness is the sort of first-person psychological judgment one would express in the words ‘I am F’. The transformation has made the self into a conceptual thought-content (vikalpa), but the expression of that thought-content uses a word, ‘I’ for example, in at most a ‘metaphorical’ sense, or at any rate some usage that is not one of genuine literal reference. (As I will point out below, upacāra is not quite metaphor, but nearer to metonymy.) Let me represent the picture schematically. The claim is that three distinct phenomena are involved in self-consciousness: 1 Conscious attention to one’s own states of mind (manovijñāna). This must have a ‘support’ (āśraya). The support is: 2 ‘Self-consciousness’ (manas)—a pre-attentive mode of being self-aware. This is subject to ‘transformation’ (pariṇāma). What it is transformed into is: 3 First-person psychological judgment—thinking ‘I am F’, for some psychological predicate F. The use of the word ‘I’ here, though, is in some sense not a genuine referring use.

SUBJECTIVITY AND SELFHOOD

131

It will help if I begin by stating the conclusions for which I want to argue. I will argue that these claims should be understood as follows. My possession of a first-person perspective, a perspective on my own mental life, has to be underwritten. What underwrites it is the fact that my mental life presents itself to me, in a primitive and pre-attentive way, as being mine.This same primitive mode of being self-aware is rendered in such a way that it seems to justify me in making assertions of the form ‘I am F’. In fact, it is never the case that assertions of such a form are true of a self. Uses of ‘I’ never literally refer to a self. I will argue that this final claim is ambiguous, and distinguish the reading Vasubandhu wishes to give it from another, in my view more promising, idea. 2.  PRE-ATTENTIVE CONSCIOUSNESS: MANAS AND MINENESS

The proposal we are examining might be expressed as the conjunction of three propositions: [1] There is a pre-attentive mode of self-awareness through which my experiences present themselves to me as mine. [2] First person psychological judgment draws upon additional conceptual resources, ones not available on the basis of [1] alone. [3] First person psychological judgments do not actually involve genuine reference to a self. Let me examine these propositions in turn. The ability, not just to have a world in view, but also to reflect upon the fact that one does, seems to be an essential part of what it means to be conscious. Sidney Shoemaker says that It is essential for a philosophical understanding of the mental that we appreciate that there is a first person perspective on it, a distinctive way mental states present themselves to the subjects whose states they are, and that an essential part of the philosophical task is to give an account of mind which makes intelligible the perspective mental subjects have on their own mental lives. (Shoemaker 1996: 157)

It is to this task that our Buddhists address themselves when they say that conscious ­attention to one’s own mental life (mano-vijñána) must have a support, which they claim is a pre-attentive mode of being self-aware (manas). I think that the point of this argument is easy enough to understand as long as we remember that it is impossible to think about one of one’s own mental states, a particular feeling of hope for example, and yet not be sure whose mental state it is. There is no question of having a first person perspective on one’s mental life, without that mental life presenting itself to one as one’s own. In a muchquoted passage, Peter Strawson says: It would make no sense to think or say:This inner experience is occurring, but is it o ­ ccurring to me? (This feeling is anger; but is it I who am feeling it?) Again, it would make no sense to think or say: I distinctively remember that inner experience occurring, but did it occur

132

JONARDON GANERI

to me? (I remember that terrible feeling of loss; but was it I who felt it?) There is nothing that one can thus encounter or recall in the field of inner experience such that there can be any question of one’s applying criteria of subject-identity to determine whether the encountered or recalled experience belongs to oneself—or to someone else. (P. F. Strawson 1966: 165)

If I cannot be mistaken about whose inner experience it is that I am experiencing, this is because no identification of a subject, and so no possibility of misidentification, is ­involved at all. What I am suggesting, then, is that our Buddhist philosophers explain the ‘immunity to error through misidentification’ (Shoemaker 1984) of s­elf-­ascriptions, by ­acknowledging that when my experience presents itself to me as my own, no ­representation of myself as a subject takes place. Asaṇga and Vasubandhu postulate, i­nstead, the existence of a primitive mode of self-awareness, a basic awareness of the contents of my inner life (my ‘store-­consciousness’) as mine. And this, in turn, is what makes it possible for me to have a first-person, rather than merely a third-person, perspective on my mental life. In Sartre’s theory of consciousness, I might note in passing, there is a proposal that is in some respects comparable. Sartre speaks of a pre-reflective self-awareness, which ‘has no need at all of a reflecting consciousness in order to be conscious of itself. It simply does not posit itself as an object’ (Sartre 1957: 45). Dan Zahavi has redescribed it as ‘an immersed non-objectifying self-aquaintance’ (Zahavi 2005: 21). Sartre argues that an infinite regress will ensue if such a mode of self-acquaintance is not acknowledged, and it is interesting to observe that we find the infinite regress argument used too by one of Vasubandhu’s immediate followers, Diṅniaga, in a defense of reflexivism (Ganeri 1999; see also the contributions by Evan Thompson and Mark Siderits to this volume). 3.  FIRST PERSON PSYCHOLOGICAL JUDGMENT

Vasubandhu speaks of a ‘transformation’ of basic self-awareness into explicit self-­ ascription, a transformation based on conceptual fabrication (vikalpa) and justifying only a ­‘metaphorical’ use (upacāra) of the language of self. What is difficult is to understand how it can be thought wrong to make the transition from being aware of oneself as being in a certain mental state to explicitly asserting that one is.Vasubandhu’s thesis is that this transition demands a new conceptual resource (one which is not in fact available). Is that thesis true? Zahavi, for one, does not see any difficulty with this transition. He says: Contrary to what some of the self-sceptics are claiming, one does not need to conceive of the self as something standing apart from or above experiences, nor does one need to conceive of the relation between the self and experience as an external relation of ownership. It is also possible to identify this pre-reflective sense of mineness with a minimal, core, sense of self. … In other words, the idea is to link an experiential sense of self to the particular first-personal givenness that characterizes our experiential life; it is this first-personal givenness that constitutes the mineness or ipseity of experience. Thus, the self is not something that stands opposed to the stream of consciousness, but is, rather, immersed in conscious life. (Zahavi 2005: 125)

SUBJECTIVITY AND SELFHOOD

133

Zahavi’s ‘minimal self ’ precisely consists in a ‘pre-reflective sense of mineness’, and it ­appears to follow that, to refer to oneself in the first person, nothing more is required than that one s experience be given ‘immediately, noninferentially and noncriterially’ (2005:  124) as mine. He also says, however, that ‘this form of egocentricity must be distinguished from any explicit I-consciousness. I am not (yet) confronted with a thematic or explicit awareness of the experience as being owned by or belonging to myself. The  mineness is not something attended to; it simply figures as a subtle background presence’ (2005: 124). So there is, after all, a transition, but it is a transition which involves only paying attention to the mineness inherent in my experience, not in the exercise of any new conceptual resource.8 I think that Vasubandhu’s response would simply be that if someone wants to use the words ‘minimal self ’ as a synonym for the manas, then nobody will object to him doing so. On the other hand, if the implication is that the ‘minimal self ’ does what a self is meant to do, then it is too minimal to count. It seems to me that the minimal self does not do one of the things that a respectable concept of self must, and that is to individuate thinkers. It is true of you and me alike that our experience is given to us with an immersed mineness.The property ‘being a thought of one’s own’ is a property like ‘being a divisor of itself ’, which is equally true of every number; the reflexive pronoun is just a place-holder. ­Zahavi says that ‘the particular first-person givenness of the experience makes it mine and ­distinguishes it for me from whatever experiences others might have’ (2005: 124). That choice of words suggests that he thinks that first-person givenness is individuative of individual selves.9 François Recanati (Recanati 2007) has an interesting account of the transition we are interested in. He draws a distinction between implicit and explicit de se thoughts, a de se thought being ‘a de re thought about oneself, that involves a particular mode of ­presentation, namely a first person mode of presentation’ (2007: 169). He continues: As Frege wrote in ‘The Thought’, ‘everyone is presented to himself in a particular and ­primitive way, in which he is presented to no one else’. I call the ‘special and primitive’ mode of presentation which occurs in first person thoughts ‘EGO’ or rather ‘EGOx’ where ‘x’ stands for the name of the person thinking the thought. (2007: 170)

Explicit de se thoughts are de se thoughts ‘the content of which involves an ­“identification component” through which the object thought about is identified as oneself ’. When a subject looks at themselves in a mirror and thinks ‘My legs are crossed’, they identify themselves under the concept EGO, and ascribe to themselves a property: I am that person whose legs are crossed. An implicit de se thought involves no such identification. Recanati says that ‘implicit de se thoughts are identification-free, and they are de se only externally: no concept EGO occurs as part of the lekton [roughly, the content]. The lekton is a personal proposition, without any constituent corresponding to the person to whom a property is ascribed’ (2007: 176). What this means is that in an implicit de se thought, one simply thinks of one’s legs as crossed, a thought that is true because it is indeed one’s own legs which are crossed. (The distinction is reflected in language: ­contrast the anaphoric construction ‘He expects that he will be late’ with the gerundival

134

JONARDON GANERI

construction ‘He expects to be late’. See also Perry 1998.) Recanati’s point is that it is precisely because no identification of a subject is involved that implicit de se thoughts are immune to error through misidentification. Recanati argues that the concept EGO involved in the notion of an explicit de se thought is itself explained by this notion of an implicit de se thought. He says: The notion of an implicit de se thought in which the self is not represented is important …, to understand the concept of self that occurs in explicit de se thoughts. Indeed, the ability to entertain implicit de se thoughts is arguably a necessary condition for anyone to evolve the concept EGO. That is so because, as suggested by Evans, Perry, and myself following them, the concept EGO is best construed as a repository for information gained in a first person way … Now a piece of information is gained in the first person way if and only if it is the content of an implicit de se thought. It follows that the first step in an elucidation of the concept of self is a correct analysis of the functioning of implicit de se thoughts. (2007: 177)

The notion of manas, as introduced by Asaṇga and developed by Vasubandhu, seems to me to be rather close in spirit to Recanati’s notion of implicit de se thought. It certainly contains no representation of the self, and it has as its ‘foundation’ (ālambana) all the information contained in the store-consciousness. It is a first personal way of thinking about (manana) that information. The description ‘repository for information gained in a first person way’ might seem like a very good description of the joint contribution of manas and store-consciousness.10 Moreover, as we have seen, Vasubandhu speaks of a ‘transformation’ of manas into a concept of self, and that echoes with the claim Recanati is making about the evolution of the concept EGO. Finally, Recanati claims that the ability to entertain implicit de se thoughts is a necessary—though not necessarily a sufficient—condition for the evolution of the concept EGO, and this too is something Asaṇga and Vasubandhu are keen to stress; indeed, it is the primary motivation for introducing the notion of manas in the first place. For, as Asaṇga says in the Yogācārabhūmi, ‘That [manas] has the mode of taking the store-consciousness as its object and conceiving it as “I am [this]” (asmiti) and “[this is] I” (aham iti)’ (quoted in Waldron 2002: 42). Note that the concept EGO, as a repository of information gained in a first person way, does individuate thinkers, for no two such repositories will be alike. This concept of self is much less ‘minimal’ than Zahavi’s. So where is the difference? Both agree that there are explicit de se thoughts involving the concept EGO, but while Recanati thinks that the truth conditions of such thoughts are such that they are often true, Vasubandhu thinks that they are always false. He thinks this because he thinks that the concept EGO, which has certainly evolved in the manner described, is an empty concept, like the concept PHLOGISTON or the concept ­PEGASUS. Our Buddhists think that the evolution of the concept EGO brings with it all manner of moral defilements, and one form of justification for that claim is that the concept rests in this way on an error. Sthiramati’s comment on the first of the 30 Verses bears the point out: he says that the concept of self presents only an apparent (nirbhāsa) referent, just as the perception of someone with an eye-disease presents only apparent hairs and circles. It is ‘metaphorically designated’ (upacaryate) because it is said to be there when it is

135

SUBJECTIVITY AND SELFHOOD

not, as if one were to use the word ‘cow when there is an ox. Sthiramati’s example, incidentally, shows that the notion of upacāra is much closer to metonymy than to metaphor, as traditionally understood.11 With P. F. Strawson’s assertion that ‘no use whatever of any criteria of personal identity is required to justify [a person’s] use of the pronoun “I” to refer to the subject’ (P. F. Strawson 1966: 165),Vasubandhu would appear to dissent. For his view seems to be that the use of the pronoun ‘I’ never refers to a subject of experience. Strawson’s point is that we don’t need any extra conceptual resource in order to make explicit self-references, and in particular we don’t need a criterion of identity. The pronoun ‘I’ is not a term which we can correctly use only if we have successfully identified its referent, because if it were then there would be the possibility of error through misidentification relative to it. Strawson infers that ‘I’ refers to its subject without there being a criterion of identity. Perhaps what Vasubandhu would do would be to agree with this argument but contrapose it. His point would then be that all genuine reference involves the identification of a referent, and given that there is no question of such an identification in the case of the first person, the pronoun ‘I’ cannot be a genuine referring term. In saying that it is instead a ‘metaphor’ (upacāra), there seems to be a gesture at the possibility of a different, non-referential, account of its use. To say ‘I feel hopeful’ is, it might to be thought, to speak non-referentially of the existence of a hope which presents itself pre-attentively as mine. I will argue that, although this is not actually Vasubandhu’s strategy, it is nevertheless a viable one. 4.  TWO USES OF ‘I’

Can one hear an echo of the Buddhist idea in the following remark? One of the most misleading representational techniques in our language is the use of the word ‘I’, particularly when it is used in representing immediate experience, as in ‘I can see a red patch’. It would be instructive to replace this way of speaking by another in which immediate experience would be represented without using the personal pronoun. (Wittgenstein 1975: 88)

In another place, Wittgenstein speaks of ‘two different cases in the use of the word “I” (or “my”)’, the ‘use as object’ and the ‘use as subject’ (1960: 66–7). The ‘use as object’ is the use to which it is put when we refer to ourselves as human beings, embodied entities in a public space, the use it has when, for example, one person says to another, ‘I am just going to the shops to get the paper’, or ‘I have twisted my ankle’. Having distinguished between these two uses, one strategy would be to identify one of these uses as the primary use, and analyze the other use as being in some way derivative upon the first. Indeed, it is more in keeping with Indian theory about non-literal language to speak in this way of primary and derivative uses, rather than in terms of a distinction between literal and metaphorical use. The derivative use is metonymic rather than metaphorical: that is, the term is used to refer to something else, which stands in some relation to the primary referent. Among the contemporaries of Asaṇga and Vasubandhu are Vaiśeṣika philosophers, who argue that the primary use of ‘I’ is to refer to a self (atman), and that its use to refer to oneself as an

136

JONARDON GANERI

embodied being, in statements like ‘I am fat’, is an act of derivative reference, that is, reference to something which stands in an ‘is the body of ’ relation to the primary reference.12 A variant on this approach is advocated by Galen Strawson. Strawson argues that the two uses are both genuinely referential, and neither is primary, in short, that ‘I’ is not univocal. One use is to refer to what he describes as a ‘thin subject’, a thin subject being ‘an inner thing of some sort that does not and cannot exist at any given time unless it is having experience at that time’ (2008: 156; cf. 2009: 331–8). The other use is to refer to ‘the human being considered as a whole’: Are we thin subjects? In one respect, of course, we are thick subjects, human beings considered as a whole. In this respect we are, in being subjects, things that can yawn and scratch. In another respect, though, we are in being subjects of experience no more whole human beings than hands or hearts: we are—literally—inner things, thin subjects, no more things that can yawn or scratch than eyebrows or thoughts … —But ‘What then am I?’ Am I two different sort of things, a thin subject and a thick subject? This is ridiculous … My answer is that ‘I’ is not univocal. We move naturally between conceiving of ourselves primarily as a human being and primarily as some sort of inner subject (we do not of course naturally conceive of ourselves as a thin subject). Sometimes we mean to refer to the one, sometimes to the other; sometimes our semantic intention hovers between both, sometimes it embraces both. (G. Strawson 2008: 157–8)

Elsewhere, G. Strawson (2007) is clear that the relation between the two uses is one of metonymy, and indeed that the underlying relation is one of whole to part: I think that we do at different times successfully use ‘I’ to refer to different things, to human beings considered as a whole and to selves. In this respect the word ‘I’ is like the word ‘castle’. Sometimes ‘castle’ is used to refer to the castle proper, sometimes it is used to refer to the ensemble of the castle and the grounds and associated buildings located within the perimeter wall, sometimes it can be taken either way. The same goes for ‘I’, but ‘I’ is perhaps even more flexible, for it can sometimes be taken to refer to both the self and the whole human being, indifferently. Our thought (our semantic intention) is often unspecific as between the two. (G. Strawson 2007: 543)

Vasubandhu, although he does not say so here, would perhaps be content to endorse as ‘conventional’ the use of the first person in statements like ‘I am going to the shops’, a use governed by the token-reflexive rule that ‘I’ refers to the speaker. When ‘I’ is used in the expression of first person psychological judgment, however, his claim is that the reference to an inner self fails, that this use erroneously imports a subject-predicate model and imposes it upon one’s inner experience. In other words, his view of this use of ‘I’ is that there is a combination of metonymy and error-theory. When ‘I’ is used metonymically to refer to the inner subject, something goes wrong, and what goes wrong is that there is nothing at the far end of the metonymic relationship for it to refer to.This, indeed, is a view which G. Strawson (2007: 543) considers and rejects: If it turns out that the best thing to say about selves is that there are no such things, then the best thing to say about ‘I’ may well be that it is univocal after all, and that the apparent ­doubleness of reference of ‘I’ is just the echo in language of a metaphysical illusion.

SUBJECTIVITY AND SELFHOOD

137

If this is right, then ‘I’ is not in fact used to refer to selves as distinct from human beings even when its users intend to be making some such reference and believe that they are doing so. On this view, the semantic intentions of ‘I’-users sometimes incorporate a mistake about how things are. I disagree.

According to the interpretation of Vasubandhu we have reached, then, ‘I’ does not function as an expression of genuine reference, but is rather one of disingenuous reference: it is a referring expression without a referent, its use creating the false impression that there is one.That is, I suggest, the best way to understand Vasubandhu’s claim that it is a ‘metaphor’. There is another possibility, though. It might be the case that the mistake which ­‘I’-users make is not a metaphysical one, but rather a mistake about the semantic role of ‘I’ itself. Perhaps what goes wrong in the use of ‘I’ in representing immediate experience is that speakers take themselves to be making a referential use of an expression, and in doing so mistake its true logical role. It would be as if someone thought that ‘perhaps’ is a referring expression, and then imagined that there must be something in the world that it designates. This view is attractive, because it does not make the argument hinge on a prior metaphysical claim about the reality of selves, which then stands in need of further, extra-linguistic, justification. Rather, once we are clear about the logical role of ‘I’, we see that looking for a referent is as misguided as looking for the referent of ‘perhaps’. And indeed, this is what Wittgenstein seems to say about the ‘use as subject’ of the word ‘I’, its use in a sentence such as ‘I have a pain’. For Wittgenstein, denying that ‘I’, when used as in first person psychological judgments, is a referring expression, is the only way to explain the phenomenon of immunity to error through misidentification: [T]here is no question of recognizing a person when I say I have a toothache.To ask ‘are you sure that it’s you who have pains?’would be nonsensical … And now this way of stating our idea suggests itself: that it is as impossible that in making the statement ‘I have toothache’ I should have mistaken another person for myself, as it is to moan with pain by mistake, having mistaken someone else for me. … To say, ‘I have pain’ is no more a statement about a particular person than moaning is. (1960: 66–7)

The suggestion that there is a non-referential account of the use of ‘I’ was developed in one direction by Anscombe (Anscombe 1975). Anscombe, however, does not distinguish two uses, and argues that the first person does not refer, even in cases like ‘I have a broken arm’. Other writers have tried, following the lead of P. F. Strawson, to argue that immunity of error does not commit one to a non-referential account of ‘I’, and indeed to reconcile immunity with the idea that ‘I’ refers univocally to the embodied human being (see for example Campbell 2004; McDowell 2009). In an insightful remark about Anscombe, Campbell suggests that the best way to understand her position is as claiming that the patterns of use involving the first person do not require justification in a semantic foundation: An alternative reaction would, of course, be to say that we ought to abandon the search for a semantic foundation for our use of the first person. There are only the patterns of use, and no explanation to be given of them.This was essentially G. E. M. Anscombe’s position in her

138

JONARDON GANERI

famous paper, ‘The First Person’, in which she claimed that the first person does not refer. This claim is generally rejected, simply because philosophers have thought that when there is a use of the first person, there is, after all, always someone around who can be brought forward as the referent. But this is an extremely superficial response to Anscombe’s point. Her claim is best understood as making the point that the ascription of reference to the first person is empty or idle; it does no explanatory work. (Campbell 2004: 18)

I have argued at length elsewhere that something along these lines is just the move made by the Mādhyamika philosopher Candrakīrti (Ganeri 2007, ch. 7). His position, I have claimed there, is that we can give a fully explanatory use-theoretic account of the role of the first person in performances of self-appropriation, an account in which it is otiose to assign a reference. Asaṇga and Vasubandhu, on the other hand, say that in the movement from a pre-attentive self-awareness to an explicit use of ‘I’, a transformation of some sort is involved, one which involves conceptual work (vikalpa), and that the use of ‘I’ is metaphorical or metonymic. Their view is that the use of ‘I’ is indeed referential, and that the use as object and the use as subject are to be understood as making reference to, on the one hand, the human being, and on the other, an inner subject of experience, this second use being derivative from the first. There is, however, no subject of experience, and so the subjective use of the first person is an error. 5. CONCLUSION

I have argued that the new theory of Asaṇga and Vasubandhu consists in an account of the first-person perspective.Their further claim that the first person itself, the word ‘I’, is used ‘metaphorically’ in reporting the contents of the first-person perspective, rests on a prior commitment to the non-existence of an inner subject of experience. Only this permits them to claim that its use is one of what I have called ‘disingenuous reference’. I have distinguished a different strategy, which is to begin with the observation that such reports are immune to error through misidentification, and to argue that it follows that in the proper account of the use in first person psychological judgments of the word ‘I’, the assignment of a referent is explanatorily superfluous. Some of the parts of this strategy are to be seen at work in various thinkers. Anscombe argues that ‘I’ is not a referring expression, but does not distinguish the two uses of the term. Candrakīrti explains the use of ‘I’ in a way that makes the assignment of a referent superfluous, through an appeal to the thought that its role has to do with self-appropriation (upā-dāna), but he does not base this claim on the phenomenon of immunity. The full strategy being defended here emerges only as a ‘fusion’ of components drawn from our various sources.13 NOTES 1. The five so-called skandhas or ‘ingredients’ that combine into individual thoughts or ­experiences: rūpa, vedanā, samjñā, samskāra, and vijñāna. For details, see Ganeri (forthcoming). This flexible doctrine is transformed in various ways by later Buddhist thinkers, beginning with Asaṇga and Vasubandhu in the c. 4th ce.

SUBJECTIVITY AND SELFHOOD

139

2. Ṣaḍvijñānadhātavaś cakṣurādyāśrayā rūpādyālambanāvijñāptayaḥ (Vasubandhu, Pañcaskandhaka 135). 3. Thereby adding a new sense for this term, which in its everyday use in Abhidharma is simply synonymous with conscious attention (citta, vijñāna). Kramer (2008) translates kliṣṭamanas as ‘notion of I’, and observes that the incorporation of this new concept represents a ­modification in the traditional system of the five skandhas, a modification that is evident in Sthiramati’s commentary on Vasubandhu’s Pañcaskandhaka. She says: ‘In particular the function of vijñāna-skandha—the original role of which was actual perception—was w ­ idened through the inclusión of subliminal forms of mind, like the ‘store mind’ (dlaya-vijñdna) and the “notion of I” (kliṣṭa-manas). The strong emphasis placed by Sthiramati on vijñāna is ­evident, for instance, when he states that ordinary people—those who have not perceived reality—regard the vijñāna as the self (ātman), whereas they view the other four skandhas as “mine” (ātmīya)’ (2008: 155). She adds that ‘[i]nterestingly, Sthiramati also mentions ­alternative concepts of the self, for example that of the Sarnkhya tradition. According to his understanding, the Sāṃkhyas only regard rūpa-skandha [matter] as ātmīya, and all the other four skandhas as ātman. He thus claims that for the Sāṃkhyas the self is not only identical to vijñāna but also consists of the other factors accompanying the mind (caitasika)’ (2008: 155). Galloway (1980) translates kliṣṭa-manas as ‘passional consciousness’, and derives interesting information about the notion from Gunaprabha’s commentary on the Pañcaskandhaka. See below. Dreyfus and Thomson (2007: 112) translate kliṣṭa-manas as ‘afflictive mentation’, and comment that ‘[t]his is the inborn sense of self that arises from the apprehension of the store-consciousness as being a self. From a Buddhist point of view, however, this sense of self is fundamentally mistaken. It is a mental imposition of unity where there is in fact only the arising of a multiplicity of interrelated physical and mental events’. 4. See Galloway (1978) for a detailed argument that as a Yogācāra technical term, manas should be translated as ‘consciousness’ rather than neutrally through its cognate in English, ‘mind’. 5. Mahāyanāsaṃgraha 1.7; trans. Anacker in Potter (2003) from the extant Chinese and Tibetan translations. 6. Galloway (1980: 18) reports from Guṇaprabhā’s commentary on Vasubandhu’s Pañcaskandhaka: [Vasubandhu:] ‘In reality, the consciousness (manas) has the storehouse perception for its phenomenon.’ [Gunaprabhā:] This means that it phenomenalizes [sees] the storehouse perception as a self. [Vasubandhu:] ‘It is that which is associated with the constant delusion of self (ātmamoha), view of self (ātmadṛṣṭi), egoism of self (ātmamāna), and lust for self (ātmarāga), and so on.’ [Gunaprabhā:] It is explained as operating always, and arises as good (kuśala), bad (akuśala), and indifferent. His saying ‘It is of one class’ means that it has a passionate (kliṣṭa) nature. ‘It is continually produced means that it is momentary. 7. Triṃśikākārikā: ātmadharmopacāro hi vividho ya pravartate | vijñānapariṇāme ’sau || Tvk la-c || tasya vyāvṛtirarhatve tadāśritya pravartate |tadālambaṃ manonāma vijñānaṃ mananātmakam || Tvk 5 | |vijñāna-pariṇāmo ’yaṃ vikalpo yadvikalpyate | tena tannāsti || Tvk 17a-c ||. The translation is from Anacker (1984), slightly modified. 8. Indeed, in Zahavi (1999), Zahavi makes it a ‘minimal demand to any proper theory of self-awareness’ that it ‘be able to explain the peculiar features characterising the subject-use of “I”; that is, no matter how complex or differentiated the structure of self-awareness is ­ultimately shown to be, if the account given is unable to preserve the difference between the first-person and third-person perspectives, unable to capture its referential uniqueness, it has failed as an explanation of self-awareness’ (1999: 13). 9. Joel Krueger (this volume) seems to share my reservation about the selfhood of the ‘minimal self ’. To put the point in an Indian vocabulary, the ‘minimal self is somewhat akin to the impersonal Advaitic ātman, present equally in all. Zahavi sometimes, however, appeals to

140

10.

11.

12. 13.

JONARDON GANERI

an embodiment criterion, rather than to first-person givenness per se, as what individuates distinct minimal selves, and that would certainly adequately distinguish the notion from the Advaita conception. The history of the Yogācāra concept of the store-consciousness is rather complicated. ­Originally conceived merely as a vehicle for the perpetuation of mental forces when the normal six types of awareness are absent, it was a technical solution to what would otherwise be a difficulty in the Yogācāra theory of individual persistence. Dreyfus and Thompson (2007: 112) say of it that ‘[t]his continuously present subliminal consciousness is posited by some of the Yogācāra thinkers to provide a sense of continuity in the person over time. It is the repository of all the basic habits, tendencies and propensities (including those that persist from one life to the next) accumulated by the individual’. Reaffirming this description in his article in this volume, Dreyfus offers the translation ‘basic consciousness’, and discusses an illuminating range of associations and resonances. The detailed studies by Lambert Schmithausen (1987) and Hartmut Buescher (2008) have revealed much greater complexity in the use of the notion in early Yogācāra than scholars had previously acknowledged. Schmithausen comments that ‘it may well be that ālayavijñāna was, initially, conceived as a kind of ‘gap-bridger’, but hardly in such a way that its occurrence in ordinary states had been denied’ (1987: §2.13.6). Items in the store do not themselves carry a feeling of mineness, but ground the feeling of mineness which attaches itself to the stream’s conscious self-attention. They comprise a sort of database for the mind, information which can be drawn upon in the activity of bringing the states of the stream into conscious attention. tam ātmādinirbhāsam rūpādinirbhāsam ca tasmād vikalpād bahirbhūtam ivopādāyātmādyüpacāro rūapādid-haropacāraś cānādikālikaḥ pravartate vināpi bāhyenātmanā dharmaiś ca | tadyathā ­taimirikasya keśondükādyupa-cāra iti | yac ca yatra nāsti tat tatropacaryate | tad yathā bāhike gauh | (Sthiramati,Triṃśikāś-vijñapti-bhāṣya; see Buescher 2007). See the discussion under Vaiśeṣika-sūtra 3.2.9–14. In his contribution to this volume, Matthew MacKenzie argues that the performativist model which I find in Candrakῑrti needs to be supplemented rather with embodied and enactivist elements. Such a move “fuses” the Buddhist theory with ideas drawn from recent phenomenological literature, a theme of many of the contributions to this volume. I have tried instead to ‘fuse’ the Buddhist theory with elements taken from the recent analytical tradition, and the existence of both such possibilities suggests to me that Indian theory might well serve to create the intellectual space for a rapprochment between those hitherto separated strands of Western thought. REFERENCES

Anacker, S. (1984), Seven Works of Vasubandhu, the Buddhist Psychological Doctor, Religions of Asia Series (Delhi: Motilal Banarsidass). Anscombe, G. E. M. (1975), ‘The first person’, in Samuel Guttenplan (ed.), Mind and Language (Oxford: Clarendon Press). Buescher, H. (2007), Sthiramati’s Triṃśikāvijñaptibhāṣya: Critical Editions of the Sanskrit text and its ­Tibetan Translation, Sitzungsberichte / Osterreichische Akademie der Wissenschaften, Philosophisch-­ Historische Klasse (Wien:Verlag der Österreichischen Akademie der Wissenachaften). ——— (2008), The Inception of Yogācāra-Vijñānavāda (Wien: Verlag der Österreichischen Akademie der Wissenachaften). Campbell, J. (2004), ‘What is it to Know what “I” Refers to?’ The Monist 87.2: 206–18. Dreyfus, G. and Thompson, E. (2007), ‘Asian Perspectives: Indian theories of mind’, in Morris Moscovitch, Evan Thompson, and Philip David Zelazo (eds.), The Cambridge Handbook of Consciousness (Cambridge: Cambridge University Press).

SUBJECTIVITY AND SELFHOOD

141

Galloway, B. (1978), ‘Vijñāna, saṃjñā and manas’, The Middle Way 53.2: 72–5. ——— (1980), ‘A Yogācāra Analysis of the Mind, Based on the Vijñāna Section of Vasubandhu’s ­Pañcaskhandaprakarana with Gumprabha’s Commentary’, Journal of the International Association for ­Buddhist Studies 3.2: 7–20. Ganeri, J. (1999), ‘Self-Intimation, Memory and Personal Identity’, Journal of Indian Philosophy 27.5: 469–83. ——— (2007), The Concealed Art of the Soul: Theories of Self and Practices of Truth in Indian Ethics and ­Epistemology (Oxford: Clarendon Press). ——— (forthcoming), Mind’s Own Nature:The Reconciliation of Naturalism with The first person perspective, ch.7. Kramer, J. (2008), ‘On Sthiramati’s Pañcaskandhakavibhāsā: a Preliminary Survey’, Nagoya Studies in Indian Culture and Buddhism: Saṃbhāsā 27: 149–72. McDowell, J. (2009), ‘Referring to Oneself ’, in The Engaged Intellect (Cambridge, MA: Harvard ­University Press). Perry, J. (1998), ‘Myself and I’, in Marcelo Stamm (ed.), Philosophie in Synthetischer Absicht (Stuttgart: Klett-Cotta). Potter, K. H. (2003), Büddhist Philosophy from 350 to 600 A.D., Encyclopedia of Indian Philosophies, vol. 9 (Delhi: Motilal Banarsidass Publishers). Recanati, F. (2007), Perspectival Thought: A Plea for Moderate Relativism (New York: Oxford University Press). Sartre, J. (1957), The Transcendence of the Ego: An Existentialist Theory of Consciousness (New York: Noonday Press). Schmithausen, L. (1987), Ālayavijñaña: On the Origin and the Early Development of a Central Concept of Yogācāra Philosophy, Studia Philologica Buddhica, 2 v. (Tokyo:International Institute for Buddhist Studies). Shoemaker, S. (1984), ‘Self-Reference and Self-Awareness’, in Identity, Cause, and Mind (Cambridge: Cambridge University Press). ——— (1996), The First-Person Perspective and Other Essays (Cambridge: Cambridge University Press). Strawson, G. (2007), ‘Selves’, in Brian P. McLaughlin and Ansgar Beckermann (eds.), The Oxford ­Handbook of Philosophy of Mind (Oxford: Oxford University Press). ——— (2008), Real Materialism and Other Essays (Oxford: Clarendon Press). ——— (2009), Selves: An Essay in Revisionary Metaphysics (Oxford: Clarendon Press). Strawson, P. F. (1966), The Bounds of Sense:An Essay on Kant’s ‘Critique of Pure Reason’ (London: Methuen). Waldron, W. S. (2002), ‘Buddhist Steps to an Ecology of Mind: Thinking about “Thoughts Without a Thinker”’, Eastern Buddhist 34.1: 1–52. Wittgenstein, L. (1960), Preliminary Studies for the ‘Philosophical Investigations’, Generally Known as the Blue and Brown Books (Oxford: Blackwell). ——— (1975), Philosophical Remarks (Chicago: University of Chicago Press). Zahavi, D. (1999), Self-Awareness and Alterity:A Phenomenological Investigation (Evanston, IL: N ­ orthwestern University Press). ——— (2005), Subjectivity and Selfhood: Investigating the First-Person Perspective (Cambridge, MA: MIT Press).

P A R T

IV SELF

CHAPTER

9

The self as a dynamic constant Ra ¯makan..tha’s middle ground between a Naiya ¯yika eternal self-substance and a Buddhist stream of consciousness-moments Alex Watson Source: Journal of Indian Philosophy, 2014, 42, 173–193. INTRODUCTION

The present article gives an account of Bhaṭṭa Rāmakaṇṭha’s (950–1000) contribution to the Buddhist–Brāhmaṇical ātman debate, by depicting him as carving out some middle ground between the Brāhmaṇical and the Buddhist views. The particular Brāhmaṇical view concentrated on is Nyāya, and ‘Buddhism’ is used in this article to refer specifically to Dharmakīrtian Buddhism with its doctrine of momentariness (kṣaṇikavāda).1 The first section analyzes what was at stake in the debate between Nyāya and Buddhism. The ­second section distinguishes Rāmakaṇṭha’s position from that of Nyāya, by giving the two classical Naiyāyika arguments for the existence of the self, and Rāmakaṇṭha’s response to them.The third section positions Rāmakaṇṭha between Nyāya and Buddhism.The fourth section elaborates the differences between Rāmakaṇṭha’s position and Buddhism. Bhaṭṭa Rāmakaṇṭha was the most prolific and influential of Śaiva Siddhānta writers in the early period of that tradition. This ‘early period’ I take to end in the twelfth century, from which time onwards Śaiva Siddhānta survived only in the Tamil-­speaking South, where it was transformed under the influence of devotionalism and Advaita Vedānta. Bhaṭṭa ­Rāmakaṇṭha, like most of the early Śaiva Siddhānta exegetes, was K ­ ashmirian; he was an older contemporary of Abhinavagupta. This article is one out of five in the present volume concerned with Śaiva philosophy and its ‘Dialogues and Debates’. All of the others apart from this are concerned not with the dualistic tradition of Śaiva Siddhānta, but with the non-dualistic tradition of Somānanda, Utpaladeva and Abhinavagupta. Thus one can see that p­ hilosophical debate in early Śaiva Siddhānta is even less studied than philosophical debate in 145

146

ALEX WATSON

non-dualistic Śaivism. The authors of those four articles, Lyne Bansat-Boudon, John Nemec, Isabelle Ratié and Raffaele Torella, have all contributed, elsewhere too, to the study of the latter, as have others.2 The number of scholars working on the former is smaller.3 ¯YA VERSUS BUDDHISM 1. NYA

To appreciate how Rāmakaṇṭha dealt with the question of selfhood differently from ­earlier Indian philosophers, we will first observe the character of the debate before his time. I identify three ways in which the self was conceived of by the Brāhmaṇical schools, in particular the Naiyāyikas, and in each case show how the Buddhists denied such a conception. 1.1  Self as unitary essence

The Naiyāyikas conceived of the self as the unitary essence of a person. It is unitary in the sense that it is one thing over time: it endures without ceasing to exist and without its nature changing in any way. For the Buddhists we have no unchanging essence: we are something different in every moment. In this moment I am an association of particular mental and bodily states, and in the next moment I am a different association of mental and bodily states. By the time of the second moment, the first mental and bodily states have ceased to exist; and there is nothing that continues to exist from the first moment to the second. This means that as a person moves across a room, it is inaccurate to speak of movement; rather than there being one thing that moves, there is a plurality of things arising in very quick succession, in neighbouring locations.4 It is like a film of a person projected on a screen, which actually consists of a plurality of separate frames, but each one follows the previous so rapidly, and resembles it so closely, that it produces the illusion of one ­continuous person. This is the Buddhist doctrine of momentariness; it encourages us to see ourselves not as unitary and permanent, but plural and momentary.5 Figure 9.1 depicts the ­contrast ­between the Brāhmaṇical notion of an enduring, unchanging self (represented by a line) and the Buddhist idea that what we are in one moment is not what we are in next (­illustrated by circles that are distinct but touch each other, as Buddhist moments are ­distinct but temporally contiguous). Furthermore, even at one point of time, for Buddhism, we are not one thing but an association of five: a bodily state and four mental states.6 See Figure 9.2.

Nyāya

Figure 9.1 

Buddhism

147

THE SELF AS A DYNAMIC CONSTANT Nyāya

Buddhism

Figure 9.2 

Thus the Brāhmaṇical self, with its permanently unchanging essence, dissolves in ­Buddhism into a diachronic and synchronic plurality. The self, since it endures ­ permanently, beyond death, is what explains reincarnation for the Brāhmaṇical schools. In other words, it is that which means we continue to be the same thing when we have a different body, in a different incarnation, or no body, between incarnations. How then can the Buddhists, in whose teachings reincarnation occupies an important place, explain the process? During life, each moment of consciousness (which is one of the four kinds of ­mental constituents of a person) is linked to the next moment of consciousness in that it causes it to arise. The same goes for the other three kinds of mental constituent, and the p­ hysical constituent. The way it works at death is similar to the way it works during life: the last ­moment of consciousness before death gives rise to a new consciousness in the first moment after death. The same goes for the other three kinds of mental constituent. But whereas during life these four mental constituents were always associated with a ­momentary configuration of the body, at death the four can separate from the bodily constituent and can reproduce themselves sequentially until such a time as they become associated with a new body, a new embryo. See Figure 9.3, in which the vertical lines represent the point of death and the point of the beginning of a new life.To the left of the first vertical line, at the bottom, is the body of the present life; to the right of the second vertical line, at the bottom, is the body of the next life. So both sides in this debate are dualists, in that for both there is a non-physical part of us that exists beyond the body and senses and is not brought to an end by death. Only the Cārvākas denied that. But the non-physical part was conceived of very ­differently: by one side as eternally unchanging and by the other as momentary (and as fourfold even in one moment). Nyāya

Figure 9.3 

Buddhism

148

ALEX WATSON

1.2  Self as substance

The Naiyāyikas and Vaiśeṣikas distinguished substances (dravyas) from qualities (guṇas), the former being property-possessors (dharmins) and the latter properties (dharmas). A thing, such as a pot or a mango, is a property-possessor, and it has five qualities—taste, smell, colour etc.—corresponding to our five senses. The thing was regarded as a separate ontological entity from its qualities, as indicated by our use of language when we say, ‘the smell of the mango’, implying that the mango is something that exists over and above its smell. Nevertheless a quality is inextricably linked to a substance. It cannot exist without one. We do not find a colour existing alone in mid-air. There must be some substantial object to which it belongs, some substrate (āśraya) that locates it. The Naiyāyikas and Vaiśeṣikas use this principle to argue for the existence of the self. Just as colours or smells presuppose substances to which they belong, so consciousness presupposes a substance to which it belongs, that substance being the self.7 The Buddhists denied the existence of a self conceived of as the substance to which consciousness belongs.This was part of a more general denial of the existence of s­ubstances over and above qualities.8 Whereas to a Naiyāyika a mango is one thing with five qualities, to a Buddhist it is five things occurring together, i.e. at the same time and in the same place.9 This is illustrated in Figure 9.4, taking the large circle to refer to a mango, and the small circles to refer to its smell, taste, colour, etc. Or the large circle can equally well ­represent a self, in which case the diagram illustrates that for Nyāya consciousness etc. belong to a self, whereas for Buddhism consciousness and the other four constituents (skandha) of a person exist together, as part of a conglomeration, without belonging to anything else.10 By disputing that colours, smells etc. belong to a substance, Buddhism calls into ­question the very concept of a quality (guṇatva). In as much as the concept itself implies the concept of a substance, being one incomplete half of a substance-quality distinction, Buddhism does away with talk of qualities.11 It refers to the things that are termed qualities by Nyāya as simply parts (deśa) of a conglomeration (samudāya, samūha, saṅghāta).12 Furthermore, whenever we use expressions that might seem to describe parts as belonging to a whole, Nyāya

Figure 9.4 

Buddhism

149

THE SELF AS A DYNAMIC CONSTANT

such as ‘the trees of the forest’ or ‘the colour of the mango’, the term for the whole should not be taken to imply the existence of anything other than the sum of the parts. It refers not to a unity, but to a conglomeration of elements: a particular group of trees in the first case, a particular group of five sensible properties in the second. 1.3  Self as agent

The Naiyāyikas and Vaiśeṣikas also conceived of the self as the agent of physical actions (kartṛ), and the agent/subject of cognitions (jñātṛ). (In Figure 9.5 the continuous line on the left to which all of the circles are attached represents the agent; the circles represent either physical actions or cognitions.) On the one hand it is that which, through the impulse of its will/effort (prayatna), initiates all of our physical actions. On the other it is the perceiver of our perceptions, the thinker of our thoughts, etc.The perception of a pot, say, lasts just for an instant but its perceiver outlives that perception and is the perceiver of the next and subsequent perceptions. For Buddhists that which brings about a physical action is just that which causes it, which for them is the intention that occurred in the stream of consciousness in the moment preceding the action. The Vaiśeṣikas had compared the self as instigator of bodily movements to a puppeteer instigating the bodily movements of a puppet below.13 Such a notion of an agent standing above the sequence of mental and physical actions is precisely what is denied by the Buddhists. The intention that brings about my present action of touching the computer keyboard was itself caused by the previous moment of consciousness, and so on. There is no part of a person standing outside this chain of mental and physical events; each event is conditioned by the previous ones and brings the next one into existence, and there is nothing over and above this causal chain that is ­unconditioned. So Buddhism, by bringing the agent down from its lofty position, dividing it up into d­ iscrete moments of intention and dispersing them into the psycho-physical stream, r­ eplaces a two-tier model with a one-tier model.14 How did the Buddhists dispute the Naiyāyika and Vaiśeṣika notion of the self as the agent/subject of cognition? For Buddhism the agent of a cognition (jñātṛ/grāhaka) is s­imply the cognition itself (jñāna/grahaṇa). That which is conscious of a pot is ­consciousness at that particular moment. So if two consecutive cognitions occur to me, verbalizable as ‘I see a pot’ and ‘I see a cloth’, the two occurrences of ‘I’ have two different referents: two different instances of consciousness. No two physical actions share a common agent, because each has its own separate prior intention; no two cognitions, or mental actions, share a common agent, because Nyāya

Figure 9.5 

Buddhism

150

ALEX WATSON

they are all their own agent. In both cases the agent of the first action exhausted itself with that action and then ceased to exist, so it is not available to be the agent of the second action. In the case of cognitions, just as in the case of physical actions, we have a two-tier model replaced by a one-tier model.The subject of consciousness is dissolved into ­consciousness itself. The existence of a thinker separate from thoughts, or a perceiver separate from ­perceptions, is denied. Neither cognitions, nor physical movements, are seen as actions that require an ontologically distinct actor, but rather simply as events that occur in a particular psycho-physical stream. The absence of a continuous agent was unacceptable to the Naiyāyikas and the other Brāhmaṇical schools, because of its corollary that the thing that performs an action is not the same as the thing that experiences the fruit of that action subsequently. This seemed unjust: why should one thing experience the positive or negative consequences of an action performed by something else? For Buddhism, that is just the way it is. A planted seed turns into a shoot, a stalk, leaves, a flower and then a fruit. No one would say that it is unjust for the fruit to accrue to the flower and not the seed. It is in the nature of things that the seed has turned into something different by the time the fruit comes along. Similarly an action is performed and by the time the fruit of that action occurs, the stream that performed it has become something different.15 The result does not occur in a different stream however. That would be unjust. Each of the three Buddhist positions that we have just observed results from applying more general Buddhist principles to the specific case of the self. The denial of a permanent, unchanging self is a special case of the conception of the momentariness of everything. The denial of the self as a substance possessing qualities is a special case of the denial of substances over and above qualities.The denial of the self as autonomous agent is a special case of the general position that nothing stands outside the chains of causes and effects that make up the world. ¯ MAKANT HA AGAINST NYA ¯ YA 2. RA ..

Having observed the two poles that occupy, in my view, two extremes in this debate, we can now begin to locate Rāmakaṇṭha between them. First we will see how he differentiates himself from the Naiyāyika extreme. He does not think that Naiyāyika arguments are capable of establishing a self. He counters them by agreeing with Buddhist arguments against them. We will look at his responses to the Naiyāyika argument for the self as substance to which qualities belong, and the Naiyāyika argument for the self as agent of cognition. 2.1  Self as substance

Rāmakaṇṭha is convinced by the Buddhist denial of substances over and above qualities. This denial focuses on the evidence of our experience. Addressing the mango example, Rāmakaṇṭha draws on Buddhist argumentation, and points out that all we can e­ xperience

THE SELF AS A DYNAMIC CONSTANT

151

there are five separate qualities.16 Through our eyes we can see a visible form (rūpa), consisting of a shape and colour, through our faculty of taste we can experience a taste, and so on. But we do not experience some further possessor of those qualities, lurking behind them. None of our faculties apprehends such a thing. Neither could they in principle, given that our eyes can only sense form, our noses smells, etc. The case of the self is analogous. We experience states of consciousness (perceptions, desires, thoughts, etc.), so we can grant reality to them; but we do not perceive some substrate of those states of consciousness, underpinning them, or some substance uniting them, in which they inhere. Such a thing is an ontological extravagance that results from going beyond the evidence of our experience and multiplying entities. Could we not infer the existence of a self as the thing to which consciousness belongs? No, for the inference would require for its validity an example illustrating the existence of substances over and above qualities, of property-possessors over and above properties; but the fact that we do not need to assume the existence of a unitary mango to which its qualities belong indicates that such examples are not forthcoming. Could we not infer the existence of a unitary mango to which the smell, taste, c­ olour etc. belong? This could then serve as the example in the inference of a substance to which consciousness belongs. A unitary mango could be inferred as, for example, the only ­plausible explanation of the fact that the mango’s smell, taste, colour, etc. occur together. They never split off from each other. Does this not indicate that they all belong to the same thing? No, argues the Buddhist, drawing on the principle of ­parsimony of ­postulation (kalpanālāghava), the Indian version of ‘Ockham’s razor’. Rather than ­postulating an ­imperceptible substance to which the five properties ‘stick’, it is more parsimonious, argues the Buddhist, to assume that they stick to each other. For how this ‘sticking to each other’ was elaborated in terms of their forming a causal complex in which they function as co-operating causes (sahakāripratyayas) for each other, see Watson (2006, pp. 57–58). To conclude: if it cannot even be proved that a mango-substance exists as the substrate of taste etc., obviously it cannot be proved that a self-substance exists as the substrate of consciousness.17 2.2  Self as agent of cognition

In the second Naiyāyika argument, we are asked to envisage a situation where someone experiences pleasure from a particular kind of object, and on seeing the same kind of object later and remembering the earlier pleasure, feels desire for the object (see Figure 9.6, where P = pleasure, S = seeing and D = desire). Unless the earlier pleasure and the later seeing of the object had the same subject, the desire would not arise. After all, points out the Naiyāyika, we do not find desire arising in one person (Y) as a result of pleasure in another (X) (Figure 9.7). So if one person were not one subject, as assumed in Figure 9.6, but a plurality of subjects (as depicted in Figure 9.8), surely desire would not arise. Why would a subsequent subject of experience desire something that caused pleasure not to it, but to some totally different subject of experience?

152

ALEX WATSON

P

S

D

Figure 9.6 

P X

Y

S

D

Figure 9.7 

The fact that people do desire things that have previously given them pleasure indicates that it is the same thing that is the agent of both the desire and the earlier pleasure. That is the Naiyāyika argument. Rāmakaṇṭha’s refutation of it is again Buddhism’s refutation of it.18 The argument gains its plausibility from likening the situations depicted in Figures 9.7 and 9.8. It rests on the claim that since desire does not arise in the first of these situations, it would not arise in the second. But the Buddhist has perfectly adequate means at his disposal for distinguishing the two. Two people are not the same as two moments within the same stream: the latter are joined by a causal chain; the former are not.Thus in the situation represented in Figure 9.8, the final subject is linked by a chain of cause and effect back to the earlier pleasure, such that it has access to memory traces (saṃskāras) of the pleasure. Person Y, by contrast, does not have access to memory traces of person X’s pleasure, and that is why desire does not arise in person Y. The validity of the argument requires that the reason desire does not arise when there are two people (Figure 9.7) is because of a lack of sameness of subject. But the Buddhist has a plausible alternative: that it is due to a lack of a chain of causation along which traces can be transmitted. So long as this alternative remains unrefuted, difference of subject will

153

THE SELF AS A DYNAMIC CONSTANT

P

S

?

Figure 9.8 

not be sufficient to logically preclude the rise of desire. Hence the occurrence of desire will not entail sameness of subject. In order for the argument to work, the Naiyāyika has to prove that desire can only arise in the same subject that experienced the earlier pleasure. He tries to do that by pointing out that when desire does occur, it is in the same subject as that of the pleasure (­Figure 9.6), not a different one (Figure 9.7). But the Buddhist just replies that a single person is not a single subject, but a plurality of different ones (as in Figure 9.8). Thus this is not an argument that forces any shift in the Buddhist position; it requires for its validity that a single person is a single subject, but that is exactly what is in question. Neither the argument from consciousness as a quality requiring a support, nor this one from desire as requiring the same subject as the pleasure that gave rise to it, oblige the Buddhist to rethink.19 ¯ MAKANT HA BETWEEN NYA ¯ YA AND BUDDHISM 3. RA ..

Rāmakaṇṭha, then, agrees with Buddhism that there is no self as substance over and above consciousness, and no self as agent over and above consciousness. For him, as for ­Buddhism, consciousness does not require something other than itself in which to inhere. He agrees with Buddhism that the perceiver of our perceptions, the thinker of our thoughts, is just consciousness (grāhaka/jñātṛ = jñāna). How then does he preserve the self? For him consciousness is the self. He equates the self and consciousness, or to put it another way, he characterises consciousness as the nature (svabhāva), of the self.20 This means that he holds consciousness to be permanent, not momentary, as it is for both Buddhism and Nyāya. Although consciousness for Nyāya belongs to a permanent self, it itself consists of discrete, momentary instances, as in Buddhism.The difference between the three views is represented in Figure 9.9. Nyāya: there is a self that is separate from consciousness. Buddhism: there is no self. Rāmakaṇṭha: There is a self but it is just consciousness. Rāmakaṇṭha crosses out the line but joins up the dots into a line. He travels down the path of Buddhist argumentation

Figure 9.9 

154

ALEX WATSON

quite a long way: he reduces the self to the stream of consciousness. But he then argues that the stream is unchanging. So between Buddhism and Nyāya it was a debate about the existence or non-existence of an entity. Between Buddhism and Rāmakaṇṭha there is agreement about what exists; it is just a question of how to classify that: whether as something plural or unitary, changing or unchanging. For Rāmakaṇṭha it is unitary and unchanging, but it is not a static entity like the self of the Naiyāyikas. It is dynamic, yet constant. Dynamic in that it is a process, the process of the shining forth of consciousness. Constant in that (1) the light of consciousness pours out always in the same form, and (2) there are no breaks in the process. Consciousness as envisaged by Rāmakaṇṭha, then, differs in two ways from consciousness as envisaged by Buddhism: it is differentiated neither qualitatively nor temporally. Consciousness for Buddhism, divided up as it is into dissimilar discrete entities, each one ceasing to exist before the next one comes into existence, resembles a light forever going on and off, and each time producing a different coloured light; consciousness for Rāmakaṇṭha resembles a light that is permanently on, forever sending out light of the same colour. This constant pouring forth of the illuminating light of consciousness is precisely what the self is, just as the sun is nothing more than a constant pouring forth of light. ¯ MAKANT HA AGAINST BUDDHISM 4. RA ..

The difference of Rāmakaṇṭha’s position from that of Buddhism will now be further elaborated. 4.1  Refutation of the momentariness of consciousness

Rāmakaṇṭha argues that as we proceed through life, experiencing various objects of ­perception, we never lose a sense that it is me who is the experiencer of those o ­ bjects.21 The Buddhist accepts that we have this sense of unbroken personal identity, but he claims that it is mistaken.To be precise, he maintains that our direct, non-conceptual (nirvikalpaka) experience of ourselves is of a sequence of distinct momentary ­perceivers, but that we superimpose the concept of oneness on to that plurality. It is like the example mentioned earlier of a plurality of distinct film-images appearing as one continuous image. Or, to use the Buddhists’ own example, like a very calm river that looks like one stable, unchanging piece of water, despite being lots of different bits of water rushing by.22 I give here two of Rāmakaṇṭha’s arguments against the coherence of this Buddhist position. 1 It is possible that with regard to a sequence of entities external to us, we might be fooled by the rapidity with which they succeed each other and the similarity of each one to the previous, into mistaking the plurality for a unity. But the Buddhist is asking us to believe that a sequence of momentary perceivers could fool themselves, as though the distinct film images, or the bits of water in the river, could deceive themselves into thinking that they are one unbroken thing, or could experience themselves as one unbroken thing. This seems much less plausible.23

155

THE SELF AS A DYNAMIC CONSTANT

2 In the standard example of superimposition, namely that of silver on to motherof-pearl, it is clear that we would not mistake the mother-of-pearl for silver unless we had experienced silver previously. But in a Dharmakīrtian universe, in which everything—both perceivers and perceived objects—is momentary, how could a perceiver ever have experienced something enduring, in order to acquire the ­concept of duration, in order to superimpose it on to what is momentary?24 4.2  The constancy of consciousness

So Rāmakaṇṭha’s view is that although the objects of our experience change, consciousness itself, the perceiver, is constant. This asymmetry was problematic for Dharmakīrtian Buddhism, committed as it was to the non-difference of perceiver and perceived objects (grāhyagrāhakābheda). This non-difference entails that any change in what is perceived necessitates a change in the perceiver. If the object of our experience changed from a pot to a cloth, and yet there was no change in consciousness, how would consciousness have registered the change in the object? The very way in which consciousness perceives an object is by being marked by that object. Rāmakaṇṭha does not see why that should be so. When light—an analogy that the Buddhists also use to elucidate the nature of consciousness—illuminates an object, we do not regard that object as colouring the light, marking it or modifying it in any way.25 So similarly for Rāmakaṇṭha consciousness is the illuminator of objects but is itself unaffected by them. Experience changes because different objects come within the range of the light of consciousness, not because consciousness itself changes.26 Rāmakaṇṭha points out that the Buddhists themselves do not hold consistently to the position that consciousness is differentiated by its objects. For they accept that at one moment a single, undivided perceiver can perceive a multi-coloured object. If in one moment consciousness can be single and yet perceive a plurality of different colours, i.e. objects, why cannot one temporally extended consciousness perceive a plurality of ­objects? If a relationship of one perceiver to many objects is possible at one point of time, what reason is there to deny the possibility of such a relationship over time?27 See Figure 9.10, where consciousness is represented above, and its perceived object(s) below. Between the two views of consciousness, Rāmakaṇṭha’s and Buddhism’s, two ­differences can be distinguished. One concerns consciousness’ temporal extension and the other its relationship to its objects.

One moment

Figure 9.10 

Temporal Sequence

156

ALEX WATSON

4.3  Temporal extension

As we saw above, Rāmakaṇṭha presents himself as aligned with the Buddhists against the Naiyāyikas. He stresses that, like the Buddhists, he denies the existence of a self over and above consciousness, because all of the phenomena that it is supposed to explain (such as desire for objects that have previously given pleasure) are explainable on the assumption of consciousness alone. He represents himself as with Buddhism and against Nyāya in regarding a self separate from consciousness as a fictional product of metaphysical speculation that is neither perceived nor inferrable, neither part of our experience nor ­necessary to explain our experience. The self that he asserts is, he maintains, simply the flow of consciousness, i.e. exactly the same thing that the Buddhists are talking about when they speak of consciousness. But if Rāmakaṇṭha and the Buddhists were really both attempting to describe the same thing, why would they arrive at such diametrically opposed characterizations of it as permanent versus existing only for a moment? For that reason it seems better to present the situation in a way other than as Rāmakaṇṭha does. Figure 9.9 reflected Rāmakaṇṭha’s presentation: the top line under ‘Nyāya’ is o ­ mitted under ‘Buddhism’ and ‘Rāmakaṇṭha’. But rather than seeing the latter two p­ ositions as ­involving the crossing out of a Naiyāyika self that is separate from c­ onsciousness, the ­denial of its existence, it is probably better to see them as both resulting from the merging of the Naiyāyika concepts of self/perceiver and consciousness (Figure 9.11). The balance of what is retained from each of these two Naiyāyika concepts differs: the Buddhist concept retains more of the features of Naiyāyika consciousness; Rāmakaṇṭha’s retains more of the features of the Naiyāyika perceiver. The direction of the reduction is opposite in each case. Buddhism reduces the perceiver to consciousness; Rāmakaṇṭha reduces consciousness to the perceiver. Buddhism ends up with a concept that is closer to Naiyāyika consciousness, but not identical to it (which is why it is not placed on the same horizontal level in the diagram), because it includes grāhakatva (the condition of being the subject of perception), which Nyāya places in the self. Rāmakaṇṭha ends up with a ­concept that is closer to the Naiyāyika perceiver, but not identical to it, because it is fl ­ owing and dynamic, not a static substance, and not ontologically distinct from consciousness. 4.4  Relation to objects

The second difference is between Rāmakaṇṭha’s conception of consciousness as removed from any influence from its objects and Buddhism’s conception of it as changed by every change in its objects. Consciousness for the former is an observer located outside the

Figure 9.11 

THE SELF AS A DYNAMIC CONSTANT

157

world of changing objects; for the latter it is a participant that is immersed in and m ­ odified by involvement with the world. The Buddhist includes a perceived object (grāhya) as one pole of consciousness. For ­Rāmakaṇṭha perceived objects lie outside consciousness, but for Buddhism within it.28 So of course consciousness is differentiated by its objects for Buddhism. But Rāmakaṇṭha frees consciousness from influence from its objects by characterizing its relationship with them as merely their illuminator, that which casts light upon them. The ontological ­distance that he puts between it and objects releases it from modification by them. Thus another way to capture this difference between the two schools of thought is to note that what for Buddhism is one thing is two things for Rāmakaṇṭha (and three for Nyāya) (Figure 9.12). What the Buddhist characterizes as an instance of consciousness (e.g. nīlajñāna, a ­cognition of blue) comprises, for Rāmakaṇṭha, the perceiver, i.e. consciousness proper, and the perceived object (the blue), which is not of the nature of consciousness. Rāmakaṇṭha separates out and excludes from consciousness not only what is perceived through the five senses, but also ‘internal’ objects such as pains.These for Rāmakaṇṭha are not forms of consciousness or instances of consciousness, but objects of consciousness. The same goes for thoughts. But after a whole thought such as ‘this is a pot’ (an example of determinative cognition, adhyavasāya) has been separated off, what is left? The answer is: the perceiver of that thought, its observer. The thought itself is not of the nature of consciousness; consciousness is that which is aware of it. CONCLUDING REMARKS

We have seen that the positions put forward by the two protagonists in this debate, Nyāya and the Buddhism, are not collectively exhaustive. Thus the falsity of one does not entail the truth of the other, since Rāmakaṇṭha provides a third alternative. Rāmakaṇṭha’s self is not a Naiyāyika substance to which consciousness belongs, a dharmin in which cognitions inhere.29 By comparison it is less static, more of the nature of an outpouring, and less ontologically distinct from the flow of experience. Occupying middle ground between a Naiyāyika self and a Buddhist stream of consciousness-moments, his position is not vulnerable to most of the arguments that the Buddhists use against the Naiyāyikas, nor to the arguments that Naiyāyikas use against Buddhists. If it is coherent, it arguably removes the Naiyāyika self as an effective player in the debate. For it achieves what the Naiyāyika wants to achieve (opposition to the Buddhist denial of a self, assertion of a permanent perceiver) while being less extravagant in its metaphysical claims and less susceptible to the familiar arguments against a self-substance in which consciousness inheres. It retains the idea of a permanent self without postulating any unperceived entity, and without resorting to the problematic postulation of a property-possessor that transcends its properties and is unaffected by changes in the latter. In the final stage of the contest then, we could perhaps dispense with the Naiyāyika position, leaving just Rāmakaṇṭha and the Buddhists to fight it out. Such at least is the way that Rāmakaṇṭha presents the situation. In a future publication (Watson forthcoming b) I ask which position out of these two is to be preferred. I argue that this new pair, Rāmakaṇṭha and Dharmakīrtian Buddhism, are just as unnecessarily polarized as the older pair of Nyāya and Buddhism, since

158

ALEX WATSON

Figure 9.12 

for Rāmakaṇṭha too the self is completely unmodifiable (avikārya). I place Kumārila ­between Rāmakaṇṭha and Buddhism, but I argue that Kumārila, Rāmakaṇṭha and Nyāya all share problematic assumptions, and that the best way to challenge Buddhism’s view of a ­momentary stream of subjects is to adopt a position that falls closer to that view than that of any of these three self-theorists. NOTES 1. The representation of Buddhism below refers also to the Abhidharmakośabhāṣya of ­Vasubandhu, the ideas of which filter through to Dharmakīrti via its influence on ­Vasubandhu’s pupil, Dignāga. 2. See for example Kawajiri (2011), Lawrence (2000, 2005, 2008, 2009), Nemec (2011, 2012), Rastogi (1986, 2004), Ratié (2006, 2007, 2009, 2010a, b, c, 2011a, b), Torella (1988, 1992, 2002, 2007a, b, c, d, 2008). 3. I am using ‘philosophical debate’ here to denote the subject matter of those writings in which theological assumptions are not taken for granted but either set aside or argued for, in which metaphysical and epistemological issues are discussed by confronting the author’s own view with that of rival traditions, and in which the scriptures of the author’s own tradition are not appealed to as authoritative. (For a longer elaboration, see Watson 2006, pp. 74–89.) The most important philosophical writings of Rāmakaṇṭha are the Nareśvaraparīkṣāprakāśa, the Paramokṣanirāsakārikā and the sixth chapter of the Mataṅgavṛtti. Publications that concern these are: Watson (2006, 2010a, b) and Watson et al. (2013). The earliest surviving Śaiva Siddhānta text that debates non-superficially with Buddhism and other rival philosophical traditions is the Mṛgendravṛtti by Rāmakaṇṭha’s father, Nārāyaṇakaṇṭha. This has been studied and translated by Hulin (1980). Rāmakaṇṭha’s philosophy has recently, for the first time, been addressed in a general book on Indian Philosophy: Bartley (2011, pp. 199–209). 4. See Abhidharmakośabhāṣya, p. 473, 20–23; Duerlinger (2003, p. 99). 5. As to why Buddhists asserted the momentariness of both mental and physical entities— why they explained change not as one thing becoming modified, but rather as a succession of distinct momentary things—see Dharmakīrti’s Pramāṇaviniścaya 2:53–55, Vādanyāya, pp. 2,1–3,13 and Hetubindu, p. 4*,6–7, p. 19,10–13; and Dharmottara’s Pramāṇaviniścayaṭīkā ad 2:53–55 and Kṣaṇabhaṅgasiddhi. See also Steinkellner (1963, 1968/1969), Mimaki (1976), von Rospatt (1995),Yoshimizu (1999) and Sakai (2010a, b, 2011). 6. These four kinds of mental state are: feelings (vedanā), ideation (sañjñā), impulses (saṃskāra) and consciousness (vijñāna); see Vetter (2000).

THE SELF AS A DYNAMIC CONSTANT

159

7. The argument involves three contentions, each of which had their own supporting arguments: (1) Qualities cannot exist without substances to which they belong; (2) Consciousness, desire, aversion, pleasure, pain, volition are qualities; (3) The self is the only possible substance to which these qualities could belong. See Nyāyavārttika ad 1.1.10, p. 62,12–18, and Praśastapādabhāṣya, p. 16,3–7. For the second stage of the argument, see e.g. Nyāyavārttika ad 3.2.18, Nyāyamañjarī, Vol. 2, p. 278,14–15 and Candrānanda ad Vaiśeṣikasūtra 2.2.28. For the third stage of the argument, see e.g. Nyāyamañjarī,Vol. 2, pp. 284,6–293,2 and Nyāyasūtra 3.2.47 with the commentaries ad loc. See also Chakrabarti (1982), Oetke (1988, pp. 255–256, 258–260, 280, 286–300, 359–360, 464), Matilal (1989, pp. 74, 77; 1994, p. 286), Preisendanz (1994, pp. 187, 209, 278–281), Kano (2001) and Watson (2006, pp. 174–184). 8. See Abhidharmakośabhāṣya, pp. 475,14–16 and 475,22–476,3; Duerlinger (2003, pp. 103 and 104). 9. The concept ‘mango’, for Buddhism, corresponds to no reality, but is a false unity that we superimpose on to something plural, like the concept ‘forest’. To use Vasubandhu’s terminology in the Abhidharmakośabhāṣya, forests and mangoes are prajñaptisat, not dravyasat (see e.g. p. 461,14ff.): they have merely conceptual, not substantial, existence. 10. The four skandhas that Buddhism groups with vijñāna (consciousness)—rūpa, vedanā, sañjñā, saṃskāra—are of course not the same as the qualities of the self that Nyāya groups with jñāna (consciousness): icchā, dveṣa, prayatna, sukha, duḥkha (Nyāyasūtra 1.1.10). 11. See Abhidharmakośabhāṣya, p.  476,1–2, Mataṅgavṛtti, vidyāpāda, p.  153,8–11 and Nareśvaraparīkṣāprakāśa introducing 1:5, p. 11,1–4: na hi rasādīnāṃ guṇatvam asmākaṃ sāṅkhyānām api vā prasiddham, rūparasādisamudāyavyatirekeṇānya sya kasya cid āmraphalāder dharmiṇo ‘nupalambhāt, ‘For it is not proved that taste and the like are qualities for us [Buddhists] or for the Sāṅkhyas either. For we do not perceive any separate property-possessor, such as a mango fruit, over and above the conglomeration of colour, taste and the rest.’ On the Sāṅkhya view that is referred to there, see Wezler (1985), Bronkhorst (1994) and Watson (2006, pp. 186–188). 12. See Nareśvaraparīkṣāprakāśa introducing 1:5, p. 11,4–6: ata evāmrasya rasa iti bhedavyapadeśaḥ samudāyaikadeśatvakhyāpanāya vanasya dhavaḥ śobhana itivad upapadyata eva, ‘That is precisely why the expression, “the taste of the mango”, [that exhibits in its formulation] difference, serves [in fact] to designate [simply] that [the taste] is a single part of a conglomerate, just like [the expression] “the Dhava tree of the forest is beautiful”.’ 13. See Praśastapādabhāṣya, p. 15,12 and Candrānanda ad Vaiśeṣikasūtra 3.2.4, p. 28,18–19. 14. See Abhidharmakośabhāṣya, pp. 476, 19–477,3; Duerlinger (2003, p. 107). 15. See Abhidharmakośabhāṣya, p. 477,11–17; Duerlinger (2003, p. 108). 16. See Nareśvaraparīkṣāprakāśa, avatārikā to 1.5 (p.  11,1–4), Kiraṇavṛtti ad 2:25ab (p.  53, 4–8), Mataṅgavṛtti, vidyāpāda, p. 153,8–11, and Watson (2006, pp. 184–192; 2010a, pp. 87–89). 17. This strategy of Buddhist argument goes back to the Abhidharmakośabhāṣya. The opponent there states that the self is required as the support (āśraya) of consciousness (citta) and traces (saṃskāra), in the way that earth supports its qualities such as smell. Vasubandhu replies that the example of earth is exactly what convinces him that there is no self. The fact that we ­perceive only a certain combination of smell and other properties, not some extra ­entity ‘earth’ supporting them, indicates that there is no such entity, and this indicates that, ­analogously, there is no self supporting consciousness and traces (Abhidharmakośabhāṣya, p.  475,14–16; Duerlinger [2003, p.  103]). On Śāntarakşita and Kamalaśīla’s refutation of the view that desire and other states of consciousness require a support, see Hulin (1978, pp. 100–101). 18. See Nareśvaraparīkṣāprakāśa introducing 1:5, pp. 9,10–10,8, and Watson (2006, pp. 138–159 and 240, note 99).

160

ALEX WATSON

19. Versions of this argument are found at Nyāyabhāṣya ad 1.1.10, p. 16,5–20, Nyāyavārttika ad 1.1.10, pp. 60, 12–63,2, Nyāyamañjarī, pp. 278, 4–284,5; see also Oetke (1988, pp. 345–352 and 256–258), Matilal (1989, pp. 74–77; 1994, pp. 286 and 289–291),Taber (1990, pp. 36–37), Preisendanz (1994, pp. 202, 306) and Kapstein (2001, pp. 146–151 and 375–383). Apart from the earliest, that of the Vrttikāra in the Śābarabhāṣya, they all appeal to the concept of synthesis (pratisandhāna, anusandhāna). Desire would not arise, it is argued, were it not for the s­ubject’s ability to synthesize the earlier pleasure with the present seeing of the object. One can only synthesize cognitions of which one is the subject. Therefore the earlier pleasure  and the present seeing must have the same subject. But, replies the Buddhist, why is it the case that only cognitions having the same subject can be synthesized? What is required for s­ynthesis, as the Naiyāyika also recognizes, is the activation of a memory trace of the earlier cognition. Why is this not enough? Why do the Naiyāyikas also insist on a further requirement, namely sameness of subject?   At this point certain Naiyāyikas give a verbalization of the synthesis, such as ‘Earlier I ­derived pleasure from this object, and now this same I am experiencing it again’; and argue that such a cognition would not arise unless I were indeed the subject of both the earlier pleasure and the present seeing. It is true, replies the Buddhist, that such a cognition implies that I sense myself as the subject of both the present and the past experiences of the object. But this sense of sameness is easily explainable as resulting from the rapidity with which consecutive momentary subjects succeed each other, and the similarity of each one to the previous; this fools us into superimposing oneness on to what is actually multiple. Some Naiyāyikas put forward synthesis as verbalized above not as a means of inferring the self, but as including a direct perception (pratyakṣa) of the self. But this is countered on the grounds that it assumes what the argument sets out to prove: the validity of such seeming experiences of an enduring subject.   Even some Naiyāyikas regard the argument as a failure (those that Jayanta refers to as svayūthya, ‘those of our own fold’), pointing out that it can only work if it asserts that ­synthesis includes direct perception of an enduring subject. But if such a subject is ­available to ­pratyakṣa, then this whole inference from desire becomes pointless (Nyāyamañjarī, p. 277,14–17).   Furthermore even proponents of the argument such as Uddyotakara and Jayanta allow their Buddhist interlocutors (pūrvapakṣins) to overcome the various Naiyāyika strategies they put forward. They see its success as dependent on an independent refutation of the coherence of momentariness. Thus Uddyotakara allows his opponent to answer each of his points until at the end he argues that a momentary entity would not be able to leave a trace on another momentary entity, whether the latter existed contemporaneously with it or ­immediately after it (Nyāyavārttika, pp. 62,19–63,2; see also Taber [2012]). Similarly, throughout Jayanta’s long discussion the opponent is able to answer all of Jayanta’s assertions, and the debate is only closed when Jayanta asserts that he will explain later in the chapter that there can be no relation of cause and effect between momentary cognitions (­Nyāyamañjarī,Vol. 2, p. 284,3–4).   For a more detailed analysis of the argument and its history, see Watson (2006, pp. 138– 157 and 159–165). Note that it is different from, though sometimes mistakenly conflated with, the arguments we find in the commentaries to Nyāyasūtra 3.1.1, which are also often put in terms of synthesis (pratisandhāna). The argument that we have been examining is based on cognitions at different points of time being synthesized by a single entity, who must therefore exist continually over that timespan. The arguments in the commentaries to 3.1.1 are based on perceptions from different sense-faculties being synthesized by a single

THE SELF AS A DYNAMIC CONSTANT

161

entity, who must therefore exist over and above the individual sense-faculties. What is aimed to be proved is not necessarily an entity that endures over time, but one that exists above and ­beyond the plurality of sense-faculties. On these arguments based on 3.1.1 see Halbfass (1976, p. 163), Matilal (1986, pp. 252–4, 372), Oetke (1988, pp. 260–269), Taber (1990, pp. 39–42), Laine (1993), Preisendanz (1994, pp. 183–187), Chakrabarti (1992) and Ganeri (2000; 2007, pp. 180–181). 20. The view that consciousness is the nature of the self may appear to some as not so different from the view that it is a quality/property of the self; talk of a thing’s ‘properties’ in English can seem more or less synonymous with talk of its ‘nature’. But in Indian philosophical discourse whereas the relation between a thing and its nature was held to be identity, sameness (tādātmya), the relation between a thing and its qualities (guṇas) or properties (dharmas) was held to be inherence (samavāya). A thing and its nature are the same thing; a thing and its qualities/properties are not—the latter belong to the thing, but are of a different nature. Furthermore the difference between the self and its properties is more pronounced than the difference between a physical object and its properties. For the self is eternal, its particular properties are momentary; it is omnipresent, they are restricted to a particular place; and in the state of liberation it is free of all of its particular qualities (sakalaguṇāpoḍha). For an account of the evolution of this tendency in Nyāya and Vaiśesika—the increasing distance that developed between the self and its qualities—see Frauwallner (1956, pp. 91–104; 1984, pp. 61–71). 21. Nareśvaraparīkṣāprakāśa ad 1:5, pp. 13, 20–14,18; Watson (2006, pp. 220–230). 22. Nareśvaraparīkṣāprakāśa ad 1:5, pp. 14,18–15,5; Watson (2006, pp. 230–236). 23. That a sequence of momentary perceivers could deceive themselves is also contradicted by the Buddhist assertion that all cognition is non-conceptual with regard to itself. For more details of Rāmakaṇṭha’s argumentation here, see Watson (2006, pp. 237–238, 245–251; 2010b, pp. 302–303). The former includes editions and translations of the relevant Sanskrit passages from the Nareśvaraparīkṣāprakāśa, Mataṅgavṛtti and Paramokṣanirāsakārikāvṛtti. For a discussion of what precisely Dharmakīrti means by cognition being non-conceptual with regard to itself, see Watson (2010b, pp. 317–319). 24. This point is an extrapolation of Rāmakaṇṭha’s thinking, rather than a close report of what he has written. His account of why superimposition would be impossible if everything were momentary (see sarveṣāṃ kṣaṇikatvena yojanānupapatteḥ at Nareśvaraparīkṣāprakāśa ad 1:5, p. 15,17–18, in a passage translated and analyzed at Watson 2006, pp. 238–245) focuses more on superimposition requiring an enduring perceiver than on it requiring experience of duration. 25. For discussion of the light analogy, see Watson (2010b, p. 305), Watson and Kataoka (2010, pp. 304–306) and especially Watson (2013). 26. See, for example, Nareśvaraparīkṣāprakāśa ad 1.6ab, p. 26,4–13: sannihitārthaprakāśakatvaṃ hy ātmanaḥ svabhāvaḥ pradīpāder iva tathānubhavāt siddhaḥ.tathā hi yo yaḥ sannihito ‘rthas taṃ taṃ svaśaktyaiva prakāśayann ayam anubhūyate.[ … ] atha kas tasya sannihitenārthenopakāraḥ kṛtaḥ, na kaścit. kathaṃ tarhi tam eva prakāśayati? tathāsvabhāvatvād ity uktam, upakāre ‘pi tadasvabhāvasya prakāśakatvādṛṣṭeḥ. ‘[Rāmakaṇṭha:] For the nature of the self, like that of lights and such like, is established to be the illuminating of objects that are present, because we experience it in that way.To explain further, the [self] is experienced as illuminating, through its own power alone, whatever objects are present. [ … ] If [you ask] what benefit is produced for [­consciousness] by the object present, [we say] none at all. [Buddhist:] Why then does it reveal only the [object that is present]? [Rāmakaṇṭha:] I have already said that it is because its nature is thus, because we do not find that things which do not have as their nature [to reveal what is

162

ALEX WATSON

present] can illuminate even if there is benefit [from the object].’ For more on this passage in particular, and on Rāmakaṇṭha’s position in general, see Watson (2006, pp. 333–382; 2010a, especially pp. 111–112). 27. For the full argument, see Nareśvaraparīkṣāprakāśa ad 1.6ab, pp. 26,19–28,11, and Watson (2006, pp. 335–348). 28. I am using ‘perceived object’ and ‘object’ to refer to the grāhya, i.e. the phenomenal object, that which is directly perceived and is the content of a perception. I am not using it to refer to the external object that is postulated by the Sautrāntikas as the cause of this p­ henomenal object; that is not directly perceived (grāhya), but rather inferred (anumeya). Hence the claim I make here that objects lie within consciousness is true for both the Yogācāra and the ­Sautrāntika strands of Dharmakīrtian Buddhism. 29. Nor is it a Sāṅkhya puruṣa that is devoid of cognition, nor a Vedāntic Self that in its motionless inactivity resembles the ether. REFERENCES Primary sources Abhidharmakośabhāṣya: Abhidharmakośabhāṣya of Vasubandhu. Ed. Pradhan, P. Tibetan Sanskrit Works ­Series 8. K. P. Jayaswal Research Institute, Patna, 1975. Aṣṭaprakaraṇam: Tattvaprakāśa, Tattvasaṅgraha, Tattvatrayanirṇaya, Ratnatraya, Bhogakārikā, Nādakārikā, Mokṣakārikā, Paramokṣanirāsakārikā. Ed. Dvivedī, Brajavallabha. Yogatantra-granthamālā 12. Vārāṇasī, 1988. Hetubindu: Dharmakīrti’s Hetubinduḥ,Teil 1,Tibetischer Text und rekonstruierter Sanskrit Text. Ed. Steinkellner, Ernst. Wien: 1967. Kiraṇavṛtti: Kiraṇavṛttiḥ. Bhaṭṭa Rāmakaṇṭha’s Commentary on the Kiraṇatantra.Volume 1: Chapters 1–6. Ed. Goodall, Dominic. Publications du départment d’indologie, Institut français de Pondichéry / École française d’Extrême-Orient 86 (1). Pondichéry, 1998. (In references to this edition, Éline n ­ umbers are not counted from the top of the page; they rather follow the line numbers printed in the edition, which start counting from the previous verse-segment.) Kṣaṇabhaṅgasiddhi: see Frauwallner (1935). Mataṅgapārameśvara: Mataṅgapārameśvarāgama, Vidyāpāda, avec le commentaire de Bhaṭṭa Rāma kaṇṭha. Ed. Bhatt, N. R. Publications de l’Institut français d’Indologie 56. Pondichéry, 1977. Mataṅgavṛtti: in Mataṅgapārameśvara. Nareśvaraparīkṣāprakāśa: Nareśvaraparīkṣā of Sadyojyotis with the Prakāśa Commentary of Rāmakaṇṭha. Ed. Sāstrī, Madhusūdan Kaul. KSTS 45. Delhi, 1989. (Original Edition Śrīnagar 1926.) Nyāyabhāṣya: Gautamīyanyāyadarśana with Bhāṣya of Vātsyāyana. Ed. Thakur, Anantalal. Nyāyacaturgranthikā 1. New Delhi, 1997. Nyāyamañjarī: Nyāyamañjarī of Jayantabhaṭṭa, with Ṭippaṇi—Nyāyasaurabha by the Editor. Ed.Varadacharya, K. S. 2 Volumes. Oriental Research Institute Series 116, 139. Mysore, 1969 and 1983. Nyāyasūtra: in Nyāyabhāṣya. Nyāyavārttika: Nyāyabhāṣyavārttika of Bhāradvāja Uddyotakara. Ed. Thakur, Anantalal. Nyāyacaturgranthikā 2. New Delhi, 1997. Paramokṣanirāsakārikāvṛtti: in Aṣṭaprakaraṇam. See also Watson, Goodall and Sarma (2013). Pramāṇaviniścaya: Pramāṇaviniścaya, Chapters 1 and 2. Ed. Steinkellner, Ernst. China Tibetology Publishing House/ Austrian Academy of Sciences Press, Beijing–Vienna, 2007. Pramāṇaviniścayaṭīkā: see Sakai (2010a). Praśastapādabhāṣya:Word Index to the Praśastapādabhāṣya: a Complete Word Index to the Printed Editions of the Praśastapādabhāṣya. Ed. Bronkhorst, Johannes and Ramseier,Yves. Delhi, 1994. Ślokavārttika: Ślokavārttika of Śrī Kumārila Bhaṭṭa with the Commentary Nyāyaratnākara of Śrī Pārthasārathi Miśra. Ed. Śāstrī, Dvārikādāsa. Prāchyabhārati Series 10.Varanasi, 1978.

THE SELF AS A DYNAMIC CONSTANT

163

Vādanyāya: Dharmakīrtis Vādanyāyaḥ Teil I. Ed. Much, M. T. Philosophisch-Historische Klasse, ­Sitzungsberichte 581. Österreichische Akademie der Wissenschaften, Wien, 1991. Vaiśeṣikasūtra: Vaiśeṣikasūtra of Kaṇāda with the Commentary of Candrānanda. Ed. Jambuvijayaji, Muni Sri. Gaekwad’s Oriental Series 136. Baroda, 1961.

Secondary sources Bartley. (2011). An introduction to Indian philosophy. London: Continuum. Bronkhorst, J. (1994). The qualities of Sāṅkhya. Wiener Zeitschrift für die Kunde Südasiens, 38, 309–322. Chakrabarti, A. (1982). The Nyāya proofs for the existence of the soul. Journal of Indian Philosophy, 10, 211–238. Chakrabarti, A. (1992). I touch what I saw. Philosophy and Phenomenological Research, 52(1), 103–116. Duerlinger, J. (2003). Indian Buddhist theories of persons. Vasubandhu’s “Refutation of the theory of a self ”. London: RoutledgeCurzon. Frauwallner, E. (1935). Dharmottaras Kṣaṇabhaṅgasiddhiḥ: Text und Übersetzung. Wiener Zeitschrift für die Kunde des Morgenlandes, 42, 217–258 [= Kleine Schriften 530–571]. Frauwallner, E. (1956). Geschichte der indischen Philosophie (Vol. 2). Salzburg: Otto Muller Verlag. Frauwallner, E. (1984). History of Indian philosophy (Vol. 2).Translated from original German into English by V. M. Bedekar. Delhi: Motilal Banarsidass. Ganeri, J. (2000). Cross-modality and the self. Philosophy and Phenomenological Research, 61(3), 639–657. Ganeri, J. (2007). The concealed art of the soul. Theories of self and practices of truth in Indian ethics and epistemology. Oxford: Oxford University Press. Halbfass, W. (1976). Zum Begriff der Substanz (dravya) in Vaiśeṣika. Wiener Zeitschrift für die Kunde ­Südasiens, 20, 141–166. Hulin, M. (1978). Le Principe de l’Ego dans la Pensée Indienne Classique. La Notion d’Ahaṃkāra. Paris: ­Publications de l’Institut de Civilisation Indienne 8 (44). Hulin, M. (1980). Mṛgendrāgama: sections de la doctrine et du yoga avec la Vṛtti de Bhaṭṭanārāyaṇakaṇṭha et la Dīpikā d’Aghoraśivācārya.Traduction, Introduction et Notes. Pondichéry: Publications de l’Institut français d’Indologie 63. Kano, K. (2001). Pariśeṣa, Prasaṅga, and Kevalavyatirekin—The logical structure of the proof of Ātman. Journal of Indian Philosophy, 29, 405–422. Kapstein, M. (2001). Reason’s traces; identity and interpretation in Indian and Tibetan Buddhist thought. Studies in Indian and Tibetan Buddhism. Boston: Wisdom. Kawajiri, Y. (2011). A critique of the Buddhist theory of adhyavasāya in the Pratyabhijñā school. In H.  Krasser et al. (Eds.), Religion and logic in Buddhist philosophical analysis: Proceedings of the fourth ­international Dharmakirti conference,Vienna, August 23–27, 2005 (pp. 281–290). Wien: Österreichische ­Akademie der Wissenschaften. Laine, J. (1993). Some remarks on the Guṇaguṇibhedabhaṅga chapter in Udayana’s Ātmatattvaviveka. ­Journal of Indian Philosophy, 21, 261–294. Lawrence, D. P. (2000). Rediscovering God with transcendental argument. A contemporary interpretation of ­monistic Kashmiri Śaiva Philosophy. Delhi: Sri Satguru Publications [State University of New York, 1999]. Lawrence, D. P. (2005). Remarks on Abhinavagupta’s use of the analogy of reflection. Journal of Indian Philosophy, 33, 583–599. Lawrence, D. P. (2008). The mythico-ritual syntax of omnipotence. On Utpaladeva and Abhinavagupta’s use of Kriyā-kāraka theory to explain Śiva’s action. In M. Kaul & A. Aklujkar (Eds.), The grammatical traditions of Kashmir: Essays in memory of Pandit Dinanath Yaksh (pp. 446–488). Delhi: DK Printworld. Lawrence, D. P. (2009). Proof of a sentient knower: Utpaladeva’s Ajaḍapramātṛsiddhi with the Vṛtti of Harabhatta Shastri. Journal of Indian Philosophy, 37, 627–653. Matilal, B. K. (1986). Perception. An essay on classical Indian theories of knowledge. Oxford: Clarendon Press.

164

ALEX WATSON

Matilal, B. K. (1989). Nyāya critique of the Buddhist doctrine of non-soul. Journal of Indian Philosophy, 17(1), 61–79. Matilal, B. K. (1994). The perception of the self in Indian tradition. In R. Ames et al. (Eds.), Self as person in Asian theory and practice (pp. 279–295). New York: SUNY Press. Mimaki, K. (1976). La Réfutation bouddhique de la permanence des choses (sthirasiddhidūṣaṇa) et la preuve de la momentanéité des choses (kṣaṇabhaṅgasiddhi). Paris: Publications de l’Institut de Civilisation Indienne 41, de Boccard. Nemec, J. (2011). The ubiquitous Śiva: Somānanda’s Śivadṛṣṭi and his tantric interlocutors. New York: Oxford University Press. Nemec, J. (2012). The two Pratyabhijñā theories of error. Journal of Indian Philosophy, 40, 225–257. Oetke, C. (1988). “Ich” und das Ich. Alt-und Neu-Indische Studien 33. Stuttgart: Franz Steiner Verlag Wiesbaden. Preisendanz, K. (1994). Studien zu Nyāyasūtra III.1 mit dem Nyāyatattvāloka Vācaspati Miśras II. Altund Neu-Indische Studien 46. Stuttgart: Franz Steiner Verlag. Rastogi, N. (1986). Theory of error according to Abhinavagupta. Journal of Indian Philosophy, 14, 1–33. Rastogi, N. (2004). Manas and Jñānendriyas in Kashmir saivism. In D. Prahladachar, M. L. N. Murthy, & R. S. Murty (Eds.), Work culture and efficiency, with special reference to Indriyas (pp. 72–116). Tirupati: Rashtriya Sanskrit Vidyapeetha, Deemed University. Ratié, I. (2006). La Mémoire et le Soi dans l’Īśvarapratyabhijñāvimarśinī d’Abhinavagupta. Indo-Iranian Journal, 49(1–2), 39–103. Ratié, I. (2007). Otherness in the Pratyabhijñā philosophy. Journal of Indian Philosophy, 35(4), 313–370. Ratié, I. (2009). Remarks on compassion and altruism in the Pratyabhijñā philosophy. Journal of Indian Philosophy, 37(4), 349–366. Ratié, I. (2010a). The dreamer and the yogin—On the relationship between Buddhist and Śaiva ­idealisms. Bulletin of the School of Oriental and African Studies, 73(3), 437–478. Ratié, I. (2010b). A five-trunked, four-tusked elephant is running in the sky—How free is imagination according to Utpaladeva and Abhinavagupta? Études Asiatiques/Asiatische Studien, 64(2), 341–385. Ratié, I. (2010c). Le non-être, une preuve de l’existence du Soi? La notion d’abhāva dans la philosophie de la Pratyabhijñā. Journal Asiatique, 298(2), 421–493. Ratié, I. (2011a). Le Soi et l’Autre. Identité, différence et altérité dans la philosophie de la Pratyabhijñā. Jerusalem studies in religion and culture 13. Leiden: Brill. Ratié, I. (2011b). Can one prove that something exists beyond consciousness? A Śaiva criticism of the Sautrāntika inference of external objects. Journal of Indian Philosophy, 39(4–5), 479–501. Sakai, M. (2010a). Dharmottaras Erklärung von Dharmakīrtis kṣaṇikatvānumāna (Pramāṇaviniścayaṭīkā zu Pramāṇaviniścaya 2 vv. 53–55 mit Prosa). Doctoral thesis, University of Vienna. Sakai, M. (2010b). Arcaṭa ni yoru Dharmakīrti no Setsunametsuronshō Kaishaku: kṛtakatvānumāṇa to ­sattvānumāna [Arcaṭa’s interpretation of Dharmakīrti’s kṣaṇikatvānumāna: kṛtakatvānumāna and ­sattvānumāna”]. Nanto Bukkyo, 93, 38–62. Sakai, M. (2011). Śākyabuddhi and Dharmottara on the inference of momentariness based on the ­absence of external causes of destruction. In H. Krasser, H. Lasic, E. Franco, & B. Kellner (Eds.), ­Religion and logic in Buddhist philosophical analysis. Proceedings of the fourth international Dharmakīrti ­conference, Vienna, August 23–27, 2005 (pp. 407–421). Wien: Österreichische Akademie der Wissenschaften. Steinkellner, E. (1963). Augenblicklichkeitsbeweise und Gottesbeweis bei Śaṅkarasvāmin. Doctoral thesis, ­University of Vienna. Steinkellner, E. (1968/1969). Die Entwicklung des kṣaṇikatvānumānam bei Dharmakīrti. Wiener Zeitschrift für die Kunde Südasiens, 12–13, 361–377. Taber, J. (1990). The Mīmāṃsā theory of self-recognition. Philosophy East and West, 40(1), 36–57. Taber, J. (2012). Uddyotakara’s defense of a self. In I. Kuznetsova, J. Ganeri & C. Ram-Prasad (Eds.), Hindu and Buddhist ideas in dialogue; self and no-self (pp. 97–114). Farnham, Surrey: Ashgate. Torella, R. (1988). A fragment of Utpaladeva’s Īśvarapratyabhijñā-vivṛti. East and West, 38, 137–174.

THE SELF AS A DYNAMIC CONSTANT

165

Torella, R. (1992).The Pratyabhijñā and the logical-epistemological school of Buddhism. In T. Goudriaan (Ed.), Ritual and speculation in early tantrism, studies in honor of André Padoux (pp. 327–345). SUNY series in tantric studies. Albany: State University of New York Press. Torella, R. (2002). Īśvarapratyabhijñākārikā of Utpaladeva with the Author’s Vṛtti, critical edition and annotated translation. [Roma, 1994], corrected edition: Delhi: Motilal Banarsidass. Torella, R. (2007a). Studies on Utpaladeva’s Īśvarapratyabhijñā-vivṛti. Part I: anupalabdhi and apoha in a Saiva Garb. In K. Preisendanz (Ed.), Expanding and merging horizons. Contributions to South Asian and crosscultural studies in commemoration of Wilhelm Halbfass (pp. 473–490). Wien: Österreichische Akademie der Wissenschaften. Torella, R. (2007b). Studies on Utpaladeva’s Īśvarapratyabhijñā-vivṛti. Part II: What is memory? In K. Klaus & J.-U. Hartmann (Eds.), Indica et Tibetica. Festschrift für Michael Hahn zum 65. Geburtstag von Freunden und Schülern überreicht. Wiener Studien zur Tibetologie und Buddhismuskunde 66 (pp. 539–563). Wien: Arbeitskreis für tibetische und buddhistische Studien Universität Wien. Torella, R. (2007c). Studies on Utpaladeva’s Īśvarapratyabhijñā-vivṛti. Part III. Can a cognition b­ ecome the object of another cognition? In D. Goodall & A. Padoux (Eds.), Mélanges tantriques à la mémoire d’Hélène Brunner (pp.  475–484). Pondichéry: Institut Français de Pondichéry/École Française ­d’Extrême-Orient, Collection Indologie 106. Torella, R. (2007d). Studies on Utpaladeva’s Īśvarapratyabhijñā-vivṛti. Part IV. Light of the subject, light of the object. In B. Kellner, H. Krasser et al. (Eds.), Pramāṇakīrtiḥ. Papers dedicated to Ernst Steinkellner on the occasion of his 70th birthday (pp. 925–940). Wiener Studien zur Tibetologie und Buddhismuskunde 70.1–2. Wien: Arbeitskreis für tibetische und buddhistische Studien Universität Wien. Torella, R. (2008). From an adversary to the main ally: The place of Bhartṛhari in the Kashmirian Śaiva Advaita. In M. Kaul & A. Aklujkar (Eds.), The grammatical traditions of Kashmir: Essays in memory of Pandit Dinanath Yaksh (pp. 508–524). Delhi: DK Printworld. Vetter, T. (2000). The ‘Khandha passages’ in the Vinayapiṭaka and the four main Nikāyas. Wien: ­Österreichische Akademie der Wissenschaften, Philosophisch-Historische Klasse, Sitzungsberichte 682. von Rospatt, A. (1995). The Buddhist doctrine of momentariness: A survey of the origins and early phase of this doctrine up to Vasubandhu. Alt-und neu-indische Studien 47. Stuttgart: Franz Steiner Verlag. Watson, A. (2006). The self’s awareness of itself. Bhaṭṭa Rāmakaṇṭha’s arguments against the Buddhist doctrine of no-self. Wien: Publications of the De Nobili Research Library 32. Watson, A. (2010a). Rāmakaṇṭha’s concept of unchanging cognition (nityajñāna): Influence from ­Buddhism, Sāṃkhya,Vedānta. In J. Bronkhorst & K. Preisendanz (Eds.), From Vasubandhu to Caitanya. Studies in Indian philosophy and its textual history. Papers of the 12th world Sanskrit conference (Vol. 10.1, pp. 79–120). Delhi: Motilal Banarsidass. Watson, A. (2010b). Bhaṭṭa Rāmakaṇṭha’s elaboration of self-awareness (svasaṃvedana), and how it differs from Dharmakīrti’s exposition of the concept. Journal of Indian Philosophy, 38, 297–321. Watson, A. (2013). Light as an analogy for cognition in Buddhist idealism (vijñānavāda). Journal of Indian Philosophy. doi: 10.1007s10781-013-9192–5. Watson, A. (forthcoming a). Two streams of Mīmāṃsaka thinking concerning how the self is perceived. In J. Ganeri (Ed.), Oxford Handbook of Indian Philosophy. Oxford: OUP. Watson, A. (forthcoming b). Self or no-self? The Buddhist–Brahmanical Ātman debate. In J.Tuske (Ed.), Indian Metaphysics and Epistemology. London: Continuum. Watson, A., Goodall, D., & Sarma, S. L. P. A. (2013). An enquiry into the nature of liberation. Critical edition and first translation of the Paramokṣanirāsakārikā and vṛtti. Watson, A., & Kataoka, K. (2010). Bhaṭṭa Jayanta’s refutation of the Yogācāra Buddhist doctrine of ­Vijñānavāda: Annotated translation and analysis. South Asian Classical Studies, 5, 285–352. Wezler, A. (1985). A note on Mahābhāṣya II 366.26: Guṇasaṃdrāvo Dravyam [Studies on Mallavādin’s Dvādaśāranayacakra II]. In Buddhism and its relation to other religions. Essays in honour of Dr. Shozen Kumoi on his seventieth birthday (pp. 1–33). Kyoto: Heirakuji Shoten. Yoshimizu, C. (1999). The development of sattvānumāna from the refutation of a permanent existent in the Sautrāntika tradition. Wiener Zeitschrift für die Kunde Südasiens, 43, 231–254.

CHAPTER

10

Arguing from synthesis to the self Utpaladeva and Abhinavagupta respond to Buddhist No-selfism Arindam Chakrabarti

Source: Irina Kuznetsova, Jonardon Ganeri and C. Ram-Prasad (eds), Hindu and Buddhist Ideas in Dialogue: Self and No-Self (London: Ashgate, 2012), pp. 199–216. Not just the effect–cause relationship, or the process of remembering, or the cancellation (exposure of erroneousness) of one cognition by another which are common to all practices by which ordinary people’s lives go on, even various specific practices such as buying and selling (mixed with defilement), or the interaction between a religious instruction and an instructed (unmixed with defilement), presuppose the existence of a single knower; for, after all, it is ­synthesis (samanvaya) which is the life-breath of any ­practice whatsoever.1 INTRODUCTION: ABHINAVAGUPTA’S AGENDA

Buddhist No-selfism is clearly not a descriptive metaphysics. It does not claim to unearth how we actually think (or are conceptually constrained to think) of ourselves at a particular moment of time or at the time of recalling a past experience or planning a future activity. Its stance about the subject’s ‘really being no one’ flaunts its revisionary character by committing nearly all sectors of linguistic and practical life – including the practice of Buddhist pedagogy and debates or interactions with others – to a series of edifying or not-so-edifying errors. But Abhinavagupta shows us that even errors and exposures of errors of any kind would not be possible without a connecting and enduring self. It is an argument that goes deeper than the following one-liner by P.T. Geach: ‘If only a self can have illusions, then it cannot be an illusion that a self exists.’2 In this chapter, after considering briefly the Kashmir Śaiva rehearsal of the Buddhist anti-self arguments (found in ch.  1.2 Īśvara Pratyabhijñā Vimarśinī), I try to reconstruct ­Abhinavagupta’s central counter-arguments – from memory, from non-apprehension, 166

ARGUING FROM SYNTHESIS TO THE SELF

167

from error, and from cancellation of error – to establish the existence of a self which not only bridges different times, but also different persons. The fundamental idea behind all these ‘self ’-establishing arguments is the need to explain the possibility of synthesis, of the reflexive reconnecting of different moments of cognition. According to Utpaladeva, whose text Abhinavagupta was freely commenting upon, granted the self-confined nature of particular fleeting cognitive episodes, their grasping common or uncommon objects and comparing or contrasting their inner contents would be impossible without an overseeing and synthesising self. This self ultimately turns out not only to be the connector of a single person’s experiences at different times, but also the universal basis of i­nterpersonal and pragmatic interactions, ending up in an equally revisionary metaphysics of an all-­ inclusive synthesising self-consciousness that is unbounded by time or individuality. It is to be ‘recognised’ as the universal Subject, the only free agent and artist, the great lord who plays out within his consciousness the objective drama of plurality, temporal s­equence, and differentiation of particular persons. The Buddhist does not hold the extreme eliminativist view that the pronoun ‘I’ is ­simply meaningless. Besides his avoidance of all extreme views in treading the middle path, this kind of abolition of the use of the word ‘I’ would make it impossible for him to say ‘I have no thesis, so I have no fault,’3 or even, ritually but sincerely, ‘I take refuge in the Buddha’4 with a verb conjugated in the first-person singular. However, just because the Buddhist keeps using the first-person singular pronoun and believes that she herself – not someone else – is working towards liberation through the therapy of her own desires, the Nyāya champions of an enduring substantial selfsoul should not prematurely celebrate a victory. The Buddhist (or a present-day proto-­ Buddhist Parfitian) ‘self ’-denier is not so easily silenced.The very presupposition that only a self can have illusions is easily rejected by an ancient Buddhist or even a contemporary connectionist. A bundle of impermanent factors (skandha-s) can keep making mistakes or be in a state of false belief that it is more than a bundle of functional complex systems, just as a soul-less computer program can become delusional and produce the sentence ‘I am not a computer, I am an eternal immaterial power.’ And in verbal transactions, ‘I’ could be used by and for what in reality is a group or a community (Microsoft could tell Apple, ‘I am no worse than you.’). Further, on the more basic question of whether we can get to the structure of reality by studying meanings of words and sentences, at least the Yogācāra Buddhist exclusion (apoha)-theorists are very clear: language is no guide to reality. Even our word-occasioned determinate cognitions are, in the ultimate analysis, convenient and conventional errors, and real reality remains untouched by words. Although we shall consider at the very end an odd Mādhyamika argument against the very semantic cogency of the indexical ‘I’, the thrust of the major Buddhist ­anti-self ­arguments was not so much against the first person, but against the first person’s ­diachronic identity, against staying the same across different times. In response, non-­ Buddhist, ­especially orthodox Vedic, philosophers of ninth–tenth-century Kashmir would keep pointing out that solid perceptual and inferential evidence for the existence of an enduring self (or many such selves) is writ large over our practical, social, and mental life: In sacred rituals, in religious duties, in agriculture and commerce, people skillfully make deliberations of the form “I did this yesterday and the next day I must do this other thing”

168

ARINDAM CHAKRABARTI

only because they can re-identify themselves as the same across times. Therefore, surely, each of them is directly aware of one constant person [dhruvam puruṣam] in all the states.5

Take, for example, the very process of philosophical debate which the Buddhist is so eager to enter into. One has to listen to and make sense of the opponent’s sometimes prolonged utterances in order to respond to them. The crucial need for a unifier comes from the necessity of what Kant called ‘synthesis’ – holding together in a single self-conscious cognitive embrace what are clearly registered as discrete packets of outer and inner manifolds. Let me take a typical sentence (written at least a century later by Abhinavagupta, also in the context of proving the unity of the self from the fact of memory): Only if the external and internal objects such as blue and pleasure, etc. carried forth by the mouths of the rivers of separate episodes of awareness all flow and rest in one single great ocean of consciousness which calls itself ‘I’, can they, then, be mutually synthesised and related to each other, otherwise how can unconscious material things, or discrete awareness-­ episodes of them which remain spatio-temporally limited and insulated inside their own existence, get connected by themselves, because … if the earlier flash of awareness and the later flash of remembering were simply two events separated from one another then there never would be any recall; hence through memory one unlimited Knower-Reality is proved.6

Now, that was an awfully long sentence. Philosophers would often write or utter such long sentences, both in Kashmir and in Königsberg. Long before Abhinavagupta, Jayanta Bhaṭṭa derived an anti-Buddhist argument for the existence of an enduring self from the very experience of comprehending such a long sentence. To translate his elegant verse on this process of understanding a long sentence: The phonemes are heard in a sequence. When the traces left behind by the earlier auditory perceptions are all re-awakened at the time of hearing the last phoneme, the partitioning of the total sound-series into words is done along with the activated memory of the learnt conventional meaning of each word. Spontaneously examining the syntactic dove-tailing of the words and the semantic congruence of their meanings, the total sentence meaning is glued together. Now, all of this would be impossible to explain without postulating a single knower.7 Śabara in Mīmāṃsā and Vātsyāyana in Nyāya take this phenomenon of recognition (­pratyabhijñā) to be the strongest evidence for a continued diachronically identical

­knower-self. Recognition, by the way, is not to be confused with remembering. It is an experience of a present object as identical with something perceived in the past. RECOGNITION SUBJECTED TO MEDIEVAL BUDDHIST DIALECTIC

But the Buddhist subjects this phenomenon to his usual dilemmatic scrutiny.8 Is this ­recognition, allegedly of the form ‘This (present Object/Self) is (identical with) THAT (the past Object/Self)’, one piece of awareness or two? If it is a single awareness, then what is its instrumental cause? If it is caused by a sense organ alone, then it could not

ARGUING FROM SYNTHESIS TO THE SELF

169

possibly grasp the past, the THAT-part, for our senses can only grasp the present, and not other times. If it is caused by the memory-traces alone, then the present part will not be accessible, since memory traces are causally irrelevant to our perception of a currently given object. The possibility that the senses and the traces together would produce this single cognition is as absurd as unlike causes such as a lump of clay and a bunch of threads together producing a hybrid object like a pot-cloth! So it must be broken up into two distinct pieces of cognition with two distinct objects, neither of which can anticipate, bring back, or have any relation to the object of the other, to judge their objects to be either the same or distinct. (Let us flag and remember this point, which Utpaladeva takes from the Buddhists and then uses against them in Īśvara Pratyabhijñā Vimarśinī 1.8.3: ‘Spatially and temporally separated discrete states which are entirely self-confined cannot get synthesised or connected without a single continuous bridging awareness; how is any connecting or correlating possible otherwise?’) If the cognitions are insulated from each other, so must be their objects. Even if, for argument’s sake, we concede that an act of recognition is a single piece of awareness, what would be the nature of its object? Is the pillar or the person that is re-identified a past one or a future one or a present one? If it is a past thing, then the awareness, strictly speaking, is a memory – which, by the way, even the Naiyāyika does not trust as a proper form of knowledge (pramāṇa). If it is about the future persistence of the object, then it is almost like a wish or decision (saṃkalpa-prāyam eva tat). If the recognition is merely of the object as it stands in the present point-instant, so much for your hope that it is going to prove the existence of an enduring object continuing from the past into the future. Thus, recognition seems to behave like a strange nomadic nun who had come with the promise to prove the permanence of things, but goes away proving their impermanence and momentariness. As for the possibility that the same object could be characterised by both pastness and presentness, the Buddhist (as represented by Jayanta) argues exactly like McTaggart against the A-series: ‘is past’ and ‘is present’ are incompatible predicates, one defined in terms of the negation of the other. How can the same thing satisfy both of them? This last bit of argumentation of course has a long history going back to Vasubandhu’s refutation in the Abhidarmakośabhāṣya of one of the four Sarvāstivādin views of the reality of past and future (See AKB, kośa-sthāna 5). THE BUDDHIST ATTACK AS SUMMARISED BY UTPALA

In Īśvara Pratyabhijñā Vimarśinī 1.2, Utpala anticipates the most serious charges against the idea of a unifying, enduring self. His most intimate enemy seems to be the Yogācāra ­Buddhist whose standard anti-self arguments are broken down into five broad parts. First, there is no self because none can be perceived. All perceptions belong to either of two possible types: non-conceptual perception of essenceless particulars (­svalakṣaṇābhāsam), or conceptually articulated perception, connected to language and practical activities. It is logically impossible for an enduring self (especially one endowed with powers of knowledge, will, and action) to be evidenced by the first sort of perception. Such a perception, according to Dharmakīrti in Pramāṇavārttika, does not illuminate any object other than the perception itself, and it perishes instantly. Apart from the fact that the second kind of

170

ARINDAM CHAKRABARTI

verbalisable and conceptual perception only gives us access to imaginary abstractions and generalities (vikalpa) and the self is not supposed to be an abstract generality, the Buddhist opponent takes all day-to-day linguistic reference to the self as reducible to references to the changing body or a bodily state or a particular mental state, and not to any ‘owner’ of the body or mental states. What words can refer to is what the second kind of p­ erception can refer to. And those are never real. So no real self can be seen by any kind of perception. Second, even the alleged inference from the phenomenon of memory fails to prove the self, because the causal linking work that the self was supposed to do is in any case done by residual traces or latent dispositional impressions (saṃskāra-s). Third, even the argument from qualities to quality-possessing substance is shown to be flawed. Fourth, the activity and voluntary movement of living people from which self was inferred in other bodies, or the so-called power of action of the Kashmir Śaiva self, is shown to be an illusion because there are no actions really in the world. What appears to be a continuous motion or action of a single body or agent is nothing but the successive emergence of distinct entities in distinct yet contiguous places. Fifth, apart from causal succession, there are no other relations; hence no permanent self needs to exist in order to account for the relatedness of terms appearing as objects of awareness (‘being an object of ’, in particular, is not a real relation). The most intriguing connection to be noted is between Buddhist No-selfism and the Buddhist attack against relation as incoherent, on the grounds that a relation is at once in two terms yet it is one. Utpaladeva must have felt compelled to write an almost word-forword response to Dharmakīrti’s short text refuting relations because he found this link between denial of self and denial of the reality of relations to be irrefutably strong. On some other occasion, we need to unravel the exact link between these apparently unlinked anti-realisms of the Buddhist. In this chapter we cannot and need not rehearse Utpaladeva-Abhinavagupta’s long and complicated answer to each of the anti-self arguments summarised in this second (‘Objections from the Buddhist’) chapter of Section One on Cognition (Jñānādhikāra) of Īśvara Pratyabhijñā Vimarśinī. Instead, in order to get to the main troublesome issue of why we need a self at all to explain recognition and memory, let us go back to the standard Nyāya (not Kashmir Śaiva) way of meeting the Buddhist diatribe against ‘recognitive, recollective ­experience as a proof of the self ’. This will lead us into the most fascinating time-straddling phenomenology of the backward-looking ‘that’ which is the heart of the recall aspect of ­recognition, for example when one discovers in amazement: ‘This old man is that friend I had in childhood!’ (or when we discover that we are touching that very object which we saw some time back). Jayanta had replied to each of the above objections against recognition’s unity of c­ ontent. Is the recognition ‘This is the same pillar as the one I saw yesterday’ one ­cognition or two? The answer is clear if we just remain faithful to our introspective report. It is a single identificatory awareness, where sense organs supply one term of the identity and re-­ activated memory trace supplies the other term. As to how these disparate sources could co-operate causally, the fact that they do is experientially undeniable, unlike the implausible analogy of a pot-cloth which no one has ever perceived. If a time-straddling content is presented to all self-conscious knowers, it needs an explanation in terms of different

ARGUING FROM SYNTHESIS TO THE SELF

171

cognitive capacities working together, which is much more intuitive and unsurprising than clay and cotton threads working together to produce a single effect. As to the nature of the object of this unitary experience, the accurate description is that it is an object delimited by the present time signalled by ‘this’ and qualified by the past time signalled by the THAT (Nyāyamañjarī (1983) 2.7 [p. 333]). There is no incoherence in its being past and present at once. The same entity was present in the past and it is past in the present. When the two states of the entity at different times are attributed to it, the difference between the states and the times is very much registered, rather than forgotten. Indeed, if the tense-difference is not registered, the two times would collapse into one and the continued endurance of the entity would not be registered. The past figures in this cognition as past. One thing must be noted here. In later Nyāya, a sense organ’s direct access to a past (now perished) phase or quality of an object is explained in terms of an extraordinary connection between our senses and absent objects where another current or recalled awareness itself works as the connector (jñāna-lakṣaṇā-sannikarṣa). Recognition is one of five or six contexts (including illusory experience, significantly for a Buddhist critic, and synaesthesia) where this kind of extraordinary connection is invoked. But Jayanta does not directly talk about this sort of special link through memory here. Instead, he follows the verdict of his intuitive analysis of recognitive experience as ‘mental’ (mānasa) and comes up with this rule: That a sense organ does not grasp a past object is not its fault like an ophthalmic disease. It is incapable of catching the past time as an independent object, but it does not have any ­inherent incapacity to grasp the past as a predicate qualifying a current object which is within its range. The sense organ is not by itself capable of accessing the past, but when it is assisted by a memory-trace it becomes so capable.9

Thus recognition is established to be entirely a variety of sense perception. Jayanta does not quite explicitly accept the idea of a non-ordinary contact between the external sense-organ and a remembered past object, perhaps with the assistance of the inner sense. In the next passage he considers the possibility that recognition is an apperceptive mental awareness wherein the continuous identity of the object, and reflexively of the subject, is evidenced in direct experience. At this point, a nagging difficulty may still be raised by the Buddhist No-selfist. The past is, after all, gone and non-existent. How can it figure, even as a qualifier, in the content of a current cognition? Jayanta meets this worry with an intriguing example of counting consciousness. Suppose a glutton has been counting as he is eating some b­ erries. When he experiences ‘I have eaten one hundred berries,’ he recognises the hundredth one as hundredth only in relation – the relation of succession – to his perception of the previous 99 as previous. He does not have to see and taste all one hundred together in order to have made that perceptual counting judgement. Although those 99 berries are no longer ‘there’, they can ‘ride on the impression’ and thus the bygone past can perfume the present perception of the one hundredth berry. If, at this point, the Buddhist falls back on his general scepticism about all qualificative judgements as mere products of conceptual imagination, and not as knowledge proper (which can be of only pure given particulars),

172

ARINDAM CHAKRABARTI

Jayanta would throw up his hands and remark: ‘Great Sir, is there anything on earth that you do not denounce as a mere figment of conceptualisation? But, long live those (realists) who trust in the veridicality of predicative perceptions!’10 Not that a Nyāya realist would consider recognitive perception to be infallible. But it is not to be dismissed as error simply because it is necessarily a concept-enriched perception. If someone re-identifies someone’s hair, which has been cut off once and has grown back again, to be that very same hair, such re-identification would be mistaken. Seeing a shaven head in the interval between the earlier and later perceptions would counter or invalidate this sort of re-identification, but in the case of a person’s identity across times, perceived by recognition, there is no counter-evidence of an intervening time when no self was witnessed at all. Intervals of deep sleep and so on are not periods of awareness of absence of self, but periods of absence of all awareness. Although memory is so crucial for the Nyāya defence of a permanent self, unlike ­recognition, which is taken as a variety of determinative perception, a simple recollection of the past has always remained epistemically suspect to the Nyāya philosophers, at least since Udayana asked the rhetorical question: ‘But how can we even call any remembering “­veridical”?’11 The reason for this is a double bind they seemed to face: on the one hand they admitted that for the memory-awareness to be correct (unembellished) there should not be anything over and above the content of the past perception whose residual trace (saṃskāra) causes the memory; but on the other hand they have to deal with this element of ‘was’ or ‘pastness’ that was absent at the time of the original perception and has to be added to the memory-content in order to adjust it to the moment of recall, or else a memory of yesterday’s experience of eating an apple would figure today as ‘I am eating an apple’ when one is not eating an apple at all. This is a rather confusing mess for the Nyāya philosophers. In Īśvara Pratyabhijñā Vimarśinī I.4.2–4 Abhinavagupta directly addresses the issue of the role of felt pastness of the remembered experience at the time of ­remembering, while unpacking the rich and complex phenomenology behind the paradigmatic verbalisation of memory: ‘That pot which I saw/which was seen by me’ or simply ‘That pot’ with a temporal remoteness, non-current-ness, built into the b­ ackward-pointing ‘That’. In direct determinate perception, the external pot can occupy centre stage, the ‘I see’ can recede into the background. When this same perception is recalled, the inward ‘I see’, adjusted into ‘I saw’, has to come out saliently because remembering – what Kant ­significantly calls ‘reproductive imagination’ – is more of ­reflective self-consciousness (­vimarśa), and less prominently a presentation of an object outside, the object being absent at the time of remembering.12 How can the absent object be presented (not inferentially, but directly) correctly? That was the puzzle for much early Nyāya. Abhinavagupta’s subtle analysis here (under Īśvara Pratyabhijñā Vimarśinī 1.4.2) is beyond the scope of this chapter, but it opens up an avenue for taking memory not as an ‘error’ of confusing the absent past with the present, but as an exercise of a power of the autonomous self to bring back and ‘see as absent, feel as a “had been”’ something that is not available at the present time (the time of recall). One form of this power of bringing up the absent is memory, the other form being productive or poetic imagination: ‘Thus it is said that the connective experience of “that” is of the nature simultaneously of two opposite connecting experiences of the earlier and the later times.’13 Indeed, unless the now-ness of the current experiencing time was also somehow reflectively felt, with

ARGUING FROM SYNTHESIS TO THE SELF

173

respect to what would the then-ness (past-ness) be established? So both times, along with their mutual incompatibility, must be wrapped up together within the self ’s wilful re-­ enjoyment of its own now-absent past experiences. FROM NON-APPREHENSION TO SYNTHESIS IN RECOGNITION

This talk of absence brings me to Utpaladeva’s and Abhinavagupta’s subtle transcendental argument from the content of perception of absence to the existence of an abiding self, as against the Yogācāra Buddhist. The Kashmir Śaiva epistemologists openly acknowledge their debt to Dharmakīrti and the Buddhist theory of the reflexive nature of a­ wareness (sva-saṃvedana):‘From amidst our opponents I have to accept a lot.’14 But, using the logic of reductio ad absurdum proofs that the Buddhists used so often, Utpaladeva tries to show that if one assumes the theory of momentary self-aware states of consciousness ­positing their own internal ‘objects’ within themselves, then one would eventually have to ­reject the more fundamental Buddhist assumption that there is no unifying self behind this stream of discrete and object-discriminating cognitive states. One would thus be ­compelled to be committed to a synthesising unifying enduring self for the purpose of explaining the common practice of claiming diachronic self-identity, communication with other people and remembering other times, and so on. Since the Buddhists, somewhat like Hume, also confess to not finding any directly perceptible self or ego besides the passing mental states, the Kashmir Śaiva epistemologists – known as the school of Recognition – analyse the phenomenon of not-finding or non-apprehension (anupalabdhi) rather closely. Of course, this is intimately connected to Dharmakīrti’s theory of inference, where establishing the invariable c­ oncomitance, especially when it is a causally grounded connection, between prover-­sign and ­property-to-be-inferred runs on the basis of observing co-absences and non-­observance of one’s presence in a place where the other is absent. Before I get into the details of their argumentation, I wish to draw attention to an ­uncanny occurrence of an argument from non-apprehension in Gottlob Frege, the f­ather of twentieth-century Analytical philosophy, where also the immediate context was the proof of a self beyond the mere subjective passing ideas. Frege’s passage starts in a very Nyāya-like vein: ‘Can there be a pain without someone who has it?’15 In answer to this insistence on a substantial subject who owns the ideas but is not reducible to them, Frege imagines that the reductionist Humean would say, ‘Can I be (just a) part of the content of my consciousness while another part is, perhaps, an idea of the moon?’ This comes very close to Dharmakīrti’s account of self-less awareness episodes which, in their reflexivity, create the illusory division within themselves of one part which calls itself the grasper ‘I’ and another part that is felt to be the external object grasped! But Frege quickly ­rejects this hypothetical error-theory of ‘ego’-concoction. He has a series of interesting ­arguments which I shall not enter into here. But one of his arguments is from non-­ apprehension. Here is the crucial line: I am not my own idea and if I assert something about myself, e.g. that I do not feel any pain at this moment, then my judgment concerns something which is not a content of my consciousness, is not my idea, that is me myself.16

174

ARINDAM CHAKRABARTI

Now, to go back to Utpaladeva’s argument against the Buddhist as explicated by ­Abhinavagupta. Suppose I have an experience that there is no pot here on the floor.What is the content of my awareness? Ontologically, neither the Buddhist nor the Śaiva admit that there is an additional objective entity characterising the floor called absence-of-pot. Even if we are talking to a Nyāya-Vaiśeṣika philosopher who does posit such an additional entity, what would be the phenomenological account of what I am seeing? The empty floor, to be sure. But if the empty floor or only the floor is no different from the floor, our experience would be the same when we saw the floor with or without other things, for even there the floor is the same. At best, the bare floor is experienced as distinct from the pot, which it was even when the pot was on it. To be distinct from is not necessarily to be devoid of. What, then, makes the experience a ‘non-apprehension of the pot’ possible? Not just an apprehension of the distinction between the pot and the floor, to be sure.The absence pops up as content of cognition only when the previous experience of the floor with an additional object which obstructs partially our sensation of the floor is remembered, compared with, and missed in the current unobstructed experience of the floor. Unless we can somehow demarcate the pot-less floor from the recalled pot-ful floor, we would never have a clear perception that there is no pot on the floor. This demarcation requires a comparing and contrasting of external situations and our experiences of them. Just as in the case of recognition one has to bring back the object of past experience and assert its identity with the object of current experience, so too in non-apprehension one must recall what it would have been to encounter a floor with a pot, and then find the current experience of the bare floor to be unlike that. Then alone would the floor enter the content of the experience as not just floor-only, but as bare pot-less floor. How would a mere self-contained unowned and un-connected momentary awareness of the floor perform this feat of comparing and contrasting? That, surely, would require re-living a previous experience (with the powers of cognition, remembering, as well as the capacity to keep apart, contrast, or exclude) and recognising the different objectual feel of the current experience. Only a single cogniser that experiences, stores, recalls, and distinguishes could do this work of synthesis. There is no point insisting that the non-apprehension of the jar is an indirect inferential outcome from the ‘prover-sign’ (liṅgo): non-availability of an apprehension of the jar that is fit to be observed. Abhinavagupta shows elaborately how, first, this sort of inference would lead to an infinite regress, for non-availability itself is a kind of non-­ apprehension; and second, even inferences need a putting together of the perception of the prover-sign, a recall of the invariable concomitance between the sign and what is to be inferred, and a final unified judgement that such a pervaded sign could not be present in the absence of the property to be proved, and so on, which gets us back to the need for an abiding self. Thus, from all possible escape routes, we are forced back to the requirement of synthesis and demarcation – comparison and unification – of objects (viṣaya-melanam), which is impossible without a single thread of a self-aware self running through, but not reducible to, this passing flow of percepts and ideas. The self is most starkly not reducible to just the set of percepts or feelings when what is perceived is a mere lack of feeling, because the lack of feeling is no actual mental state or percept. Even if apprehensions can exist without an apprehender, Abhinavagupta shows us that self-conscious non-apprehension cannot.

ARGUING FROM SYNTHESIS TO THE SELF

175

What is true specifically of not apprehending a pot is also true of our negative introspective reports of not feeling any pain or, as K.C. Bhattacharya poignantly calls it, the feeling of a lack of feeling. And I would suggest that, for the sake of argument, assuming that Hume was honestly reporting his failure to find any self, we could subject even Hume’s negative claim that when he searched, he could not find any substantial self within himself to this phenomenology of non-apprehension. As a result (somewhat like the Buddhist critique of recognition I have quoted above from Jayanta Bhaṭṭa), we could remark that non-­ apprehension (of a pot, or a feeling, or an impression of a pure ego) which was invoked to prove the non-existence of the self ends up proving the opposite: the existence of a transcendental unifier of experiences. When generalised about all states of object-directed awareness, not just nonapprehension, the full chain of transcendental arguments would run roughly as follows: 1 There cannot be a state of consciousness which is not self-conscious and intentional, that is to say, which is not aware of itself as grasping a definite object (a basic ­Yogācāra assumption). 2 Self-consciously grasping a definite object – distinct from or the same as another object – cannot be explained without the capacities of demarcation (from objects of other experiences) and remembering or recognising (as the same object as of another experience). 3 Demarcation and re-identification are impossible without mutual comparing, connecting or synthesis of individual cognitive states. If the states were insulated within themselves, then they would have no inkling of the contents of one another. Mere causal ordering or impact on one another would not explain how two of them could be about the same object or recognised to be of related or distinct objects. Just being temporally ordered or causally connected would not suffice for the sparks of awareness (saṃvit) to become ‘conscious about something’: they would have to be recognitions of relations. 4 No (more than causal) connecting of momentary cognitive states is possible without a stable connector which can run through, know all about, distinguish, and hold together all the different cognitive states across times and places, hence not itself a momentary self-confined cognition or their mere collection. Therefore, to close the chain, there would be no states of consciousness unless there were a permanent ­single knower-self (and not just because states require a substratum which owns them). What if this sort of idea of the continuous self was a congenital illusion of these cognition episodes? In every case where the Buddhist resists the idea of commonness, either of a continuant across times or of a property across instances, his last resort is an error theory, a theory that diagnoses our ineliminable tendency to see sameness in difference as a ‘mistake’.The agnostic about the self superciliously turns diagnostic about the common man’s use of the concept of an abiding self. Even the above transcendental argument, they may say, rests on an innate error of positing a synthesiser of the series over and above the series. In response, Abhinavagupta would run the same transcendental argument from the very possibility of an error. Nothing counts as an error unless it could in principle be

176

ARINDAM CHAKRABARTI

detected or exposed. The fact that we can even make a correction to an error shows that in us there is more than a mere succession of discrete mental states. Suppose we call a particular perceptual experience an illusion. Could the invalidation of the previous experience (that it is a piece of silver) happen on the basis of an inference? Since at the time of the subsequent, veridical and corrective, experience (that it is merely mother of pearl) the previous cognition is gone and unavailable as a subject-term (pakṣa) of the inference (‘That cognition was erroneous because …’), any such inference would suffer from the fallacy of unestablished subject-term (Īśvara Pratyabhijñā Vimarśinī (1986) 1.7.13). Suppose, for the sake of argument, that there is simply a causally connected succession of two distinct and un-owned (Yogācāra Buddhist sort of) representational cognitive states of the form ‘That is water over there’ (C1), and then ‘There is just a dry sun-heated surface at a distance’ (C2). How is the second going to cancel the first and prove it to have been an error? If the correction or cancellation happens simply because when C2 arises C1 has perished, then every subsequent awareness will count as a correction or falsification of the previous one. But for a cognitive episode to end is not for it to be cancelled as mistaken. Nor is there any natural epistemic enmity or contrariety between a previous awareness and a later one simply because one is earlier and the other later. Unless there is a single knower who owns them both and refers them both to a single object (eka-viṣayatayā vinā), recognises that something cannot be both a pool of water and a dry heated surface at the same time, and also evaluates C2 to be more accurate than C1, detection of error would be impossible (Īśvara Pratyabhijñākārikā 1.7.6–13). Thus, any perceptual error, or any error for that matter, presupposes the existence of a non-momentary perduring self. Of course, the thread of memory runs through all of the above arguments, from ­recognition to non-apprehension to error-correction. Of the three powers enriching the self listed by Utpaladeva – the power of awareness, the power of remembering, and the power of distinguishing or excluding – it is the middle one about which Abhinavagupta announced: ‘It is in the power of remembering that the self ’s ultimate freedom consists. I am free because I remember!’17 SYNTHESIS GOES BEYOND THE INDIVIDUAL: FROM OTHER TIMES TO OTHER SUBJECTS

Holding together (samanvaya), then, is the lynchpin of the Kashmir Śaiva argument for the existence of the self (Īśvara Pratyabhijñā Vimarśinī 1.7.2–3 [p. 357], and also 1.7.13). But this unification goes beyond the embodied individual selves. Jayanta was interested only in proving a permanent individual soul, a distinct one in each body. But Abhinavagupta is not content with that alone. Once he has the transcendental argument from synthesis of cognitions to a constant unifier, he uses it to extend that unity beyond a single knower to include the consciousness of so-called other knowers (santānāntara, ­pramātrantara). A  Nyāya  metaphysician would find this outrageously over the top (­atiprasaṅga). But ­Abhinavagupta claims that he is simply generalising the same argument from synthesis or reconnecting (samanvaya or anusandhāna). Until now we seem to have proven that all that a single person comes to be conscious of must be somehow interconnected with a self-enjoying creative I-consciousness, which weaves them into some sort of oneness that is tolerant of a projected plurality of objects

ARGUING FROM SYNTHESIS TO THE SELF

177

and places and times. But what about the distinction between one knower and another, between myself and others? Abhinavagupta gives a very subtle argument to overcome even that basic otherness. First, let it be admitted that my own consciousness is known to me directly. I know what it is like to be self-aware and aware of objects. And if, say in the context of an effort at empathy with a friend, I feel acutely that I am not quite feeling this friend’s own emotions, as a missed feeling don’t I have to subjectively be aware of those emotions, in however inadequate a fashion? (Īśvara Pratyabhijñā Vimarśinī 1.1.4 [pp. 75–6]) In that sense, could not the unfelt pleasures and pains of another person need to become objects of my direct awareness as what I fail to feel, just as a remembered even is experienced by the same experiencer as what is not now happening? What is it that I ruefully miss when I confess that I cannot feel even my closest friend’s own pain as he feels it? Compare me at other times with the other self now. Just as in order to remember (and miss) my past as my past (not as my present), I had to connect and re-live the bygone experience with the present one, so too in order to think of the other person as another subject I have to connect his subjectivity with mine. Notice that even distinguishing requires connecting or relating. If A and B cannot be considered simultaneously, they cannot be distinguished, especially when A is gone whenever B comes or A has absolutely no access to B. My self-awareness manifests itself through my bodily life-activities, and I notice others’ bodily activities just as immediately as I notice my own, though there are differences of clarity of access set up by our habitual walls of individuality. Now, observable actions of sentient beings are quite distinct from mere physical movements. As Abhinavagupta remarks in Īśvara Pratyabhijñā Vivṛti Vimarśinī (1.1.5. [p. 105]), the observed going of a living being is not like the observed movement of water or the rolling of a stone; neither is the motionless sitting by a person similar to the motionlessness of a stone. It is undeniable that we notice the actions of others as movements or postures enlivened by the self-sentient ‘feels’, even if they are not ‘my feels’. Just as ‘He knows’ is said as an abbreviation of ‘He is in a position to say “I know,”’ similarly, ‘He walks’ is asserted by me to the extent that I can feel what it is for him to make himself aware that ‘I am walking.’ Thus, even others’ actions are observed (not inferred) by us to be tingling with the same subjectivity as I feel behind my voluntary actions.We must reject the suggestion that our knowledge of other minds is merely an analogical inference. The word used by Utpala in the context of our awareness of consciousness in other bodies is ūhyate (Īśvara Pratyabhijñākārikā 1.1.4). And Abhinavagupta clarifies; ‘ūhyate’ does not mean that others’ sensations are merely inferred. To do ‘ūha’ is to intuitively extrapolate or to generalise through substitution (as in the original Mīmāṃsā context of ritual substitution of objects) or to directly postulate from ‘otherwise inexplicability’ to make it highly likely. Here, the process is partly a function of our sense-organs: we see that the other is in pain, we feel their pleasure (sometimes more than at other times). Thus, the word ūha here signifies ‘direct acquaintance’.18 When we are thus sensuously aware of the power of activity in others’ bodies as something cognitive and conscious, this awareness inside others does not appear to us as a ‘this’, as a mere inert material property. To be a ‘this’ is to be not of the nature of light; a mere this obstructs the light as non-conscious mere object. Whatever the

178

ARINDAM CHAKRABARTI

modern brain-mind identity theorist may say, when I say and see that my friend is suffering or my daughter is singing happily, I do not mean thereby that she is undergoing some physical objective, even in her C-fibres or in her amygdala or somewhere else in her body. I mean exactly the same sort of thing as what I mean when I say that I am in pain or I am singing (something as subjectively feelable as that), even if I do not feel it as mine in this instance. If a state of consciousness appears as a ‘this thing out there’, then it is not appearing as a state of consciousness at all, hence it is as good as not appearing. But others’ states of consciousness are ‘seen’ in their faces and postures and felt through sympathy (to make a Wittgensteinian point, minus Wittgenstein’s allergy to the ‘inner’). Therefore, even others’ mental states appear to us as subjective, as connected to the ‘I’. The otherness only belongs to the adjuncts and dividers such as these outer bodies, but the consciousness ascribed to them, qua consciousness, rests on the I-ness of the knower-in-general, as much as my own consciousness rests on the I-ness. Self-ascription of conscious states may well be dependent upon other-ascription of those states. (I am wittingly reading Strawson into Abhinavagupta here, with no claim of an accurate interpretation of either.) Unless we are interpreters of others’ conscious behaviour – both linguistic and non-linguistic – we cannot be self-interpreters. At the very foundation of self-consciousness lie these two kinds of integrative capacities: the capacity to link, compare, and place on a par the subjectivity in others with the subjectivity in myself, and the capacity to integrate the bits of my own states over time into one centre of experience. Thus the argument from the requirement of synthesis, in Abhinavagupta, works for other people as well as for other times. Here is his final dramatic clinching of the argument: That inner vibrant active power – as self-luminous as the power of awareness …, when it is realised through the medium of another body, etc. makes its own awareness-essence known. An awareness never manifests itself as a ‘this’ (mere inert object). To be a ‘this’ is to be un-awareness, to be foreign to illumination or manifestation.Whatever manifests itself has the shape of an ‘I’. Hence, even another person’s awareness is one’s own self, the otherness is only of the adjuncts such as bodies; even that, when examined rationally, is not really outside subjective consciousness. Thus, at the level of ultimate truth, all knowers of the world are only one knower, and he alone exists.19

Besides our limited – for him, illusory – individual self, Abhinavagupta ‘recognises’ a real universal Subject, an infinite ‘I’, where all self-regarding and other-regarding concerns rest, the stopping place for all reason-seeking and all conscious craving and relishing and ­marvelling. When the universe is recognised to be contained within this non-­ individualistic (yet divinely personal) subjectivity of this ‘all that I am’, divisive egoistic craving and existential suffering cease and life becomes a perpetual festival. The process of arriving at this boundless unified Self through logical analysis of remembering, re-­ identifying, distinguishing, non-apprehension, inferring, and error-correction is rather complex. And my hunch is that the argumentation process itself is part of the meditation. After all, ­Abhinavagupta does insist (in Tantrāloka, Book 6) that good reasoning (sattarka) is the best of all methods of Yoga!

ARGUING FROM SYNTHESIS TO THE SELF

179

FROM REJECTION OF THE INDIVIDUAL INDEXICAL ‘I’ TO ‘I AM MY WORLD’

This brings me back to one Mādhyamika Buddhist argument that questions the very idea of the first-person as incoherent. Here is Āryadeva’s ingenious argument: This is also why the self, essentially, does not exist. Had the self been essentially real, then, just as it is the basis/support of one person’s ego-usage [ahaṃkāra] it would have been also the basis/support of everyone’s. Since in this world, hotness is the essential nature of fire, we do not find some fires which are not hot (but cold). Just like that, if the self essentially ­existed, then it would be the self for all, and would be the target of everyone’s use of ‘I’ (object of everyone’s ego-usage). But it is not so, hence it is said: ‘What is your self is my non-self (for me it is an other/not-self), therefore this self is not necessarily (universally, as a rule) a self.’ Isn’t it simply an imaginary super-imposition of titles (such as ‘I’,‘the self ’,‘the p­ erson’, etc.) on impermanent entities?20

If we go back to the original metaphysical formulation of the Mādhyamika Buddhist argument, it simply makes the following point. If something is real, it must be the same for all. But the so-called self is not the same for all. So it must be unreal, a figment of ­imagination. Even after we have exposed some ambiguity or even fallacy in the argument, we cannot get away from the haunting attraction of this metaphysical insight: If, objectively, there were a self, would it not be a self for everyone? Perhaps there are two distinct insights behind this. The first is a general constraint on ontological commitment: only that entity could be said to exist with a real objective ­nature, which would be recognised to be of that same nature from an impersonal view from nowhere. The second is a phenomenological observation about the sense of self: the self is only a myself from a single person’s unshareable egocentric point of view. From every other point of view, it is an other, a non-self. We could question both these initially appealing claims. It is easier to challenge the first austere criterion of objective existence. A mother does not have to be everyone’s mother in order to really be a mother! (Of course the counter-challenge could be that being the first-person is not a relational property like being a mother. Or is it?) But let us try to explore a radical rebuttal of the second, nearly universally accepted, phenomenological claim about the unshareability of the sense of self, in the light of the surprising set of arguments given by Abhinavagupta in order to establish an enduring, relation-­enabling, ­experience-reconnecting, synthesising self, against the Yogācāra Buddhist theory of a mere succession of self-aware but mutually insulated, momentary mental states. Yes, indeed, if what is the I = Self for you were totally an Other = Non-self for me, then the category of the I = Self would be metaphysically suspect. Let us grant Āryadeva that (though Jayanta or any other Naiyāyika would never concede this premise). But is it correct that your self is my non-self, that your awareness is something I have no first-person subjective access to? Could we not revisit the question, not just semantically, but phenomenologically, if your ‘I’ could be my ‘I’ too? When I directly feel your pain from inside, and even when I register my failure to enter into your experience in that

180

ARINDAM CHAKRABARTI

uniquely subjective way (Failure to experience what? Your subjective feel, right? How do I know that I am missing that?), I have to have an identificatory absence-­recognising inkling of that common (but distinction-tolerant) same subjecthood, the same I-ness that you, I, God – or, according to Abhinavagupta, even a worm – feels by virtue of ­simply feeling alive. (The intricate argumentation from recognition of alive-ness in others to the recognition of their selfhood is rehearsed in Īśvara Pratyabhijñā Vivṛti Vimarśinī 1.1.5. [vol. I, pp. 104–9].) For Abhinavagupta, to think of consciousness or cognition is to think of the ‘I’ (­nothing to do with the limited ego). This ‘I’ is the same for me and you and others. Its very all-­ encompassing relation-making self-distinguishing reflexive phenomenal character resists genuine multiplicity or total otherness. Therefore, this all-unifying universal Self exists objectively – that is, for all of us – but not as an object. All of us and all objects that we re-identify and co-perceive exist within this self-conscious Self. Indeed, in all its glory and playfulness of projecting numerous others within and against itself, this universal consciousness makes a small ‘I’ of each arbitrarily and magically constructed embodied knower, and a ‘You’ of all the rest. Pretending that all those others are mere ‘those living things over there’ and then re-identifying them back in oneself, through aesthetic and spiritual experience, this all-encompassing subjectivity may be all that there really is. All the knowers of the world unite in this single wavy ocean of self-awareness. This metaphysics of Kashmir Shaivism may be just as revisionary as Buddhist No-­ selfism, extreme denial of the first-person being responded to by an equally extreme all-embracing affirmation. But it is a fascinating philosophical example of using radical cognitive atomism – where everything falls apart into linkless windowless particulars – to swing back dialectically to an opposite extreme of cognitive holism, binding everything with a single thread of synthesis.21 NOTES

1. 2. 3. 4. 5. 6. 7. 8.

9. 10. 11. 12. 13. 14.

Īśvara Pratyabhijñā Vimarśinī 1.7.13–14. Geach (1979: 38). ‘nāsti me pratijñā, tato nāsti me doṣaḥ’; Vigrahavyāvartanī by Nāgārjuna (2002) 2.29. ‘Buddham śaranam gacchāmi’. Nyāyaniañjarī (1983) Part 2, p. 295. Īśvara Pratyabhijñā Vimarśinī 1.7. 2–5. Nyāyamañjarī (1983) Part 2, p. 295. The following diatribe is my free translation/summary of Jayanta Bhaṭṭa’s dramatisation of standard Buddhist polemic against the Nyāya argument from recognition that seeks to ­establish the diachronic identity of a permanent self. See ibid.: 308–35. Ibid.: 334. Ibid.: 332. Quoted by Gaṅgeśa in the last sentence of Savikalpakavāda of Tattvacintāmaṇi, Part One on perception. See the translation in Phillips (2004). See Torella (1994: 106, fn. 12). ‘viruddha pūrvāpara parāmarśa svabhāva eva “sa” iti parāmarśa ucyate’; Īśvara Pratyabhijñā Vimarśinī 1.4.1 [p. 153]. ‘pūrvapakṣamadhyāt mayā bahu angīkartavyam’; ibid.: 1.3.1.

ARGUING FROM SYNTHESIS TO THE SELF

181

15. Strawson (1967: 33). 16. Ibid. 17. ‘smaraṇa śaktir eva hi paramam svātantryam’; Īśvara Pratyabhijñā Vivṛti Vimarśinī 1.4.1 [vol. 2, p. 4]. 18. ‘atra aṃśe indriyavayaparaṇam api asti … tataś ca sākṣātkaram upalakṣayati “uhaḥ”’; Īśvara Pratyabhijñā Vivṛti Vimarśinī 1.1.5 [vol. 1, p. 101]. 19. ‘āntarī kriyāśaktiḥ sa ca para-śarīrādisāhityena avagatā svaṃ svabhāvaṃ jñānātmakam ­avagamayati, no ca jñānam idantayā bhāti … bhāti ca yat tad eva aham ity asya vapuḥ iti para-jñānam svātmā eva … viśvaḥ pramātṛvargaḥ paramārthataḥ ekaḥ pramātā, sa eva ca asti’; Īśvara Pratyabhijñā Vivṛti Vimarśinī 1.1.4 [vol. 1, p. 76]. 20. Catuḥśataka (1974) 10.3. 21. My understanding of the immensely complex argumentation in Īśvara Pratyabhijñā Vimarśinī and Īśvara Pratyabhijñā Vivṛti Vimarśinī is still in the making. But whatever clarity I have reached after my own struggle with the texts is partly thanks to reading Prof. R ­ affaele ­Torella’s excellent book and expository papers on this subject and to discussion, with d­ ecades-long gaps, with Prof. Navjivan Rastogi. BIBLIOGRAPHY Sanskrit works Abhidharmakōśa and Abhidharmakōśabhāṣya of Vasubandhu, Pradhan, P. and Haldar, A. (eds) (1975), ­Abhidharma-kōśa with Abhidharma-kōśa-bhāṣya (2nd edn), Patna: K.P. Jayaswal Research Institute. Catuḥśataka of Āryadeva, Bhattacharya,V. (ed.) (1974), Catuḥśataka by Āryadeva, Calcutta:Visva Bharati. Īśvara Pratyabhijñākārikā, Torella, R. (1994), The Īśvara Pratyabhijñākārikā of Utpaladeva with the ­Author’s Vṛtti, critical edn and annotated tr., Rome: Istituto Italiano Per il Medio ed Estremo Oriente (­reprinted 2002, Delhi: Motilal Banarsidass). Īśvara Pratyabhijñā Vimarśinī, Pandey, K.C. and Iyer, K.A.S. (eds) (1986), Īśvara Pratyabhijñā Vimarśinī of Abhinavagupta: Doctrine of Divine Recognition, vol. I, Delhi: Motilal Banarsidass. Īśvara Pratyabhijñā Vivṛti Vimarśinī, Shastri, P.M.K. (ed.) (1938) Īśvara Pratyabhijñā Vivṛti Vimarśinī, 3 vols, Bombay: KSTS (reprinted 1987 Delhi: Akay Book Corporation). Nyāyamañjarī of Jayanta Bhaṭṭa,Varadacharya, K.S. (ed.) (1983), Nyāya-Mañhjarī, with the editor’s ṭippaṇī, 2 vols, Mysore: Oriental Research Institute. Vigrahavyāvartanī, Bhattacharya, K., Johnston, E.H. and Kunst, A. (eds) (2002), Delhi: Motilal Banarsidass.

Non-Sanskrit works Geach, P.T. (1979), Truth, Love and Immortality, Berkeley, CA: University of California Press. Phillips, S. (2004), Epistemology of Perception: Gañgeśa’s Tattvacintāmaṇī, Jewel of Reflection on the Truth (about Epistemology): The Perception Chapter (Pratyakṣa-khaṇḍa), New York: American Institute of Buddhist Studies. Strawson, P.F. (ed.) (1967), Philosophical Logic, Oxford: Oxford University Press.

CHAPTER

11

‘I am of the nature of seeing’ Phenomenological reflections on the Indian notion of witness-consciousness Wolfgang Fasching

Source: Mark Siderits, Evan Thompson and Dan Zahavi (eds), Self, No Self? (Oxford: ­Oxford University Press, 2010), pp. 193–216. 1. INTRODUCTION

Irrespective of the often considerable differences between their metaphysical doctrines, many of the major philosophical schools of India agree in their basic assumption that, in order to become aware of one’s own true nature, one has to inhibit one’s self-consciousness in the usual sense, namely one’s ‘ego-sense’ (ahaṃkāra, literally ‘I-maker’).The normal way we are aware of ourselves—that is, our self-awareness as a distinct psychophysical entity with particular characteristics and abilities, formed by a personal history, standing in manifold relations to other things and persons, etc.—is in this view really the construction of a pseudo-self that obscures what we really are. One has to come to realize with regard to all aspects of one’s personality that ‘this is not mine; this am not I; this is not the Self of me’, as the Buddha puts it (Saṃyutta Nikāya XXII.59, Rhys Davids/Woodward 1972–79, vol. III: 60) and as, for example, Advaitins and proponents of classical Yoga could affirm without reservation. Yet, whilst for Buddhism this means that the spiritual aim is to realize that it is an ­illusion that something like a self exists at all, for ‘orthodox’ schools such as Advaita Vedānta or Sāṃkhya and Yoga, liberation lies, on the contrary, in becoming aware of the true self (ātman or puruṣa). In this paper, I would like to cast, from a phenomenological point of view, some ­reflections on what this overcoming of the ego-sense strived for by these traditions could possibly mean, and will try to vindicate the view of Advaita Vedānta that it does not amount to a dissolution of oneself into a mere flux of substrate-less transient phenomena, but rather to a realization of one’s self as something that changelessly underlies this flux.1 182

WITNESS-CONSCIOUSNESS

183

This ‘self ’ is of course radically different from what we normally experience as ‘­ourselves’: It has no qualities at all, can never become an object of consciousness (but is nonetheless immediately self-revealed), is identical neither to the body nor to the mind (­qua mental goings-on we can introspectively observe), and neither does, nor wants, anything. What should this be? It is characterized as the ‘seer’ (draṣṭā) or ‘witness’ (sākṣin)—that is, as that which sees (that which is conscious). Yet this is not supposed to mean that the self is a ‘something’ that performs the seeing or is in a state of seeing: Rather, it is, as Advaita Vedānta (just as, e.g. Sāṃkhya) stresses, in explicit contrast to Nyāya and Vaiśeṣika, nothing but seeing (consciousness) itself. ‘The perceiver’, as, for example, the classical Advaitin Śaṅkara says, ‘is indeed nothing but eternal perception. And it is not [right] that perception and perceiver are different’ (Upadeśasāhasrī II.2.79, Mayeda 1992: 241; addition in brackets by Mayeda). Sākṣin is, as Tara Chatterjee formulates, ‘the never-to-be-objectified principle of awareness present in every individual’ (Chatterjee 1982: 341). So the claim against the Buddhists is not that there has to be some entity in addition to, and behind or beyond, our experiential life as its substrate, but that there is a stable element within it—yet not as some invariant content or content-constellation we could experience (such a thing is indeed not to be found), but as the very process of experiencing itself, as the permanence of ‘witnessing’, in which everything we experience has its being-experienced, and which is the constant ground of our own being. It is this notion of witness-consciousness that I wish to make some sense of in the following.2 2.  SELF VS NO-SELF: THE BUDDHIST CHALLENGE

The central question of Advaita Vedānta is that of the nature of one’s own self as the subject of experience. I evidently have manifold constantly changing experiences at each moment, and it is no big problem to observe them introspectively; but who am I who has all these successive experiences? It is the nature of this experiencer of the experiences that the whole thinking of Advaita revolves around—not in the sense of some reputed ‘­experience-producer’ (so that today one could be tempted e.g. to assume the brain is the ‘true self ’), but in the sense of a subject-‘I’ as belonging to the nature of experiencing as such, however it may causally come about. Buddhism famously denies the existence of such an experiencing ‘I’. In Saṃyutta Nikāya XII. 12, for example, the Buddha answers the question of who it is who feels by ­saying: ‘Not a fit question … I am not saying [someone] feels. And I not saying so, if you were to ask thus: “Conditioned now by what, lord, is feeling?” this were a fit question’ (Rhys Davids/ Woodward 1972–79, vol. II: 10; bracketed addition by the translators). So, in the ­Buddhist perspective, the mental life is to be characterized as a flux of permanently changing ­substrate-less mental events, each caused by some other, previous event, rather than in terms of a persisting experiencing self (an ātman). Experiences take place, but there is no one who experiences them. It goes without saying that in the various schools of Buddhism the anātman doctrine has seen numerous interpretations (not all implying an outright denial of the existence of a self;3 indeed, in Mahāyāna and Tibetan Buddhism one can find views that are quite compatible with the Advaitic concept of witness-consciousness4). However, for the sake of contrast I here construe the no-self thesis primarily in the sense of a strictly reductionist theory, as espoused by the Abhidharma schools. Even in this reading, the denial of the

184

WOLFGANG FASCHING

existence of an experiencing subject is not meant to deny, at least on a conventional level, the existence of something like a unitary ‘person’ (pudgala), just as Buddhists would not deny that there are chairs or states. Yet a chair is wholly constituted by its parts and the way they are assembled, and is nothing over and above this, and similarly, the existence of a person does not involve the existence of a self over and above the manifold ephemeral phenomena that form, if sufficiently integrated, what we call ‘one person’. A person is, in the Buddhist view, nothing but a certain ‘psychophysical complex’, that is, an ‘appropriately organized collection of skandhas’5 (MacKenzie 2008: 252). A person in this sense ‘has’ her experiences only in the sense that a whole ‘has’ parts, and not in the sense of some self-identical ‘I’-core as the ‘bearer’ of its experiences, that is, as an experiencer.6 The account of persons in classical Indian Buddhist Abhidharma texts is, in its r­ejection of a substantial self, quite in accord with the (at least by implication) dominant modern Western view on this topic (cf. Siderits 2003): it corresponds to what Derek Parfit calls the ‘reductionist view of personal identity’, that is, the thesis that a person’s ­enduring ­existence consists in (and is therefore reducible to) more fundamental facts, namely ­certain relations of connectedness and continuity between physical and mental events (Parfit 1987: 210–214)—and that in no way is the transtemporal identity of a person due to the continued existence of something like a ‘self ’ as a ‘separately existing entity’ (ibid.: 210). The many experiences of one person are not unified by each being connected to one enduring subject, but by being connected with one another, and the very ‘oneness’ of the subject is, the other way around, constituted by this (longitudinal) unification of the experiences. This sounds plausible enough: What should there be in addition to the physical and mental events and their interrelations? What else should a person be but a ­‘psychophysical complex’ of some sort? Nevertheless the ‘orthodox’ schools of India vehemently ­challenge the Buddhist anātman (‘no-self ’) thesis, and insist on precisely what the Buddhists r­ eject: that there is more to the existence of a person than this complex of skandhas, that there ­exists a ‘self ’ in addition to the body and the experiences, which is the ‘who’ of experiencing. Is this more than just a dogmatic assumption? Can anything be said in favor of this view? I think, on closer consideration, one has indeed to admit that it is hard to avoid ­feeling a certain unease about a purely ‘selfless’ account of one’s own existence. Is it ­really true that there is no experiencing ‘I’? Are there really only experiences, but no one who experiences them? Undeniably there seems to be a clear difference between an experience being experienced by me and an experience not being experienced by me. Speaking only of mental events, connected by some interrelations on the basis of which a permanent ‘I’ is constructed, deals with experiences more or less as if they were just objective occurrences, without taking their subjective mode of being—their ‘first-person ontology’ (Searle 1992: 16)—sufficiently into account. After all, experiences do not just lie about like stones or chairs, equally accessible in principle to everyone: Experiences only exist in being subjectively experienced, and that seems to mean: in being experienced by a respective subject. And obviously, all of my experiences, no matter how different they may be, have this one thing in common: that I experience them. In this sense, experiences are not thinkable as being ‘ownerless’: they are essentially experiences of an experiencing ‘I’. And the big question of Advaita Vedānta is precisely what this ‘I’ that experiences its experiences (this ‘first person’ of their ‘first-person ontology’) is.

WITNESS-CONSCIOUSNESS

185

Yet, the anātmavādin (denier of a self) might reply that even if one concedes this subjective character of experience, this does not at all necessitate positing an additionally ­existing subject. Rather, the subject searched for (the ‘experiencer’) is simply the experience itself and not something ‘behind’ it (cf. e.g. Strawson 2003). The experiences, as the taking place of subjective appearance (as ‘events of subjectivity’, as Strawson puts it: 2003: 304), constitute the respective ‘inner dimension’ of a subject, and are therefore not ‘had’ by an additionally existing self. This is indeed the position advanced by Yogācāra Buddhism and the school of Dignāga: This line of Buddhist thought expressly acknowledges the subjectivity (the  being-­ subjectively-experienced) of experience, but rejects interpreting this fact as the ­experience’s being experienced by a subject—rather it is supposed to refer to its self-givenness (svasaṃvedana) belonging to the very nature of experience:7 An experience, in revealing its object, is simultaneously revealing itself, ‘self-illuminating’ (svaprakāśa) (just as a lamp does not need to be illumined by a second lamp in order to be visible). ‘Svasaṃvedana thus provides a continuous, immediate, and internal first-person perspective on one’s own stream of experience’ (MacKenzie 2008: 249), without presupposing a ‘first person’ in addition to experience itself. The stream of experience is given to itself and not to a self.8 Of course I do not experience myself qua experiencer as just being the present ­experience experiencing itself, but as someone who, as one and the same, lives through permanently changing experiences, and hence is to be distinguished from them. Yet for the Buddhist/reductionist account, this apparent diachronic identity of the subject is wholly constituted by relations between the experiences (most prominently memory-­ relations): The experiential life of a person is, in this view, a series of causally connected mental events without any underlying enduring self, and an important part of the relevant causal connections that constitute the unity of one person is that the contents of one experience leave memory-traces in the succeeding one. Nothing more (especially not an enduring self) is necessary to account for my remembering ‘my’ past experiences (and hence my experience of my continued existence) (cf. Dreyfus this volume p. 133; Siderits this volume pp. 314–15; Watson 2006: 153–165). It is true that I do not just remember that, anonymously, experiences have occurred, but my past experiencing them9—but this is simply due to the fact that the very meaning of the sameness of the self, of ‘one person’, is co-constituted by these very memory-connections (cf. Siderits 2003: 25): I remember my experiences as mine not because I remember my ‘I’ experiencing them, but because they are mine precisely insofar as I can remember them (i.e. insofar as they stand in the right form of causal connection to my present experience). Advaita Vedānta, in contrast, insists that the subjectivity of experience refers to an ­experiencing subject. Just like Yogācāra Buddhism, it rejects the Nyāya thesis that an ­experience of an object only becomes itself manifest by becoming the object of another, subsequent experience (comparable to modern ‘higher-order representation theories’): rather, for an experience, to be means to be conscious.10But at the same time they reject the Yogācāra idea that it is each experience that is conscious of itself, ‘self-illuminating’ (svaprakāśa)—rather I, the subject, am immediately aware of my experiences as they come and go (cf. Timalsina 2009: 20–21).11 For example, if I am in a melancholy mood, this mood is not conscious of itself—for Vedānta this does not make much sense—rather the mood exists in virtue of my experiencing it (cf. Chatterjee 1982: 343).

186

WOLFGANG FASCHING

And indeed, one might question whether it is really sufficient to account for the subjectivity of experience—its being-experienced-by-me (respectively)—in terms of its phenomenal self-givenness (svasaṃvedana), as ‘an awareness of what one’s experience is like both in the sense of how the experience represents its object and how it feels to undergo the experience’ (MacKenzie 2008: 249). The question is for whom there is something it is like to be in a particular mental state. And it is far from clear that it really makes much sense to say that it is for the mental state itself to be (in) this state. This ‘who’ of experiencing is an additional fact with regard to the experience and its phenomenal character: No facts whatsoever about an experience or its ‘what-it-feelslike-ness’ can ever imply its being experienced by me (except, precisely, that it is I who experiences it). It appears to be perfectly conceivable that this very experience with all its relations to other experiences of the same stream of consciousness, to this body and to the rest of the world, could have existed without the ‘I’ which experiences it being me. This seems to be a contingent (and even, as Thomas Nagel states, ‘outlandish’ (Nagel 1986: 55)) fact (as I argue in Fasching 2009; cf. also Madell 1981; Klawonn 1987). This quite enigmatic additionality of the being-experienced-by-me with regard to all other properties of an experience, changes, I believe, the perspective on the question of the diachronic unity of the subject. It seems that what happened once can happen again: that an experience happens as the taking place of me. This refers to something radically ­different from the question of whether there are experiences that are connected to, or continuous with, my present one. When I ask whether I will still exist tomorrow, I do not ask whether there will be experiences that, for example, have a first-personal access to my present one. I do not refer to any aspects of the contents of some experiences in the future at all, but simply and irreducibly to the question of whether these experiences will be experienced by me.12 And this seems to be logically compatible with a complete loss of memory or any other kind of psychological change (cf. Williams 1973). 3.  SELF AS CONSCIOUSNESS

What, then, is this ‘me’? Interestingly, for Advaita Vedānta13 the true ‘I’ is in no way some trans-experiential entity (as is the view of Nyāya and Vaiśeṣika), but is in a certain sense nothing but experience itself. For Advaita, ‘the self is the object-experiencing …, i.e., ‘experiencing of something’, and is not only becoming manifest in it as something which stands, as it were, behind or beyond it’ (Hacker 1978: 275). So in this view experience does not take place for a subject, but simply as the subject. Where, then, is the dissent from Buddhism and its rejection of an experiencing self in addition to experience? The crucial difference is that ‘experience’ is meant here in the sense of consciousness (cit or caitanya), which in Advaita Vedānta is strictly distinguished from the mind (in the sense of the changing mental states). When, for example, Advaitins speak of jñāna (‘cognition’ or, in the terminology of this paper, ‘experience’) as being the essence of the self, they expressly distinguish it from what they call the vṛtti-jñānas, that is, the manifold transient mental states (Chatterjee 1982: 342; cf. also Hiriyanna 1956: 344 and Timalsina 2009: 17).14 So, in Advaita Vedānta, consciousness is not equated with the single ephemeral ­experiences or with some property of them. Rather, it is understood as something

WITNESS-CONSCIOUSNESS

187

that abides as that wherein the coming and going experiences have their manifestation (being-­experienced). Consciousness is, so to speak, the witnessing (experiencing) of the experiences, and while the experiences change, experiencing itself abides. After all, the succession of the experiences consists precisely in one experience after another becoming experientially present, which presence as such therefore does not change. Just like Yogācāra Buddhism, Advaita Vedānta espouses the idea of the ‘self-­luminosity’ (svaprakāśatva) of experience—yet not as a feature of the individual mental states— these are things that become manifest in experience (qua consciousness)—but rather of ­consciousness itself (cf. Chatterjee 1982: 342–344, 349).15 Consciousness, like light, is the medium of visibility of all things and does not have to be illuminated by another light (i.e. become the object of consciousness) in order to be revealed—it is the shining itself as the principle of revealedness.16 Light is not visible in the way illuminated objects are, but at the same time, it is not concealed:17 It is present, and it is precisely its presence that is the medium of the presence of everything—first and foremost of the experiences whose very existence consists in their being-present.18 Hence for the Advaitins—although they hold that mental states are manifest essentially, and not by virtue of being the object of some further, higher-order mental states—it is not adequate to say that they are immediately self-aware. Rather, they exist in manifesting themselves in the medium of the luminosity of consciousness, which is i­mmediately self-­ revealed. Experiences have their very being in their being-consciously-present (in ­being manifest in ‘primary presence’, as Erich Klawonn (1987, 1998) calls it), and while these ­experiences are permanently fleeting, conscious presence as such abides (Klawonn 1998: 59; cf. Zahavi 1999: 80; Zahavi this volume: p. 59). So, in this view, the manifold transient experiences have their manifestation in one consciousness. Yet why should we assume this? Why should we draw a distinction between the individual experiences and consciousness, thereby obviously hypostasizing consciousness into a ‘something’ in addition to experience? Why should we assume an irreducible sameness of consciousness, if, quite evidently, constantly new consciousness-events are transpiring? Conscious experiences admittedly share the feature of being conscious, but it seems to be an obvious fallacy to speak here of something like a persisting consciousness-­ entity. So is there any justifiable sense in which consciousness is to be distinguished from the individual experiences and in which a multitude of experiences can be the taking place of the same consciousness? I think there is. Already in a purely synchronic perspective, consciousness comprises many experiences, that is, I am actually seeing, hearing, thinking, etc. manifold things at the same time.The question is how one should account for this oneness of the experiencing ‘I’ across its manifold simultaneous experiences (i.e., what binds these experiences together as ‘mine’). Naturally, the reductionist cannot explain the synchronic unity of experiences by their being experienced by one subject (by me). For, in her view, there is simply no subject one could presuppose as explanans; rather, it’s the other way around: just like diachronic unity, the unity of being-experienced-by-one-subject at a time is to be explained by the being-unified of the experiences by unity relations that hold between them. Yet what sort of relations could these be? One must not forget that it is not just any relation, any unity between experiential contents that is at stake here, but the unity of being-present-in-one-consciousness. Certain experiential contents can be more strongly

188

WOLFGANG FASCHING

associated than others, and thereby bound together to form experiential ‘fields’ (experiential unities) in contrast to a background; they can be coordinated as constituting one coherent space, and the like—but all such relations that might bind together experiences into ‘total experiences’ actually presuppose their being-present-together (cf. Dainton 2006: 240–244). Only what is co-present in this sense can be associated. And this presence is nothing other than the being-experienced of the experiences in which, in the sense of their ‘first-person ontology’, the experiences have their being. So they do not exist and additionally become somehow unified. Rather, it is their very being (namely their being-experienced) wherein they have their unity. One could counter that it was inadequate to speak of many simultaneous experiences in the first place. Rather, it is one total experience with an inner complexity.19 But the crucial question is precisely wherein the unity of this ‘one total experience’ lies. Nothing on the content-side can do this job. So one obviously has to distinguish between the one experience and the many experiential contents that manifest themselves within it. And if one wishes to call the latter ‘experiences’, it is important to understand that the one experience is not a sum or a composition of these many experiences (qua experiential contents), but rather it is ‘experience’ in the sense of the experiencing of the experiences (cf. Zahavi 1999: 80): that wherein they have their being-experienced, their primary presence—quite in the sense of the Advaitic notion of jñāna (or sākṣi-jñāna, as it is occasionally called) in contrast to the vṛtti-jñānas. So when we speak of many simultaneous experiences, their difference lies in what is present, not in presence itself. This, I would suggest, is how the talk of ‘witnessing’ in Advaita Vedānta should be interpreted: We stated that the ‘witness’ (sākṣin) is not understood as an observing entity standing opposed to what it observes, but as the very taking place of ‘witnessing’ itself, and ‘witnessing’ is nothing other than the taking place of the experiential presence of the experiences, in which the experiences have their very being-experienced and thereby their existence. In this sense, consciousness can be understood as the existence-dimension of the experiences (cf. Klawonn 1987; Zahavi 2005: 131–132; Zahavi this volume: p. 58; Fasching 2009: 142–144). A dimension comprises a multitude of elements that stand in manifold relations to each other, yet it is not the sum of these elements or a result of their interrelations, but what makes them, together with all their relations, possible in the first place. In this sense, ‘the self ’ qua consciousness is to be distinguished from its experiences, but not as a ‘separately existing entity’—just as space is not a separately existing entity in addition to the spatial objects, yet also not identical to them or reducible to their relations (since any spatial relations presuppose space).20 So the unity of being-experienced-together is irreducible to the many experiences and their relations, being rather that wherein they have their being, and this is nothing other than what Advaita Vedānta calls the ātman as ‘the immediately co-experienced unity of experiencing’ (as Paul Hacker characterizes Advaita’s ātman, using a formulation of ­Scheler’s about the ‘person’: Hacker 1978: 274; cf. ibid.: 275). When Advaita Vedānta equates the self with consciousness, this is not supposed to mean that the subject is composed of the many contents of consciousness. I qua consciousness am not an agglomeration of phenomenal contents, properly organized, but rather their thereness, their presence (and that is the one presence of the manifold contents).21

WITNESS-CONSCIOUSNESS

189

Now the question is: What is the nature of the temporal abiding of experiential presence through the permanent succession of experiences? Does a new presence with new contents not take place each moment? Is there a succession of presences together with the succession of contents (after all, the presence-of-this now and the presence-of-that then are obviously different presence-events)? Or is it, rather, not one and the same consciousness, in which the experiences have their coming and going? In other words: Can two presence-events at different times be the taking place of the same presence, that is, is there an irreducible sense in which two such presence-events can be the taking place of (one and the same) me? Vedānta insists that what changes when one experience follows the other (presence-of-this being succeeded by presence-of-that), are actually the contents of consciousness, not consciousness itself (cf. Sinha 1954: 329). And indeed, as soon as one distinguishes consciousness from the experiences, the assumption that the diachronic identity of consciousness has to consist in unity relations between the experiences appears less compelling. And if one takes a closer look at the nature of the presence of the momentary experience, it becomes outright implausible:The ‘primary presence’ (the current being-experienced) of an experience always and essentially is the presence of the temporal streaming of experience transpiring right now. And that means: presence is irreducibly presence of the current taking place of temporal transition. (Otherwise no time-experience, no experience of change and persistence, would ever arise.) So the indubitable evidence of my experiences in their very being-experienced is always their evidence as passing the thereby ‘abiding dimension of first-personal ­experiencing’ (Zahavi 2005: 131). And, therefore, the absolute evidence of my present existence is the evidence of my present living through these streaming experiences. The being-­experienced of the Streaming experiences as streaming implies the permanence of the actuality of experiencing itself, which is the being of my ‘I’.22 Therefore I, qua consciousness, am not the passing experiences, but rather their manifestation as passing, which does not pass with them: the abiding experiencing of the changing experiences (Fasching 2009: 144–145).23 So the question of whether the subject is something that can exist, in an irreducible sense, as one and the same at different times, must, I believe, be answered in the affirmative: It only exists as now-transcending from the start; in contrast to the fleeting experiences it abides as the presence of the streaming experiences as streaming. Experiences only exist in being experienced, that is, experientially present, and they are essentially present as streaming, which implies the abidance of this presence itself. This abidance cannot be constituted by relations between momentary ‘experience-stages’, because there simply are no experience-stages that would not have their primary presence as temporally passing. That is: There is no experiential evidence prior to the evidence of the ‘standing’ of the experiencing ‘I’. This abidance of the ‘I’ cannot properly be conceived of as the enduring of an object in time that derives its persistence from unity relations between its temporal stages. For presence is not so much something that takes place in the respective present, but rather it is this very present itself—not in the sense of the objective time-point that is now present and then sinks into the past, but in the sense of the presentness of the respective present moment:24 What marks a particular moment as being now is no objective feature of this

190

WOLFGANG FASCHING

special point on the timeline (cf. Nagel 1986: 57), rather it is the ‘now’ only in relation to the experiential presence of the subject (cf. Husserl 2006: 58, 390, 406). Consequently, the abiding of the ‘I’ is not so much the enduring of an inter-temporal object (with its coming and going temporal ‘object-stages’), but should rather be conceived of in terms of the ‘standing’ of the present itself, in which the very passing of time (the permanent becoming-present of ever-new time-points and object-stages) consists:25 that is, of the phenomenon that it is always now. While the temporal stages of an object one is conscious of continually sink into the past, consciousness itself does not elapse: ‘… even though the object of knowledge changes’, says Śaṅkara, ‘the knower, being in past, future, and present, does not change; for his nature is eternal presence’ (i.e. the presentness of the present) (Brahmasūtrabhāṣya II. 3.7, Deussen 1920: 389). So the evidence of the abiding of the subject is not the experience of some object-persistence, but the condition of the possibility of any experience of persistence.26 4.  THE PRESENCE OF THE WORLD AND THE SUBJECT IN THE WORLD

What Advaita Vedānta soteriologically aims at as the realization of the ‘self ’ is nothing other than becoming aware of experiential presence (consciousness) as such. So far, we have characterized this consciousness as the presence of the experiences. Yet this should not be misunderstood as meaning the presence of merely mental contents, of some subjective interiority in contrast to the outer world. Rather, consciousness exists as the presence of anything we could ever refer to, be it ‘inner’ or ‘outer’. The presence of the experiences is the presence of the world.27 The presence of sensuous contents, for example, is ipso facto the sensuous presence of the respective perceived object; my thinking is nothing but the ­successive presenting-itself of some meaning-constellation (some ‘thought’ in the ­noematic sense); the moods I live through are aspects of the way the world is there for me, etc. So experiences are manifestation-events—they exist as appearing-of-something, and appearance as such has its existence in its being-present (being-experienced)—namely its existence as appearance-of. The presence of experience means that appearance-of-­something takes place, and so the presence of experience is ipso facto the presence of this something. Therefore my being qua presence means that all sorts of things are present to me. I can investigate these things given to me in manifold ways, and I can also reflect on their modes of givenness.Yet what can be said about the presence itself as such of what is present to me? This presence (consciousness) is notoriously elusive. It has no observable properties of its own, is no particular and distinguishable content we encounter, and can never stand before us as an object.Therefore the self in the Advaitic sense is not one of the ‘seen’ things but the ‘seeing’ itself: ‘I am neither this object, nor that, I am That which makes all objects manifest’ (Śaṅkara, Vivekacūḍāmaṇi verse 493, Prabhavananda/Isherwood 1978: 115). Presence is not a phenomenon of its own that I could find in addition to other phenomena, but simply the taking place of thereness of any phenomena.This is the sense of the so-called ‘transparency’ of consciousness, that is, the fact that when one tries to attend to the consciousness of an object, one can hardly help ending up attending to what it is conscious of.28 So the presence itself of what is present can never be an observable object, yet at the same time it is the most familiar thing in the world: It is that wherein everything we

WITNESS-CONSCIOUSNESS

191

e­ xperience has its being-experienced, the medium of all phenomena (the taking place of their phenomenality). And the soteriological aim of Advaita Vedānta—the realization of ātman—is nothing but simply becoming explicitly aware of this taking place of presence as such. Of course, in a certain sense we are constantly conscious of our being conscious. After all, we are not living in a permanent state of complete self-forgetfulness, fully ­absorbed in the objects: we are not only conscious of the objects we see but also—at least ­implicitly—of our seeing them. But evidently, this is not the form of self-awareness ­Advaita Vedānta strives for—rather, it is, in their view, precisely a form of self-forgetfulness, that is, of o ­ bscuration of the fundamental dimension of our own being qua subjectivity: It is ahaṃkāra (‘ego-sense’), the awareness of a distinct ‘I’ (aham) as an inner-worldly subject with particular empirical (psychophysical) properties (a jīva, ‘person’). To say that I am not only conscious, for example, of this desk I see, but also of my seeing it actually means that I am aware of myself sitting here and looking at the desk and of the fact that the desk appears in this particular way precisely because it is given to me as someone viewing it from this particular angle, with these particular sense-organs, and so on. So that which I am aware of here, is my localization in the world, and of my own body to which I relate the rest of the appearances. When I experience an unchanging object in changing modes of givenness, I experience this change as being due to the changing relations between the experiencing subject and the experienced object: that is, together with the object, experienced in changing modes of givenness, a ‘subject’ is experienced for whom the manifestations are manifestations, a ‘subject’ which is itself something that is objectively located within the objective world, standing in manifold—physical and ­psychical—relations to other things. And this is essential to object-givenness in general. Objectivity means appearance-­ transcendence: We apprehend the subjective appearance as not being the object itself, but as only being an aspect of this object, that is, this object as seen from a certain viewpoint, in certain respects. Hence the from-where of seeing is necessarily co-constituted with the seen object—co-constituted as a ‘subject’ that is itself part of the objective world (cf. ­Husserl 1952: 56, 109–110, 144; Albahari 2006: 8–9, 88). So in a way the experience of objects is ipso facto also self-experience, in the sense of the self-localization of the subject within the realm of the objects. This not only holds for our being conscious of ourselves as a body, but also with regard to the mental aspects of what we experience as our ‘I’: For example, the field of givenness is never a mere homogeneous plane, but features an attentional relief which indicates a mental ‘I’ to which certain things are attentionally ‘nearer’ than others: I can direct my attention to this or to that within the field of what is consciously there for me, so that my ‘I’ is obviously to be distinguished from this field (cf. Husserl 1952: 105–106), an ‘I’ with particular personal interests and the like. This ‘self-experience’ as a particular psychophysical being means that we identify a certain special sphere of what is experientially given to us ‘ourselves’: that is, we constantly distinguish within the realm of phenomenonal contents between what belongs to ‘ourselves’—one’s body, one’s thought and so on—and what is located ‘outside of ourselves’ (cf. Albahari 2006: 51 56–60, 7329). This is what is called adhyāsa (‘superimposition’) in Advaita (see below).

192

WOLFGANG FASCHING

So object-givenness implies the givenness of the subject (an indicated and experienced from-where of experiencing) as a necessary moment of the structure of the field of the objectively given. Now the point is that the experiencing itself—consciousness—is not a structural moment of what is given, but is the very taking place of givenness itself. The whole inner/outer (self/not-self) distinction constitutes itself within the realm of experiential contents—and consequently experiencing itself is not located within some ‘inner sphere’. My consciousness is not to be found on one side of this inner/outer distinction in which what we experience is necessarily structured, but is, again, the taking place of experience itself. The viewpoint is part of the structure of the field of presence and therefore not presupposed, but constituted by it. Hence consciousness of myself as an ‘inside’ as opposed to an ‘outside’ is not a way of being aware of consciousness as such, which is not a special inner realm opposed to the outer objects, but the thereness of these objects, the appearing of what appears (be it ‘inside’ or ‘outside’).This ‘pre-interior’ consciousness is what Advaita Vedānta means by ‘self ’: ‘[T]he self which is of the nature of consciousness … [is] the witness of both the seer and the seen’ (Śaṅkara, Ātmajñopadeśavidhi III.7, quoted in Gupta 1998: 38), therefore it is ‘the pure “subject” that underlies all subject/object distinctions’ (Deutsch 1969: 49), ‘the “field” of consciousness/being within which the knower/knowing/known distinctions arise’ (Fort 1984: 278). 5.  THE PROCESS OF DE-SUPERIMPOSITION

In order to become aware of the self in this sense it is necessary to stop identifying oneself with what presents itself as ‘I’ and ‘mine’: the ‘annihilation of the ego-sense’ (Ramana in Osborne 1997: 19). Normally we are not explicitly aware of consciousness as such, since we are totally lost in the objects of consciousness and can also understand ourselves only as one of the objects. This erroneous self-understanding as one of the objects is what is called adhyāsa in Advaita Vedānta, the ‘superimposition’ of self and not-self: Certain ­experienced contents are appropriated as belonging to one’s own self, as an ‘inner’ opposed to what is located ‘outside’ the self (cf. Fort 1984: 278) as it articulates itself in our ‘saying, for ­example, “that am I”, “that is mine”’ (Śaṅkara, Brahmasūtrabhāṣya, Introd., Deussen 1920: 3). Accordingly, the way of becoming aware of one’s true nature consists in a ‘process of desuperimposition’ (Indich 1980: 16, 10), that is, a process of de-identification from ­anything one objectively encounters as one’s purported self (cf. Fasching 2008). One stops identifying oneself with the inner-objective ‘subject’, the psychophysical entity (jīva) one normally takes oneself to be. Instead of identifying certain configurations of experienced contents as being ‘oneself ’, one begins to experience oneself as the abiding experiencing itself (the taking place of presence) of any contents. De-superimposition means radically distinguishing oneself from all objects by no longer delimiting oneself (as an ‘inside’) as opposed to the objects ‘out there’. One stops considering anything as being ‘oneself ’ or ‘one’s own’: ‘He to whom both “I” … and “my” … have become meaningless, becomes a knower of Ātman’, as Śaṅkara puts it (Upadeśasāhasrī I.14.29, Mayeda 1992: 138). In the ‘de-identified’ mode of experiencing that is strived for, one completely lets go of ‘oneself ’ and becomes nothing but ‘seeing’, without any distinct ‘seer’ standing apart from

193

WITNESS-CONSCIOUSNESS

the ‘seen’. This amounts to a profound transformation of one’s self-experience and of the way of being in the world.To experience oneself as the ‘witnessing’ leads to a sense of detachment, a loosening of one’s involvement in the concerns, desires, and fears of the ego.30 One experiences oneself as an inner stillness in the midst of all motion, and as non-acting even when engaged in action.31 Instead of simply identifying oneself with a particular configuration of experiential contents, standing in permanently changing relations of activity and passivity to other experienced contents, one simultaneously experiences oneself as the abiding experiencing itself, as the motionless and non-acting dimension of manifestation of all movements and activities, of any ‘inside’ and ‘outside’ and all relations between them. One no longer apprehends oneself as a subject-pole in opposition to an object-pole, being affected by it, reacting to it, dealing with it, but as the event of presence of any subject and object. ‘I (= Ātman) am of the nature of Seeing, non-object …, unconnected [with anything], changeless, motionless … [Touch] does not produce for me any change of gain and loss …, since I am devoid of touch, just as a blow with the fist and the like [does not produce any change] in the sky’ (Upadeśasāhasrī III. 3.115, Mayeda 1992: 252; bracketed additions by Mayeda). With the dropping of the notions of ‘I’ and ‘mine’, in a way nothing remains for ­oneself, and in this sense it could be seen as a dissolution of the self. Yet for Advaita Vedānta this ‘nothing’ actually just means no-thing, that is, non-objectivity. The modern Advaitic ­author Arvind Sharma answers the question of whether ‘the sense “I am” [is] real or unreal’ with the words: ‘Both. It is unreal when we say: “I am this, I am that”. It is real when we mean “I am not this, nor that’” (Sharma 1993: 96–97).32 One becomes aware of oneself precisely when one ceases to find oneself anywhere. 6. CONCLUSION

In opposition to Buddhism, Advaita Vedānta insists on the existence of an abiding self, a self which consists in nothing but consciousness (‘seeing’ or ‘witnessing’) and as such is the non-object kat’ exochen, since seeing is not itself something visible. I have argued that this view does indeed capture something essential about the nature of experience. Buddhism and Advaita Vedānta agree that it is necessary to inhibit the identification with the ‘I’ and the clinging to what is ‘mine’ to achieve liberation. The theoretical interpretation of this process is where they disagree. Now it is my opinion that the notion of witness-consciousness allows for a more faithful description of what actually happens in this process than the idea of no-self (at least in its reductionist interpretation). Buddhism invites us to reflect on our own being and holds that what we will find are all kinds of transient phenomena (the five skandhas), but nothing like a stable ‘self ’. With regard to each of the skandhas one should understand;‘this is not mine; this am not I; this is not the Self of me’ (Saṃyutta Nikāya XXII. 59).This insight leads us to the liberation from the illusion of self.Yet the question is: If there is nothing but these transient phenomena that constitute our being (in other words: if this simply is what we are)—who is it then that is not identical to all this? Who is it who can say of her body, her thoughts, etc. ‘this am not I’? This ‘who’ is, I wish to suggest, nothing but the experiencing consciousness in which all the passing phenomena have their manifestation and which Advaita Vedānta regards as our ‘self ’.33

194

WOLFGANG FASCHING

NOTES 1. Although my main point of reference is the Advaitic understanding of the self, I will primarily focus on aspects it shares with Sāṃkhya/Yoga and many other Indian schools, i.e. independent of its monistic commitments. (For an attempt to make sense of the Advaitic idea that ultimately only one self exists, cf. Fasching 2010.) 2. I must stress that I intend to pursue, as the subtitle says, ‘phenomenological reflections’ on the Advaitic understanding, and not engage in a staunch exegesis of the details of the various Advaitins’ theories. I wish to discuss philosophically what I take to be a basic intuition about the nature of consciousness that seems to provide something like a foundation of the Advaitic speculations. 3. For example, MacKenzie (this volume) argues that the Madhyamaka school holds—in contrast to the reductionism of the Abhidharma—that the self is not reducible to more basic phenomena, but ‘is an emergent phenomenon that, while real, is not a substantial separate thing’ (ibid.: p. 258). (Whether or not this makes a crucial ontological difference naturally depends on the precise definition of ‘emergence’.) 4. Miri Albahari even interprets the Pali Canon as implicitly, but centrally, assuming the existence of a witness-consciousness—‘a reading’, as she admits, ‘that aligns Buddhism more closely to Advaita Vedānta than is usually acknowledged’ (Albahari 2006: 2; cf. also Albahari this volume). 5. The term skandhas refers to the five types of phenomena (dharmas) that constitute the person according to Buddhism. 6. Expressions like ‘Devadatta’s desire’, as the Buddhist argues in Bhaṭṭa Rāmākaṇṭha’s Nareśvaraparīkṣā, do not imply that there is something beyond the desire as its agent, but are ‘just indicating that [the desire] is connected with a particular stream of cognition, like [such expressions] as “the flow of the Vitastā [river]”’ (Watson 2006: 190; bracketed additions by Watson). 7. Cf. MacKenzie 2007: 47–49; MacKenzie 2008; Dreyfus 1997: 339–340, 400–402; and the contributions of Dreyfus, Krueger and Thompson in this volume. 8. This view is comparable to non-egological accounts in phenomenology, for example by Sartre, Gurwitsch and the Husserl of the Logical Investigations. 9. As Śaṅkara stresses against the Buddhist view: Brahmasūtrabhāṣya II.2.25, Deussen 1920: 353–354. 10. Cf. Chatterjee 1982: 342: ‘The Advaitists say, that when we have an awareness of an object, the object is indeed manifested, but it is not the only thing revealed; here we have an automatic awareness of the awareness too.The two awarenesses are simultaneous, but they are not of a similar structure, in fact they are the two aspects of the same awareness.’ 11. Śaṅkara argues against the Yogācārins that even if the experience, like a lamp which need not be illuminated by a second lamp in order to be visible, is revealed by itself, it still has to be revealed to a subject (otherwise it would be ‘like lamps, and be they thousands, burning in the midst of a mass of rocks’ (Brahmasūtrabhāṣya II.2.28, Deussen 1920: 361–362), i.e. without anyone seeing them). Cf. Ingalls 1954: 301. 12. Cf. Brahmasūtrabhāṣya II.2.25 where Śaṅkara stresses that my continued existence refers to strict numerical identity and not to some similarity (Thibaut 1962: 415; this sentence is missing in Deussen’s translation)—and observes that while, with regard to external things, it is admittedly possible to mistake similarity with identity, this is impossible with regard to oneself as the subject (which today is called the ‘immunity to error through misidentification’).

WITNESS-CONSCIOUSNESS

195

13. Just as for Sāṃkhya and Yoga, and, by the way, for the Śaiva Siddhāntin Bhaṭṭa Rāmakaṇṭha: cf. the interesting study by Alex Watson (2006). In his Nareśvaraparīkṣāprakāśa, Bhaṭṭa ­Rāmakaṇṭha initially lets the Buddhist win over Nyāya and Vaiśeṣika, which assume the existence of a self as a further entity beyond cognition. But while Buddhism concludes that there actually is no self, only the cognitions, Rāmakaṇṭha holds that cognition itself is the self (ibid.: 213–217). He thereby repeats earlier debates between Buddhism and Sāṃkhya (whose view of the self he largely inherits) (ibid.: 93). 14. Quite similarly, Bhaṭṭa Rāmakaṇṭha differentiates between two meanings of jñāna, namely on the one hand the many transient cognitions, and on the other, the one abiding cognition which is our very self (and which he also terms, when it comes to contrasting the two senses of jñāna, prakāśa = ‘illumination’ or saṃvit = ‘consciousness’): the latter being a permanent witnessing or experiencing of the passing cognitions (Watson 2006: 354–373). 15. In Brahmasūtrabhāṣya II.2.28, Śaṅkara lets the Buddhist ask whether, with his stressing of the self-revealedness of the cognizer, he is not actually adopting, only in other words, the Buddhist’s own view of the self-givenness of cognition, and answers: ‘No! Because cognition is to be distinguished [from the cognizing subject] insofar as it is originating, passing away, manifold, etc.’ (Deussen 1920: 362, addition in brackets by Deussen). 16. Cf. e.g. Śaṅkara’s Upadeśasāhasrī I.15.40–41: ‘[It] has the light of knowledge as Its nature; [It] does not depend upon anything else for [Its] knowledge. Therefore [It] is always known to me. The sun does not need any other light for its illumination’ (Mayeda 1992: 145–146, bracketed additions by Mayeda). 17. Cf. Upadeśasāhasrī I.15.48 and 50: ‘Ātman Itself … is by nature neither knowable nor not knowable’. ‘Just as there is neither day nor night in the sun, since there is no distinction in the nature of light, so is there neither knowledge nor ignorance in Ātman, since there is no distinction in the nature of knowledge’ (Mayeda 1992: 146). ‘Though light is an ­illuminator, it does not illumine itself [since it has in itself no difference as between illuminator and ­illuminated] … In like manner Ātman [which has homogeneous knowledge] never sees ­Itself ’ (ibid.: I.16.12, Mayeda 1992: 150, bracketed additions by Mayeda). Cf. Ram-Prasad this volume, section 5, and Ram-Prasad 2007: 78–79. 18. For an insightful discussion of the understanding of consciousness as ‘luminosity’ in Indian philosophy, cf. Ram-Prasad 2007: 51–99. 19. Cf. the suggestion of Bayne and Chalmers in Bayne, Chalmers 2003: 56–57. 20. Cf., e.g. Upadeśasāhasrī II.2.58 and I.14.50: ‘Ātman, like space, is by nature not composite’; ‘there is no distinction at any time in the Seeing which is like ether’ (Mayeda 1992: 237 and 140–141). 21. In her very lucid paper on the concept of witness-consciousness, Miri Albahari (2009) ­r igorously distinguishes it from the ‘for-me-ness’ or ‘mineness’ (i.e.‘first-personal ­g ivenness’), which Dan Zahavi posits as the core sense of self.While mineness is a property of ­experience, witness-consciousness is ‘the modus operandi of the subject that has them’ (Albahari 2009: 68), i.e. of a ‘separate me’ (ibid.: 73) (whereas for Zahavi ‘the self … does not exist in ­separation from the experiences, and is identified by the very first-personal givenness of the experiences’: Zahavi 2005: 132). I agree that experiences and consciousness have to be distinguished in a certain sense (this being the very idea of ‘witness-consciousness’), yet I disagree with breaking them apart as if they were separate existences, as Albahari seems to do. ­Witness-consciousness is, according to Albahari, the ‘mode-neutral awareness’ that is supposed to account for the experiences’ accessibility to reflection, and for the unity of ­consciousness across manifold experiences (Albahari 2009: 71–72), thus obviously our pre-­ reflective awareness of our own experiences.Yet this is precisely what Zahavi calls ‘mineness’

196

22.

23.

24. 25. 26.

WOLFGANG FASCHING

qua first-personal givenness. To ‘witness’, according to my understanding of the term, does not literally mean that the subject ‘observes’ the experiences (as Albahari formulates: ibid.: 68), as if the witness were a separately existing entity that watches experience-objects ­existing outside of it. Rather, it should be understood as the experiencing of the experiences in which they have their very being. It is simply not the case that the being-present (first-­ personal givenness, for-me-ness) of the experiences and the witnessing as the modus operandi of the subject are two different things. And according to the Advaitic (and my) understanding, the ‘me’ of the for-me-ness (i.e. the self) is—quite in agreement with Zahavi—not something to be posited in addition to this presence (for-me-ness), but something that consists in nothing other than the witnessing/experiencing itself. Furthermore, Albahari holds that this for-me-ness is, as an aspect of experience, something introspectively detectable, and also in this respect stands in contrast to witness-consciousness which, as ‘built into the very act of being aware’, can never become an object of awareness (Albahari 2009: 68–69). I have my doubts about the former claim. ‘Mineness’ is about as much a ‘real predicate’ as is ‘being’ according to Kant. It is in no way a content towards which one could direct one’s attention, no introspectively examinable quality (no ‘feeling’: Albahari 2009: 70) my experiences have in addition to other qualities (such as the specific character of my pain) (cf. also Zahavi this volume: p. 59)—it is rather precisely the first-personal thereness-for-me of my experiences, together with all their qualities. Cf. Śaṅkara in Upadeśasāhasrī II.2.75 (in answering the question of how the perceiver, ­perceiving now this, then that, can be said to be changeless): ‘If indeed you were subject to transformation, you would not perceive the entire movement of the mind … Therefore, you are transcendentally changeless’ (Mayeda 1992: 240). ‘There must be some constant continuous principle to see their [the cognitions’] origin and destruction … And this continuous consciousness is sākṣin’ (Chatterjee 1982: 349). Along comparable lines, Rāmakaṇṭha argues against the Buddhists that there is no need to assume that the change of the objects of consciousness implies a change of consciousness itself: For even the Buddhists cannot deny that many objects are conscious in one single consciousness at one point in time (and it is of no help for the Buddhist to hold that this is due to a unifying cognition: it is still necessary to appeal to the possibility of a single cognition having many objects). So, Rāmakaṇṭha argues, if the multiplicity of objects at one time does not affect the singleness of consciousness, why should the multiplicity of objects over time? It is the contents of consciousness that change, not consciousness itself (Watson 2006: 335–348). Cf. Husserl 1966: 333: ‘… the now-consciousness is not itself now’. Cf. UpadeśasāhasrĪ I.5.3: ‘Just as to a man in the boat the trees [appear to] move in a direction opposite [to his movement], so does Ātman [appear to] transmigrate …’ (Mayeda 1992: 114; bracketed additions by Mayeda). In my interpretation of the notion of witness-consciousness, I owe much to Dan Zahavi’s views on consciousness and the self. However, I am not sure whether we fully agree regarding the nature of the diachronic identity of the self. Zahavi states ‘that it is the shared mode of givenness that makes two experiences belong to the same subject, i.e. … it is their exposure in the same field of primary presence which makes different experiences of one and the same self ’ (Zahavi 1999: 144). I find this formulation ambiguous. It could mean that a past experience is mine insofar as it is in my present experiencing given in a first-person mode, ‘as mine’, or it could mean that the experience, when originally experienced, had its manifestation in the same field or dimension of first-personal givenness. Hence, the question is: Is a past experience mine insofar as, and because, it is first-personally accessible to me (in the present), or is it first-personally accessible to me because it was experienced by me? The

197

WITNESS-CONSCIOUSNESS

27.

28.

29.

30. 31. 32.

33.

first reading ultimately amounts to a reductionist view in Parfit’s sense, and Zahavi appears to tend towards this approach (cf., for example, the final section of his contribution to this volume). On the other hand, he speaks of an ‘abiding dimension of experiencing’ (Zahavi 1999: 80), which would allow a view of the ‘sameness’ of the ‘same field of givenness’ in the second sense (the view I favor). Formulations such as, ‘Not only is the first experience retained by the last experience, but the different experiences are all characterized by the same fundamental first-personal givenness’ (this volume: p.  58) may also be interpreted along these lines. Cf. Śaṅkara’s Brahmasūtrabhāṣya II.2.28: ‘We are obliged to assume objects apart from cognition, namely on the ground of cognition itself. For no one cognizes a post or a wall as mere cognition, but as objects of cognition everyone cognizes the post or the wall’ (Deussen 1920: 359). Consciousness is no object we could find anywhere and is in this sense ‘invisible’. But this does not mean that it is concealed. The transparency of consciousness does not mean that the cognitive processes through which we represent objects are not themselves again represented and thereby normally unknown to us (as Metzinger understands it: Metzinger 2003: 163–177), but rather that there simply is nothing to represent, because consciousness is nothing but the thereness of whatever it happens to be consciousness of and nothing beyond that: It is not an object we fail to be conscious of, but no object at all. Albahari 2006: 57:The identification of the subject with certain aspects of the body or mind involves ‘the subject—the witnessing as it presents from a psycho-physical perspective— identifying with those very khandhās [= skandhas] (objects of awareness) that contribute … to the impression of a hemmed-in perspective from which the world is witnessed’. Cf. e.g. Śaṅkara’s descriptions in Vivekacūḍāmaṇi verses 428–442, Prabhavananda/Isherwood 1978: 104–106; Osborne 1997: 31. ‘… he who, though acting, is actionless—he is the knower of Ātman’ (Upadeśasāhasrī I.10.13, Mayeda 1992: 124); cf. also Osborne 1997: 32. Cf. Upadeśasāhasrī I.6.6: ‘The learned should abandon the “this”-portion in what is called “I”, understanding that it is not Ātman’ (Mayeda 1992: 116). (Formulations like this seem to contradict the claim, such as Ram-Prasad (this volume) makes, that in Advaita Vedānta the term ‘I’ does not at all refer to the ātman.) This article was conceived and written in the framework of the Austrian Science Fund (FWF) research project Experiential Presence (P21327). I wish to thank Himal Trikha, Chakravarthi Ram-Prasad and the editors for their comments and suggestions. REFERENCES

Albahari, M. (2006), Analytical Buddhism:The Two-Tiered Illusion of Self (Basingstoke: Palgrave Macmillan). ——— (2009), ‘Witness-Consciousness: Its Definition, Appearance and Reality’, Journal of Consciousness Studies 16/1: 62–84. Bayne, T. and Chalmers, D. J. (2003), ‘What Is the Unity of Consciousness?’ in Axel Cleeremans (ed.), The Unity of Consciousness: Binding, Integration, and Dissociation (Oxford: Oxford University Press). Chatterjee, T. (1982), ‘The Concept of Sākṣin’, Journal of Indian Philosophy 10: 339–356. Dainton, B. (2006), Stream of Consciousness: Unity and Continuity in Conscious Experience (London: Routledge). Deussen, P., trans. (1920), Die Sutra’s des Vedānta oder die Çārīraka-Mīmāṅsā des Bādarāyaṇa nebst dem ­vollständigen Commentare des Çaṅkara (Leipzig: Brockhaus). Deutsch, E. (1969), Advaita Vedānta: A Philosophical Reconstruction (Honolulu: University of Hawaii Press).

198

WOLFGANG FASCHING

Dreyfus, G. B. J. (1997), Recognizing Reality: Dharmakīrti’s Philosophy and Its Tibetan Interpretations (­Albany, NY: State University of New York Press). Fasching, W. (2008), ‘Consciousness, Self-Consciousness, and Meditation’, Phenomenology and the ­Cognitive Sciences 7: 463–483. ——— (2009), ‘The Mineness of Experience’, Continental Philosophy Review 42: 131–148. ——— (2010), ‘Fremde und eigene Gegenwart: Über Anderheit, Selbst und Zeit’, in Matthias Flatscher and Sophie Loidolt (eds.), Das Fremde im Selbst-das Andere im Selben (Würzburg: Königshausen & Neumann). Fort, A. O. (1984), ‘The Concept of Sākṣin in Advaita Vedānta’, Journal of Indian Philosophy 12: 277–290. Gupta, B. (1998), The Disinterested Witness: A Fragment of Advaita Vedānta Phenomenology (Evanston, IL: Northwestern University Press). Hacker, P. (1978), Kleine Schriften, ed. Lambert Schmithausen (Wiesbaden: Franz Steiner). Hiriyanna, M. (1956), Outlines of Indian Philosophy (London: George Allen & Unwin). Husserl, E. (1952), Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie. Zweites Buch: Phänomenologische Untersuchungen zur Konstitution, ed. Marly Biemel (The Hague: Nijhoff (Husserliana IV)). ——— (1966), Zur Phänomenologie des inneren Zeitbewußtseins (1893–1917), ed. Rudolf Boehm (The Hague: Nijhoff (Husserliana X)). ——— (2006), Späte Texte über Zeitkonstitution (1929–1934): Die C-Manuskripte, ed. Dieter Lohmar (­Dordrecht: Springer (Husserliana Materialien VIII)). Indich, W. M. (1980), Consciousness in Advaita Vedānta (Delhi: Motilal Banarsidass). Ingalls, D. H. H. (1954), ‘Śaṃkara’s Arguments against the Buddhist’, Philosophy East and West 3/4: 291–306. Klawonn, E. (1987), ‘The “I”: On the Ontology of First Personal Identity’, Danish Yearbook of Philosophy 24: 43–75. ——— (1998),‘The Ontological Conception of Consciousness’, DanishYearbook of Philosophy 33: 55–69. MacKenzie, M. D. (2007), ‘The Illumination of Consciousness: Approaches to Self-Awareness in the Indian and Western Traditions’, Philosophy East and West 57/1: 40–62. ——— (2008), ‘Self-Awareness without a Self: Buddhism and the Reflexivity of Awareness’, Asian ­Philosophy 18/3: 245–266. Madell, G. (1981), The Identity of the Self (Edinburgh: Edinburgh University Press). Mayeda, S., ed. and trans. (1992), A Thousand Teachings: The Upadeśasāhasrī of Śaṅkara (Albany, NY: State University of New York Press). Metzinger, T. (2003), Being No One (Cambridge, MA: MIT Press). Nagel, T. (1986), The View from Nowhere (New York: Oxford University Press). Osborne, A., ed. (1997), The Collected Works of Ramana Maharshi (Boston, MA: Weiser Books). Parfit, D. (1987), Reasons and Persons (Oxford: Clarendon Press). Prabhavananda, S. and Isherwood, C., trans. (1978), Shankara’s Crest-Jewel of Discrimination (­Viveka-Chudamani) (Hollywood, CA:Vedanta Press). Ram-Prasad, C. (2007), Indian Philosophy and the Consequences of Knowledge: Themes in Ethics, Metaphysics and Soteriology (Aldershot: Ashgate). Rhys Davids, C. and Woodward, F. L., ed. and trans. (1972–79), The Book of the Kindred Sayings (­Saṃyutta-Nikāya), 5 vols. (London: Routledge & Kegan Paul). Searle, J. R. (1992), The Rediscovery of the Mind (Cambridge, MA: MIT Press). Sharma, A. (1993), The Experiential Dimension of Advaita Vedānta (Delhi: Motilal Banarsidass). Siderits, M. (2003), Personal Identity and Buddhist Philosophy: Empty Persons (Aldershot: Ashgate). Sinha, D. (1954), ‘The Concept of Sākṣin in Advaita Vedānta’, Our Heritage 2: 325–332. Strawson, G. (2003), ‘What Is the Relation between an Experience, the Subject of the Experience, and the Content of the Experience?’ Philosophical Issues 13: 279–315. Thibaut, G., ed. (1962), The Vedānta-Sūtras with the Commentary by Śaṅkarācārya, 2 vols. (Delhi: Motilal Banarsidass).

WITNESS-CONSCIOUSNESS

199

Timalsina, S. (2009), Consciousness in Indian Philosophy:The Advaita Doctrine of ‘Awareness Only’ (London: Routledge). Watson, A. (2006), The Self’s Awareness of Itself: Bhaṭṭa Ramākaṇṭha’s Arguments against the Buddhist Doctrine of No-Self (Wien: Sammlung de Nobili). Williams, B. (1973), ‘The Self and the Future’, in Problems of the Self: Philosophical Papers 1956–1972 (Cambridge: Cambridge University Press). Zahavi, D. (1999), Self-Awareness and Alterity: A Phenomenological Investigation (Evanston, IL: Northwestern University Press). ——— (2005), Subjectivity and Selfhood: Investigating the First-Person Perspective (Cambridge, MA: MIT Press).

P A R T

V

METAPHYSICS

CHAPTER

12

– ya-Vais´esika theory The Nya . of universals Kisor Chakrabarti

Source: Journal of Indian Philosophy, 1975, 3:3–4, 363–382. I.  STATEMENT OF THE THEORY

The Nyāya theory of universals is unambiguously realistic and its realism is evident from the way a universal (sāmānya or jāti) is defined. A universal is defined as “an entity that is eternal and inseparably inherent in many entities”.1 Since a universal is an eternal entity that is present in many particulars, it is clear that a universal is conceived to have an ontological status that is distinct from and independent of that of the particulars themselves. Thus in the Nyāya view there is an identical cloth-ness in all pieces of cloth, an identical humanity in all human beings; and the entitative status of these universals is not affected by the origin and destruction of particulars to which they may be related. The relation between a universal and its particulars (called samavāya, which for want of anything better, we have translated as the relation of ‘inseparable inherence’) is held to be of a unique kind and sharply distinguished from ordinary relations like that of contact (saṃyoga) between two substances, relations of magnitude (parimāṇa), spatial and temporal relations, etc. This relation is held to be an ‘inseparable relation’ (ayutasiddhavṛtti) in the sense that one of the two relata, viz., the particular, has necessarily to remain related to the other relatum, viz., the universal, until it is destroyed. The necessity and inseparability of the relation is, therefore, only in one direction and not in both directions; and in fact it has been specifically mentioned that a universal would persist, even if all objects belonging to the class happened to be destroyed.2 There is no reason to confuse this relation with an internal relation as has been done by modern interpreters,3 because the necessity of the relation is only in one direction and not in both directions.

203

204

KISOR CHAKRABARTI

As for the evidence in justification of the above theory, one common argument is d­ erived from ‘the notion of sharing a common character’ (anugatapratīti). Thus in the Nyāya view, though individual human beings differ from one another in innumerable respects, we discern a common element of humanity running through all of them, though it may not be always possible to specify in words what that common element is. It is ­argued that in the absence of any compelling evidence to the contrary, such knowledge of identity (which presumably is present in every human mind) has to be regarded as true, whence the admission of universals as real entities becomes necessary as the objective basis of such knowledge of identity. On the ground of this knowledge of identity, we classify objects (in a broad sense) into different natural kinds, e.g., man, horse, colour, sound, etc. Without the admission of universals, such classification has to be regarded as ultimately arbitrary and conventional. A second argument concerns the significance of general words in our language. A ­general word like ‘horse’ or ‘colour’ refers to an unlimited number of objects and an explanation is needed as to how a word is capable of doing so. An explanation may be found if universals are admitted to be objectively real. A distinction has to be made between the śakya or denotation and śakyatāvacchedaka or connotation4 of a word. Then it may be said that a general word connotes a universal and it is able to denote an unlimited number of objects, because the universal is present in all of them. Without admitting universals it would be difficult to explain how the relationship between a word and the unlimited number of objects it may denote happens to be established.5 A third argument concerns the justification of causal laws. The statement of a ‘causal law’ (sāmānyakāraṇatā) implies that every object of a certain kind (and nothing else) is the cause of a particular kind of effect.Thus when we determine fire to be the cause of smoke what we mean is that all fires (of a certain kind) are causes of smoke and that nothing but fire is the cause of smoke. As the Nyāya would say, ‘the property of being the cause of smoke’ (dhūmakāraṇatā) is possessed by all fires and by nothing else. But how should we explain (from the philosophical point of view) the fact that all objects of a certain kind become the cause, and nothing else becomes the cause? According to the Nyāya, the only way to explain this rationally is to suppose that all those objects share a common characteristic by virtue of which they can become the cause, and since this characteristic is not possessed by other kinds of objects, they do not become the cause. The admission of such common characteristics shared by many objects, however, amounts to the admission of universals.6 A fourth argument, closely related to the previous one, concerns the justification of ‘inductive generalisation’ (vyāptigraha). A causal law is a general statement which involves a generalisation from the observation of a limited number of particular instances. The problem is to justify the ‘inductive leap’ from the observation of a limited number of cases to the generalisation which covers not only cases observed, but also all possible cases, past, present and future. It is argued that a solution to the ‘problem of induction’ is found by admitting universals as real entities. Then it may be said that from an observation of particular instances we are able to see the connection between the universal properties involved; and the generalisation to all cases would be justified, because identical properties are present in all cases.7

THE NYĀYA-VAIŚEṢIKA THEORY OF UNIVERSALS

205

On the basis of the arguments (rather superficially) outlined above, the Nyāya considers it necessary to admit the existence of universals as objectively real entities which are distinct from, and independent of, particulars themselves. The range of universals admitted to be objectively real is rather comprehensive. There are first of all what may be called generic universals, i.e., universals whose instances are individual substances (dravya), for example, man, horse or stone. There are, secondly, the qualitative universals, i.e., universals whose instances are qualities (guṇa), for example, colour, sound or shape. The notion of quality (guṇa) is very broad in Nyāya metaphysics. Thus in addition to ordinary qualities of physical objects, there are qualities of the ‘soul’, such as pleasure, pain, desiring, hating etc., and there are universals corresponding to these qualities as well as to physical qualities. Moreover, many relations, such as relations of magnitude (parimāṇa), the relation of contact (saṃyoga) between two substances, e.g., the book is on the table, relations of spatial or temporal proximity or distance etc., are brought under the head of qualities; and universals pertaining to them are also admitted to exist. Thirdly, there are universals pertaining to motion (karma), such as contraction, expansion, etc. Russell8 has noted that though philosophers have recognised universals named by adjectives and substantives, those named by verbs and prepositions have been usually overlooked. A look at the above list will show that this complaint certainly could not be brought against the Nyāya theory. With respect to extension, universals are classified into three kinds.9 There is first the universal of largest extension (parasāmānya), viz., existence (sattā) in which every other universal is included. There are, secondly, universals of smallest extension (aparasāmānya), e.g., cowness, horseness etc. in which no other universal is included. Most universals, however, are such that while they are included in universals of larger extension, they themselves include universals of smaller extension within them. For example, the class of animals is included in the class of substances and itself includes classes like man, horse etc. within it. This classification corresponds exactly to the Aristotelian view that except for the highest genus (summum genus) and the lowest species (infimae species), whatever is a species is at the same time a genus and whatever is a genus is also a species. Further light may be thrown on the Nyāya theory by comparing it with similar ­theories in Western philosophy. The Nyāya theory should be clearly distinguished from the Ante Rem or Ideal theory of the Platonic type. According to the latter theory, Ideas or Forms which are the reality behind natural objects, exist eternally in the world of Being and have their shadowy manifestations in the world of Becoming. We know the ultimate Forms by a priori intuition and so are able to recognise their incomplete manifestations in sensible things. In the Nyāya view, however, particulars are as much real as universals themselves, and the relation between the former and the latter is that of ‘inseparable inherence’. On this point the Nyāya theory is much closer to the Aristotelian theory. Thus, like Aristotle, the Nyāya would deny that there is (as stated in the tenth book of Plato’s Republic) an unseen and eternal bed, or quintessential bedness, by participation in which any bed ­becomes a bed; but still would hold, like Aristotle, that there is an identical bedness in all beds, an identical humanity in Smith, Brown and Jones, and the relation between such identities and particulars to which they belong is not merely accidental, but necessary (in the sense specified above).10

206

KISOR CHAKRABARTI

Moreover, Aristotle distinguishes between primary and secondary beings (substances) and holds that though an individual man or horse is a primary being, the species man or horse is a secondary being. Thus the universal, for Aristotle, could not exist as an individual, a primary being, exists. The individual cannot be repeated, and cannot be common to many. Consequently, that which is universal and common to many cannot be in the sense in which the individual is. Turning now to the Nyāya theory we find that a distinction is made between sattā and bhāvatva, which may be translated as ‘existence’ and ‘being’ respectively. It is held that though universals have bhāvatva or being, they do not have sattā or existence; while particulars have both ‘being’ and ‘existence’. The distinction between ‘existence’ and ‘being’ is not at all clear and would require careful research to be spelled out exactly. But one thing that is clear is that though universals and particulars are both real, Nyāya is making a distinction between the ontological status of the two, a distinction which is a permanent part of Aristotle’s teaching. It may also be noted that Russell11 ­distinguished between ‘existence’ and ‘subsistence’ and held that while particulars exist, universals subsist or have being, where subsistence is opposed to existence as being timeless. For Aristotle, however, a universal is an abstraction and therefore non-sensible. Both it and its relations are apprehended by reason. (In Plato’s view also Ideas or Forms are non-sensible and can be grasped only by a priori intuition.) But in the Nyāya view most universals (with the exception of universals like ‘atomness’, ‘soulness’ etc.) are known through sense perception.Thus in the Nyāya view, when three different pots are presented together in perception, we may observe that all of them share a common character, viz., pot-ness. The Nyāya does not deny that dispositional elements are present, both in the recognition of the common character and in the observation that two or more objects have the same character. But in either case, none of this makes the discovery less of a discovery. Again, in the sense of singling one element out of what is given, abstraction is necessarily present in the discovery of a common character like colour. We concentrate upon the colour in noticing that the colour in three different objects is one and the same, and single it out for observation. But such observation does not involve abstraction in another sense, viz., framing an abstract complex idea. In the Nyāya view a universal is not an abstraction in the latter sense, and this is true not only for qualitative universals like colour, sound, etc., but also for generic universals like horseness and cowness. Thus cowness is a simple, indivisible element that is present in every cow and is directly apprehended at the moment an individual cow is perceived. The Nyāya view on this point is quite uncompromising; the reason behind this view is that predicates such as ‘cow’, ‘horse’, etc. are regarded as simple and unanalysable, as are qualitative predicates like ‘red’, ‘blue’, etc. In fact in the Nyāya view it is only a simple predicate that could name a universal. If a predicate were complex and made out of simpler concept, it could not name a universal because the universal is a unitary entity devoid of any parts. Thus cowness is not a complex idea formed by putting together the common features of different individual cows. On this point the Nyāya view is diametrically opposed to the Lockeian view that general ideas such as ‘cow’, ‘man’, etc., are framed through a process of conceptual combination. As already said, if a general idea is formed through a process of conceptual combination, then in the Nyāya view, it cannot be the name of a universal. On the other hand cowness, humanity, etc. are simple properties and discovered directly through sense perception.

THE NYĀYA-VAIŚEṢIKA THEORY OF UNIVERSALS

207

This also shows how different the Nyāya view is from the Aristotelian view regarding the relation between a genus and its species. In Aristotle’s view the essence of the species man, for example, is constituted by the genus animality together with the specific difference rationality. Thus in Aristotle’s view the relation between a genus and its species is necessary, the genus constituting a part of the essence of the species. In the Nyāya view, however, both humanity and animality are simple concepts.The relation between the two is not necessary in the Aristotelian sense, because animality is not a part of the concept of man. This difference has resulted in a basic difference between the theories of definition of Nyāya and Aristotle. In Aristotle’s view a definition must be the statement of essence, the statement per genus et differentiam. In the Nyāya view, such essential definition of terms like cow and man is impossible. On the other hand, according to the Nyāya, a definition is the statement of a ‘unique characteristic’ (asādhāraṇdharma), i.e., a characteristic which is present in every member of the class of objects to be defined and not present in anything else.12 The genus is never included in a definition since a genus is not a unique characteristic in the above sense. Moreover, such a unique characteristic does not have to be essential in Aristotle’s sense, but may very well be accidental. For example, cow is defined as that which has a dewlap. Now dewlap is a unique characteristic of cows in the sense specified, and hence this definition is satisfactory from the Nyāya point of view. But a dewlap is still an ‘accident’ and hence for Aristotle this definition is not satisfactory. In Aristotle’s view the aim of a definition is to state what the definiendum really is. In the Nyāya view the aim of definition is much more modest. The aim is to differentiate the definiendum from everything else and be able to use words unambiguously. Both of these two aims are fulfilled by the statement of a unique characteristic. Another peculiarity of the Nyāya theory must now be stated. To a Western realist a quality is a repeatable character, such as the colour blue. Nyāya, however, distinguishes between particular qualities (guṇa) which are not repeatable and common qualities which are repeatable. Thus each blue substance is connected with its own blue colour which is not shared at all and is as particular as the substance itself; all these distinct blue colours are connected in turn to the shared universal blueness. Just as there is a class of men so there is a class of ‘blues’ and the members of this class are not blue substances, but the distinct blue colours themselves. A similar distinction is made for relations. There are relation-­ particulars as well as relation-universals; and relation-universals have as their members the relation-particulars and not the relata themselves. It may be noted that whether qualities and relations of particular things are particular or universal is a controversial question in contemporary philosophy.13 The Nyāya position in this controversy would be that in one sense all qualities and relations of particular things are also particular. But particular qualities and relations may themselves be classified under common qualities and relations and these latter are all universal. II.  OBJECTIONS TO THE THEORY BY OTHER SCHOOLS

The Nyāya theory of universals came under serious criticism from many other schools of Indian philosophy; criticisms often followed the pattern of criticisms of realistic theories in Western philosophy. The Nyāya argued that different individuals are known to be

208

KISOR CHAKRABARTI

identical in certain respects and universals must be admitted as the basis for this knowledge of identity. The Sāṃkhya challenged what is being assumed in this argument, viz., that different individuals are known to be identical in certain respects. It was objected that different individuals are not known to be identical but only known to be similar in some respects. Three different pots may be known to be similar to one another because all of them have a similar colour, say, red. And the colour in respect of which these are similar is also not the identical colour in all of them. The colour of the first pot is as particular as the pot itself and is only similar to the colour of the second pot. At this point it was held by Śālikanātha14 of the Mīmāṃsā school that what different individuals really have in common is similarity or resemblance. Hence it is enough to admit resemblance (sādṛśya) itself as a distinct entity and thus get rid of the innumerable universals hypostatised in the Nyāya theory. The controversy of similarity vs identity was carried on indefinitely and we will only briefly indicate how the Nyāya defended its position. It was argued that mere similarity is not enough because “individuals of different kinds are also known to be similar”.15 The point here is that ‘belonging to the same class’ or ‘being of the same kind’ is a much stricter notion than being merely similar. For example, a chimpanzee may be said to be similar to an orangutang, but it would be improper to say that the same relationship holds between a chimpanzee and another chimpanzee. It is also unwarranted to regard similarity as a distinct entity, because similarity between two entities always reduces to their having certain characteristics in common.16 When, however, we come to consider the common characteristics themselves, it goes contrary to experience to insist that these characteristics are merely known to be similar and not identical. Let us consider two pots having exactly the same shade of red. In such a case the redness in the first instance of red is experientially indistinguishable from that of the second instance of red. One may say that the first red is distinguishable from the second red, because they are after all in different places. But that is completely beside the point. Nyāya grants that as an instance of red the first red is different from the second. What is indistinguishable are not the instances of red, but what they are instances of.When we consider what they are instances of, we have to ask whether they are instances of the same shade or not? If they are not, the point is irrelevant. If they are, then what makes them instances of the same shade must be the presence of that shade in each. To insist on mere similarity with respect to the shade is to insist on a difference that is not discernible and, therefore, has no empirical evidence to back it up. The most persistent critics of the Nyāya theory were the Buddhists who pointed to a series of difficulties in the Nyāya theory. Granted that the universal is a distinct entity, it has to be asked whether it is present in a particular in its entirety or only in part. Neither alternative is acceptable. If the universal is present in its entirety in one of the particulars, it could not be present in other particulars. Nor can we say that it exists in a particular only in part, because it, ex hypothesi, is simple and partless. It may be noted that this objection is basically the same as that raised by Parmenides against the doctrine of Forms in the dialogue Parmenides of Plato.17 Again, if the universal is distinct from particulars, it becomes necessary to explain how the universal comes to be related to a new-born particular, e.g., a cow. We cannot say that the universal has moved from the place where it already existed to the place where the cow is born, because a universal is not a substance; and according to the Nyāya, only

THE NYĀYA-VAIŚEṢIKA THEORY OF UNIVERSALS

209

substances are capable of motion. Nor can we say that cowness already existed at the place where the cow was born, because then it should have been perceived there even before the cow was born. It is also not possible for us to say that cowness also originated at the same place along with the cow, because, by hypothesis, it is eternal. In a word, the admission of universals as distinct entities raises difficulties about the relationship between the universal and particulars.18 Moreover, the Nyāya argued that universals must be admitted on the basis of the notion of belonging to the same class. It was pointed out that this argument is inconclusive, because there are cases where we do have the notion of belonging to the same class but no common simple character may be found to be possessed by all members of the class. For example, let us consider the class of cooks. As Ashoka Pandita said: “Though with reference to many individual cooks, there is a common notion of all of them being a cook, there is nothing which may be common to all of them.”19 Ashoka Pandita disposes of the suggestion that what individual cooks have in common is the act of cooking and shows by detailed analysis that there is no common simple character possessed by all cooks. Whatever may be the merit of the actual example chosen, the important point scored by the Buddhist is that if we could speak of the class of cooks without requiring a common character, why could we not do so with respect to other classes like man, horse etc.? The Buddhists held that all that exist are particulars and that universals are nothing but ‘conceptual constructions’ (vikalpa). A radically different explanation of the significance of general words was also given.When we apply a general word, such as cow, to an individual cow, we do not mean that it is of the same kind as other cows. What we really mean is that it is different from everything that is other than a cow. By applying the word cow, we rule out the application of other words such as horse, man, etc. and differentiate the object from every non-cow. Thus a general word primarily has a negative meaning signifying ‘differentiation from others’ (anyāpoha) and not a positive meaning signifying ‘belonging to the same class’ as in the Nyāya view.20 The Nyāya replied that before we can differentiate cow from every non-cow, we must already know what is meant by cow. The concept of cow is part of the concept of differentiation from every non-cow and is logically presupposed by the latter. The Buddhist analysis of the meaning of general words does not succeed in dispensing with the positive meaning of them, but rather confirms it. In fact the analysis is vitiated by circularity. As Jayanta remarks, “If the meaning of cow is to be ascertained through the negation of noncow, circularity is inevitable. ‘Cow’ is to be ascertained through negation of non-cow, but negation of non-cow is possible only through an ascertainment of what cow is.”21 As to the difficulties raised, the Nyāya position is that they are due to a misunderstanding of the nature of universals. When it is asked whether a universal resides entirely or partially in a particular, it is presupposed that a universal must reside either entirely or partially in a particular. But “Cowness is neither a substantial whole (avayavī), nor an aggregate; the word ‘part’ applies to members of an aggregate or to elements of a substantial whole; the word ‘entire’ applies to such members or elements when all of them are taken together without a remainder. Cowness is neither an aggregate nor a substantial whole; hence the words ‘entire’ and ‘partial’ are not applicable to it.”22 Thus by a subtle analysis the objection is removed by showing that the alternatives contemplated are not applicable to universals at all.

210

KISOR CHAKRABARTI

The other difficulty is also due to not realising the basic difference between a universal and a particular. A particular cannot exist at more than one place at the same time; but a universal, by hypothesis, is capable of residing at many places at the same time. So the natural thing to say is that a universal ‘resides in all objects belonging to the class’ (svaviṣayasarvagata). So the universal also resides in a new member that happens to be added to the class by being produced and there is nothing mysterious or problematic about it. When an object is produced, ‘the sum total of causal conditions’ (kāraṇasāmagrī) determines its nature and thus to which class it should belong or what kind it should be. So if somebody chooses to wonder how the relationship between a universal and a new member of the class is established, the answer is that it is the causes producing the object that establish the relationship between it and the universal.23 As to the remaining objection the Nyāya firmly held to the position that the admission of universals for man, cow, etc., was justified on the basis of experience.Thus at the time of perceiving two individual men, we directly observe that both share the common property of humanity and this kind of direct experience cannot be nullified by an appeal to cases like the class of cooks.The Nyāya, however, conceded the point that there are cases where in spite of the ‘notion of belonging to the same class’, no universals could be admitted.We must distinguish between general terms such as man, cow, etc. which name universals and other general terms such as cook, father, etc. which do not name any universals. In fact the Nyāya held that before a universal is admitted on the basis of the knowledge of identity, we must ascertain that no violation has been made of any of the ‘restictive conditions for universals’ (jātibādhaka) to the discussion of which we now turn in the following section. III.  RESTRICTIVE CONDITIONS FOR UNIVERSALS

Granted that there are universals, an important problem for the realist is to determine whether there are universals answering to every common name, and if not, how the ­population of the world of universals can be limited to what is reasonable. This problem has figured prominently both in Plato’s philosophy and in Nyāya. We may first briefly indicate Plato’s position regarding this problem. It is asserted in the Republic24 that there is an Idea corresponding to every common name, but there are substantial grounds for doubting whether Plato ever seriously upheld this doctrine. There is first the question whether in Plato’s view there are ideas corresponding to negative terms such as not-man, not-beautiful etc. Plato’s view on this point is not altogether free from ambiguity. In the Sophist he seems to be saying that there is an Idea corresponding to not-beautiful as much as there is one corresponding to beautiful. But in the Politics he says in a definitive way that though terms like not-Greek, ‘not-tenthousand’ stand for parts of the genera man and number, they do not stand for species of them, implying that there is no Idea of not-Greek or of not-ten-thousand. Secondly, in the Parmenides, upon being asked by Parmenides Socrates says that though he recognises Ideas of goodness, beauty, justice, etc., he is more hesitant about the recognition of Ideas of man, fire, water, etc. and that he definitely does not admit ideas of mud, dirt and hair. This is followed by Parmenides’ remark that Socrates’ refusal to recognise the latter Ideas is due to philosophical immaturity and that he should recognise them following the general principle. This exchange of opinion between Parmenides and Socrates indicates

THE NYĀYA-VAIŚEṢIKA THEORY OF UNIVERSALS

211

that though Plato found it necessary to admit ideas for ethical and aesthetic values, he was probably more doubtful about extending the theory of Ideas to such ‘unpleasant’ and ‘trivial’ things as mud, dirt, etc. It seems that the basic reason for the theory of Ideas was to provide an objective foundation for art and morality as well as an objective foundation for knowledge (which prompted admission of Ideas for mathematical objects). From this point of view Plato was doubtless less concerned with things like mud and dirt, because they are, after all, not objects of knowledge in the stricter Platonic sense of knowledge that is necessarily true. Thirdly, it becomes clear from the remarks of Aristotle that some Platonists held that there are Ideas only of substances and that there are no Ideas of such artificially produced things as house, ring, etc. All this evidence suggests that according to Plato and some Platonists there is not a universal answering to every common name – a view very clearly and unambiguously upheld by the Nyāya. One basic restrictive condition as already mentioned is that a universal must be simple, i.e., it must not be analysable into other properties or property components. For this reason a general term like cow or man would stand for a universal, but not terms like white man or black cow. The latter terms represent ‘complex properties’ (sakhaṇdopādhi) and do not name additional, ontologically distinct entities. That is to say, the property of being a black cow is not an additional entity over and above blackness and cowness, but is ­reducible to them. For the same reason negative terms such as ­not-man, not-red, etc., do not stand for universals. Any negation presupposes an idea of what is negated and hence cannot be regarded as simple. Moreover, many terms such as deafness, blindness, etc. which do not have the form of a negative term, but are negative in m ­ eaning, also do not stand for universals. (These terms are comparable to privative terms of Aristotelian logic.) Similarly, relative terms like father, brother, etc. cannot be said to stand for universals, ­because from the very nature of the case they cannot be regarded as simple.25 A second restrictive condition is that whenever a universal is admitted, the relationship between it and its particulars must be ‘necessary’ or ‘inseparable’ in the required sense. For example, let us consider the class of cooks mentioned earlier. Though we can speak of a class of cooks, ‘cookness’ or the property of being a cook is not a universal because the relationship between it and the individuals is not necessary, but accidental. A third restrictive condition is as follows.Two universals may be so related that the first is included in the second, i.e., every member of the first is a member of the second, e.g., humanity and animality. Again, two universals may be so related that neither is i­ncluded in the other, e.g., cowness and horseness. It is held that in the latter case two such ­universals could not have any members in common; in other words, there must not be any cross-­ division or over-lapping of universals. So two universals may have some members in common if and only if the first is included in the second or vice versa.26 A fourth restrictive condition is that no universal can be admitted where the admission would result in violation of the essential nature of members. Thus no universal can be admitted to be shared in common by the so-called ‘ultimate individuators’ (viśesa) which are held to be self-differentiating and which are hypostatised to differentiate qualitatively indistinguishable atoms from one another. (Atoms cannot be held to be different from one another without such hypostatised individuators, because it would have followed by an application of the principle of identity of indiscernibles that atoms which are qualitatively indistinguishable are all one and the same.) Objects of the same kind can always

212

KISOR CHAKRABARTI

be differentiated from objects of other kinds through their difference in kind. Thus a cow can be differentiated from a horse through the universal cowness possessed by the former and not possessed by the latter. If all ultimate individuators are held to be of one kind sharing a common universal, we would also have to hold that they could be differentiated from other entities through the universal possessed by them. But that would be contrary to the essential nature of ultimate individuators, viz., that they are self-differentiating. (It is also necessary to hold that they are self-differentiating. If we ask how an ultimate individuator can be differentiated from another ultimate individuator and postulate that they are distinguished from each other through another ultimate individuator, we would be hurried off into a vicious infinite regress.) Hence the admission of no such universal can be allowed. According to a fifth restrictive condition, if two universals have exactly the same ­members, they are not two different universals, but one. For example, potness and ‘conchshell-like-neck-ness’ (kambugrīvādimatva) are instantiated in exactly the same entities, viz., pots, and hence there are not two different universals corresponding to them, but one. It is obvious that this restrictive condition states what is basically the same as the thesis of extensionality for sets in modern set theory. This restrictive condition was introduced by Udayana, the last great logician belonging to the school of early Nyāya, and in all probability he is the first logician to have formulated the extensionality thesis27 This ­restrictive condition, however, reflects an important change in the concept of a universal. While in the early Nyāya view a universal is understood as a class-property or a repeatable character, because of this restrictive condition a universal ( jāti), for Udayana, comes to be regarded as a class in extension. In fact Udayana held that as an ‘abstract property’ (upādhi), ‘conch-shell-like-neck-ness’ is different from potness. This shows that Udayana distinguished between a universal ( jāti) or a class in extension and an abstract property (upādhi), a distinction which is never made in earlier Nyāya philosophy. It appears that in Udayana’s view corresponding to the same class in extension, there may be more than one abstract property, a view which is also upheld by many modern logicians.28 In fact, in the process of discussing the restrictive conditions, Udayana threw new light on the problem of universals and opened up a whole new perspective as may be seen from the account of the following restrictive conditions. A sixth restrictive condition is that no universal can be admitted to exist, the admission of which would lead to a vicious infinite regress. For this reason there can be no universal of which every universal is a member; for if we had any such universal, then, by hypothesis, we have a given totality of all universals that exist and all of them belong to this big universal. But this universal is itself a universal and hence (since it cannot be a member of itself, because in Udayana’s view no universal can be a member of itself) it too along with others must belong to a bigger universal and so on ad infinitum. What is said here has interesting analogues in modern set theory in which it is held that a set of all sets (i.e., a set to which every set belongs) does not exist. And the reason why the set of all sets cannot exist is similar to that given here, viz., that if such a set existed, it must itself belong to a still larger set and this process will be continued ad infinitum; so that in a certain sense it must be larger than itself, which is a contradiction.29 Udayana of course has no objection to infinite regress as such, but does object to infinite regress of a certain kind which he considers to be vicious. An infinite regress is

THE NYĀYA-VAIŚEṢIKA THEORY OF UNIVERSALS

213

vicious if it makes the ‘basis of the regress’ (mūla) impossible.Thus the infinite regress arising in the case of the universal of all universals is vicious, because what the regress essentially proves is that we could never have such a universal of all universals. There are other cases of infinite regress which are not vicious in the above sense. For example, if one asks whether the tree is the cause of the seed or the seed is the cause of the tree, one would find oneself thrown into a regress that is unstoppable. This regress is not vicious, because in this case the regress starts only after supposing that the tree or seed already exists. The situation is different with the universal of all universals. Here we could not suppose that it already exists because the very concept of a universal of all universals commits us to an infinite regress. In fact if we could suppose that the universal of all universals already exists, the regress would not have started at all.30 A seventh restrictive condition is that no universal can be admitted where the relation of ‘inseparable inherence’ (samavāya) between the putative universal and its members could not be admitted to be possible. We know that the relation between a universal and its members is called the relation of ‘inseparable inherence’. We could ask whether there is a universal which has inseparable inherence as a member, and the answer is said to be no. The reason why the answer is no is that the relation of inseparable inherence could not be admitted to be possible between the universal and inseparable inherence itself. For suppose that the relation of inseparable inherence is possible in a given case.Then we have to suppose that inseparable inherence is related to the universal by inseparable inherence. But the latter is also a member of the universal and hence it also must be related to it by inseparable inherence and so on to a vicious infinite regress. The vicious infinite regress makes the admission of the relation of inseparable inherence impossible in the given case and that in turn rules out the admission of the universal. It is probable that there are cases where the relation of inseparable inherence would be impossible for reasons other than vicious infinite regress and probably that is why this restrictive condition has been stated separately as an additional restrictive condition and not a sub case of the previous one. On this point too we can see an interesting parallel in modern set theory. The relation between a set and its elements is the relation of membership. As it was asked whether there is a universal whose member is inseparable inherence, so it may be asked whether there is a set whose member is membership and the answer is no. This set may be expressed in set-theoretical terms as the set of all ordered pairs such that x ¨ y, which may be written with the standard notation as follows: { | x ¨ y}. Assuming that this set exists, we can talk about its domain D: D { | x ¨ y} = {x | ∃ y (x ¨ y)}. But given any set x, it is a member of some set, e.g., {x} From this it follows that ∀x ∃y (x ¨ y); from which it follows that ∀ x x ¨ D. Thus D is proved to be the universal set which as already seen in the discussion of the previous restrictive condition, could not exist; and hence it follows that the above set also could not exist. Finally, a property that could not be instantiated in more than one object is not a universal. The very basis of admitting a universal, viz., the notion of belonging to the same class, is impossible in such a case. For example, diktva or spaceness is not a universal, because space is one and infinite, and hence the property spaceness is instantiated only in one object. Spaceness, however, is a ‘simple property’ (akhaṇḍopādhi) and like a universal is held to a real entity not ontologically distinct from its locus, viz., space.

214

KISOR CHAKRABARTI

By implication a property which could be instantiated in no objects at all is also not a universal; and precisely for the same reason as above, viz., that the notion of belonging to the same class is impossible in such a case. For example, general terms such as ‘rabbit’s horn’, ‘son of a barren mother’, ‘turtle’s hair’, etc., do not represent universals. Udayana devoted considerable attention to the analysis of such ‘non-referring expressions’ and at the risk of a slight digression we may briefly state the gist of it. We may begin with an examination of statements like ‘the rabbit’s horn does not exist’; in nine out of ten cases the subject term of a sentence denotes an object. But to hold that the subject term of our sentence denotes an object is clearly unsatisfactory, because the very meaning of the sentence is that there is no such object as a rabbit’s horn. Again let us consider the statement ‘the rabbit’s horn is not sharp’. It may seem that the statement is true; since the rabbit’s horn does not exist, it is false to say that it is sharp and hence the statement that it is not sharp ought to be true. But Udayana points out that both the statements ‘the rabbit’s horn is sharp’ and ‘the rabbit’s horn is not sharp’ are false and proper analysis will show that there is nothing paradoxical about it. The statement ‘the rabbit’s horn is sharp’ is much more complex than it may apparently seem and ought to be analysed as follows: ‘the rabbit exists, the horn exists, the horn is sharp and the rabbit possesses the horn’. Now it is clear that the statement is false, because one of the conjuncts, viz., that ‘the rabbit possesses the horn’ is false. For the same reason the statement ‘the rabbit’s horn is not sharp’ is also false. Similarly, the statement ‘the rabbit’s horn does not exist’ should be analysed as ‘no rabbit has a horn’ in which case the apparent difficulty of not finding any denotation for the subject term is removed.31 It should be noted that this analysis of ‘non-denoting expressions’ is remarkably similar to that of Russell.32 Russell also held that both the statements ‘the present king of France is bald’ and ‘the present king of France is not-bald’ are false and gave a similar analysis of these two statements. Russell rejected33 Frege’s view that definite descriptions always have denotation on the ground that this creates problems for expressions like ‘the present king of France’ where the denotation appears to be absent. Nyāya, likewise, would not subscribe to the position that all definite descriptions are denoting expressions. In the Nyāya terminology a definite description like ‘the son of Yajñadatta’ stands for a ‘complex predicate’ (sakhaṇḍopādhi).The expression of a complex predicate is a denoting expression when every component of it is the name of a real entity. Thus in our example ‘the son of Yajñadatta’ is a denoting expression, because every component, viz., ‘son’, ‘Yajñadatta’ and the relation ‘of ’, is the name of a real entity. But ‘the rabbit’s horn’ is not a denoting expression, because one of the components, viz., possession of horn by rabbit, is not the name of a real entity. We may now bring together the results we have so far obtained.There are first of all terms like man, cow, etc., corresponding to which there are universals which are ­ontologically real entities distinct from individual objects. Secondly, there are terms like spaceness, etc., corresponding to which there are no universals, but are called ‘simple ­properties’ which are also ontologically real and distinct from their loci.Thirdly, there are terms like ‘black cow’,‘white man’, etc., which represent complex properties and do not stand for any new, additional universals over and above those represented by their simple components.The peculiarity of the Nyāya view is that both indefinite descriptions like ‘a Brahmin from Kashi’ and definite

THE NYĀYA-VAIŚEṢIKA THEORY OF UNIVERSALS

215

descriptions like ‘the son of Yajñadatta’ are thrown into the class of complex properties. Fourthly, there are terms like ‘rabbit’s horn’, ‘son of a barren mother’, etc., which, unlike those listed above, are ‘non-denoting expressions’. Finally, there are terms like ‘the universal of all universals’, ‘the universal whose member is inseparable inherence’, etc., which could not represent any universals because of the special difficulties involved.34 To conclude: We have seen that the Nyāya and Vaiśeṣika believed that universals are real entities ontologically distinct from particulars and inseparably inherent in them. The distinctive feature of the theory is that universals are held to be simple as well as sensible. Another distinctive feature is an elaborate study of various restrictive conditions that ­prevent the admission of universals answering to every common name. Our discussion shows that though one may not necessarily agree with either the main tenet of the theory or specific conclusions drawn on particular points, as a theory it is worked out extremely well to minute detail requiring subtle philosophical analysis. For all these reasons the Nyāya-Vaiśeṣika theory should be regarded as an interesting counterpart of the theories of universals in Western philosophy. NOTES 1. See (20), pp.  51–2. The number in parentheses refers to the number of the entry in the bibliography. Though the theory of universals presented in the paper was developed by the Vaiśeṣika as much as by the Nyāya school, we have avoided writing down the word Vaiśeṣika every time a reference to the Nyāya view has been made. 2. See (20), p. 59. 3. See (8), section on samavāya. 4. We have used the word connotation for the want of anything better. It should here be ­understood to mean ‘the mode under which an entity becomes presented through the word’ and is comparable to Frege’s Sinn. For Frege’s view, see ‘On Sense and Nominatum’ reprinted in (2). 5. The first two arguments have been concisely stated by Śrīdhara as follows: “… the nature of each object being different from others, how could there be knowledge of a common form in them (without the universal), and how could they be designated by the same word? (Without the universal) the relation of a word with the unlimited individual objects of a class could not be established.” See (17), p. 32. 6. See (15), pp. 80 ff. 7. See (15), pp. 458 ff. 8. See (12), p. 94. 9. The classification is also often said to be twofold, viz., (a) the universal of largest extension and (b) universals with an extension smaller than that of another universal. Raghunātha of Navya-Nyāya, however, held that there is no universal of the largest extension like sattā or ‘existence’ in which every universal is included. See (20), pp. 59–60. 10. Radhakrishnan writes: “The universals on this view, answer to the separate, supra-sensual arch-typal forms. … Praśastapada’s view is akin to Plato’s realism, according to which sensible things are what they are by participation in the universal forms of Ideas which are eternal and self-subsistent.” See (10), vol. 2, pp. 211–2. Our remarks show that such assimilation of the Nyāya view to Plato’s view is a mistake. 11. See (12), p. 100.

216

KISOR CHAKRABARTI

12. See (20), p. 52. A definition must be non-circular and free from the defects of being too wide, too narrow and impossible. (Impossibility here means not applying to any definiendum at all.) A definition must also be ‘simple’ and avoid superfluous qualification. 13. See (1), chap. IX. 14. See (13), chap. on Pramāṇapārāyaṇa. 15. “Mere similarity is not enough as held by Sāṃkhya, because individuals of different kinds are also known to be similar. …” See (7), p. 204. 16. See (20), p. 27. 17. See (3), vol.VI, p. 215. 18. See (17), p. 755. For a discussion of this and other Buddhist objections also see (15), chap. IX. 19. See (16), p. 94. 20. On this point the Buddhist view is comparable to Spinoza’s view that all determination is negation. For a fuller account of the Buddhist view see (14), Introduction. 21. See (4), p. 278. 22. See (18), p. 669. 23. See (17), p. 754. 24. For the necessary references in this paragraph see (11), chap. XI. 25. See (5), p. 73. 26. For this and following restrictive conditions see (4), pp.  73–7. This particular restrictive ­condition and the next restrictive condition were, however, rejected by Raghunātha see (20), p. 56. 27. Udayana’s date is held to be 10th or 11th century A.D.The author is not aware of any earlier logician who could be claimed to have formulated the extensionality thesis. For Udayana’s date see (10), vol. 2, pp. 152–3. 28. See (9), p. 88. 29. See (6), p. 170. 30. The distinction between an infinite regress that is vicious and an infinite regress that is not vicious is a very important logico-mathematical discovery and the author proposes to ­unravel the full significance of this distinction in a future paper. 31. See (5), pp. 133–40. 32. See Russell’s ‘On Denoting’ reprinted in (2), pp. 108, 112. 33. See (2), p. 107. 34. A very significant development that took place in post-Udayana Nyāya philosophy (usually known as Navyanyāya) was the generalisation of the concept of upādhi already introduced by Udayana. Navyanyāya logicians used the term upādhi to mean any class of objects, irrespective of the number of objects in the class, irrespective of whether objects in the class shared a simple, common property, irrespective of whether the relation between the objects and the class was ‘inseparable’. Everything that exists came to be regarded as an upādhi in this generalised sense. It appears that the Navyanyāya concept of upadhi was a close approximation to the modern concept of a set. A full discussion of the Navyanyāya theory of upādhi is, however, beyond the scope of this paper and has to be left to be the subject-matter of a series of papers in future. BIBLIOGRAPHY Aaron, R. I., The Theory of Universals (Oxford: Clarendon Press, 1952). Feigl, H. and Sellars, eds., Readings in Philosophical Analysis (Appleton-Century-Crofts, 1949). Fowler, H. N., ed., Plato (William Heinemann, 1917). Jayanta, Nyāyamañjarī (Benares, Chowkhamba: Kashi Sanskrit Series No. 106, 1936).

THE NYĀYA-VAIŚEṢIKA THEORY OF UNIVERSALS

217

Maitra, S. K., Fundamental Questions of Metaphysics and Logic, Calcutta, 1956. Matilal, B. K., Epistemology, Logic and Grammar in Indian Philosophical Analysis (The Hague, Mouton, 1971). Mendelson, E., Introduction to Mathematical Logic (D.Van Nostrand, 1964). Mishra, U., History of Indian Philosophy,Vol. II (Allahabad, Tirabhukti Publications, 1966). Pap, A., An Introduction to Philosophy of Science (New York, Free Press, 1962). Radhakrishnan, S., Indian Philosophy,Vols. I and II, (London, George Allen and Unwin, 1929). Ross, D., Plato’s Theory of Ideas (Oxford, Clarendon Press, 1963). Russell, B., The Problems of Philosophy (Oxford University Press, 1912). Śālikanātha, Prakaraṇapañcikā (Benares, Hindu University, 1961). Sharma, D., The Differentiation Theory of Meaning in Indian Logic (The Hague, Mouton, 1969). Shastri, D. N., Critique of Indian Realism (Agra University, Agra, 1964). Shastri, H., ed., Six Buddhist Nyāya Tracts (Calcutta, Bibliotheca Indica, 1910). Śrīdhara, Nyāyakandalī and Padārthadharmasaṃgraha of Praśastapāda, ed. with a Hindi translation and introduction by Jha, D. (Benares Sanskrit University, 1963). Tarkatirtha, A. and Tarkatirtha, T., eds., Nyāyadarśana with Bhāsya, Vārttika, Tātparyatīkā and Vṛtti ­(Calcutta Sanskrit Series, 1936). Udayana, Kiraṇāvali, eds. Sarvabhauma, S. and Vedantatirtha, N. C. (Calcutta, Bibliotheca Indica, 1956). Viśvanātha, Bhāṣāpariccheda and Siddhāntamuktāvalī, ed. with a Bengali translation by Shastri, P. (Calcutta, Sanskrit Pustak Bhandar, 3rd. ed., 1969).

CHAPTER

13

Objectivity and proof in a classical Indian theory of number Jonardon Ganeri

Source: Synthese, 2001, 129:3, 413–437. 1.  MATHEMATICS AND THE PHILOSOPHICAL THEORY OF NUMBER

The theory of number serves as a highly instructive example of the nature of rationality, theory-­change and philosophical thinking in India.  I  will show how the processes of theory rejection and theory modification are driven by perceived explanatory failure rather than by discovery of foundational incoherence. I will argue too that the source of structure, order and pattern is seen to reside in the natural world, and not in any domain of abstract objects or in the empirically unconstrained working of a priori reflection. The discussion of number broaches fundamental questions about the nature of objectivity and the correspondence theory of meaning, and it explores the basic principles underlying the construction of a formal ontology. At the end, I will make some comparisons between Indian philosophers’ theory of number and the Indian mathematicians’ description of the epistemology of mathematics and the nature and function of mathematical proof. The philosophical study of number in India specifically, the theory of the realist Nyāya-­ Vaiśeṣika school,1 belongs with the analysis of the structure of empirical knowledge. This theory of number is an account of the semantics and epistemology of contingent numerical judgements, judgements whose subject matter concerns questions of numerical quantity. Such judgements are typically expressed by statements like ‘there are two pots in this room’ or ‘there are nine planets in the solar system’. The study of number is therefore an enquiry into the role of number within empirical and scientific knowledge. Like much classical philosophy in India, the Nyāya-­Vaiśeṣika theory of number is broadly empiricist: facts about the numerical properties of objects do not differ in kind from facts about their colour, shape or extension. Number is part of the natural world.While these philosophers 218

OBJECTIVITY AND PROOF

219

generally do not discuss explicitly the nature of mathematical truth, they implicitly favour an empiricist philosophy of mathematics, according to which the laws of arithmetic are of the same kind as the laws of nature, differing only in the degree of generality they possess. Numerals have three distinguishable grammatical roles. They function as adjectives in sentences like ‘there are nine planets’, as predicates in sentences like ‘the planets are nine’ and as substantives in sentences like ‘nine is the square of three’. In keeping with their disinterest in mathematical truth, these authors do not much discuss the substantial use of the numerals, which is the use most relevant to arithmetic. Because the numerals are not treated as names of numbers, but rather as numerical predicates, there was no reason to introduce numbers into the ontology as objects; there is no tendency, for example, to analyse number set-­theoretically. The adjectival and predicative role of the numerals, on the other hand, are analysed in depth. It is a result of modern semantic theory that the adjectival use of the numerals can be expressed in a first-­order language with identity, and this result is partly anticipated by Bhāsarvajña. The dominant approach, however, was to treat numbers as properties of objects, exactly as colour adjectives stand for colour properties. A strict semantic realism among these philosophers led to the reification of anything which appears in thought or language. However, in their fascinating treatment of vagueness and the term ‘many’, the Nyāya-­Vaiśeṣika philosophers reach the limits of their semantic realism. 2.  NUMBER AS A PROPERTY OF OBJECTS 2.1.  The ‘number of qualities’ problem

Numbers are mentioned by Kanada (c. ad 100) in Vaiśeṣikasūtra 1.1.9 as one member in a list of seventeen kinds of ‘quality’ (guṇa).The Vaiśeṣika concept of a quality is a technical one. Qualities are distinguished, on the one hand, from substances and movements, and on the other, from universals and the so-­called ‘particularities’ (viśeṣa). A quality is a non-­ repeatable property-­particular, a trope.The specific shade of grey a cloud has at a moment in time is in this sense a quality, something which is different from, and yet belonging to the same general type as, every other shade of grey. The Vaiśeṣika qualities are defined further as properties of substances alone. An immediate corollary (VS 2.1.16) is that qualities do not themselves have qualities. The category of quality is not restricted, however, to monadic property-­particulars; some qualities, for example contact and disjunction are relation-­particulars. Vaiśeṣika theory assimilates numbers into the category of qualities. There is, in a given pair of objects, a specific quality of duality. This duality is different from the duality in any other pair, and yet all such dualities belong to the same universal kind twoness. As one is a number, every object possesses a unit-­quality, distinct from but of the same kind as every other unit-­quality. The theory is based on a reasonable intuition that when a number of objects are collected together, the number of objects in the collection is a fact about the objects over and above the facts about the individual objects. It is, however, highly problematic, as the Nyāya-­Vaiśeṣika authors were well aware. Following the terse discussion in VS 7.2.1–9, the difficulty most strongly emphasised is the problem of accounting for our apparent ability to speak of numbers of things other than substances:

220

JONARDON GANERI

7.2. 1 Because of its difference from colour, taste, smell and touch, oneness is a different [kind of] thing. 2 Similarly, separateness [is a different kind of thing]. 3 The absence of oneness and separateness-­of-­one in oneness and separateness-­of-­one is explained by minuteness and magnitude. 4 Oneness is not present in everything, because motions and qualities are devoid of number. 5 That [application of ‘one’ to motions and qualities] is an error. 6 In the absence of oneness, there is no derivative application [of the word ‘one’ to motions and qualities]. 7 Because of the absence of oneness and separateness-­of-­one in cause and effect, unity and separateness-­of-­one are not present. 8 This is explained for the two non-­eternals [viz. non-­eternal number and separateness]. The principal objection rehearsed here is that individual substances each have the numerical quality of being one, and yet, since qualities do not inhere in qualities or motions, individual qualities and motions cannot be said to be numerically one. We might call this the ‘number of qualities’ problem. The Vaiśeṣikasūtra solution to the problem is disarmingly simple. It consists in denying that it is literally and properly correct to speak of a number of qualities. It would be an error to do so. The Vaiśeṣikasūtra explanation for the use of the term ‘one’ to objects other than substances is that it is a derivative use, a metonymic extension of the term beyond the domain of its proper application.VS 7.2.6 is to be understood in this context as pointing out that if the term ‘one’ were not properly applicable somewhere, i.e., to substances, then it could not be used metonymically either. According to the standard analysis of derivative meaning (lakṣaṇā), a term may refer derivatively to objects that are in some way related to the literal reference of the term (see Raja 1963: 231–233, Ganeri 1999: 91–94). One may say ‘The platform is shouting’, meaning that the people on the platform are shouting. The point of 7.2.6 is then that without a base use of the numeral ‘one’ to substances, there could be no metonymic application of the term to other categories of thing.2 At face value, this attempt to solve the problem has little to recommend it. There is nothing about the use of numerals in statements about quality and motion to distinguish it from the use in application to substances.3 The expression ‘two’ has precisely the same meaning in ‘There are two pots on the ground’ as in ‘There are two shades of blue in this painting’ or ‘There are two movements in that dance-­step’. Later writers took the idea to be that such statements are indirect ways of talking about the number of substances. Bearing in mind that qualities are nonrepeatable properties, it follows that there are exactly as many shades of blue as there are blue things. We might hope, then, to paraphrase a statement about the number of blue qualities as a statement about the number of blue substances. The idea is that the phrase ‘two (shades of ) blue’ does not attribute a quality (two) to a quality (shade of blue), but attributes both qualities to a substance. It is paraphrased as ‘two blue (substances)’. The relation between numbers and qualities is not one of inherence but rather one of ‘co-­inherence in a single substratum’.

OBJECTIVITY AND PROOF

221

2.2.  The problem of self-­inherence

This is, perhaps, a solution to the ‘number of qualities’ problem. However, as later authors were well-­aware, the problem is only a special case of some more general difficulty. For Śaṅkara Miśra (c. 1430), the real problem is one of self-­inherence.The trouble is with talking about the number of numbers, and it presents the same difficulty as talking about the movement of movements or the quality of qualities. He therefore explains VS 7.2.3 in the following way (1923: 174): [Objection:] Since one can say such things as ‘oneness is one’ or ‘separateness is separate from colour and so on,’ surely oneness is in oneness and separateness is in separateness. Because of this, it is said [7.2.3] that ‘the absence of oneness and separateness-­of-­one in oneness and separateness-­of-­one is explained by minuteness and magnitude’. [The meaning is that] just as minuteness and magnitude do not possess minuteness and magnitude, the application of which [characteristics] to them is derivative, so oneness and separateness do not possess oneness and separateness, the application of which to them is derivative. Two later sūtras, viz. “by motions, motions” [7.2.24], and “by qualities, qualities” [7.2.25] exhibit the same derivative meaning as this earlier one.The meaning is that just as motions do not possess motions and qualities do not possess qualities, so too oneness and separateness do not possess that [oneness and separateness]. Śaṅkara Miśra’s argument is that because nothing can inhere in itself, apparently acceptable talk of self-­inherence must be non-­literally construed. In particular, when we say that there is one number one, we are not really saying that the number one inheres in itself, any more that when we say that minuteness is minute, we are saying that minuteness inheres in itself.The point does not apply to numbers other than one, since we would not say that there are two number twos, even if we want to say that there is one number one. So for Śaṅkara Miśra, the non-­literal use of numerals is restricted solely to the use of ‘one’ to itself. Śaṅkara Miśra is right to be worried about the idea of self-­inherence. It is a straightforward matter to modify Russell’s paradox and show that the property of non-­self-­ inherence is incoherent. For let R be the property of being a non-­self-­inherent property. Is R self-­inherent or not? If it self-­inherent, then (by the definition of R) it is a non-­self-­ inherent property – which is a contradiction. If it is not self-­inherent, then (by the definition of R) it is not a non-­self-­inherent property – again contradictory. Russell’s solution in his Theory of Types was to block the paradox by ruling as insignificant the application of a property to itself, so that the incoherent property cannot be formulated (cf. Sainsbury 1979: 315, 1995: 111). It is a little surprising that none of the Nyāya-­Vaiśeṣika authors discovered the paradox. There is, however, some hint of it in Gaṅgeśa’s account of ‘non-­ pervasive’ location (see Matilal 1975: 455–7). As has recently been shown, the celebrated grammarian Bhartṛhari hit upon a paradox of a rather similar type, when he argued that the denotation relation is undenotable, on the grounds that no relation can have itself as a relatum (H. & R. Herzberger 1981; Houben 1996).

2.3.  The problem of cross-­categoricity

The self-­inherence problem is not a grave difficulty for the general theory that numbers are qualities, for most numbers are not self-­inherent, even if it forces us to regard the claim

222

JONARDON GANERI

that there is one number one as having only non-­literal significance. More serious is the problem that numbers are cross-­categorial, that entities in any category can be enumerated. This is seen by Raghunātha (see below) as a fatal flaw in the doctrine. If we can speak about the number of things in any category, then in what category do we place numbers themselves? We may note here in passing that in contemporary discussions of number, the theory that numbers are properties of objects is considered refuted by the simple observation that, in a statement like ‘there are two people in the room’, no one person is said to be two (cf. Sainsbury 1991: 159). This observation, however, refutes only the thesis that numbers are unary properties; it leaves open the possibility that they are many-­place relations. It is for the same reason that the statement ‘the people in the room are friends’ does not imply that each person is a friend. I discuss this further below. 3.  NUMBER AS A RELATION OF NON-­IDENTITY

Bhāsarvajña (c. 950) is a radical Naiyāyika who rejects the classical Vaiśeṣika theory that numbers are qualities. His discussion of number is very astute and insightful, and contains some important innovations. His theory is that number adjectives do not stand for qualities of substances, but for relations of identity and difference (1968: 159): Identity and difference depend on sameness and distinctness in colour and so on, and so are not considered to be qualities. Further, it is a tautology to say “the one is identical” or “the many are different.”

Bhāsarvajña’s idea is simplest to see in the case of the predicative use of numerals. The statement ‘a and b are one’ is, he claims, synonymous with ‘a = b’. To say that Hesperus and Phosphorus are one is just to assert the identity ‘Hesperus = Phosphorus’. On the other hand, the statement ‘a and b are two’ asserts that a ≠ b.The adjectival use of numerals admits of a similar analysis. Indeed, it is now standard to formalise sentences of the form ‘there are n Fs’ by means of non-­identity. The analysis varies, depending on whether we read the sentence as asserting that there are at least n Fs or as asserting that there are exactly n Fs. Thus, ‘there are at least n Fs’ is paraphrased by (see Sainsbury 1991: 160): (∃x1) (∃x2) … (∃xn) (Fx1 & Fx2 & Fxn & x1 ≠ x2 ≠ x3 & x1 ≠ x3 & … & xn–1 ≠ xn The problem of the ‘number of qualities’ presents no particular difficulty for this account. For to say that there are two shades of blue in the painting is to assert that the two shades are different from each other. As long as we can refer to and quantify over qualities, the analysis is the same as before. The analysis of statements involving ‘one’ as identity statements is the source of some unclarity. Udayana (c. 1000) objects to the account on the grounds that identity is something unique to each thing, but we can say ‘one pot’, ‘one cloth’, and so on. Since the identity of the pot is not the same as the identity of the cloth, it cannot be the meaning of the adjective ‘one’ here (Udayana 1956: 442; cf. Potter 1977: 597). The objection seems to be that if ‘a is one’ means ‘a = a’, then the predicate ‘… is one’ means ‘… = a’, and ‘b is one’ then comes to mean ‘b = a’ rather than ‘b = b’. This objection has some force if Bhāsarvajña’s claim is that ‘one’ means identity in the sense of the individuating essence

223

OBJECTIVITY AND PROOF

of a thing. But we need not read him this way. It is not after all very clear what we are doing when we say that something is one, but Bhāsarvajña’s proposal is that we are not attributing a property to the object, but asserting, trivially, that it is identical to itself. It is interesting to note that Bhāsarvajña’s analysis is echoed, very much later, in Gadādhara’s comments on the meaning of the word ‘one’. Gadādhara (c.  1650) is discussing the adjectival use of the term, and he states that the meaning of ‘one F’ is: an F as qualified by being-­alone, where ‘being alone’ means ‘not being the counterpositive of a difference resident in something of the same kind’ (Gadādhara 1929: 189). In other words, ‘one F’ is to be analysed as saying of something which is F that no F is different to it. If this is paraphrased in a first-­order language as Fx & ¬(∃y) (Fy & y ≠ x) then it is formally equivalent to a Russellian uniqueness clause: Fx & (∀y) (Fy → y = x) This idea, that as used adjectivally ‘one’ expresses uniqueness, seems to be in the spirit of Bhāsarvajña’s claim that it expresses the identity of a thing. In any case, it is clear that, for Gadādhara, ‘one’ has a role similar to that of the definite article. After presenting his radical analysis, Bhāsarvajña goes on to claim for it a much greater degree of generality than possessed by the old account (1968: 159): Now the application of ‘twoness’ to qualities etc. is unproblematic. Moreover, even its use in arithmetic, as when one says “a hundred and a half a hundred by six hundred” goes through. So it is said that one is the initial integer, two is that [one] together with another identical, three is [those] two together with another identical, four is those [three] together with another identical, and so on. The convention for the grounds for the use of ‘twoness’ etc., established by the treatises on arithmetic, is explicable [in this way].

Bhāsarvajña reveals here a clear awareness that the integers can be defined recursively, and, unusually for the Nyāya-­Vaiśeṣika authors, displays some familiarity with the mathematical literature. He says that one is the first whole number, two is one together with another identical, three is two together with another identical, and so on. The Sanskrit term for integer is ‘abhinna’, which means ‘undivided’ or ‘whole’, as well as ‘identical’, and Bhāsarvajña plays on the ambiguity in order to switch from the adjectival to the substantival use of numerals. 4.  NUMBER AND COLLECTION

The anomalous status of numbers in the traditionalVaiśeṣika system is perceived very clearly by Raghunātha Śiromaṇi (c. 1510). In another radical departure, he declares numbers to belong to a new category altogether (1957: 86–7, Potter’s translation, slightly emended): Number (is) a separate category, not a (kind of) quality, for (we make) the judgment (that there is) possession of that (namely, number) in qualities, etc. And this (judgment we make that qualities have number is) not an erroneous one, for (there is) no (other) judgement (we make which contradicts (it). If (you argue) that judgments of this kind (occur when

224

JONARDON GANERI

there) is inherence of two qualifiers in one individual (substance), (I say) no, for inherence and inherence-­of-­two-­qualifiers-­in-­one-­substratum (are) two (quite) different (connectors), from which (one) cannot derive the homogenous idea of possession. And just as the judgement of oneness in potness, etc., (is to be explained by the opponent) through inherence-­ of-­two-­qualifiers-­in-­one-­substratum, so (there ought to be)  a  judgment of twoness in colourness, etc., (explainable) through inherence-­of-­two-­qualifiers-­in-­one-­substratum, and that (is) not possible.

Raghunātha puts his finger in exactly the right place: the is-­the-­number-­of relation is not reducible to the relation of inherence or any relation constructed out of it. The fact that each quality is unique to the substance it inheres in means that there is a one-­to-­one correspondence between substances and qualities. So we might hope to paraphrase talk of the number of qualities with talk of the number of substances. But this will not solve the cross-­categorial problem, since we can also count universals. On the other hand, merely to say that numbers belong in a category of their own does not yet tell us very much about them. Raghunātha’s main contribution to the Nyāya-­ Vaiśeṣika theory of number lies in his account of the relation between numbers and things numbered and its distinction from the inherence relation. He observes that while inherence is a distributive relation, the number-­thing relation has to be collective. The distinction occurs in the context of sentences with plural subjects, for example, ‘the trees are old’. An attributive relation is distributive if it relates the attribute to every subject: if the trees are old, then each individual tree is old. A relation is collective if it relates the attribute to the subjects collectively but not individually. Thus, ‘the trees constitute a forest’ does not imply that each tree constitutes a forest. Number attributions are collective. If one says that there are two pots here, one does not imply that each pot is two. Inherence, however, is a distributive relation, and so cannot be the relation of attribution for numbers. This relation is called the ‘circumtaining’ relation (paryāpti) by Raghunātha (p. 38): the ‘circumtaining’ relation, whose existence is indicated by constructions such as “This is one pot” and “These are two”, is a special kind of self-­linking relation.

His commentator Jagadīśa explains: It might be thought that the circumtaining relation is [in fact] nothing but inherence … So Raghunātha states that circumtaining [is a special kind of self-­linking relation]. … In a sentence like “This is one pot”, circumtaining relates the property pot-­hood by delimiting it as a property which resides in only one pot, but in a sentence like “These are two pots”, circumtaining relates the property twoness by delimiting it as a property which resides in both pots. Otherwise, it would follow that there is no difference between saying “These are two” and “Each one possesses twoness”.

Thus, the number two is related by the circumtaining relation to the two pots jointly, but to neither individually. If it were the case that collective relations have to be analysed as relations to classes, and not to the members of classes, then Raghunātha’s idea would resemble to some degree Frege’s definition of numbers as properties or classes of classes (cf. Ingalls 1951: 77; Shaw 1982). This is, however, premature, for we are not forced to analyse all collective relations as relations to classes (cf. Mohanty 1992: 246; Ganeri 1996). Indeed,

OBJECTIVITY AND PROOF

225

the idea that numbers relate to objects collectively is quite consistent with Bhāsarvajña’s idea that number statements are statements of the mutual distinctness of a plurality. The significance of Raghunātha’s innovation is that the domain of numbers is seen now to relate to the world of objects in a way very different from that of the domain of universals and qualities. Numbers thereby acquire a distinctive ontological status, although there is no reason to suppose that Raghunātha takes them to be abstract or Platonic objects. 5. PRAS´ASTAPADA’S ‘EIGHT MOMENTS’ THEORY

Praśastapāda (c. 530) supplements the Vaiśeṡikasūtra account of number with a detailed description of the ontogenesis of numbers, and of the psychological process by which they come to be cognised. The entire process is supposed to take eight moments of time (1923: 48–50): When [1] the eye of the cogniser contacts with a pair of substances of a similar or a dissimilar type, a cognition arises [2] of the generic oneness which inheres in those [qualities] which inhere in the objects of contact, and then [3]  a  single awareness arises of the two unit-­ qualities from the cognition of the relation between them and the generic oneness. Next [4]  twoness begins to form from the two unit-­qualities in their substrata, in dependence of that [single awareness]. But then [three things happen] at a single time [5]: a cognition of the universal twoness arises in this; the start of the decline of the combinative cognition (apekṣābuddhi) due to that cognition of the universal twoness; and the beginning of the emergence of a cognition of the duality-­quality from the cognition of the relation between it and the generic twoness. And now, [three more things happen] at a single time [6]: due to the destruction of the combinative cognition, the duality-­quality begins to decline; the cognition of the duality-­quality causes the destruction of the cognition of the universal twoness; and there begins to emerge a cognition of the two substances, in the form ‘[there are] two substances’ from the duality-­quality and its relation to the cognition. Following this, [four things happen] at a single time [7]: the [full] arising of the cognition of the two substances in the form ‘[there are] two substances;’ the destruction of twoness; the beginning of the decline of the cognition of the duality-­quality; and the beginning of the emergence of a dispositional memory trace from the cognition of the two substances. Following this [8], [there is] the destruction of the cognition of the duality-­quality from the cognition of the two substances, and the [full] dispositional memory state also from the cognition of the two substances.

This intricate description of the process is distilled by later writers into the following sequence of eight events (cf. Udayana 1956: 449, Śaṅkara Miśra 1923:177): 1 Sensory connection with the two objects. 2 Awareness of the general property oneness. 3 Combinative cognition. 4 Production of twoness in the objects. 5 Awareness of the general property twoness. 6 Awareness of a duality-­quality. 7 Knowledge of the two substances as qualified by such a duality. 8 Dispositional memory trace.

226

JONARDON GANERI

Praśastapāda’s theory of the production of number and numerical knowledge is as follows. One first perceives two separate objects. On perceiving an object, one can become aware of the sorts of qualities it possesses. Technically, this type of awareness is called saṃyukta-­ samaveta-­samaveta, awareness of what inheres in what inheres in the object directly seen. In this case, what one becomes aware of is the universal property oneness, a property which inheres in every unit-­quality (property-­particular), and so in the particular unit-­qualities which inhere in the two perceived objects. The third step is a mental act of counting or enumeration, a ‘combinative cognition’.The thinker collects the two objects into a single awareness, judging ‘(here is) one and (here is) one’.The mental counting up of the objects produces in those objects a new feature, the property of being two. Although produced by the cogniser’s counting, and depending for its continued existence on the continuing existence of that event, the number is an objective property of the objects themselves. Once the number two has come into existence, it is possible for the cogniser to become aware of it. He becomes aware, first of all, of the universal property twoness which inheres in all duality-­qualities, then of the particular duality-­quality which inheres in the two perceived objects, and finally of the two objects as qualified by that particular duality-­quality. This episodic knowledge-­event gives way to a dispositional memory trace, which preserves the knowledge for the cogniser after the episodic knowledge-­event has faded away. A detailed examination of the psycho-­dynamics of this account would be beside the point here. I will make a brief comment before turning to examine the important role attributed to the ‘combinative cognition’ or mental act of counting in the theory, and to its epistemology of number. The account depends on the Vaiśeṣika doctrine that mental events have a life cycle of three moments. Created in the first moment, they develop into full-­blown and causally efficacious events at the second moment, and then decline to their final destruction at the third.This leads to a difficulty in the eight-­moments theory, for the number two in the objects, created at moment (4) will have ceased to exist by the time the cogniser becomes aware of it at moment (7). The Nyāya-­Vaiśeṣika authors expend considerable energy on this rather sophistical puzzle. Potter (1977: 290) accurately observes that Praśastapāda’s account tries to reconcile the theory with the doctrine that ‘in order for one stage in a process to destroy the previous one that previous one must be capable of destruction’. Older theories apparently ran into a problem here, for the knowledge that the objects are two, which arises at moment (7), ought to but cannot destroy the number two, since that is already destroyed at moment (6). Praśastapāda’s rather straightforward solution to the problem is to give the combinative cognition a life-­span of four, rather than three, moments (cf. Faddegon 1918: 201). 6.  OBJECTIVITY IN THE THEORY OF NUMBER

The doctrine that numbers are created by mental acts of counting is important but difficult. Potter regards the doctrine as ‘a very dangerous admission’ for a school that ‘wishes to maintain a surefooted realism about the external world’ (1977: 120). Among the Nyāya-­Vaiśeṣika, Śrīdhara (c. 990) is most aware of the danger, and he reasons as follows (1977: 27–5): This [combinative] cognition produces the twoness of the two objects which are the substrata, from the two unit-­qualities which [each] depend on [the] universal oneness. The two

OBJECTIVITY AND PROOF

227

substrata are inherent causes, the two unit-­qualities are non-­inherent causes, and the cognition of both is another causal factor. … The thesis that an object can be produced by a cognition is not an unworldly one, for the production of pleasure etc. from [cognitions] is seen [to be similar]. Nor is it right to say that there is a disanalogy [between numbers and pleasures] in that the production of an external object in such a way is not seen, for there is no distinction here between the two cases in the manifestation of the correlation [between cognition and production of object]. [Objection:] Although it has been shown that the causal foundation (ālambana) of the two qualities is what makes manifest the twoness, it is not the case that the cognition is its immediate antecedent cause. [Reply:] No, [if that were so then there would be] no regulation. If twoness were not produced by the cognition, then other people could cognise it, as with colours and so forth, because there would be no regulating causal factor. When produced by the cognition, however, this regularity arises: that which is produced by a person’s cognition is ascertainable by him alone. The demonstration runs as follows – 1 Twoness is produced by cognition, 2 Because it is restricted to being thought of by a single cogniser. 3 That which is restricted to being thought of by a single cogniser is produced by cognition; as, for example, a pleasure. 4 Twoness is restricted to being thought of by a single cogniser. 5 Therefore, it too is produced by cognition.

This is a very interesting passage. A number is an objective entity, even though it is both private (capable of being cognised by at most a single person) and also mind-­created. The inference from privacy to being mind-­created is relatively unproblematic – it is hard to imagine how something which can only be cognised by a single person can be entirely mind-­independent. But how does this give any support to the claim that numbers are objective? The analogy with pleasures and pains is apt, for the Nyāya-­Vaiśeṣika entertain the curious theory that pleasures and pains are distinct both from their (mental) causes as well as from the perception of them (cf. Matilal 1986: 295–308). Pleasures and pains are also private to individuals, but, unlike numbers, they reside in the soul of the person and not in the outside world. In saying that numbers have mental acts of counting as their causes, the Nyāya-­Vaiśeṣika give voice to an intuition that there is an element of arbitrariness or a ‘strain of subjectivity’ (Matilal 1987: 19) in the numerical description of the world. Faddegon suggests that the arbitrariness lies in choosing a ‘standard of counting, measuring, weighing’ (1918: 206–7). What is certainly true is that the question ‘How many things are there here?’ has no determinate answer until we specify the sort of thing to be counted. As Frege said (1950, §22), ‘while I am not able, simply be thinking of it differently, to alter the colour or hardness of a thing in the slightest, I am able to think of the Iliad either as one poem, or as 24 books, or as some large number of verses’.Yet the Nyāya-­Vaiśeṣika do not seem to have the sortal dependency of counting in mind, when they say that numbers are produced in objects by mental acts of counting. For Praśastapāda explicitly said that the two things can be of similar or dissimilar kinds. Their point seems rather to be that the mental act is one of collecting together a number of objects, thereby making a particular group salient to the cogniser. From all the objects in the surroundings, the cogniser picks out in his mental act a selected subset. Looking at a garden full of trees, one might think ‘these four

228

JONARDON GANERI

trees are tall’.What makes it true that the subject in such a thought is a group of four trees, when there are many more trees around, is that those objects have been selected mentally by the cogniser. In other words, counting up the objects produces the number because it produces the group to which the number belongs. This in no way detracts from the objectivity of the number of objects being thought about, nor does it make the truth-­value of the thought depend on anything other than the state of those objects. Mental acts of counting and processes of selective attention seem to have a place only in judgements with demonstrative plural subjects. If one asks how many planets are there in the solar system, on the other hand, the number is fixed in advance of any counting up. In many cases, it is more natural to say that the number of things falling under a description is discovered and not created. The Nyāya-­Vaiśeṣika broach this problem with their account in the course of their discussion of indefinite pluralities, to which we shall turn. Before doing so, we may note another argument that Śrīdhara has for the existence of numbers as external entities (1977: 270). It is that if numbers did not exist, then nothing would regulate our cognition of them: if the forms [cognitions] of ‘one’, ‘two’ etc. were not compliant with external objects, they would not appear according to a condition of occurring from time to time. So number too is to be admitted, because there is no difference between the arising of both [cognitions of number and cognitions of other external things].

It is central to the concept of objectivity that external objects are thought of as capable of being perceived and of existing unperceived. A familiar neo-­Kantian thought is that in order to make sense of the idea that one can perceive what can also exist unperceived, one must think of perception as having certain spatio-­temporal ‘enabling conditions’, such that in order to perceive something one must be appropriately located – both spatially and temporally – with respect to it.4 One can then make sense of the fact that a perceivable object is not actually perceived by thinking that the enabling conditions for its perception are not satisfied. Śrīdhara’s argument is similar: if numbers were not objective, then what we perceive would not be accountable to what there is to be perceived in our surroundings. Even though numbers have mental causes in the Nyāya-­Vaiśeṣika theory, the perception of a number is a distinct mental event with external ‘enabling conditions’. 7.  THE EPISTEMOLOGY OF NUMBER

The first four steps in the ‘eight moments’ theory describe the origin of numbers greater than one as properties of a group of objects collected together by a mental act of counting. The second four steps describe the origin of our knowledge of those properties, once they have come into being. According to Vaiśeṣikasūtra 4.1.12–13, numbers are themselves perceptible when they reside in perceptible substances. In seeing a group of objects, we see also their properties, one of which is their number.The claim that we come to know of the number of objects perceptually is a little surprising, for one might have thought that it would be more natural to think of the act of counting itself as our means of coming to know how many objects there are. We have already seen, however, that it is crucial to Vaiśeṣika realism to maintain a sharp distinction between the creative act of counting and the epistemology of

229

OBJECTIVITY AND PROOF

numerical knowledge. Presumably when the collected objects are not perceptible, then the act of counting them is our primary epistemological access to their number. Mathematical knowledge, as distinct from empirical knowledge of the number of things, is considered by western philosophers of mathematics to have a distinctive epistemological status. It is thought to be possible to know a priori, without recourse to any empirical enquiry, that for example 17 × 27 = 459. Interestingly, the Nyāya-­Vaiśeṣika did consider the hypothesis that mathematical truths are known by  a  special means, in the course of their discussion of the source of knowledge called ‘inclusion’ (saṃbhava). Inclusion is said to have been recognised as an independent source of knowledge by the Paurāṇikas. Vātsyāyana defines it as ‘the cognition of the existence of one thing from the cognition of the existence of another thing never absent from the first’ (1985: 574). He cites, as an example, coming to know of an āḍhaka from a droṇa. Āḍhaka and droṇa are both measures of grain, the former being one quarter of the latter. So the point is that one is in a position to know that one has a quantity x of grain if one knows that one has a quantity of grain that includes x. Later writers give an example more relevant to our present purposes: the inference that there are a hundred of a certain thing because there are a thousand. We must take ‘there are a hundred’ to mean ‘there are at least a hundred’ and not ‘there are exactly a hundred’. However, although they mention ‘inclusion,’ the Nyāya authors fail to recognise in it any distinctive epistemological features. They reduce it to the standard pattern of quasī-­inductive inference, in which one infers that something is F because it is G, on the basis of one’s knowledge that F never occurs without G. Knowledge of the invariable concomitance between F and G is, however, taken to be derived from a search for counter-­examples, cases where F is present without G. As long as one can find no such counter-­example, one has a good reason to believe that the properties are concomitant. The inference that there are a hundred because there are a thousand, on the other hand, is based on the inclusion of a hundred in a thousand, something which one can come to know without searching for counter-­examples or performing any empirical enquiry.This example shows how the scientific empiricism of the Nyāya-­Vaiśeṣika authors deterred them, rightly or wrongly, from accepting the existence of a priori mathematical truth. 8.  THE MEANING OF ‘MANY’

An interesting and revealing discussion arose with respect to the meaning of sentences like ‘There are many trees in the forest’ or ‘the army is composed of many soldiers’. In such statements, the adjective ‘many’ has, formally, a role quite like a numerical adjective. Śrīdhara is said indeed to argue that ‘many’ is itself a number, distinct from any number in the cardinal series. As Śaṅkara Miśra (1923: 179) reports him: ‘Since in the case of a forest or an army, there is no definite combinatory cognition [i.e. counting up of the number of trees or men], a mere manyness comes into being, but not the number a hundred or a thousand’ – this is the view of the teacher Śrīdhara.5

If numbers are created by mental acts of counting, then the uncounted army or forest has no definite number of men or trees. The perceiver merely surveys the entirety, and

230

JONARDON GANERI

judges ‘there are many men’ or ‘there are many trees’. Śrīdhara’s view, it seems, is that in such a case an ‘indefinite’ act of counting produces in the objects the ‘indefinite’ number, many. That Śrīdhara is forced into such a move indicates the presence of a deep fault-­line in the Nyāya-­Vaiśeṣika theory. The problem arises whenever one has a clearly delineated but uncounted collection of objects. The collection might, as in these cases, be a perceptible aggregate, or it might be picked out under a description, for instance ‘planets in the solar system’. There seems in such cases to be an inconsistency between the common-­ sense intuition that the collection has some definite if unknown number, and the Nyāya-­ Vaiśeṣika thesis that number is the result of a mental act of counting up the objects in the collection. Udayana ridicules Śrīdhara’s attempt to solve the puzzle by substituting a definite if unknown number with the known but indefinite number ‘many’. He says (1956: 458): Perhaps [as Śrīdhara says,] twoness is produced by a pure combinative cognition, threeness by a [combinative] cognition of a oneness with a twoness, fourness by a [combinative] cognition of a oneness with a threeness, etc. How about the army and the forest? It might be thought that there is here no [combinative] cognition of a oneness with some definite number, and moreover that no definite number such as a thousand comes into being. But still a mere ‘manyness’ comes into being without a specific universal [numerical property] from a combinative cognition, because it can have as its foundation an indefinite [number of] onenesses. This [theory of Śrīdhara] is false. One could not then have a doubt [about the number of things] without there being a specific number of [things qualified by] pothood or [things] delicately distinguished, nor could there be such statements as ‘the large army is larger’ at the time of becoming aware of the addition of another group of men. Śaṅkara Miśra summarises Udayana’s argument: “‘No doubt could arise as to whether [the army] has  a  hundred or  a  thousand parts, nor could one judge ‘the large army is larger’; so this is not right’ – this is the view of the teacher Udayana” (1923: 179). Udayana insists on the validity of common-­sense: a doubt about the number of men in the army can arise only if one thinks of it as having a definite but unknown number. Still more devastating is his objection that if many is a number, we could not compare one group of many with another, or judge one group to be larger than another – for both would have the same number, many. There is, however, a way to make sense of the idea of a definite but unknown number, if we clearly distinguish between the ontogenesis of numbers and their epistemology. In a normal case, the mental act of counting which produces the number will also manifest that number to the cogniser. That is to say, the means by which one comes to know how many objects there are will be precisely that act of counting them which produces the number in them. This need not always be the case. It is conceivable that some definite number can be produced in the objects by an indefinite mental act of counting, an act which will not itself supply the cogniser with any knowledge of the number of things. A definite or precise mental act of counting is both the cause and the ‘manifestor’ of the number. An indefinite or imprecise mental counting is the cause but not the manifestor of a number (Udayana 1956: 459). There is room in the theory for a gap between the creation and the discovery of numbers.

OBJECTIVITY AND PROOF

231

The meaning of the term ‘many’ presents another puzzle. Nyāya-­Vaiśeṣika semantic theory is by and large realist or referential: the meaning of a word is the object for which it stands (see Ganeri 1999: 82–107).The word ‘many’ cannot however stand for the actual number in the army (a thousand, say), for the sentence ‘there are many men in the army’ does not mean that there are  a  thousand men in the army. So ‘many’ must denote, as ­Uddyotakara clearly states, the distinct number manyness (1985: 211): What does the phrase ‘large army’ mean? The phrase ‘large army’ designates those very ­elephants etc. which are the abode of number, when increasing with the addition of new groups of men. So when the elephants etc. are described as ‘large’, the word ‘large’ denotes the number manyness which is the cause of the increase. Śaṅkara Miśra grapples with the same problem (1923: 180):

Is manyness nothing but number [beginning with] three and ending at parārdha,6 or a number different from them? Not the first, since even in the case of an army and a forest, there must be a definite number such as a hundred or a thousand. Nor the second, since a manyness distinguished from three etc. is never observed. So [Udayana says that] manyness is indeed a number such as a hundred, produced by a combinative cognition which does not have as its foundation a definite [number of] units; but [the number] a hundred is not revealed there because its manifestor is absent. We, on the other hand, say that manyness is indeed a number different from but colocated with three etc., and produced by the combinative cognition that produces the threeness etc. It is different, because there is a difference in prior absence. How else could one say ‘There are many, but I do not know if there are precisely a hundred or a thousand’? If asked the question ‘Shall I bring a hundred or a thousand mangoes?’, one replies ‘Bring some large amount – why do you want to know the exact number?’ Three is due to a combinative cognition [of one] with two; four is due to a combinative cognition [of one] with three; and so on higher and higher and higher. When manyness arises without  a  combinative cognition, it does not depend on specific earlier numbers. In the case of the army or the forest, manyness alone comes into being and not any other number. So the above dilemma is one with false horns. Śaṅkara Miśra’s point is that what ‘many’ denotes is neither  a  particular number in the cardinal series, nor the series in totality, but rather a property that co-­exists with the definite number the objects possess. It is, perhaps, the property of being large in number. ‘Many’ is in fact a vague predicate, and it is easy to see how a version of the Sorites paradox can arise. For if there are many men in the army, and one is taken away, there are still many men in the army. Repeating the process, however, one would eventually reach the situation where there are just two men in the army, and two is not many. Uddyotakara does indeed link ‘many’ with processes of increase, but none of the Nyāya-­Vaiśeṣika authors formulate the Sorites paradox in this form. In the discussion of the semantics of ‘many’, the Nyāya-­Vaiśeṣika authors explore the limits of the attempt to treat numerals as predicates, or names of properties. Frege’s remarkable insight (1950) was that numerals are not predicates but quantifier expressions. Statements like ‘there are seven F s’, really belong in the same semantic category as ‘there

232

JONARDON GANERI

are no F s’ and ‘there are some F s’. If ‘/F/’ stands for the cardinality of the extension of ‘F’, then the semantics for such sentences follow a uniform pattern: ‘there are no F s’ is true iff /F/ = 0 ‘there are no F s’ is true iff /F/ > 0 ‘there are at least seven F s’ is true iff /F/ ≥ 7 ‘there are at least seven F s’ is true iff /F/ = 7 When the numerals are recognised to be quantifier expressions, ‘more than’ no longer causes any special difficulty: ‘there are more F s than G s’ is true iff /F/ > /G/ The Nyāya-­Vaiśeṣika authors, like their medieval counterparts in the West, approached semantics with a name-­bearer model of the meaning relation, and do not develop an theory of quantification adequate to the expression of such terms. 9.  THE THEORY OF NUMBER AND INDIAN CONCEPTS OF RATIONALITY AND PROOF

Among the most important features of the Indian theory of number, we may select the following: 1 No distinction is drawn between scientific-­empirical and mathematical truth. Mathematical objects are thought of as properties of the natural world, not as abstract or Platonic entities. Mathematical knowledge (e.g., that 17 × 27 = 459), in so far as it is considered at all, is not taken to be different in kind from empirical knowledge about the number of things of a certain type. 2 Numbers are regarded as created, not discovered. In spite of this, they are thought to be genuinely objective properties of the natural world. The world is not thought of as intrinsically numbered, but numbers are nevertheless the external correlates of thought. The mental act of counting that creates a number is usually but not necessarily our source of knowledge about the number. 3 The theory of number developed by the philosophers is not based on any great awareness of the work of mathematicians, nor is its purpose to study the philosophical foundations of mathematics or scientific practice. 4 Neither Russell’s paradox nor the Sorites paradox is recognised, even though they could easily have been formulated in the terms of the discussion. 5 Numbers are not reduced to sets. Indeed, sets are not introduced into the ontology. The ontology is intensional, in that among its basic constituents are properties and relations rather than sets. We can see, then, that the ‘official’ philosophers in India did not try to analyse the philosophical basis of mathematics, but were more interested in describing the semantics

OBJECTIVITY AND PROOF

233

of contingent numerical statements and the epistemology of empirical numerical knowledge. They tackled some problems in their theory (the ‘number of qualities’ problem, the treatment of ‘many’) but did not uncover other paradoxes and inconsistencies in their conceptual framework. They sought for a definition of objectivity which did not make it rest on mind-­independence, but allowed for the mind to have a causal role in the creation of objective entities. The account of number affords an instructive insight into the mechanisms of theory-­ change and internal criticism in Indian philosophical thinking. We have examined two cases where the theory was modified under pressure from within: the rejection of the assimilation of number to the category of quality by Bhāsarvajña and Raghunātha, and the jossling between Śrīdhara, Udayana and Śaṅkara Miśra over of the correct account of the term ‘many’. What drove the revision in the first case was a perceived explanatory failure in the older theory. The account of numbers as qualities did not seem able to explain their cross-­categorial behaviour, nor to give a satisfactory explanation of the relation between numbers and the world of objects. The old theory failed because it supplied no explanation of a range of observed facts about the actual use of the concept of number, in particular that the concept is applicable to objects in any ontological category. In the case of ‘many’, a failure to explain the observed application of the concept is again what drives the theory-­modification. The facts in need of adequate explanation here are (i) that one can judge that there are many things of a certain type without knowing how many, and (ii) that different ‘manys’ are commensurable and capable of being ordered.The process of rejection and revision in Indian philosophical theory is driven, then, by a perceived mismatch between the older theory of a concept and the actual use of that concept in a range of cases. Such  a  process of theory-­rejection might be juxtaposed with another, according to which one looks for inconsistency or incoherence in the basic concepts and postulates which form the foundation of the theory. When successful, this method typically results in the discovery of paradox and absurdity. It has been a familiar element in philosophical thinking in the west since the Greeks (Zeno, Eubulides), but was not without its promoters in India (Sañjaya, Nāgārjuna, Śrīharṣa). Now both Russell’s paradox and the version of the Sorites paradox involving ‘many,’ point to a difficulty with fundamental presupposition in the Nyāya-­Vaiśeṣika theory, that every intelligible condition has a determinate application (pravṛttīnimitta). While the condition ‘being a substance’ has a determinate application to substances, and ‘being two’ has a determinate application to groups of two, neither the condition ‘being many’ nor the condition ‘being non-­self-­inherent’ have determinate applications – one is vague, the other is self-­contradictory. My present point is that if the Nyāya-­Vaiśeṣika authors did not discover these paradoxes it was because the search for such paradoxes is not what drove their philosophical practice. The underlying idea in their philosophical theorising is that a theory is rationally acceptable if, or to the extent that, it explains actual use – the greater the range of uses explained, the better the theory. Coherence was, of course, a constraint on the adequacy of a theory, but the idea of rational acceptability as foundational coherence was not the active principle behind the rejection or revision of philosophical theory. The Nyāya-­Vaiśeṣika theory of number reveals another important aspect of philosophical theorising in India: the importance of empirical grounding. The general importance

234

JONARDON GANERI

attached to empirical grounding is to be seen in the role of the ‘example’ in the Nyāya informal logic, and in the subordinate status accorded to ‘hypothetical reasoning’ (tarka) (Matilal 1986: 78–80, Mohanty 1992: 117–8). Matilal suggests that a priori reasoning is not admissible because ‘[p]urely a priori certainty is not an acceptable certainty in a properly empirical philosophy. … [W]hen no empirical evidence is forthcoming, the reductio will lose its utility. In this way, I think, the pramāṇa theorists were well equipped to resist the attempts of the sceptic-­dialecticians who generally used the reductios (prasaṅga) to refute any philosophical position’ (ibid.). Purely a priori reasoning is destructive. The pure working of the mind can create what it likes and destroy what it likes without regulation. Regularity, order, structure – these come from the natural world. Constructive reason is reason grounded in empirical evidence. Even the theory of number, apparently an entirely a priori discipline, would be baseless and arbitrary unless empirically grounded. Numbers, therefore, have to be observable features of the natural world, and not abstract Platonic objects, and counting does not create numbers in the abstract, but rather numbered collections of observable objects. What is rejected here is the idea that there could be order among mathematical objects unless those objects were part of the natural world. The ‘official’ philosophers in India were not the only ones to raise and discuss philosophical problems. Matilal (1989) claims, for example, that while the philosophers or pramāṇa theorists offered no account of moral theory, their failure to do so was compensated by the moral theorising of the epic authors, the later narrators who retold the epic stories with new nuances, the writers on dharmaśāstra and their commentators. The same is true in mathematics. The mathematicians themselves did reflect on the nature of their own enquiry into mathematical truth. Bhāskarācārya II states that ‘utpatti (mathematical demonstration) is the means (pramāṇa) of establishing truth in mathematics’ (reported in Srinivas 1992: 45). Bhāskara’s commentator Ganeśa Daivajña defines an utpatti as follows (see Srinivas ibid: 44): Whatever be stated in the vyakta or avyakta branches of mathematics without utpatti will not be rendered free from error. It will not acquire any standing in an assembly of scholarly mathematicians. The utpatti is directly perceivable like a mirror in the hand. It is for this reason, as also for the elevation of the intellect, that I proceed to enunciate utpatti in its entirety.

The purposes of mathematical demonstration are thus: (i) to eliminate uncertainty, (ii) to convince an assembly of experts, (iii) to make the mathematical result transparent, and (iv) to improve the intellect or scholarship of the reader. What is important to note here is how different this description of mathematical demonstration is from the concept of mathematical proof in the west, namely formal deduction in an axiomatic system. According the Indian mathematicians, the goal of a mathematical demonstration is to bring the reader to the point where they really understand how and why theorem or rule is correct. A particularly nice example is Ganeśa’s geometrical demonstration of Pythagoras’ Theorem (a theorem that is already explicitly stated in the Kalpasūtras). It consists in the following diagram, along with an accompanying explanation: The area of the large square is z2, that of the small square is (y–x)2, and the area of each 1  1 2 triangle is xy . So z 2 = ( y − x ) + 4  xy  = x 2 + y 2. After reflecting on the diagram and 2  2 accompanying explanation, it is supposed to become obvious that Pythagoras’ Theorem is true.

235

OBJECTIVITY AND PROOF

Frege begins the Grundlagen with the remark that ‘[i]n arithmetic, if only because many of its methods and concepts originated in India, it has been the tradition to reason less strictly than in geometry, which was in the main developed by the Greeks’ (1950: §1). There is more than a little racial stereotyping and ignorance in this comment. A contrast can indeed be made between the concepts of proof in India and Greece. What we must remember, however, is that the contrast is not between two ways of doing mathematics, but rather between two theories of what ideal proof consists in. The actual practice of mathematicians, whether in the west or in India, is to produce informal proof-­sketches with many ‘obvious’ steps missing. Such real proofs are a far cry from what Frege’s logicism led him to imagine an ideal proof should be. If the Indian theory as to what mathematical demonstration consists in is ‘less strict’ than Frege’s ideal, this does not show that they reasoned less strictly, but rather that they conceived of the aim of rationality differently. Rationality in mathematics has for its goal what one might call ‘real understanding’ of a result, and not simply a proof that the result has to be true. For one can follow the steps of a formal proof, and thereby realise that the result has to be true, without gaining thereby any ‘real’ understanding of what the result shows, how it is or should be applied, or where its significance lies. For that reason, a proper understanding of a result includes, in part, an understanding of how it relates to actual examples. On this point – that proof must and argument must make an essential appeal to paradigms, examples and exemplars – the philosophers and mathematicians in India agree.7

x y

z

Figure 13.1

236

JONARDON GANERI

NOTES 1. I  restrict my attention to the philosophers of the analytical stream of Indian philosophical thought, the Nyāya-­Vaiśeṣika. The main innovations are due to Kaṇāda (c. AD 100), Praśastapāda (c.  530), Bhāsarvajña (c.  950), Śrīdhara (c.  990), Udayana (c.  1000), Śaṇkara Miśra (c. 1430) and Raghunātha (c. 1510). 2. VS 7.2.7 is apparently directed against the Sāṃkhya theory that there is an identity between cause and effect. 7.2.8 is due to the doctrine that, while all numbers greater than one are non-­eternal, the number one is eternal when it resides in eternal substances (the atoms), and only non-­eternal in non-­eternal substances. We can ignore this rather obscure doctrine. See Faddegon (1918: 200–1). 3. The point is made by Udayana (1956: 459). 4. See, in particular, the discussion of objectivity in Evans (1985: 261–2). 5. It is difficult to find the view ascribed here to him. All he seems to say on the matter is this (1977: 270): ‘If it is said that there is no number apart from substances, because a difference is not grasped, then that is not right. For when a close arrangement of trees is seen from a distance, even though oneness etc. is not grasped, one can still grasp their nature. Thus too is explained the difference [of number] from colour etc. Although from a distance the colour is not grasped, there is still an awareness of the substance [= the forest?]’. 6. Praśastapāda states (1977: 270) that, after one, the numerical qualities run from two to a large but finite number parārdha. The number parārdha is mentioned in many texts as the highest decimal place name. The precise value of parārdha varies: in the Taittirīya Saṃhitā (4.40.11.4, 7.2.20.1) and the Maitrāyanī Saṃhitā (2.8.14), it is given as 1012, while the Kāṭhaka Saṃhitā records both 1012 (17.10) and 1013 (39.6). Among the mathematicians, it is always 1017.There are names for higher decimal powers in Buddhist and Jaina texts (Abhidharmakośa 3.93–94 names powers up to 1059). Only the Nyāya-­Vaiśeṣika take parārdha to be the highest number, and not merely the highest named place value.There is no room for the idea of a maximal finite number if one thinks of the number series as generated by recursive application of the successor function, but among the Nyāya-­Vaiśeṣika authors, only Bhāsarvajña attempts so to construct the number series. Within a conception of numbers as qualities of substances, indeed, it seems that there has to be a largest number, if the number of things in the cosmos is finite. 7. For more on the role of examples in philosophical proof procedures and Indian problem-­ solving heuristics, see Ganeri (2001). REFERENCES Bhaduri, S.: 1947, Studies in Nyāya-­Vaiśeṣika Metaphysics, Bhandarkar Oriental Research Institute, Poona. Bhāsarvajña: 1968, Nyāyabhūṣaṇa. Edited by Svami Yogindrananda, Saddarsan Prakasan Pratisthanam, Varanasi. Dvivedi, Sudhakara: 1933, Lives of Hindu Astronomers (Gaṇaka-­tarañgiṇī), Benares. Faddegon, B.: 1918, The Vaiśeṣika System, Described with the Help of the Oldest Texts, Johannes Müller, Amsterdam. Frege, G.: 1950, The Foundations of Arithmetic, translated by J. L. Austin. Blackwell, Oxford. Gadādhara Bhaṭṭācārya: 1927, Śaktivāda, with Kṛṣṇa Bhaṭṭa’s Mañjūṣa, Mādhava Bhaṭṭācārya’s Vivṛtti and Sāhitya Darśanācārya’s Vinodini. Edited by G. D. Sastri. Kashi Sanskrit Series no.57, Benares. Ganeri, J.: 1996, ‘Numbers as Properties of Objects: Frege and the Nyāya’, Studies in Humanities and Social Sciences,Vol. 3: Epistemology, Logic and Ontology after Matilal, Indian Institute of Advanced Studies, Shimla 3.2: 111–121.

OBJECTIVITY AND PROOF

237

Ganeri, J.: 1999, Semantic Powers: Meaning and the Means of Knowing in Classical Indian Philosophy, ­Clarendon Press, Oxford. Ganeri, J.: 2001, Philosophy in Classical India, Routledge, London. Gautama: 1985, Nyāyasūtra, with Vātsyāyana’s Bhāsya, Uddyotakara’s Vārttika, Vācaspati Miśra’s ­Tātparyaṭīkā and Viśvanātha’s Vṛtti. Edited by T. Nyāyatarkatīrtha and A.Tarkatīrtha. Calcutta: ­Calcutta Sanskrit Series nos. 18–19. Second Edition, Munshriram Manoharlal, Delhi. Guha, D. C.: 1979, Navya-­nyāya System of Logic, Second Edn, Motilal Banarsidass, Delhi. Herzberger, H. and R. Herzberger, 1981, ‘Bhartṛhari’s Paradox’, Journal of Indian Philosophy 9, 3–32. Ingalls, D. H. H.: 1951, Materials for the Study of Navya-­Nyāya Logic, Harvard University Press, ­Cambridge, MA. Kaṇāda: 1961, Vaiśeṣikasūtra of Kaṇāda, with the Commentary of Candrānanda, critically edited by Muni Sri Jambuvijaya, Oriental Institute, Gaekwad’s Oriental Series 136, Baroda. Kaṇāda: 1911, The Vaiśeṣika Sūtras of Kaṇāda, with the Commentary of Śaṅkara Miśra, translated by Nandalal Sinha, The Panini Office, Bhuvaneswari Asrama, Allahabad, pp. 233–244. Matilal, B. K.: 1975, ‘On the Navya-­Nyāya Logic of Property and Location’, in Proceedings of the 1975 International Symposium of Multiple-­valued Logic, Indiana University, Bloomington, pp. 450–461. Matilal, B. K.: 1985, ‘On the Theory of Number and Paryāpti in Navya-­Nyāya’, Journal of the Asiatic Society of Bengal 27, 13–21. Matilal, B. K.: 1986, Perception: An Essay on Classical Indian Theories of Knowledge, Clarendon Press, Oxford. Matilal, B. K.: 1989, ‘Moral Dilemmas: Insights from Indian Epics’, in B. K. Matilal (ed.), Moral Dilemas in the Mahābhārata, Indian Institute of Advanced Study, Shimla, pp. 1–19. Mohanty, J. N.: 1992, Reason and Tradition in Indian Thought. An Essay on the Nature of Indian Philosophical Thinking, Clarendon Press, Oxford. Perrett, R.: 1985, ‘A Note on the Navya-­Nyāya Account of Number’, Journal of Indian Philosophy 13, 227–234. Potter, Karl: 1977, The Tradition of Nyāya-­Vaiśeṣika up to Gaṅgeśa, Encyclopedia of Indian Philosophies, Vol. II, Motilal Banarsidass, Delhi. Potter, Karl and Sibajiban Bhattacharyya: 1993, Indian Philosophical Analysis: Nyāya-­Vaiśeṣika from Gaṅgeśa to Raghunātha Śiromaṇi, Encyclopedia of Indian Philosophies,Vol.VI, Motilal Banarsidass, Delhi. Raghunātha: 1957, The Padārthatattvanirūpaṇam of Raghunātha Śiromaṇi, edited and translated by Karl H. Potter, Harvard University Press, Harvard Yenching Institute Studies, Vol.  17, Cambridge, MA, pp. 86–87. Raghunātha: Avacchedakatvanirukti with Jagadīśa’s Jāgadīśī, Dharmananda Mahabhaga (ed.), Kashi Sanskrit Series 203,Varanasi. Raja, K. K.: 1963, Indian Theories of Meaning, The Adyar Library, Madras. Ramaiah, C.: 1976–7, ‘Nyāya-­Vaiśeṣika Theory of Numbers (Sānkhyā)’, Indian Philosophical Quarterly 4, 129–134. Sainsbury, M.: 1979, Russell, Routledge & Kegan Paul, London. Sainsbury, M.: 1991, Logical Forms, Blackwell, Oxford. Sainsbury, M.: 1995, Paradoxes, Second Edition, Cambridge University Press, Cambridge. Śaṇkara Miśra: 1923, Vaiśeṣikadarśana, with Praśastapāda’s Padārthadarmasaṃ graha and Śaṅkara Miśra’s Upaskāra, Dundhiraja Sastri (ed.), Chaukhamba Sanskrit Series Office, Kashi Sanskrit Series 3, ­Varanasi, pp. 173–180. Sen, S. N.: 1995, ‘History of Science in Relation to Philosophy and Culture in Indian Civilization’, in D. P. Chattopadhyaya and R. Kumar (eds), Science, Philosophy and Culture in Historical Perspective, PHICPC Monograph 1, Munshiram Manoharlal, Delhi. Shaw, J. L.: 1982, ‘Number: From the Nyāya to Frege-­Russell’, Studia Logica 41, 283–291. Shukla, K. S.: 1959, The Pāṭigaṇita of Śrīdharācārya with an Ancient Sanskrit Commentary, Kripa Shankar Shukia (ed.), Lucknow University Department of Mathematics and Astronomy, Lucknow.

238

JONARDON GANERI

Śrīdhara: 1977, Praśastapādabhāṣya with the Commentary Nyāyakandalī by Śrīdhara Bhaṭṭa, edited with

Hindi translation, by Durgadhara Jha, Sampurnanand Sanskrit Vishvavidyalaya, Ganganatha Jha ­Granthamala,Varanasi,Vol. 1, pp. 267–214. Śrīdhara: 1907–8, Padārthadharmasaṃ graha with Śrīdhara’s Nvāyakandalī, translated by Ganganatha Jha, The Pandit, n.s.Vols. 29–30, pp. 241–284. Srinivas, M. D.: 1992, ‘The Methodology of Indian Mathematics and its Contemporary Relevance’, in G. Kuppuram and K. Kumudamani (eds), History of Science and Technology in India,Vol. II, pp. 29–86. Udayana: 1956, Kiraṇāvalī, with Vardhamāna’s Prakāśa and Vādīndra’s Rasasāra, Narendra Chandra ­Vedantatirtha (ed.), The Asiatic Society, Bibliotheca Indica no. 200, fasc. 4, Calcutta, pp. 441–462.

P A R T

VI EPISTEMOLOGY

CHAPTER

14

A realist view of perception Bimal Krishna Matilal

Source: P. K. Sen and R. R. Verma (eds), The Philosophy of P. F. Strawson (New Delhi: ­Indian Council of Philosophical Research, 1995), pp. 305–326. It is impossible, says Richard Rorty, to step outside our skins—the traditions, linguistics and other, within which we do our thinking and self-criticism—and compare ourselves with something absolute.1Apparently a realist tries to do just that—step outside his skin and take what Putnam has called the externalist perspective,2 a God’s-Eye point of view. This is an old legacy which what we call ‘science’ has inherited in order to play the role of the divine observer. The realist believes that the world consists of some mind-­independent objects, even discourse-in-dependent objects. He believes himself also to be part of that world. But here complexity already set in. He views his perception as an awareness-­episode due to ‘interaction’ between his senses and objects or facts ‘out there’ (in the sense of being in space ‘outside’ or in the sense of being whatever we understand by ‘mind-independent’). He also believes that whatever he is aware of, objects or facts, is external to the awareness itself. Such a view of perception was explicitly and somewhat rigorously formulated in classical India, in Nyāyaśūtra 1.1.4, some time between 100 bc and ad 100.3 The aphorism in question adds three further qualifications: (a) the awareness must be non-verbal, though it may be verbalizable; (b) it must be non-deviating, that is, not falsifiable; and (c) it must be non-doubting, that is, not vacillating between alternatives. All these qualifications were important in the sense that they contained the ingredients for building up a strong realistic theory of perception, and thereby a sort of metaphysical realism. The problems that the tradition of Nyāya faced over the ages may simply be of historical interest today. Metaphysical realism seems to have fallen into disgrace among modern philosophers. Of course we can mention a few notable philosophers today who have ­resisted the temptation of a tactical retreat from realism into old-style relativism or 241

242

BIMAL KRISHNA MATILAL

p­ ragmatism. However, a sort of pragmatism, which is sometimes distinguished from old idealism or phenomenalism as well as from relativism, has leaped into popularity today, for it brings the ‘liberating’ message of insouciance and freedom from worries about ‘unsolvable’ ­philosophical problems. Hard-headed realism has often been discredited but it is hard to subdue it. It has been argued that metaphysical realism’s assumption of a correspondence theory of truth and a God’s-Eye, or a No-Eye, view of truth, is incoherent in spite of its powerful predominance for two thousand years. Michael Dummett makes the ­acceptance of bivalence as the all-important criterion for distinguishing realism (or ­‘reductionism’ in his terminology) from non-realism or anti-realism.4 A reductionist– realist may subscribe to the correspondence theory of truth, but what is important here is acceptance of the ‘bivalence’ strategy concerning statements within a given type of subject matter (these may include statements about the past, modal statements and moral statements, as well as mathematical statements). The realist accepts the principle of bivalence for such statements, and thereby assumes their truth-conditions to obtain independently of our knowledge or capacity for knowing that they do. He (the realist) endorses ‘the belief that for any statement there must be something in virtue of which either it or its negation is true’: it is only on the basis of this belief that we can justify the idea that truth and falsity play an essential role in the notion of the meaning of a statement, that the general form of an explanation of meaning is the statement of the truth-conditions.5 The anti-realist, on the other hand, determines the meaning of such statements in terms of the conditions which we recognize as establishing their truth and falsity. Although metaphysical realism has strong adherents even today, in view of the above, strong faith in its viability has been apparently undermined not only by neo-pragmatists like Rorty but also by such non-pragmatists as Goodman, Putnam and Dummett. The question is seriously asked whether realism is an old ‘historical relic’ or a straightforward belief-system of the common man which philosophers over two millennia have mistakenly tried to defend. The history of philosophy, however, is different from the history of, say, chemistry, for here the notion of a ‘historical relic’ is not easily available. Old problems raise their heads again, for problems are less seldom solved than resolved or rather shelved for the time being when the proponents get tried or bored. As Thomas Nagel has said, ‘It may be that some philosophical problems have no solutions. I suspect this is true of the deepest and oldest of them. They show us the limits of our understanding. … Unsolvable problems are not for that reason unreal.’6 The insouciance of some modern philosophers seems to emulate the spirit of the old sceptics. The modern pragmatist knowingly or unknowingly sympathizes with scepticism. He is a sceptic without his epoché and he obtains his ataraxia by getting rid of all worries. He is not blind but he avoids the paradoxicality of the sceptical position by closing his eyes when necessary. Transparency or (metaphysical) realism, it is perhaps rightly urged, demands some sort of what has been dubbed a correspondence theory, and/or reference theory underlining the ‘tie’ between words or thought-signs and the world, i.e. external things or sets of things. To be sure, some sort of correspondence or ‘faithfulness’ of the awareness with the world outside or the word–world marital tie should be the bedrock for distinguishing perception from misperception or ‘factual’ discourse from fictional discourse. Almost all the different versions of the correspondence theory have been found fault with. But I am not sure whether all of them have been totally refuted or proven to be

A REALIST VIEW OF PERCEPTION

243

beyond any reformulation.The Indian Nyāya school defended a formulation which I shall call the theory of ‘non-promiscuity’ (or faithfulness between the world and its awareness that is generated in us). Non-promiscuity is not correspondence—in the strict sense of the term. Correspondence needs two terms—one that corresponds and the other to which it corresponds. And in this respect it is of a piece with such relations as comparison, similarity, even contradiction or opposition. Non-promiscuity, on the other hand, is a natural or general expectation in a marital tie, or it may even be likened to the quality of transparency of a transparent medium, if you look at it from another point of view. A retreat from realism can take various forms, some of which are articulated as relativism amounting to incommensurabilism, guarded scepticism, a sort of pragmatism or solidarity within a cultural milieu (Rorty), context-dependence or even the internalist perspective (Putnam), although I am not sure whether the last named is really a significant retreat. Nor is it absolutely clear whether the anti-realist’s rejection of bivalence only (and not tertium non datur) would not be compatible even with the realist’s framework. Nāgārjuna (in India), for example, was explicitly prepared to give up tertium non datur and hence his retreat was a significant retreat from realism. It is difficult to have a common denominator for all these non-realist positions. We may dub them only negatively as ‘non-realism’. We shall try to present the view of Nyāya realism. In this context we need not sharply distinguish between epistemology and semantics, but follow the Russellian line of ­epistemologizing semantics. Besides, in our modern readings of ancient philosophers we do often attribute to them theories and views which are not expressed or formulated by them in exactly the same terminology as we prefer to use today. While epistemological issues such as that of sensation mediating between perception and the external world are well-entrenched in the ancient tradition, the semantic problem of ‘reference’ is of a comparatively recent origin. Nevertheless, the problem of the relation between thought-signs and what they stand for is quite old, which provides the rationale for the discussion of classical views in modern terms.This will be done here with regard to the classical Indian philosophers: The crucial qualifier in the Nyāyaśūtra characterization is ‘non-deviating’ or ‘non-promiscuous’, which equals faithfulness. The word for ‘deviating’ is also the word for promiscuity in Sanskrit, and hence ­negatively, the perceptual awareness-episode must be non-promiscuous, i.e. not unfaithful to the world out there. There need not be a ‘tie’, positively speaking, between the awareness and the relevant bits of the world, but there should be lack of discord between them. Discord marks off misperceptions from perceptions. One way of further understanding the notion of deviation or discord is to see the following. Each perceptual event selects some feature or features of the percept. It is conceded that such features may or may not be located therein or may not be included in the powers of the percept. A misperception misascribes, hence its verbal report is a false statement. The above is the received doctrine of the alleged explanation of a misperception. But a Nyāya realist may put it in a slightly different way. It is, of course, true that each perceptual event selects some feature or features of the percept. However, it need not be conceded that such features, or at least some of such features may not be located in the percept itself. Before the event of a misperception, something else must happen, as a realist should insist. Our mind misconnects some of these features in a way not given in the relevant percept.

244

BIMAL KRISHNA MATILAL

But the realist must also insist that this does not always happen, not before the veridical perceptual event. Such an event is an automatic happening without any further intervention. Lack of this misconnection means that we have a proper perception. The intriguing element in this explanation is the notion of the deviant feature or the accidental misconnection: He is the intruder in the proposed marriage between the awareness and the world, he is responsible for the resulting promiscuity. Wherefrom does this deviant feature come? We shall talk about it presently. But we may note that, according to this view, this deviant feature is not a mental image or a superimposition of a mental representation which lacks ‘hooking up’ with the world. (A pertinent question is: Wherefrom does a mental image come if not from the world itself?) It enters into the awareness-episode and thereby into its verbal report, from the world itself. (This is, at least, the argument of the Nyāya realists.) It is just another feature present elsewhere or at another time, much as another real person, the intruder in a marriage. The qualifier ‘non-doubting’ says that our perception or our judgement of perception should not vacillate between alternatives, i.e. given mutually exclusive alternative features, A and B, it must rest content with just one of them. This presupposes that confrontation between the world and the senses (physics and physiology) may generate a vacillating awareness, or uncertainty, the verbal report of which would be ‘x may be A or it may be B’, when the evidence is not adequate, i.e. there is only a common evidence shared by both A and B. This also underlines the fact that perception on this view would be an inference from evidence—a point which the Buddhists were never slow to point out. This, as well as the other qualifier ‘non-verbal’, led to a vortex of controversy, eventually suggesting a radical distinction between conception-free or word-free (nirvikalpa) perception and conception-loaded, word-loaded (savikalpa) perception. In underlining the ­distinction in terms of such expressions as ‘word-free’ and ‘word-loaded’, the classical ­Indian philosophers mix epistemology with semantics, as I have already noted. But perhaps, this is not a serious drawback. For several centuries this issue became the most important chapter in the history of Indian philosophy of perception and marked the great divide between ­realism and idealism. We may also note that while the adjective ‘non-­ promiscuous’ ensured physical soundness of perception, in the sense that it would not be destroyed or falsified by ‘the way the world is’, the adjective ‘non-doubting’ ensured its psychological soundness so that it may be decisive. Absolute certainty is not what is necessary, but it must be uninfected with doubt. Here again, the realist may add another point. Not all veridical perceptions occur after a doubting state of mind, as has been prescribed here. Only in some cases a doubting state may intervene. If a correct perception arises afterwards, on the basis of correct evidence, then the Nyāya realist must be drawing a very thin line of distinction between such ­perception and an inference. It may finally boil down to a terminological dispute, but I am not going to enter into it here. I shall now comment upon the qualification ‘non-verbal’. Its early meaning was ­obscure. But with the rise of a view championed by Bhartrhari in fifth century ad, a view that asserted that all our awareness of the world is necessarily interpenetrated with the ­operation of words, or concepts, in fact, that it is a linguistic affair, the claim was made by the realists that wordless, concept-free perception is possible. This came very close to an early admission of the epistemological point about the distinction between receptivity

A REALIST VIEW OF PERCEPTION

245

and spontaneity, between passive and active parts in us, the former receiving the ‘world as such’ which the latter ‘interprets’ using concepts or words. The point (sometimes ascribed to Kant) was however differently taken by the Buddhists than by the Naiyāyikas, and neither of them was Kantian. Diṅnāga (the leader of the Buddhist pramāna school), for example, regarded the word-free, conception-free sensory awareness as perception proper, for here the ‘world’ is supposed to shine in its pristine purity, without distortion or manipulation (cf. kalpanā), and as far as Diṅnāga was concerned, interpretation with concepts or words is necessarily discourse-dependent and hence perhaps a distortion, though it is a convenient way for it is instrumental for our getting along with this world. It is not clear whether this attitude originated in the fear that discursive thought, or free flow of the mind’s activity, if uncontrolled, will play us false and make us lose touch with the real, the ultimately real—a fear which Heidegger, it is said, has claimed to be typical of the ­Western philosophical tradition. (How can one claim something to be typical of the Western ­tradition without knowing much of the non-Western tradition?—a point to which my attention was drawn by my friend, Ninian Smart.) It should be noted here—a linguistic point but significant enough—that the word kalpanā or vikalpa in Sanskrit is also the word for poetic and playful fancy, the arbitrary use of language, or as the Yogasūtra defined it much earlier in the tradition, it is what is ‘word-inspired’ as well as ‘lacking in reality’. Kumārila asserted: words provide our access to fictional and utterly non-existent objects. Bhartrhari’s point was that our verbal report or language ‘carves up the world for us’, our perceived world is also a manipulated/interpreted world, and therefore the word– world tie is one of identity, or rather, the word ‘creates’ its own object (carves it out of the world) to refer to it. We do not ‘deviate’ but we ‘create’. This was an attractive idea to Diṅnāga, the Buddhist. It was acceptable to him, for then one could easily devalue the assumed reality of the interpreted world captured in our normal, veridical perceptual ­judgements. For the Buddhist goal of diverting ourselves from desire-laden activity to quietude, it was all the more welcome. The Naiyāyikas, however, criticized this theory which, they rightly feared, would lead to idealism or a Lockean kind of dilemma in representationalism, and the S­ autrāntika ­apparently came very close to such a quandary.The problem before them was threefold, the same as seem to confront even the modern realist. First, we must have an ­acceptable ­theory of perceptual illusion, a theory that does not countenance mental images, ­representations, sense-data, etc., but like a Dickensian Scrooge (in A Christmas Carol), would be able to say to the hallucinated ghost: ‘I do not believe in my senses because a slight disorder of the stomach makes them cheat, you may be an undigested bit of beef, a blot of mustard, a crumb of cheese, a fragment of underdone potato. There is more of gravy than of grave about you, whatever you are?’ Second, we need a version of ‘correspondence’ or the word–world tie that would be intelligible and defensible. Third, while conceding the constructed nature of the structure of reality captured or revealed in our veridical perceptual judgements, one should try to undervalue, if not reject altogether, the privileged position enjoyed by structureless, construction-free perception, and along with it, one may try to avoid conceptual relativism, and deny the possibility of alternative interpretative frameworks mutually incompatible and totally unintertranslatable. For if we concede that our concepts or what is within us provides the structure for the neutral, structureless, material

246

BIMAL KRISHNA MATILAL

world, then we may side with the anti-realist. However, one can save realism by not only underplaying or not emphasizing the passivity of the mind receiving the structureless world, but also by recognizing at the same time the mind’s active role in construction. If the mind’s active role is externally controlled under all normal circumstances, then most beliefs thus acquired would be true beliefs, and a form of realism can be vindicated. This external control, if admitted, would also presuppose, however, that there are at least some mind-independent, non-discourse-dominated, non-desire-laden universals, i.e. objective universals of natural kinds. The later Naiyāyikas have tried to tackle all the three issues with varying degrees of success. I shall comment primarily on the first issue, and to some extent on the other two, for they are interconnected. Connecting epistemology with semantics we will have occasion to refer to the standard Russellian solution of the problem about universals as well as fictional discourse with his theory of description and ‘knowledge by acquaintance’.These problems have a distinct, though ‘distant’, Nyāya resonance, although in details Russell and Nyāya differed a lot, while both trying to save realism. A theory of error is also a theory of reality, for they are two7 sides of the same coin. This view gradually became the conviction of most Indian philosophers from ad 800 onwards and hence a proliferation of different theories of error—i.e. about the status of the object appearing in perceptual illusion—followed, which underlined both realism and idealism in their various ramifications and combinations. During the time of Vācaspati (ninth to tenth centuries ad) there were at least five such theories prevalent (five hundred years later, about fourteen such views were recorded). Of these five, two were ascribed to the Buddhists, one to the Vedāntists and two to the realists, Naiyāyikas and Prābhākaras. The Buddhist view alternates between regarding ‘objects’ figuring in perceptual error as ­internal to and/or created by the awareness itself and regarding them as unreal or non-­existent entities for which there is no objective support or foundation in the external world. Both, in Kumārila’s description, are nirālambana-vāda, the view that ­regards all awareness as foundationless, i.e. lacking objective support in the outer world. The ­Buddhists may concede, as the Sautrāntikas did, that there is somehow a mind-­ independent world, for that is virtually presupposed in most of our rational behaviour,8 and in this our experience may be somehow grounded. But we can form no absolute conception of this world for any attempt to characterize it or talk about it would lead to nonsense.Talk of the ‘correspondence’ between awareness and reality would be impossible on this view, for we have no independent access to reality to check such correspondence, and hence non-realism (which the Buddhists preferred) wins the day. The third, Vedāntic, view is somewhat ‘pragmatic’ in the modern sense. It regards the object in perceptual illusion as neither real nor unreal, as neither part of the outer world nor an empty, airy nothing (for otherwise how can they be experienced?), but as having an indescribable, and inexplicable status. It is much like the object in Brentano’s doctrine of psychological verbs, ‘intentionally inexistent’. On the whole, however, this does not save realism; it only develops a tolerant attitude towards metaphysical realism to which I now wish to turn. (A realistic reading of Vedānta is, to be sure, possible, but that is not my concern here.) One point of some significance, however, can be made here.This view also accepts the absolute duality of the perceived world (or the interpreted world, according to Bhartrhari)

A REALIST VIEW OF PERCEPTION

247

and the world itself, i.e. the world that we perceive and the world that confronts us. The Vedāntin, it seems to me, wishes to point out that we generally use a set of terms to describe the bits and pieces of the world itself and then try uncritically to describe, with the same set of terms or sentences, the bits and pieces of the perceived world. Hence, the Vedāntin seems to argue, it is better to prefix such sets of terms or sentences of ordinary language or discourse to warn us that an illicit transition has been made in our linguistic act. The Nyāya realists would counter this argument by saying that the alleged duality is based upon an illicit generalization. There is practically no question of duality (leading to possible distortion) as long as the natural transparency of the media (perception or language) remains intact. If this quality is destroyed by a defect, it would be illegitimate to accuse all the individual members of the media. It is like accusing every spouse, even if only some of them are proven, as unfortunately some of them often are, guilty of promiscuity. Both the Prābhākaras and the Naiyāyikas were metaphysical realists. Both assumed that the world out there consists of some fixed totality of mind-independent objects and that the structure of those objects can be revealed or grasped in our veridical perceptual judgement. Such perceptions, in other words, can say ‘the way the world is’, and misperceptions are regarded or explained either as ‘confusions’ of one thing for another or ‘misplacement’ of something real in another (something real). The Prābhākaras opt for the ‘confusion’ or ‘fusion’ theory while the Naiyāyikas prefer the ‘misplacement’ theory.9 The Prābhākaras believed that, practically speaking, there cannot be any illusion, only mixing or confusion of one mode of awareness (perceptual) with another (memory). That resists giving any separate status to representation.There is direct grasp of objects, there is perfect transparency. The world is a transparent text. Of course, there are institutional or cultural constructs which inform our perceptions. But there is, it is claimed, an infrastructure common to all rational humans which directly figures in our perceptions. If the verbal report of a socalled perceptual illusion is ‘that is a snake’, then both the terms ‘that’ and ‘snake’ have ties with the world we know best: ‘that’ refers to the object in front, the piece of rope, and ‘snake’ to a member of the snake-community inhabiting this world.The so-called illusory awareness does not create or grasp any ‘form’, image or ‘appearance’ of the snake (not even the sense-datum of the snake-appearance). It directly refers to (using ‘refer’ in the old, less entrenched sense) the world. The verbal report of the so-called illusory awareness is a blending of two different modes, a sensory perception and a m ­ emory-event, although their distinction remains unrevealed in verbalizations. There may be this distinction here between the cognizing faculty and the verbalizing faculty; the later may miss something which the former does not, hence the latter confuses or gives mixed signals which the former does not.The mind passively receives both the visual object and the revived memory, but does not receive enough to see that they are in fact two, not one. That is, their distinctness or lack of connection remains unrecognized. Or better still, we may say that the mind probably receives from them as they are, i.e. that they are two, not one, but the faculty of generating the verbal report misses this lack of connection and hence there is some fusion or confusion of the two as one. Notice that here we do not talk at all about sense-data, or sensory images, or mental representations, and thereby a lot of tricky questions regarding the ontological status of sense-data, etc., which have for long plagued the theories of perception, are avoided.

248

BIMAL KRISHNA MATILAL

Gaṅgeśa (ad 1300, in his Tattvacintāmaṇi)10 has quoted a verse which summarily notes five main reasons for justifying the ‘confusion’ or ‘no-illusion’ theory of the ­Prābhākaras. The rather sweeping Prābhākara thesis is that all awareness without exception is ­object-corresponding (yathārtha), and that hence no explanation is needed for the so-called illusory appearance of objects, like the snake-image, etc., an explanation being needed only for the use of the term ‘illusion’ for a certain class of awareness. (At  the ­semantic level, an explanation is needed for the application of the term ‘fictional’ to ­fictional objects such as Sherlock Holmes.) The five reasons stated in support of this thesis are as follows: 1 Sākarā-pātāt: If an awareness (perceptual) can arise about such an object as ‘this snake’ when there is no snake present, then we have to admit the ontology of mental representations or mental images (ākāra—cf. those who admit ākāra (= form) would face similar problems about sense-data as faced in modern Anglo–­American philosophy). The Prābhākaras do not admit the idea of ‘mental representation’. Hence they argue that there cannot be any awareness that does not correspond or refer to any object. An awareness, on this view, refers to some object or other, an object belonging to the world itself much in the same way as a proper name (non-connotatively) refers to an individual. I have made a tongue-in-the-cheek use of the verb ‘refer’ in this context. If there is scepticism about the use of this rather well-entrenched term, we have to settle for the more metaphorical ‘grasp’ or ‘reveal’. The idea is that there is direct grasp of the object by the awareness (­naive realism), nothing ­intervenes. Since there cannot be any false awareness on this view, only ‘fusion’ of the two modes reflected confusedly in the verbal report, there cannot be any talk about the ‘content’ being divorced from the object. It seems that something happens between cognition and its verbal report (we should note that the verbal report may be ­inarticulate or indistinct or simply a concept-formation, as Bhartrhari reminded us). The percept and the memory-object are fused together in such verbalization. 2 Asato na bhānāt.The Prābhākaras do not accept that an awareness ‘refers’ to or grasps a non-existent (unreal) object. Revived memory remembers the past snake and perception sights the rope; both being real, we cannot talk about any unreal objects. For if we do, we immediately subvert the claims of realism. If illusory objects are ‘perceived’ once, this could happen always. There would thus be no convincing argument against ‘illusionism’ which maintains that the whole world is an illusion (Vasubandhu) or an empty dream, or that we are simply brains-in-a-vat, or that all our experiences are being manipulated by Descartes’ evil genius. 3 Saṃvid-virodhāt. A false awareness is usually explained as an episode which grasps an object (a snake) that is not present in the given spatio-temporal situation in addition to its grasping the object (the rope) that is present in the given situation. It is rather counter-intuitive (saṃvid-virodha) to maintain that when we are aware of A, a piece of rope in front, there is a contrary object, B, a snake, which is also captured in the same awareness. What captures B is a different awareness (remembering), and it is also object-corresponding in its own particular way, for we can remember that there were pieces of silver previously experienced.

A REALIST VIEW OF PERCEPTION

249

4 Hetvabhāvāt. To admit that the senses can be wrong and/or that the mind wrongly interprets the data received is to say that they occasionally ‘misbehave’. There is no causal explanation for this misbehaviour of the senses or of the mind. It is the nature of the sensory faculty or of the mental faculty that they receive/grasp only what is presented by the world.Why should they act in a way that is contrary to their nature and grasp distant and past objects? It is impossible for something to change its own nature. Therefore, it does not stand to reason to suppose that our faculties should be supposed to have changed their own nature and inform us wrongly. The Prābhākaras believe that the defect (doṣa) does not affect the causal structure of the cognition but that of verbalization. 5 Dhiyām anasvasabhayāt. The real possibility of error or illusory perception must be denied. Our conscious activities are caused by our beliefs, awareness and knowledge. But such activities prompted by knowledge or awareness would be impossible. For if an awareness is not object-corresponding, but promiscuous or guilty of misdemeanour in the way illusions are supposed to be, what trust would we have in any awareness to prompt us to act in any way? All our actions would be infected with doubt and uncertainty. To wit: Since our actions are caused by our beliefs, all our beliefs prompting actions must be trusted to be true, i.e. object-corresponding. Our failures in action show not that the beliefs or the awarenesses were wrong or not object-corresponding, but that more knowledge (true-belief) was needed for our success. As I rush to pick up the piece of silver but fall, I realize what I did not know before, viz. the distinctness of the two cases of awareness or belief. I had an awareness in the past of the piece of shell as well as a revived memory (recurrence of an episodic belief) about a piece of silver. Our action fails because there was a gap in our knowledge, not because we had no knowledge. Not illusion but ignorance (lack of knowledge of the distinctness) causes this failure. We had a partial knowledge and hence we acted and failed; we did not know enough—such is the Prābhākara position. (Notice that it is much stronger than Davidson’s point, according to which most of our beliefs must be held true, to use Davidson’s own terminology.) This also explains the usage: we apply the term ‘illusion’ when activity fails for no fault in our effort but due to ignorance of the kind already explained. The Naiyāyikas accept all these points (and reject the ‘mental representation’ theory as well as ‘universal illusionism’ on similar grounds), but they admit, unlike the Prābhākaras, that perceptual illusions are facts, not just confusions and play of ignorance, for otherwise it would be highly counter-intuitive to call perceptual illusions nothing but mixing or co-occurrence of two modes of awareness. Instead of holding that there cannot be any false awareness or illusion, the Naiyāyikas agree with commonsense in holding that perceptual illusions are facts, not just confusion of verbal reports. A perceptual illusion is a unitary awareness-event in which we grasp the object in front, the piece of shell, as the location (dharmin) or the subject, and ascribe to it a feature that is absent there, but presented to the mind by memory (receiving it from past experience).The property of being a piece of silver is part of the furniture of this world just as much as a piece of shell is. No mental image is necessary for such presentation, for mind is not ‘a wax tablet’ here. Mind is like an instrument, by which we capture an object or a fact. Objects do not enter into

250

BIMAL KRISHNA MATILAL

our mind. However, past experience, a cognitive event, disappears, leaving behind another dispositional property in the person—this is called saṃskāra on this view.This dispositional property is retained by the person and it produces another cognitive event, this time it is called remembering what was grasped before or what happened before, provided adequate stimuli are present. This is a good place to note briefly how both the realists, the Naiyāyikas and the Prābhākaras would react against a theory that is current and sometimes dominant in the Western world, the theory called causal realism. Very roughly this is the view that we directly perceive only sensations and can only infer material objects through some sort of a ‘dubiously explained’ causal relation. Even in its sophisticated formulation, in terms of ‘sense-data’ reductionism (even in the form: no physical object statements are true without some sense-data statement being true), the unsatisfactory nature of this position is not completely removed. One version of causal realism is rather widely accepted. This I ascribe to P.F. Strawson. It says: Since X’s perceiving M involves X’s sense-impressions of the appropriate kind, the fact that it sensibly seems to X just as if X perceives M is causally dependent upon some state of affairs involving M. The Nyāya critique would be that the connection between X’s perceiving M, and X’s having a sense-impression of the appropriate kind is only a contingent one, whereas the causal theory wrongly assumes it to be a necessary one.To put the matter in another way, the causal theorist may explain the connection to be causal in the sense that ‘X sees M’ is true in virtue of X’s seeing some visual datum N. Nyāya, however, contends that X may see M simply in virtue of the presence of N, where N is nothing but part of the physical reality, but it is not necessary to assume that seeing of M is always preceded (hence causally conditioned) by a sense-impression of N. In our most adult perception we see M directly. A sense-impression of N, if it is to be a cognitive fact as distinct from some pure physiological reaction, may occur and may very well precede our perception of M in some (initial) cases. But it does not happen always. In other words, ‘X perceives M’ may be true without ‘X has a sense-impression of N’ being true in many cases.This is how I understand Gaṇgeśa’s point about ‘construction-free’ perception being possible at ādya-pratyaksa, child’s perceptions, etc. (or seeing something after deep sleep, fainting spells, etc.). Pure sensory perception without any intrusion of concept (reception without interpretation) would, on this view, be impossible except, perhaps, the initial acquaintance of the neophyte with this world. Hence it accepts the supposedly Kantian point that each effable perception will have at least a minimal injection of concepts.There may be distinction between perception of things unqualified (e.g., perception of objective universals, such as cowhood, which is regarded on this view as a logical prerequisite for the perception of something as a cow) and perception of facts or propositional perception. At the semantic level, this implies that a word, say ‘water’, may be said to refer to the actual stuff water by virtue of its universal essence, waterhood. This means that when ‘water’ is uttered, the hearer understands the stuff qualified by waterhood, but the qualifying universal ‘waterhood’ must be grasped by the observer as unqualified! It must be directly ‘captured’. In the Nyāya view, if the verbal report of a perceptual cognition is ‘water’ or ‘it is water’, then, the cognition captures the stuff water under the mode of waterhood, but waterhood itself is captured directly. The verbal report usually bears out what is directly captured in the awareness. If, however, waterhood is captured under another mode of presentation, the

A REALIST VIEW OF PERCEPTION

251

verbal report would mention ‘waterhood’, as, for example, ‘waterhood is a universal’. This was believed to be enough to save the kind of realism they tried to defend. The word–world link, in this view, is not exactly a ‘physicalistic’ or a metaphysical relation of reference (that was the view of Bhartṛhari, which was refuted here), but a very well-entrenched (cf. samayika in the Vaiśeṣikasūtra), determinate unique relation (well-­ entrenched because, as they put it, it was fixed by God). Words are invested, on this view, with a power (śakti) or dispositional property, to generate, as soon as they are uttered, an awareness of the objects they refer to in the hearer because both the speaker and the hearer share the well-entrenched conventions of a speech-community. Truth is not ­‘correspondence’, but lack of promiscuity, as indicated earlier. Truth is regarded as indistinguishable from knowledgehood or, we can say figuratively, knowledge ‘captures’ or ­‘reveals’ truth. A sentence is said to be true when and only when it generates knowledge in the (ideal) hearer. An ideal hearer is one to whom the information given by the sentence is transparent enough, i.e. unambiguous enough, and in this case contingent factors such as inattention, lack of sensory ability, hypocrisy, would not present any problem. Episodic knowledge does not ‘correspond’ to reality, it grasps reality by being non-­promiscuous or ‘non-deviating’. It may be said that there are other cases of perceptual illusion for which some sort of a ‘representation’ theory or ‘visual data’ theory has to be conceded, e.g., the blue dome in the sky, the blur at a distance and hallucinations. Hallucination (Macbeth’s dagger etc.) could be explained by the Naiyāyikas as gripping memories of some traumatic experience which still haunts the perceiver. So this would be memory masquerading as perception, for generally the perceptual causal factors are present (eyes open, vividness of experience/ memory, etc.). For blue domes, blurs, apparent elliptical forms of discs, etc., the following explanation may be suggested. We may regard them as part of physiological–physical data. In other words, we may admit an ontology of a class of ‘objective’ external particulars. These objective external particulars are momentary, being dependent upon the percipient’s perceiving. It is claimed here that an objective external reality is created de novo in these cases by several factors, and it is only misattributed to a real locus. They are created by a set of physical and non-physical causal factors. But among these causal factors, a mental event (sensing) is also included. This mental event is regarded as a crucial factor in the set of causal factors, crucial in the sense that the said external reality (which is newly created) lasts only as long as that mental event lasts.They cannot exist where no observer is present, but the physical (external) realities have the power or disposition in them to generate such properties with appropriate factors.This may seem to be very similar to the acceptance of external sense-data, or even secondary qualities, but not quite. They are not properties of the surface of opaque physical objects, but only attributed to them. Besides, such external particulars are not admitted to be present in every case of sensory perception. Only in the cases mentioned above are such particulars believed to have been generated. We posit such particulars only to explain the oddity of such erroneous perceptions. They may be like ‘intentional’ properties, provided it is admitted that an intentional property is created as long as it is intended by someone, i.e. as long as the episodic intention lasts. It is not all the time there to be intended, nor is it created simply by somebody’s attitude, for other physical factors must and do cooperate, and it must disappear or die when nobody is intending it. Besides, these particulars may be shared by the agents while it is difficult to say

252

BIMAL KRISHNA MATILAL

that intended objects are shared by the agent. It may be pointed out that while we have pooh-poohed the theory of data, the immediately given, the given which are certainly known, we have now made a volte face and accepted a set of queer entities (which must be called mental). In reply, we must note that these so-called queer entities are sometimes intersubjectively verifiable. One would see the same or similar blur from the same vantage point as I do from here and now. And they are not mental (at least, in one important sense of the word ‘mental’) on this view. Fictional objects or all objects of fictional discourse, on this view, are regarded as composite; they can be broken down to simples and to further simples until we reach the real elements of the actual world we know best. The mind’s creative or active power needs materials to work upon and such materials cannot but finally belong to this world. But receptivity and spontaneity are always blended together. It would be futile to seek a distinction. We have seen that the Naiyāyikas believe in a sort of fallibilism and differ from the Prābhākaras in holding that perceptual illusions are facts. They also differ accordingly in their ‘action’ theories. According to Nyāya we act because of our beliefs or certitudes, even when they are false certitudes. Besides, even doubt may not deter us from acting. This answers the Prābhākara objection that if we admit perceptual errors to occur in our awareness we would not be able to act for we will have no trust in such awareness. For the Prābhākaras, there is no false awareness and hence our action in the case of so-called perceptual errors is prompted not by knowledge, but by the lack of it. We rush to grab a piece of silver due to our lack of knowledge of the distinctness (call it ‘confusion’) between what lies in front and what is desired by us. Hence, on this view, you have to accept the oddity that what generates actions is not a factual belief but the lack of it, for in the so-called errors, some right belief, belief about the disconnection, is missing. For Nyāya, however, false certitudes, i.e. false beliefs, or even doubts, generate action, not lack of them. Finally, what could be said from the ‘misplacement’ theory’s point of view about universal illusionism and universal scepticism? What if the entire external world is just within us, an empty dream? What if we are all simply brains-in-a-vat? Putnam has argued that some plausible assumption about the nature of reference shows that all sentient beings cannot be brains-in-a-vat. This anti-sceptical conclusion is welcome. But Putnam further thinks that for the metaphysical realist, or externalist, this ‘brain-in-a-vat’ is a real ­puzzler while his (Putnam’s) internalism can dismiss it easily. I think it is possible even for a metaphysical realist to refute universal illusionism. According to the sort of metaphysical realism I wish to defend, the ‘brain-in-a-vat’ situation is indeed a possibility (it is not incoherent or self-refuting, as some philosophers apparently suggest). But the Naiyāyikas would say that this possibility has to be parasitical upon an actual world which must be rich enough, or at least contain a minimal number of materials required, for the successful construction of the overall experiences that we as brains-in-a-vat do experience. To put the matter another way: In order to imagine a horned rabbit (a favourite example of Indian philosophers) we need at least rabbits and the materials called horns, or an alternative list of actual items out of which a horned rabbit can be constructed (if, hypothetically, we live in a world where there are neither rabbits nor horns). To imagine a rabbit dressed as an ­Oxford don looking at his watch every five minutes, we need a world inhabited by rabbits, Oxford dons, watches, etc., not an empty world, that is, we must admit, at least a world with people, gowns and clocks, but may be no Oxford dons. This is the sum

A REALIST VIEW OF PERCEPTION

253

and substance of what the Naiyāyikas call the vat-kāra katham argument (‘how this asif-ness’ argument). In a poor world where there are only vats and brains and a particular ­neuro-scientist with his instruments, how can I have the experience of tasting a mango, or the juicy tomato of H.H. Price, unless it contained in addition other materials to constitute such experience of a neuro-scientist who has had such experience or can imagine having such an experience? The existence of another actual world, whatever items it may or may not contain, is thus presupposed by this ‘brain-in-a-vat’ or ‘life is an empty dream’ argument and thereby its sceptical force is undercut. As Gilbert Ryle once put it, ‘A country which had no coinage would offer no scope to counterfeiters. … Ice could not be thin if ice could not be thick.’11 What is distressing about this sceptical suggestion is that it is a distinct possibility. It may just be that a god, or a mad scientist, is sitting there and playing tricks on us. Given our circumstances, we would never know that it is not so. It is a coherent idea. But, the realist may point out, coherence is not a sufficient ground for accepting it as true or actual. Hence we need not accept it. In either case, existence of some external world is not rejected, although our presumed knowledge of it may be infected with doubt from time to time. Scepticism is what we learn to live with, just as somebody said, ‘Paradoxes are what we learn to live with’. A metaphysical realist believes the world to be not only consisting of particulars but also to be already divided into natural kinds of particulars. Here we come to our third problem which must be resolved, if we have to defend this sort of realism. There is this much truth in the traditional view that a realist must also admit reality of some universals. This is what Putnam has called the claim about the world that it contains Self-identifying Objects (the term was used by David Wiggins), i.e. ‘the world, not thinkers, sorts things into kinds’.12 I do not find this claim to be absurd or unjustified. One can believe the world to be divided into natural kinds without knowing for certain exactly which are these natural kinds. We do, of course, try to sort them into kinds, but often go wrong, and, hence, Nyāya accepts fallibilism.We may not have the God’s-Eye view, but we cannot deny that there may be such ‘a view-from-nowhere’. Realism will be saved if we admit that there are objects in the world that are ‘self-identifying’ in this sense, even without our fully succeeding to tell exactly what they are and how many kinds there are. This may open up another possibility. The old realist doctrine that there is only one true and complete description of ‘the way the world is’ may be taken with a pinch of salt. There is one world, although it is possible to have several fairly faithful descriptions of it. For the language in which such descriptions are bound to be given, is a social-cultural fact and hence each description will necessarily tinge the objects described with different socio-cultural colourings or values. This much may be conceded without giving in to full-fledged (Putnamian) internalism where ‘objects’ are as much products of our conceptual invention as of the objective factor in experience. This is also to be distinguished from conceptual relativism or admission of alternative conceptual schemes. For the Davidsonian argument against what he calls the ‘third dogma’ of empiricism, namely, the dualism of scheme and content, of organizing system and something waiting to be organized, is acceptable to our realist. For it has not been conceded that any language will distort reality and that the mind can have a wordless grasp of the uninterpreted reality (a Buddhist position rejected by the Naiyāyikas). I believe there is a latent connection between this third dogma and the

254

BIMAL KRISHNA MATILAL

idea of the undistorted, wordless grasp of the ‘uninterpreted’ reality. We have argued that the active and the passive parts of the mind are not separable. Again, description of the world with different socio-cultural tinges does not amount to admission of completely or partially unintertranslatable, separate conceptual schemes. The realist’s concession for different socio-cultural colourings is made to make room for effects of social conditioning, explainable disagreements and knowledge of explicable error. As Davidson has said, ‘In giving up the dualism of scheme and world, we do not give up the world, but re-establish unmediated touch with the familiar objects whose antics make our sentences, and opinion true or false’.13 The above is also a far cry from the kind of pragmatism that Rorty has held. For one thing, it regards realism-idealism controversies as interesting and the coherence and correspondence theories not as non-competing trivialities. Besides, it is open to the full force of scepticism and tries to cope with it. The modern version of pragmatism trivializes scepticism, or it maintains a position akin to scepticism without the sceptical worries and emotions. The metaphysical realist (I prefer to call him the ‘old-fashioned’ realist) has the world, as Davidson underlines, and this is not a ‘world well lost’. In fact, he can even joke as a cartoonist does, ‘To be perfectly honest, I, sometimes, wish the news were not so factual.’ NOTES

1. 2. 3. 4.



5. 6. 7. 8. 9.

10. 11. 12. 13.

Richard Rorty, The Consequences of Pragmatism, Harvester Press, Brighton, 1982, p. xix. Hilary Putnam, Reason,Truth and History, Cambridge University Press, Cambridge, 1981. The Ṅyāyaśūtras of Akṣapāda Guatama, edited by G. Jha. See also his Translation. Michael Dummett, ‘Common Sense and Physics’, in Perception and Identity, edited by G.F. Macdonald, Macmillan, London, 1979. Ibid., p. 14. Thomas Nagel, Mortal Questions, Cambridge University Press, Cambridge, 1979, p. xii. B.K. Matilal, Perception, Clarendon Press, Oxford, 1986, Chapter 5. Cf.Vyavahāra. I have used this terminology in my-book, Perception. A better terminology for writers in English may be welcome. T.C. Pratyakṣakhaṇḍa,Vol. I. with Māthuri ed. Kamakhyā Nātha Tarkavāgisá, Calcutta, Asiatic Society, 1897, p. 474. Gilbert Ryle, Dilemmas, Cambridge University Press, Cambridge, pp. 94–95. Putnam, Reason, Reason,Truth and History, p. 53. Donald Davidson, Inquiries into Truth and Interpretation, Clarendon Press, Oxford, 1984, p. 198.

CHAPTER

15

– ya perceptual theory Nya Disjunctivism or anti-­individualism? Anand Jayprakash Vaidya

Source: Philosophy East and West, 2013, 63:4, 562–585. I. INTRODUCTION

Misperception is part of the human condition. Consider a classic case of coming to confirm that one has had a misperception. On a stroll through the woods you see, in the distance, what seems to be a person. As you draw near, what looked like a person now appears to be a wooden post with a hat on it. On arrival you touch the post to confirm that it is not a person. From a pre-­theoretical perspective, what has happened? On your approach you judged that there was a person, based on what you saw. When near, you judged that it was a post and not a person, and then by touch you confirmed that what you initially saw was a misperception. In examining cases of misperception it is important to ask: what role does concept possession play in explaining the misperception? The conceptualist answer is that a necessary condition on x misperceiving a post as a person is that x possess both the concept of a post and the concept of a person, so that x can be in a cognitive position to misperceive the post for a person. The guiding idea is that if one possesses neither the concept of a post nor the concept of a person, it is unintelligible how one could have such a misperception. For how could one judge on the basis of one’s perception alone that on approach it was a person but that upon arrival it was a post, if one failed to possess these concepts? The conceptualist maintains that a subject’s misperceptions are controlled by the conceptual scheme that they possess. In general, our conceptual schemes determine a limit boundary to what we can misperceive and what others can take us to have misperceived.1 Philosophers both East and West have taken note of the fact that at least in some cases the incoherence of misperception without concept possession yields to the view that error 255

256

ANAND JAYPRAKASH VAIDYA

or false cognition depends on truth or true cognition. The view that error metaphysically depends on truth, but truth does not metaphysically depend on error is called the asymmetric dependence of error on truth or the parasitism of error on truth. The guiding idea of asymmetric dependence is that the possibility of false cognition depends on a prior true cognition, and as a consequence error is parasitic on truth. If one never perceived correctly then one could not misperceive. In his illuminating comparative essay Parasitism and Disjunctivism in Nyāya Epistemology, Matthew Dasti (2012) carefully argues that the Nyāya school of classical Indian philosophy offered many arguments for the parasitism of error on truth, and that the positions they held relative to these arguments anticipate the epistemological disjunctivism of contemporary Western epistemology, especially that of John McDowell (1996, 2009).2 Disjunctivism, in general, is, roughly, a denial of the claim that there is a common kind of experience between misperceptions and perceptions that is of robust explanatory value for the purposes of the philosophy of perception. At a high level of abstraction, disjunctivism claims that perceptions and misperceptions are similar in the same way that superficially similar, but chemically distinct, compounds are similar. For example, gold and fool’s gold or jadeite and nephrite are macroscopically similar but chemically distinct substances. By analogy, at a deeper level of explanation, disjunctivism claims that perceptions and misperceptions, though phenomenologically similar, are fundamentally distinct and should not be categorized as being of the same epistemic kind. Here I argue that on the assumption that Nyāya perceptual theory does advance arguments for parasitism, we should not conclude so quickly that their work entails or anticipates McDowell’s specific form of epistemological disjunctivism. There are two main reasons for this. First, I argue that the Burge-­McDowell debate over disjunctivism puts pressure on the idea that epistemic disjunctivism is a plausible thesis. Second, Burge’s own perceptual anti-­individualism provides a plausible alternative that has the benefit of being consistent with arguments for asymmetric dependence, as well as with contemporary research in the vision sciences. The overall approach of this essay is exploratory. It aims to provide a constructive engagement between temporally distant and culturally unrelated philosophical traditions for the purposes of enhancing philosophical discussion. On the one hand, it seeks to bring to the table conceptual resources from contemporary analytic epistemology and the vision sciences that can help shed light on what potential options are available for understanding Nyāya perceptual theory and epistemology. On the other hand, it seeks to enrich the pool of ideas from which contemporary analytic epistemology should draw when theorizing about perception by positioning for further discussion the subtle and novel account of misperception advanced in Nyāya epistemology. The plan of this essay is the following. In the second section below, I present the central components and arguments involved in the asymmetric dependence of error on truth, Burge’s anti-­individualism, and McDowell’s epistemological disjunctivism. In the third, I present and analyze the Burge-­McDowell debate over epistemic disjunctivism as a way of defending the claim that epistemic disjunctivism does not follow from asymmetric dependence. In the fourth, I use the analysis in the third to analyze Dasti’s argument and to question whether the Nyāya arguments from parasitism genuinely anticipate

NYĀYA PERCEPTUAL THEORY

257

epistemological disjunctivism. In the fifth, I present the Nyāya misplacement theory of illusion, or MTI. In the sixth, I argue that MTI (a) offers a metaphysical distinction between perceptions and misperceptions that falls short of epistemological disjunctivism, but (b) is consistent with perceptual anti-­individualism. In conclusion, in the seventh section, I discuss how Nyāya perceptual theory should be of interest to those working in contemporary epistemology and perceptual theory. II. ASYMMETRIC DEPENDENCE, DISJUNCTIVISM, AND ANTI-­INDIVIDUALISM Asymmetric dependence

At least with respect to natural-­kind terms and empirical concepts, such as water, human, shell, and dog, error or false cognition asymmetrically depends on truth or true cognition.3 It is possible for one to possess and learn a natural-­kind concept and never be in error or have a false cognition with respect to deploying it. For example, one could learn the concept shell through observation of a teacher who uses it correctly to pick out shells from stones, and then in subsequent use never misapply the concept. However, it is conceptually impossible for one to possess a concept and always be in error with respect to deploying it. The possibility of misapplying an empirical concept that is learned from one’s environment presupposes (a) that the individual possess the concept in question and (b) that their concept possession itself be a function of at least some correct cases of application. Again consider the case of the shell. Suppose a child is being taught the concept shell through ostention from a teacher, and each time the teacher attempts to get the student to apply the concept to a set of diverse objects, the student fails. It is plausible in this case that the student does not possess the concept or even understand it.That is, the student has not acquired the concept through ostention because it is never correctly applied. The failure to ever apply the concept correctly leads to the judgment that the student does not possess it. In another case, suppose one misapplies the concept cow to a zebra that one sees off in the distance, because one is not able to distinguish adequately between a cow and a zebra from the distance one is at. For one to misapply the concept cow to a zebra, one has to possess the concept cow to be able to misapply it to a zebra. The issue of misapplication requires discussion of concept possession, so we must ask: what does it take for one to possess the concept cow? If the concept cow refers to cows, then a subject that is in an environment cannot possess the concept cow unless there is some causal chain terminating in cows through which the person could have learned the concept. The central idea of asymmetric dependence can be unpacked as a commitment to two claims, one about concept possession and one about misapplication: 1 Possession. If an individual A possesses a concept C at time t, then, prior to t, A must have correctly applied C. Concept possession requires correct application. 2 Misapplication. A necessary condition on an individual A misapplying a concept C is that A possess C. If A does not possess C, then A cannot misapply C.

258

ANAND JAYPRAKASH VAIDYA

The conjunction of possession and misapplication allows for an interesting account of concept possession and learning. To possess an empirical concept, one must learn it. To learn a concept one must attempt to use it. But in attempting to use it, one can make a mistake with respect to applying the concept only when one can be said to possess the concept to a sufficient degree. If one fails to use it correctly far more often than not, one cannot be credited with possession of the concept. If one does not possess a concept, then one cannot be said to misapply it.4 Tyler Burge on anti-­individualism

Tyler Burge (1979 and 1986) defends anti-­individualism with regard to mental content. Anti-­individualism with regard to mental content can be stated broadly as a thesis about the individuation of content. Burge’s anti-­individualism maintains that 1 For an individual to possess a certain class of concepts, natural kinds and social kinds in particular, it is necessary that the individual be in a certain kind of physical and social environment. 2 It is possible for two individuals to be intrinsic duplicates of each other, while possessing distinct concepts because they are in, and come from, distinct physical and social environments. 3 The physical and social environment plays an essential role in the individuation of what empirical concepts an individual can or cannot have. 4 For certain kinds of concepts, an individual cannot possess a concept C if the individual has no causal connection either physically or socially to an environment that contains C. Although anti-­individualism with regard to mental content is not uncontroversial, it is supported by a host of thought experiments and considerations. Burge (1986) advances the thesis by considering terms such as ‘arthritis’ and ‘sofa.’ Although Burge’s anti-­ individualism is not identical to Putnam’s (1973) semantic externalism, both theses gain support from the Twin Earth thought experiment propounded initially by Putnam and examined later by Burge.5 In his thought experiment, Putnam invites us to consider two individuals in distinct possible worlds that are intrinsic duplicates of one another. Oscar and Twin Oscar live in physical and social environments that are exact duplicates of each other in every way, except for one feature. Oscar lives on Earth in an environment where H2O is present, and ‘water’ refers to H2O. Twin Oscar lives on Twin Earth where XYZ is present, and ‘water’ is used to refer to XYZ. XYZ is a substance that is distinguishable from H2O only at the level of microstructure and not at the level of macroscopic features such as taste, color, and boiling point, or functional features such as drinking, bathing, and washing. Moreover, everything that H2O is used on Earth, XYZ is used on Twin Earth, and vice versa. Putnam asks us to imagine Oscar traveling to Twin Earth and to consider Oscar’s use of ‘water’ on Twin Earth upon seeing a river. Were Oscar to say, “There is water!” upon seeing XYZ in a river, would his utterance of ‘water’ refer to the XYZ flowing in the river?

NYĀYA PERCEPTUAL THEORY

259

Anti-­individualists take the stance that Oscar’s utterance of ‘water’ would not refer to XYZ, since Oscar has no prior interaction with XYZ and no historical connection to it through a community of users of the term ‘water’ that are in physical contact with XYZ. Moreover, although H2O and XYZ have the same functional role in each environment, Oscar cannot mean XYZ by his use of ‘water,’ and Twin Oscar cannot mean H2O by his use of ‘water.’ The fundamental idea of anti-­individualism is that the individuation conditions for mental content depend on factors that go outside the head of the individual. Oscar and Twin Oscar are intrinsic duplicates. Part of what makes it the case that ‘water’ for Oscar refers to H2O and not XYZ is the fact that Oscar is part of an H2O, and not an XYZ, environment and community. Part of what makes it the case that ‘water’ for Twin Oscar refers to XYZ and not H2O is the fact that Twin Oscar is part of an XYZ, and not an H2O, environment and community. John McDowell’s epistemic disjunctivism

McDowell’s epistemic disjunctivism, ED, is offered as an alternative to the highest common factor view of experience.The highest common factor view of experience, HCF, maintains that veridical and non-­veridical cases share a common kind of mental state. The HCF is motivated in part by the argument from illusion that has been prevalent in many discussions and debates on perception, from Descartes and Hume to Ayer and Austin. McDowell describes the HCF and the line of reasoning leading to it: [T]he argument is that since there can be deceptive cases experientially indistinguishable from non-­deceptive cases, one’s experiential intake—what one embraces within the scope of one’s consciousness—must be the same in both kinds of case. In a deceptive case, one’s experiential intake must ex hypothesi fall short of the fact itself, in the sense of being consistent with there being no such fact. So, that must be true, according to the argument, in a non-­deceptive case, too. One’s capacity is a capacity to tell by looking: that is, on the basis of experiential intake. And even when this capacity does yield knowledge, we have to conceive the basis as a highest common factor of what is available to experience in the deceptive and non-­deceptive cases alike, and hence as something that is at best a defeasible ground for the knowledge, though available with a certainty independent of whatever might put the knowledge in doubt. (McDowell 2009, p. 80)

McDowell’s characterization of the HCF line of reasoning is as follows: 1 Veridical perception of a yellow lemon caused by a yellow lemon Y under normal viewing conditions and a non-­veridical perception whose content is of a yellow lemon but is caused by something other than a yellow lemon are first-­person phenomenologically indistinguishable. 2 If two states are first-­person phenomenologically indistinguishable, then they should be categorized as falling under a common epistemic kind. 3 If two states fall under the same epistemic kind, then they provide the same epistemic warrant.

260

ANAND JAYPRAKASH VAIDYA

4 So, veridical and non-­veridical perceptions provide the same warrant. 5 If two states have the same kind of warrant, then they provide a subject with the same experiential intake. 6 So, one’s experiential intake is the same in veridical and non-­veridical cases. The central idea in the HCF, in McDowell’s rendering, is that conceiving of experience according to the HCF view leads to the idea that the ultimate basis of our beliefs about the external world lies in mere appearances that cannot acquire more warrant than what is provided by what is in common between veridical and non-­veridical cases. The alternative view McDowell endorses is the disjunctive conception of experience. His epistemic disjunctivism, ED, has four main components: 1 Perception is a capacity for knowledge: A perceptual capacity … is a capacity—of course fallible—to get into positions in which one has indefeasible warrant for certain beliefs. That is what the capacity is a capacity to do, and that is what one does in non-­defective exercises of it, exercises in which its acknowledged fallibility does not kick in. For instance, a capacity to tell whether things in one’s field of vision are green is a capacity—of course fallible—to get into positions in which the greenness of things is visibly there for one, so that one has indefeasible warrant for believing that they are green. (McDowell 2011, p. 245)

2 Perceptual appearances are metaphysically distinct: The conception of [experience] I have found … can be put, in opposition to [the highest common factor conception], as a disjunctive conception of perceptual appearance: perceptual appearances are either objective states of affairs making themselves manifest to subjects, or situations in which it is as if an objective state of affairs is making itself manifest to a subject, although that is not how things are. (McDowell 2008, p. 381)

3 Perceptual appearances have asymmetric warrant: Experiences of the first kind [objective states of affairs making themselves manifest to subjects] have an epistemic significance that experiences of the second kind do not have. They afford opportunities for knowledge of objective states of affairs. According to the highest common factor conception, appearances can never yield more, in the way of warrant for belief, than do those appearances in which it merely seems that one, say, sees that things are thus and so. (McDowell 2008, p. 381)

4 Perceptual experience is non-­factorizable: [I]t is part of the point of my disjunctive conception of experience that having an aspect of objective reality perceptually present to one entails having it appear to one that things are a certain way. But that is not to say that having an aspect of objective reality perceptually present to one can be factored into some non-­mental conditions and an appearance conceived as being the mental state it is independently of the non-­mental conditions. The factoring fails; the state is the appearance it is only because it is a state of having something perceptually present to one. (McDowell 2011, p. 251)

NYĀYA PERCEPTUAL THEORY

261

Epistemic skepticism with respect to a potential domain of knowledge is the view that human subjects do not have knowledge of potentially knowable items in that domain. At least part of McDowell’s (2008) motivation for advancing epistemic disjunctivism is his belief that it provides resources for a transcendental argument against epistemic skepticism about the external world. His strategic argument can be seen to be the following: 1 If epistemic disjunctivism is true, then there are transcendental reasons for rejecting epistemic skepticism. 2 Epistemic disjunctivism is true. 3 So, there are transcendental reasons for rejecting epistemic skepticism. As a consequence of the strategic argument, McDowell’s position can be evaluated at two distinct points. On the one hand, one can evaluate whether or not epistemic disjunctivism is true. On the other hand, one can evaluate whether or not the truth of epistemic disjunctivism provides a transcendental argument for rejecting epistemic skepticism about the external world. Finally, in clarifying ED it should be noted that the target form of epistemic skepticism about the external world that it aims to undercut is a radical form of epistemic skepticism. That is, ED attempts to undermine a form of epistemic skepticism that threatens the idea that our thoughts can be about objective reality. ED does not entail the view that perception is infallible. Rather, perception is a capacity to know in the sense that we can get into positions where perception yields knowledge. III.  BURGE AND MCDOWELL ON DISJUNCTIVISM

Burge’s criticism of McDowell’s epistemic disjunctivism is an attempt to evaluate the truth of epistemic disjunctivism and not the claim that it provides transcendental reasons for rejecting epistemic skepticism about the external world. His assessment and evaluation of epistemic disjunctivism rests on an examination of perceptual psychology and vision science. His meta-­theoretical account of the philosophy of perception maintains that any theory of perceptual content must pay respect to perceptual psychology and the vision sciences. It is inconsistency with vision science that renders a theory implausible. In his work on perceptual psychology, Burge has argued that perceptual anti-­ individualism is consistent with contemporary perceptual psychology and that, indeed, the vision sciences presuppose the truth of it. Perceptual anti-­individualism, PAI, is the thesis that a constitutively necessary condition on perceptual representation by an individual is that any such representation be associated with a background of some veridical perceptual representations (Burge 2005, p. 1). In contrast to PAI, Burge argues that epistemic disjunctivism and naive realism about perception are untenable theses. The core claims of these views cannot be made consistent with contemporary work in perceptual psychology: Disjunctivism is implausible. Not only common sense but [also] the scientific knowledge [in the vision sciences] support this initial evaluation. Disjunctivism is incompatible with the Proximality Principle, which is basic in nearly all scientific study of perception.

262

ANAND JAYPRAKASH VAIDYA

Given that different distal causes can yield proximal stimulation that is relevantly the same, perception of entities in the distal environment is fallible.The Proximality Principle, together with this empirical fact, entails that the same type of perceptual state can be veridical or non-­ veridical, perceptually referential or non-­referential. (Burge 2005, p. 27)

His basic argument against ED is as follows: 1 ED denies that there is any important explanatory kind between veridical and non-­ veridical states. 2 The constitution of the perceptual system requires the truth of the Proximality Principle. 3 The Proximality Principle requires that perception involve an ability-­general kind in common between veridical and non-­veridical states. The ability-­general kind is inconsistent with the claim that there is no important explanatory kind between veridical and non-­veridical states. 4 So, ED is false. The Proximality Principle, PP, maintains that holding constant the antecedent psychological set of the perceiver, a given type of proximal stimulation (over the whole body), together with the associated internal afferent and efferent input into the perceptual system, will produce a given type of perceptual state, assuming that there is no malfunctioning in the system and no interference with the system.

A set of relevant cases where PAI, through PP, and ED disagree is shown by the following series: Suppose that one sees an object. Then as one blinks, the object is removed and replaced by a duplicate that one cannot discern from the original in the context. As one blinks again, the duplicate is removed. One is induced by an abnormal confluence of light to have a visual illusion as of an object that is indiscriminable from the originally seen object.The light array hitting the retina is, we shall suppose, type-­identical in the three cases—or at least sufficiently similar that the perceptual system cannot make use of the difference. (Burge 2005, p. 26)

PP requires that one have a general ability to use the information in common between the three cases. In Burge’s account the ability is explanatory for how we come to have a perceptual system at all. Were we not to have a general ability to use the information in common between the three states, we could not have evolved to have a perceptual system. Of course, the possibility of the system evolving also requires that there are veridical states. ED, by contrast, denies that there is an explanatorily relevant kind in common between the three states.While PAI individuates perception at a type-­level commonality, ED does so by virtue of the veridicality conditions at the token level. Given that in the three cases the perceptual state is only phenomenally similar, and not target similar, ED maintains that the states are different in an explanatorily robust manner relevant to the classification of epistemic kinds.

NYĀYA PERCEPTUAL THEORY

263

In understanding PAI, it is of central importance to take note of the kind of account that Burge believes that the proximality principle delivers. In his account, PP is supposed to deliver states that are not merely of the sub-­personal visual processing system. Rather PP governs the level of perceptual states that are attributable to individuals as conscious perceivers. This theoretical stance on PP is important because of a potential objection that one can make to the relevance of vision science to the philosophy of perception. If PP only explained sub-­personal visual processing below the level of phenomenal consciousness, it would be possible for an epistemic disjunctivist to respond as follows: since ED is a thesis that applies at the personal level of perceptual theorizing, and not at the sub-­ personal system-­processing level, the mechanism by which personal-­level perception is delivered is irrelevant to the disjunctivist thesis. In effect, the disjunctivist would block the significance of PP as providing a problem for the plausibility of ED. The issue can be seen to lead to a potential conflict of methodologies. On the one hand, the epistemic disjunctivist does not attend to theories in the philosophy of perception that engage perception at the level of sub-­personal processing. On the other hand, PAI attempts to bring into the philosophy of perception the relevance of work in the vision sciences and perceptual psychology. I believe that Burge’s work is useful at a theoretical level where it can be deployed as a mechanism for disentangling the relations between asymmetric dependence, anti-­individualism, and epistemic disjunctivism. In the account I will offer, McDowell and Burge can be seen to be in agreement over the importance that the asymmetric dependence of error on truth plays in a theory of perception.That is, they agree that veridicality is a necessary condition for the possibility of perception. Concept possession and perceptual capacities are enabled by veridicality. However, what they disagree on is what follows from asymmetric dependence. In short, PAI maintains that veridicality is necessary for a perceptual system to arise, but that veridicality is not an essential property of a perceptual type. For the purposes of distinguishing further between ED and PAI, consider the following cases: a b c d

Rick misperceives a rope for a snake. Varsha mis-­identifies a piece of tofu for a piece of chicken. Manjula misperceives regular coffee for decaffeinated coffee. Zuleica mis-­identifies F sharp with C sharp.

In each of these cases something goes wrong insofar as the person mistakes x for y. Furthermore, given asymmetric dependence, each of the individuals can only make the mistake that they make because they possess the relevant concepts. However, we might further ask: what follows from asymmetric dependence? There are three arguments relating asymmetric dependence to epistemic disjunctivism. Each argument moves from the premise concerning asymmetric dependence to some claim concerning epistemic disjunctivism. One might legitimately ask if (2) through (4) below follow from (1): 1 Non-­veridical perception requires veridical perception (asymmetric dependence). 2 So, a perception is either a mere appearance or a presenting of an objective fact (metaphysical distinctness).

264

ANAND JAYPRAKASH VAIDYA

3 So, the epistemic warrant in a veridical case is not the same as the epistemic warrant in a non-­veridical case (asymmetric warrant). 4 So, a veridical perception cannot be factored into an appearance and the objective fact that makes it a veridical perception (non-­factorizability). In analyzing the argument, we need to take note of the nature of each of the claims. Asymmetric dependence is a metaphysical claim about the relation between truth and error as they apply to the things that can be the bearers of truth and error—truth-­evaluable contents.The metaphysical distinctness of veridical and non-­veridical states is a metaphysical thesis about the proper taxonomy of perceptual states. The asymmetric warrant between veridical and non-­veridical states is an epistemic principle concerning epistemic warrant for belief. And the non-­factorizability of veridical states is a metaphysical claim about the components involved in a veridical state. The distinction between Burge’s PAI and McDowell’s ED can be understood as a questioning of each of the inferences from (1). While McDowell is far more open to inferring from asymmetric dependence various components of ED, Burge is far more cautious. In Burge’s account, asymmetric dependence is amenable to anti-­individualism, understood as the idea that perception only makes sense against the background of veridical states. This metaphysical claim, which is constitutive of perception for Burge, entails neither that there is no common factor of explanatory importance between veridical and non-­veridical perception nor that veridical and non-­veridical states have asymmetric warrant.6 The Burge-­McDowell debate leaves us with the following question: is perceptual anti-­individualism or epistemic disjunctivism a superior platform for further theorizing about perception and for categorizing Nyāya epistemology and perceptual theory? IV.  DASTI ON DISJUNCTIVISM FROM PARASITISM

Matthew Dasti maintains that “Nyāya [epistemology] privileges veridical truth-­entailing mental states and considers error conceptually parasitical upon knowledge.” And that “This [asymmetric dependence] entails a disjunctive account of pramana and non-­pramana states” (Dasti 2012, p. 3). In his account of Nyāya epistemology and philosophy of mind there are three forms of argument from parasitism: 1 Epistemic parasitism. Recognizing an error is parasitical upon knowing truth. 2 Causal parasitism. Any concept V that one deploys in various sorts of error states ultimately depends on one’s original veridical apprehension of some instance of V. 3 Parasitism of content or meaning. Divorced from connection with external reality, concepts would be drained of content, as would the words whose meanings are tied to the concepts they express. As an example of epistemic parasitism Dasti cites Uddyotakara’s response to a Buddhist interlocutor who contends that everything exists in a state of flux, and therefore all cognitions of enduring things are false: False cognitions are imitations of correct cognitions. Therefore, you must provide some ­example of correct cognition. (Dasti 2012, p. 4)

NYĀYA PERCEPTUAL THEORY

265

As an example of causal parasitism Dasti cites an argument by Vātsyāyana: The mis-­cognition of something depends on an original. The cognition of a post—which is not a person—as a person depends upon an original. Indeed, there is no experience as of a person regarding something that is not a person, if a person was never experienced in the past. (Dasti 2012, p. 6)

As an example of meaning parasitism Dasti cites Uddyotakara: He must be asked how consciousness arises in that very form (the form of specific objects). If consciousness takes the form of blood, then you must explain what blood is. Similarly, the form of water and river must be explained. In the sentence,“they see a river of pus,” each word, when examined individually, is found to be meaningless, if there are no real external objects. (Dasti 2012, p. 7)

Dasti argues that the three forms of parasitism lead to important features of epistemic disjunctivism: 1 Default trust. Arguments from parasitism show that the default epistemic position one should take is trust and not doubt. 2 Denial of HCF. Arguments from parasitism tend to block the need to find a common state between veridical and non-­veridical perception. 3 Metaphysical distinctness. Non-­veridical states are fake perceptual states; they are only phenomenally indistinguishable from genuine or veridical perception. In evaluating the merits of the Nyāya perceptual theory and Dasti’s argument for the claim that their account anticipates McDowell’s epistemic disjunctivism it is important to look at the complexity of the Burge-­McDowell debate over disjunctivism and anti-­ individualism. The debate provides additional conceptual resources for categorizing and thinking about Nyāya perceptual theory. Recall that Burge maintains that “A closely associated thesis [of anti-­individualism] is that a constitutively necessary condition on perceptual representation by an individual is that any such representation be associated with a background of some veridical perceptual representation” (Burge 2005, p. 1; emphasis added). I take Burge, in making this claim, to be endorsing the view that error depends on truth, but truth does not depend on error, at least with respect to the generation of perceptual states. For in saying that a condition on any perceptual representation is that there be a background of veridical representation, Burge is maintaining that the possibility of having a representation at all depends on veridicality. The core claim of parasitism in Nyāya, as Dasti argues, is that error and non-­veridical cognition are conceptually parasitic on truth.The passages he cites show this clearly to be the case. However, Burge argues that PAI is consistent with the Proximality Principle, and that ED is not. Given the argumentation between Burge and McDowell, two questions arise. First, if Burge is correct in arguing that ED is inconsistent with the vision sciences, we are left with an evaluative question: should we take seriously the Nyāya perceptual theory as a worthy theory of study for theorizing further about perception in contemporary philosophy? One might wonder, what can we learn from a theory that is inconsistent with a principle that is the benchmark for contemporary research on perception? Second, given

266

ANAND JAYPRAKASH VAIDYA

that Nyāya perceptual theory, Burge’s PAI, and McDowell’s ED all endorse the asymmetric dependence of error on truth, we are left with an interpretive question: given that Burge and McDowell disagree over the consequences that follow from asymmetric dependence, might there be room to investigate critically whether Nyāya perceptual theory also leans more toward PAI than ED? With respect to the interpretive question, four components of Dasti’s argument are important. First, ED does not give us a better footing for understanding an epistemic position as being one of default trust rather than doubt. It is possible that PAI can provide for a default position of trust, and perhaps even a more accurate account of the relevant kind of trust through how it explains the way that veridicality is a necessary condition on perception. We need a more robust account of the notion of default trust in order to evaluate the claim that ED provides a better position for the basic notion than any other competitor theory, such as PAI. Second, neither epistemic, causal, nor meaning parasitism show that there is no common kind of mental state between veridical and non-­veridical states that plays an important explanatory role. At best these forms of parasitism, as Dasti points out, show a resistance to finding a highest common factor from a certain philosophical frame of investigation.That is a philosophical frame that does not attempt a theory of perception based on an attempt to answer a totalizing form of epistemic skepticism. Given that Nyāya perceptual theory aims at providing an etiological account of the sources of knowledge, it may not need to find an internal component of a given source that is also at play in cases where knowledge is not produced. Moreover, it may be the case that were the certain components of their view satisfied, the theory would be amenable to the existence of an important explanatory factor between veridical and non-­veridical cases of perception. Third, the three forms of parasitism that Dasti draws attention to—epistemic, causal, and meaning—do not lean directly toward ED. On the one hand, epistemic parasitism appears to be a thesis about what is a necessary condition for identifying and recognizing a false case of perception. It does not require that there be no common kind of element between veridical and non-­veridical states. What it requires is that knowledge of F is implicated in recognizing that something is a non-­F. The claim of epistemic parasitism is quite innocuous: to identify that a zebra is not a cow, one must know what a cow and a zebra are (think here of the Meno problem). On the other hand, causal parasitism appears to be no more than the thesis of asymmetric dependence, while meaning parasitism appears to be an outright example of anti-­individualism about meaning. Recall that the central thesis of anti-­individualism with regard to meaning is that an individual’s use of a term depends on factors outside their personal psychology. These factors include social and physical facts about their environment. In the classic Twin Earth cases, Oscar and Twin Oscar are claimed to be skin-­deep duplicates of each other with identical skin-­deep histories living in environments that are exactly similar, except for the fact that Oscar’s contains H2O and Twin Oscar’s contains XYZ. Although the substances H2O and XYZ play the same water role in their respective environments, Oscar’s use of ‘water,’ and not Twin Oscar’s use, refers to H2O. The generally accepted explanation of this fact is that (a) Oscar lives in an environment where people refer to H2O samples when using ‘water,’ and (b) the fact in (a) played a key causal role in Oscar’s use of ‘water.’

NYĀYA PERCEPTUAL THEORY

267

Here, (a) and (b) explain how Oscar’s use of ‘water’ comes to mean H2O and cannot mean XYZ. The standard upshot of the Twin Earth examples of natural and social kinds is that a cannot mean y by x unless a has causally interacted in some way with y, no matter how much similarity there is taking in other non-­causal factors, such as resemblance or matching of descriptive content. Keeping in mind the central thesis of anti-­individualism, parasitism of content or meaning maintains that divorced from a connection with external reality, concepts would be drained of content, as would the words whose meanings are tied to the concepts they express. The thesis leans strongly toward the central idea of anti-­individualism: content and meaning are determined by factors outside the psychology of the individual, because what is emphasized is the role of the external world in determining content. Parasitism of content and meaning does not lean toward ED more than PAI. The passage that Dasti cites from McDowell as a way of making the connection between disjunctivism and the kinds of parasitism found in Nyāya epistemology does not announce the full robustness of the kind of disjunctivism that McDowell favors: But suppose we say—not at all unnaturally—that an appearance that such-­and-­such is the case can be either a mere appearance or the fact that such-­and-­such is the case making itself perceptually manifest to someone. As before, the object of experience in the deceptive case is a mere appearance. But we are not to accept that in the non-­deceptive cases too the object of experience is a mere appearance, and hence something that falls short of the fact itself. (McDowell 2009, p. 80)

In the passage above McDowell appears to be announcing ED via the non-­factorizability and metaphysical distinctness of veridical and non-­veridical mental states. However, his full account is tied to the additional theses that perception is a capacity to know, and that veridical and non-­veridical states have asymmetric warrant.The robustness of McDowell’s ED should lead us to ask critically whether the parasitism found in Nyāya epistemology really reaches as far as ED. Finally, it is should be noted that McDowell’s ED is in part motivated by an attempt to provide a transcendental argument against epistemic skepticism. In terms of a philosophical point of departure, epistemic disjunctivism appears as a response to the Cartesian skeptical frame. It is generated as a response to a tradition of philosophical theorizing in the Modern period of Western philosophy that turned toward taking the skeptical argument and the method of doubt as a starting point for philosophical reasoning. One of Dasti’s key comparative insights is that Nyāya epistemology starts from the default position of trust as opposed to doubt. If his understanding of this claim is correct, then the classical Indian frame for philosophical reasoning would not be similar to the Cartesian skeptical frame, and our interpretation of any disjunctive-­like components of Nyāya should be read in an appropriately adjusted manner. In the model of interpretation that I favor, I maintain that a motivation for the Nyāya account is the search for criteria by which one can explain how perception is an instrument of knowledge by looking at how the causal processes involved in misperception are distinct from those involved in perception. I believe that this component also makes their view amenable to Burge’s PAI.

268

ANAND JAYPRAKASH VAIDYA

V.  MISPLACEMENT AS A ROUTE TO METAPHYSICAL DISJUNCTIVISM

The Nyāya misplacement theory of illusion, MTI, is one of the most interesting accounts of illusion in philosophy both East and West. The Nyāya MTI allows one to attribute to the Nyāya tradition of epistemology two claims. First, veridical and non-­veridical states are metaphysically distinct because of the causal processes that go into each state. Second, the causal difference between veridical and non-­veridical states renders MTI different from McDowell’s ED, but consistent with Burge’s PAI. For the purposes of understanding the scope of MTI it is important to take note of the fact that non-­veridical states can be further classified as being misperceptions or hallucinations. MTI is propounded primarily as a theory that applies to misperceptions. Misperceptions are classified as cases where an object is seen to have a property it in fact does not have. Classical cases of this are seeing a snake as a rope, seeing a stick submerged in water as bent, seeing a white shell as being yellow, or seeing an object a in the distance as F when it is G (where F and G are incompatible). In each of these cases an object a is seen to have a property, F, that it in fact does not have. Hallucinations, on the other hand, occur when there is no object that is the foundation for false property attribution. Classic cases of hallucination do not have a particular worldly object as the ground of predication. Waking hallucinations of this kind are similar to cases of dreaming in the following sense: when one’s eyes are not being stimulated through interaction with an external environment there is no particular of the external environment that is the ground of false predication. Although it is tempting in cases of hallucination to think that there is an object that is misperceived, this is a mistake usually made on the basis of the fact that in theorizing we maybe over-­focused on the fact that in both cases one’s eyes are engaged with their external environment. To separate misperception from hallucination, contrast seeing a person as a post with hallucinating an elephant. In the later case, if one moves one’s field of vision by moving one’s head, one continues to see the elephant. By contrast, in seeing a person as a post, when one moves one’s field of vision one does not continue to see something about which one is in perceptual doubt, since the object that is seen to be either a person or a post is fixed in the external environment. In cases of hallucination, the external environment merely facilitates the hallucinated object by providing a background. The background does not involve an object that causes the hallucination. The core of MTI as an account of misperceptions, and not of hallucinations, can be grasped through an extended examination of a case of coming to confirm that one has had a misperception. Suppose that upon approaching from a distance one sees a snake, but as one comes near it is revealed to be a rope. Furthermore, as one leans forward to grab it one confirms that it is a rope and not a snake. In this succession of events we have a misperception that yields to a perception that is then confirmed via a distinct perceptual modality—­ tactile perception. The standard Nyāya analysis of this sequence of events is as follows. First, on the approach to the rope, for one to see the rope as a snake and then as a rope, one must first possess the concept of a snake and the concept of a rope. For if one has never seen a snake before, then one cannot see the rope in front of one, first as a snake and subsequently as a rope. Second, in seeing the rope as a snake, what has happened is that the normal causal process by which one would see the rope as a rope has been interrupted by a memory.The memory of a snake has arisen in one and has been imposed into awareness. By contrast, when one comes closer to the rope and sees the rope as a rope, no memory

NYĀYA PERCEPTUAL THEORY

269

has intervened into the causal stream that brings about the awareness. Rather, the causal conditions that give rise to the awareness are truth productive. On a further elaboration of MTI, Bimal Krishna Matilal explains the view by claiming that the misperception is itself based on objective features of the situation. More precisely he claims that it is very unlikely that a subject can misperceive A as B if there are no features in common whereby A can be confused with or seen to be B by an imposition from memory. In the case of the snake-­rope misperception, it is because a rope can look like a snake that it can be misperceived as a snake by the imposition of the memory of a snake into the causal stream of the perceiver. The point is that while it is likely that a memory can intervene and cause a rope to be seen as a snake, it is unlikely that a memory can intervene and cause, for example, an ocean to be seen as a spider. The latter case would likely be a hallucination of a spider imposed on an ocean, rather than a misperception. Moreover, the objective properties of the relevant objects play a key role in explaining the possibilities for misperception through the perceptual system. MTI allows two important factors in accounting for misperception. On the one hand, there is the subjective profile of the individual that includes the memories and concepts that the individual possesses. If an individual does not possess the concept of a snake, the individual cannot perceive a rope as a snake. On the other hand, the objective properties F and G of objects A and B are such that one can misperceive A and B because of F and G. Moreover, in the snake-­rope misperception, it is because snakes and ropes satisfy an objective sufficient-­similarity condition, that it is possible for one who possesses both the concept of a snake and a rope to misperceive the rope as a snake. The Nyāya MTI is quite natural and insightful.The rope has the dispositional property to be misperceived as a snake because it has some characteristics in common with a snake that allow the causal nexus, which includes the person approaching, and the person’s conceptual repertoire, to misperceive it as a snake. The misperception proceeds by way of triggering a memory of a snake that is then imposed into the cognitive stream whereby the rope is seen as a snake. By contrast, in the veridical case, as one comes near, the rope has the ability to be seen as it is by one who has the concept of a rope. No memory intervenes in the causal stream between sense organ and object. Returning to our question: what is the feature that makes veridical states metaphysically distinct from misperceptions of the snake-­rope kind? MTI maintains that the causal pathway and proper functioning by which the cognition arises in the individual explains the difference. In misperception a memory has intervened in an inappropriate way. In veridical perception there is no such intervention. Gautama, a founding contributor to the Nyāya, defines perceptual cognition in his Nyāya-­sutra as: [a] cognition [that] arises from the contact of sense faculty and sense object, [which] does not depend on language, is inerrant, and is definite. (Nyāya-­sutra 1.1.4)

Matilal (1986) explains the definition by showing that it is intended to rule out certain kinds of cases where perception is absent. He lists three main cases: 1 Perceptual doubt. One sees from a distance something that looks like it could either be a man standing or a tree trunk. One does not know which it is and has a perceptual doubt.

270

ANAND JAYPRAKASH VAIDYA

2 Misperception. One sees a snake when there is only a rope before one, or a white shell as yellow. 3 Non-­identification. One sees something but does not know what it is, since one has never seen it before or heard it described. The three cases are all cases in which we fail to perceive. In the first case, we do not perceive anything because there is epistemic indeterminacy. We are neither certain that it is a man nor certain that it is a tree trunk. In the second case, we fail to perceive because we misperceive. There is only a rope, and it is misperceived as a snake. The third case is a failure to perceive because by definition one does not know what one sees.7 Stephen Philips also offers an eloquent exposition of MTI. In explaining the case of misperceiving a rope for a snake he addresses the key feature that is important for understanding the difference between misperceptions and perceptions in Nyāya perceptual theory: Here we touch the heart of Nyāya realism. … Snakehood is available to become illusory predication content through previous veridical experiences of snakes. It gets fused into a current perception by means of a foul-­up in the normal causal process through the arousing of a snake-­hood memory formed by previous experiences of snakes.The content of an illusion is to be explained causally as generated by real features of real things just as veridical perception is too, although illusion involves the projection into current perception of predication content preserved in memory whereas at least in some cases (for example, those where an indeterminate perception furnishes the qualifier) veridical perception is not shaped by memory. (Phillips 2004, p. 111; emphasis added)

The ground of the distinction is the recognition of “a foul-­up in the normal causal process.” Veridical cases for the Nyāya are metaphysically cases of proper causal functioning across all causally relevant factors. Non-­veridical (at least cases where a subject is perceived to see an object with a property it does not have) are cases where there is an error in the normal causal processing. The metaphysical distinction between the two cases comes from the main factor that produces each, the causal nexus. In the Nyāya account veridical perception comes about when our memory does not intervene in the production of a cognition that arises from sense contact with an object. In non-­veridical perception, at least in misperception, our memory intervenes and introduces into the causal pathway a content that is improper. VI.  MTI, ED, AND PAI

McDowell’s ED includes four theses: (1) perception is a capacity to know, (2) veridical and non-­veridical states are metaphysically distinct, (3) veridical and non-­veridical states possess asymmetric warrant, and (4) veridical states are non-­factorizable. MTI grounds the claim that Nyāya epistemology and their theory of perception endorse a metaphysical distinction between veridical and non-­veridical states. However, I will argue that MTI falls short of ED, and that MTI is compatible with PAI. My strategy will be to discuss MTI relative to each of the theses that are part of ED.

NYĀYA PERCEPTUAL THEORY

271

It is unclear whether Nyāya epistemology would endorse (1). While it may be the case that their epistemology is interested in the idea that perception is an instrument for knowledge, it is not at all clear that their account conceives of perception as a capacity to know in the way explained by McDowell: “A perceptual capacity … is a capacity—of course fallible—to get into positions in which one has indefeasible warrant for certain beliefs.” The main issue is that a perceptual capacity in McDowell’s account is a capacity one has in order to get into a position where one has indefeasible warrant. In order for the Nyāya account to be similar to McDowell’s account it would have to be the case that defeasible and indefeasible warrant are important factors within Nyāya epistemology and perceptual theory. Additionally, the claim that perception is a capacity to know is too broad. In particular, Burge’s PAI does not preclude perception from being a capacity for knowledge. It simply explains the details by which perception is a capacity for knowledge in a distinct manner. What is important in identifying perception as a capacity for knowledge is how a theory frames fallibility. Burge and McDowell agree that perception is fallible; what they disagree on is the proper way to understand fallibility in perception.8 There is strong evidence that Nyāya epistemology would endorse (2), the claim that veridical states and non-­veridical states are metaphysically distinct. However, it should be noted that their reason for accepting this claim is based on the idea that the causal processes that go into veridical perception are distinct from those that go into non-­veridical perception. In advancing this metaphysical account they differ from the disjunctivist insofar as the disjunctivist emphasizes that the difference between veridical states and non-­ veridical states stems from the fact that they are only superficially or phenomenologically similar, and that phenomenological similarity is not sufficient for categorizing epistemic kinds. In general, the fact that two accounts offer a metaphysical distinction between veridical and non-­veridical states entails neither that both accounts offer the same distinction between veridical and non-­veridical states nor that they offer the distinction for the same reason. It is likely that Nyāya epistemology would also deny (3), the claim that veridical states and non-­veridical states have asymmetric warrant, for reasons similar to those present in the rejection of (1). The idea of justification as a component of knowledge provided by perception is not operative in the frame of perceptual and epistemic theorizing that is present in Nyāya epistemology. Moreover, Nyāya epistemology does not appear to engage the internalist intuition that justification as a necessary condition of knowledge requires the capacity to articulate reasons in argumentation. In particular, given that (a) the idea of asymmetric warrant is proposed against the background of the Cartesian frame of skepticism and the argument from illusion, in which internalism operates, and (b) these components are absent in Nyāya epistemology, it is unlikely that their account would endorse (3). Finally, it is unclear whether Nyāya epistemology would endorse (4). The idea that perception is non-­factorizable can be approached in two distinct ways. On the one hand, MTI allows one to claim that a misperception can be factored into object, sense organ, and memory, and perception can only be factored into object and sense organ. As a consequence, MTI allows for the view that veridical perception is non-­factorizable because it denies the presence of the memory state, which is present in non-­veridical perception.

272

ANAND JAYPRAKASH VAIDYA

On the other hand, McDowell’s claim that perception is non-­factorizable is a denial of the claim that having an aspect of objective reality perceptually present to one can be factored into some non-­mental conditions and an appearance conceived as being the mental state it is independently of the non-­mental conditions. (McDowell 2010, p. 251)

And it is an endorsement of the claim that [a] state is the appearance it is only because it is a state of having something perceptually present to one. (McDowell 2010, p. 251)

The non-­factorizability claim amounts to a position on the elements of explanatory relevance for a state being veridical. The state is a veridical appearance because it is an occurrence based on something being perceptually present to one. Objective reality being present is what explains the veridicality of the perception. The veridical state cannot be factored into the non-­mental conditions that bring it about and the mental conditions that bring it about. The structure of the objective world is an essential ingredient in explaining the veridicality. The structure of McDowell’s account of non-­factorizability does not allow for a clean location of the view in Nyāya epistemology because while MTI does give us an account of misperception it does not give us a direct positive account of perception.The conditions that Nyāya impose on perception, for example being non-­erroneous, are stated as necessary conditions, and not as positive explanatory conditions. Perceptual anti-­individualism offers a better option for categorizing MTI.There are two central reasons why. First, PAI and MTI require that there be objective features of the perceiver’s environment that can play an explanatory role in how a misperception is produced. Recall that PAI maintains that all perception requires a background of veridical perception. MTI also maintains that a misperception requires a background of veridical perception whereby the misperception can occur. One way to see this point is by looking at how objective similarity in the perceiver’s environment is used to explain misperception. In order to misperceive the rope as a snake one must correctly perceive objective features of the rope whereby it can be misperceived as a snake. If one were to fail to see the coiled rope as coiled in a specific way, it is unlikely that one would misperceive it as a snake.The objective similarity between coiled rope and coiled snake enables the possibility of misperception. Second, the proximality principle, PP, is consistent with MTI. PP states that holding constant the antecedent psychological set of the perceiver, a given type of proximal stimulation (over the whole body), together with the associated internal afferent and efferent input into the perceptual system, will produce a given type of perceptual state, assuming that there is no malfunctioning in the system and no interference with the system. As noted earlier in Philips’ explanation of MTI, the key feature of the theory is that “Snakehood is available to become illusory predication content through previous veridical experience of snakes. [This is because] it gets fused into a current perception by means of a foul-­up in the normal causal process through the arousing of a snake-­hood memory formed by previous experiences of snakes” (Philips 2004, p. 111; emphasis added).

273

NYĀYA PERCEPTUAL THEORY

What both theories take into consideration is the importance of proper functioning and non-­interference with the causal system. PP maintains that a certain type of perceptual state is the output of the relevant causal process as long as there is no malfunction or interference. MTI maintains that misperception is a consequence of an interference with the causal system. Thus, Nyāya epistemology can maintain that in cases of perception the causal system across all relevant factors is functioning properly, but when a misperception occurs it is because interference has occurred in the normal causal processing.9 Ultimately, it is the emphasis on objective properties in the world and causal processing that makes PAI a safer positioning for MTI than ED. VII. CONCLUSION

There are at least two distinct kinds of comparative philosophy. On the one hand, there is comparative philosophy that aims to compare two separate traditions and debate which traditions can lay claim to the ownership of a philosophical idea. For example, a comparative question of this kind is: is the kind of pragmatism found in the work of William James and John Dewey only to be found in the West as a product of prior thought on European philosophy, or are its basic principles also found in the East, in Chinese, Japanese, or Indian philosophy? On the other hand, there is comparative philosophy that seeks constructive engagement for the purposes of continued theorizing on a philosophical issue. For example, a comparative question of this kind is: what can contemporary projects in logical theory, such as work on logical pluralism, learn from an examination of the Jain theory of sevenfold predication? In contrasting these two kinds of comparative philosophy, one should legitimately ask what kind of comparative exploration has been presented here. So far I have aimed to establish a comparative point of the first kind. That is, I have aimed to show that a careful understanding of the difference between perceptual anti-­individualism and epistemic disjunctivism should lead us toward the view that Nyāya perceptual theory can be understood as a theory that is amenable to PAI rather than ED. I have tried to show that the evidence does not clearly lean toward ED, and that there is consistency between MTI and PAI. However, I have not argued what the wider significance would be for contemporary epistemology and perceptual theory, were this thesis correct. Moreover, one might ask: what insight for epistemology and perceptual theorizing can we gain through a comparative analysis of contemporary epistemic disjunctivism, perceptual anti-­individualism, and Nyāya parasitism? I believe that there are several advantages that contemporary epistemology and perceptual theory can gain from a comparative investigation of Nyāya epistemology. First, there is the methodological point concerning (1) the relation between a thesis and the frame of inquiry from which the thesis is advanced, and (2) what effect a frame of inquiry has on the development of a specific thesis.The fact that contemporary epistemic disjunctivism, at least in the work of McDowell, finds its roots in a reaction to ­Cartesian skepticism makes it the case that it has a quite different frame of inquiry from the theorizing that goes into the work of Nyāya epistemology. The latter’s advancement of a metaphysical distinction between veridical and non-­veridical states does not depart from a concern for refuting skepticism of the Cartesian kind. In this regard Dasti is correct to

274

ANAND JAYPRAKASH VAIDYA

point out that Nyāya epistemology departs from a default position of trust as opposed to Cartesian doubt.The departure point changes how we should understand the import and consequences of the form of the perceptual theory rendered. Second, while it is true that in contemporary analytic philosophy of mind there are perceptual theories of emotions and discussions of the epistemic role of emotions, there appears to be little or no discussion in analytic epistemology over the role of emotions in perception.10 That is, no philosophical discussion of how the emotions one is having at a certain time can affect how one sees objects in their environment. Neither McDowell nor Burge engage in how the emotions may play a substantive role in how our misperceptions come about. By contrast, the Nyāya MTI account highlights this factor as one account of how misperception can come about. By providing such an account, they offer several questions for epistemic and perceptual theorizing, such as: (1) which emotions are likely to cause misperceptions and (2) how exactly do emotions cause misperceptions? While it may be true that Nyāya epistemology is not the only tradition to investigate the relation between emotion and perception, their discussion of it along with other traditions presents a new opportunity for furthering research in epistemology and perceptual theory. Third, much of the contemporary debate concerning skepticism in analytic philosophy concerns either Cartesian skepticism or discussions of Putnam’s brain-­in-­a-vat hypothesis. In both of these cases we are concerned with a totalizing form of skepticism. In the former, the strong form comes about because an evil demon is hypothesized to be producing our perceptions; in the latter the strong form comes about because a mad scientist has plugged our brains up to a computer. Both of these forms of skepticism depart from an unnatural source. While both are logical possibilities that the strongest form of anti-­skepticism must respond to, neither are natural possibilities that a weaker form of anti-­skepticism would respond to. In terms of providing a naturalized account of how to explain misperception, the Nyāya MTI account focuses on a feature of misperception that, although non-­scientifically presented, is nevertheless insightful. The account focuses on the way in which some misperceptions must be a function of a generalized form for misperception: x can be misperceived for y by agent A because x has some properties objectively in common with y that, along with the emotions and concepts that A has, provide for an account of a natural disposition on A’s part to misperceive x for y in certain circumstances. Hopefully, future work in epistemology will aim to engage in comparative work of the constructive kind attempted here with the aim of enriching research and widening the pool of sources from which a philosophical theory can be constructed. NOTES I would like to thank Purushottama Bilimoria, Karin Brown, Manjula Rajan, Krupa Patel, and the students in my Spring 2012 Philosophy of Mind Seminar for inspiration, discussion, and criticism of this work.The failures that remain are due to my own inability to understand the intricate details of the theories involved. My hope, of course, is eventually to understand them. This essay is an extended version and treatment of an argument that I discuss in my much longer treatment of perception, “Perception and its Content: An Examination of

NYĀYA PERCEPTUAL THEORY

275

Disjunctivism and Conceptualism about Perception from the Perspectives of Phenomenology,Vision Science, NYĀYA, and Buddhist Epistemology.” 1. It is important to note that even in cases where a subject may lack robust concepts, such as cow and zebra, it is possible for the misperception of a cow for a zebra to be voiced in terms of demonstrative concepts. For example, one might on approach think, “that object looks like that,” where the first demonstrative picks out a cow and the second a zebra, in a pasture where both zebras and cows are present and the statement is understood to be a comparative judgment along a set of properties. However, on arrival, one might recoil from one’s initial judgment and think, “That object does not look like that,” recognizing that from a distance one had a misperception of similarity and that the two animals in one’s perceptual field are not similar because they are not both cows or both zebras. 2. While it is true that many philosophers have taken some kind of disjunctivist turn in recent Western epistemology, it would be incorrect to note this without pointing out that there are a great number of philosophers that have not. 3. In Anglo-­American philosophy the term ‘asymmetric dependence’ is most often associated with Jerry Fodor’s work in semantics (1987). While there are some components in common between that usage and the use I make here of the term, I want to note that Fodor’s account is not the one that is under discussion here. Rather, the simple idea that getting things right is prior to getting them wrong is what is of central importance to the use of ‘asymmetric dependence’ in this essay. 4. It is important to note that there are issues surrounding the relation between concept possession and the complete and incomplete understanding of a concept that are relevant to the issue of asymmetric dependence that go beyond the scope of this essay. For example, it appears to be possible for one to incompletely understand a concept one possesses. One can possess the concept of a right triangle without grasping that the Pythagorean Theorem holds. However, it appears impossible to possess the concept of a right triangle and deny that it is a three-­sided closed-­plane figure. Thus, the following question arises: which judgments about a given concept are necessary so that one can be said to possess the concept so as to be able to misapply it in a given case? 5. It is important to recognize that Burge and Putnam differ in their understanding of what exactly follows from the Twin Earth thought experiment. Anti-­individualism is not the same as semantic externalism. At least one key difference between the two views is concerned with the relation between sense and reference. It is a bit odd to present Burge’s Anti-­individualism through the use of Putnam’s Twin Earth case; however, I do so because of the popularity of the example and the fact that it can be used to establish the basic point that factors outside an individual’s internal psychology are relevant for individuation of mental content. 6. There is another way to separate the difference between perceptual anti-­individualism and epistemic disjunctivism. The difference comes in looking at an analogy between chemical kinds and epistemic kinds. In the basic case, because they are both externalist sorts of theories perhaps PAI and ED agree over the fact that chemical kinds are individuated by factors outside the individual. So, for example, because chemical theory states that chemical composition is essential to the categorization of chemical kinds, and XYZ and H2O are different chemical compounds, XYZ and H2O are distinct no matter what superficial similarities they possesses. However, in the case of perception, ED maintains that since veridical states are true and non-­veridical states are false, no matter what phenomenological differences there are in common between the two states, they are epistemically distinct. PAI, in contrast to ED, maintains that while it is impossible for concept possession to occur without veridicality, it is possible for two perceptual states to be identical even though one is veridical and the

276

ANAND JAYPRAKASH VAIDYA

other is not. Moreover, PAI denies that perceptual theory will maintain that truth and falsity are the relevant individuating factors for perceptual states in the same way that chemical theory maintains that chemical composition is the relevant factor for categorizing chemical compounds. In looking at Burge’s example of viewing two distinct but phenomenologically similar objects over time—dime1 and dime2—the difference is apparent. Disjunctivism maintains that the two perceptual states are distinct perceptual states because their veridicality conditions are distinct. PAI maintains that the two perceptual states are similar even though they have distinct veridicality conditions. 7. It is important to recognize that the definition given by Gautama and discussed by Matilal appears as if it supports disjunctivism. The mere fact that perception, in the definition given, has three necessary conditions that separate seeing from perceiving allows for the possibility that one could be in a state phenomenologically similar to a perceptual state, yet not be enjoying a perception.This reading gives the disjunctivist position an initial positive grounding. For example, one could argue that in each of the cases Matilal’s explanation shows that a state that is phenomenologically similar to a genuine perceptual state is not a perceptual state because some factor of relevance is missing. But given that the factors are not phenomenological, phenomenological similarity is not sufficient for a state to be a perceptual state.While the argument is clearly available, the counter-­considerations that I discuss offer another way of interpreting the overall perceptual theory offered. 8. For extensive discussion of this issue see Burge 2011. 9. It is important to note that the claim being made here is that both PAI and MTI look carefully at the causal role of the environment and the make-­up of the subject in the production of a perceptual state. What is not being claimed is that both accounts offer the same causal story. It may very well be the case that PAI and MTI disagree on the correct causal story, and even that MTI is incorrect from a scientific standpoint over what the causal story is. However, MTI is closer to PAI than ED because it looks at the causal story in rendering an account of misperception and perception, just as PAI does. ED is in essence a denial of HCF and does not look at causation in explaining the difference between veridical and non-­ veridical states; it simply looks at the veridicality conditions. 10. While a brief survey of the literature shows no serious investigation of emotional states and perception in analytic epistemology, I doubt that the same would be true of European philosophy. I am inclined to think that European philosophers have discussed in some depth the relation between emotional states and perceptual states. REFERENCES Burge,Tyler. 1979. “Individualism and the Mental.” In Peter A. French,Theodore E. Uehling, and Howard K.Wettstein, eds., Midwest Studies in Philosophy (Minneapolis: Minnesota University Press) 4, no. 1: 73–121. ———. 1986. “Individualism and Psychology.” Philosophical Review 95: 3–45. ———. 2005. “Disjunctivism and Perceptual Psychology.” Philosophical Topics 33, no. 1: 1–78. ———. 2011. “Disjunctivism Again.” Philosophical Explorations 13, no. 3: 43–80. Dasti, Matthew R. 2012. “Parasitism and Disjunctivism in Nyāya Epistemology.” Philosophy East and West 62, no. 1: 1–15. Fodor, Jerry. 1987. Psychosemantics: The Problem of Meaning in the Philosophy of Mind. Cambridge, MA: MIT/Bradford. Matilal, Bimal Krishna. 1986. Perception: An Essay on Classical Indian Theories of Knowledge. Oxford: ­Clarendon Press. McDowell, John. 1996. Mind and World. Cambridge, MA: Harvard University Press.

NYĀYA PERCEPTUAL THEORY

277

———. 2008. “The Disjunctive Conception of Experience as Material for a Transcendental Argument.” In Adrian Haddock and Fiona Macpherson, eds., Disjunctivism: Perception, Action, Knowledge, pp. 376–389. Oxford: Oxford University Press. ———. 2009. “Criteria, Defeasibility, and Knowledge.” In Alex Byrne and Heather Logue, eds., Disjunctivism: Contemporary Readings, pp. 75–91. Cambridge, MA: MIT Press. ———. 2011. “Tyler Burge on Disjunctivism.” Philosophical Explorations 13, no. 3: 243–255. Phillips, Stephen. 2004. “Perceiving Particulars Blindly: Remarks on a Nyāya-­Buddhist Controversy.” Philosophy East and West 54, no. 3: 389–403. Putnam, Hilary. 1973. “Meaning and Reference.” Journal of Philosophy 70, no. 19: 699–711.

P A R T

VII LANGUAGE

CHAPTER

16

The context principle and some Indian controversies over meaning* B. K. Matilal and P. K. Sen

Source: Mind, 1988, 97:385, 73–97. I

Recent discussions show that there are several ways to flesh out and interpret Frege’s remarks on the ‘context principle’. Besides, we believe that the issue is only the tip of an iceberg. Whichever way one wishes to resolve the problem and interpret the principle, it has some important consequences for our epistemology and ontology. It has even been suggested that the controversy between the ‘context’ and the ‘composition’ principle may be broadened to set the background for learning philosophical lessons about the controversy between holism and atomism, even between realism and relativism. But there was also an ‘Indian version’ of the problem discussed and debated extensively by the exponents of the two sub-­schools of classical Indian Mīmāṃsā. In his Two Dogmas of Empiricism, W.V. Quine insisted upon the primacy of the sentence over word or term as the vehicle of meaning and attributed this doctrine to Frege’s context principle in Grundlagen 60. It is significant to note that Quine in his Word and Object referred to the view of ‘Indian Grammarians’ in support of his view. In fact, it was the view of Bhartṛhari, a sixth-­century ad grammarian of India, about whom J. Brough wrote an article in the Transactions of Philological Society, Oxford, 1951. Quine cited the article by Brough in his footnote. Quine’s holistic interpretation of the Fregean doctrine was based upon a mistake (as Michael Dummett and others have argued), but Quine remained faithful to his interpretation and described it as an ‘important re-­orientation in semantics’ over the old term by term empiricism of Locke and Hume. Even in his latest book, Theories and Things, Quine labels this semantic primacy of sentence as one of the ‘Five milestones of Empiricism’, which gave us contextual definitions. Quine said that this is like the 281

282

B. K. MATILAL AND P. K. SEN

Copernican revolution in Astronomy, in being a shift of centre (p. 69). It made obvious that the words or terms are only like ‘grammatical particles’ for the meanings of words are abstractions from the truth-­conditions of the sentence that contains them. Whatever might be the correct interpretation of Frege’s ‘context principle’, it is widely believed that he rejected it in his later writing implicitly. H. Sluga1 has argued that Frege retained the principle in later writings, while M. Resnik2 has said that he rejected it. Dummett has claimed that Frege retained a weaker version of it implicitly in his later writings. The modern exegetical literature on Frege is vast and varied. Recently Peter Milne3 has pointed out further difficulties that one would face if one takes Frege to have understood the context principle as a thesis about reference along the lines advocated by Dummett and C. Wright. In this paper we propose first to review very briefly the exegetical literature on the interpretation of the context principle, and then we will introduce the Indian formulations of what appears at least to be a similar problem. Having first explicitly acknowledged that the types of questions which led to the controversy between the ‘context’ and the ‘composition’ principles in the exegesis on Frege were not exactly the same as led to the classical Indian philosophical debate over the anvitābhidhāna ‘connected designation’, and the abhihitānvaya ‘designation before connection’ theory,4 we aim to show also that the two controversies were not totally irrelevant to each other from the point of view of the philosophy of language. The purpose here is to bring out the points of contact and parallels between the discussion of Frege’s context principle on the one hand and the long-­lasting Indian debate on the other. It is believed that such endeavour may throw further light on the thorny issue and hence will be philosophically fruitful. As a very modest beginning, we may formulate three very simple questions, to show the divergence of the ways the two controversies, one in the modern West and the other in classical India, developed. i Does a word have a meaning only in the context of a sentence? ii Is it possible to know what a word means only by considering sentential contexts in which it occurs? iii Given that we know the meanings of words constituting a sentence, how do we come to know, understand, the meaning of the sentence as a connected whole (of meaning)? It seems that the modern Fregean controversy is connected with (i) and perhaps also indirectly with (ii), while the classical Indian controversy is directly concerned with (iii). We would like to suggest further that the Indian philosophers were also indirectly concerned with (i), and perhaps also with (ii). That it was so is suggested by our study of Bhartṛhari who criticized and rejected both sides of the (Indian) controversy between the ‘connected designation’ and the ‘designation-­before-­connection’ theories. Bhartṛhari put forward a third alternative, which we may call ‘sentence-­holism’. He suggested a strong affirmative answer to question (i) and accepted its extreme consequence, namely, words have no meaning outside, or isolated from, the context of a sentence just as ‘rat’ in the word ‘Socrates’ has no meaning.This would show at least that Quine’s (mis-­)interpretation of the Fregean doctrine has a philosophical basis and on similar considerations Bhartṛhari

283

THE CONTEXT PRINCIPLE

was led to his sentence holism. These three questions are however not unrelated. In order to answer question (iii) satisfactorily we need to know whether the words have meanings in isolation, independently of the context of the sentence in which they occur. II

If we give a strong affirmative answer to question (i) above, we get a strong interpretation of the doctrine. Here the principle is to be understood in close connection with the idea of a contextual definition. An affirmative answer to question (ii) leads to the weak interpretation. It is understood as simply a methodological policy: we should ask what a particular sub-­sentential expression means only in the context of a sentence. It seems that some of Frege’s remarks favoured the strong, some the weak interpretation. There is however an intermediate interpretation. These three interpretations differ in the following manner. In the strong interpretation, words and other sub-­sentential expressions have no meaning of their own. In the weak interpretation, these sub-­sentential expressions are not denied meanings of their own. In the intermediate interpretation, the meaning which the sub-­sentential expressions have is only the contribution which they make to the meaning of the sentence in which they occur. The strength of this third interpretation lies in the fact that unlike the other two interpretations this one presents the principle itself in a favourable light. Textual evidence here is indirect (see Appendix for different formulations of the principle by Frege). What is distinctive in the strong interpretation of the principle is that it invites us to understand the principle as saying that every expression which is short of a sentence must be defined contextually. But it is also this connection with the contextual definition which makes the principle, on this interpretation, unacceptable. Let us first be clear about the exact significance of a contextual definition.Take a classic example of the use of the contextual definition: Russell’s theory of (definite) descriptions.5 Russell’s definition of (definite) descriptions, given in his famous theory, is not actually designed to say what such an expression means; it is rather designed to show that we need not suppose that a descriptive phrase has any meaning of its own; it is designed to show that it is not a self-­subsistent semantical unit at all, for the analysis the theory offers shows how such a phrase can be eliminated altogether from a sentence. If it is said that every definition, not only the contextual is a device for eliminating the phrase which is defined, we will show the basic difference between two sorts of eliminating devices— one, in the case of non-­contextual, that is explicit, definition, and the other, in the case of the contextual. The contextual definition, unlike the explicit definition, does not actually say what an expression means. A definition can be regarded as saying what an expression means, rather than being a mere device for eliminating the expression in question if and only if it can be construed as an answer to the question ‘What does a mean?’ where a is any expression. It can be stated in the form a means b. But this is the form of an explicit definition.What then does the contextual definition do? If we want to maintain, what we have just said against the supposition notwithstanding,

284

B. K. MATILAL AND P. K. SEN

that it also gives the meaning of the expression, how does it do so? It seems that it can give the meaning, if at all, only in the following way: a means that by virtue of which S(a) means that p, where S(a) is a sentence containing the expression a and ‘p’ is replaced by some sentence giving the meaning of the sentence S(a) but not containing the expression a. But this does not really tell us what that meaning is by virtue of which this semantical equivalence between the two sentences holds. We can indeed put the point in the form of a paradox. The contextual definition tells about the meaning x of the expression a that it is the x such that it makes S(a) mean the same as ⎡p⎤. But according to Russell’s own principle—a principle regarding what he calls knowledge by description—we can know this without knowing what this x is. What has been said above shows, we think, that, treated as an account of the meaning of individual expressions, the contextual definition is very poor. So we should not perhaps treat the contextual definition as attempting that. What then does a contextual definition do? A very plausible answer is that it shows that we should not ask the question: what does the expression mean? (Just as we should not ask the questions: what does ‘sake’ mean in ‘for the sake of ’? We should ask: what does ‘for the sake of ’ mean? or, perhaps, what does ‘Philip gave his life for the sake of his country’ mean?) Our opposition to the idea that the contextual definition can be treated as a way of giving the meaning of a word and other sub-­sentential expressions can be misunderstood. So let us be quite clear about the nature of this opposition. Our main objection to this idea is that the contextual definition does not tell us what the expression means. It tells us at most that the expression means something by virtue of which a certain equivalence, namely the equivalence between the definiendum and the definiens, obtains, without telling us what that something is. Note that this is not an objection to the device called ‘contextual definition’ as such, but to the view that by this device we give the meaning of the expression defined. This objection need not be withdrawn in the face of standard contextual definition of a sentential connective, for example ‘&’, wherein truth-­conditions of sentences, formed by the help of such a connective, are specified. Here too, the only thing that the definition does by way of giving the meaning of the sentential connective is that it tells us that the connective means something by virtue of which sentences formed by their help have the truth-­conditions they have. In fact, in the case of a sentential connective, more than in any other case, it is clear that the contextual definition need not be taken to give the meaning of a connective; it can be taken to do just what it does, namely, give the truth-­conditions of sentences formed by using the connective. For the purposes of logic, after all, nothing else is necessary. (What does it matter if we do not know what the connective means, once we know the relevant truth-­ conditions? For if we know what the truth-­conditions of sentences formed by using the sentential connectives are, we can define validity of formulas containing them, the relation of logical consequence among such formulas, and all other similar concepts which have any importance for logic.)

THE CONTEXT PRINCIPLE

285

One may ask, at this point, ‘What is meaning but truth-­conditions?’ But we can answer this question without entering into the larger issue of whether we can indeed identify meaning with truth-­conditions. The answer is that even if meaning is nothing but the truth-­conditions, it is only the meaning of the sentence which is constituted by truth-­ conditions, and not the meaning of the sentential connective which the sentence contains. Truth-­conditions can constitute the meaning at most of what they are truth-­conditions of; and that is the sentence and not the sentential connective. Now, one may still say that the context principle should be understood in any case in terms of the contextual definition, which can then be understood in the manner just suggested. But even this would not really be plausible at all. For, to make use of the idea of contextual definition in our account of the context principle, we shall have to make a complete generalization first. That is, we shall have to say that all expressions which fall short of complete sentences must be defined contextually, for the context principle is a principle which is valid for all such expressions. But to make this generalization is to obliterate a distinction on which the very plausibility of the idea of a contextual definition rests. It has never been a part of the thesis that all terms can be defined only contextually, but only that some terms have to be so defined. In fact, the idea of contextual definition thrives on the distinction which Russell had in his mind, namely the distinction between complete and incomplete expressions. It is not the case that all expressions are incomplete without exception; only some expressions are so. It is only in the case of incomplete expressions that the contextual definition would be appropriate; it would not be appropriate in the case of complete expressions. There is indeed some intrinsic difficulty in the idea that the contextual definition be applied to all terms, for the very idea behind the kind of definition Russell has offered is that the descriptive phrase, for which the definition is paradigmatically apt, can be eliminated altogether from our language, and in favour of the expressions which are not incomplete: we need not have these incomplete expressions in our language, we can make do with the complete expressions alone. There is yet another reason why we should not try to understand the context principle in terms of the contextual definition. If we take the strong interpretation of the principle which this would force upon us, the context principle would be utterly incompatible with the composition principle. If the individual expressions—words and other sub-­sentential expressions—do not have any meaning of their own, there cannot be any question of deciphering the meaning of a whole sentence by (a kind of) step-­by-­step construction from the meanings of words. But Frege is not prepared to give up the composition principle, as Dummett so rightly emphasizes (see below). One may say that our objections to the interpretation of the context principle in terms of the idea of contextual definition are based on a questionable assumption. It is the assumption that contextual definition was understood by Frege in the way in which it was by Russell. But, it may be said, this assumption itself is wrong; and that it is wrong is shown by the fact that, while for Russell the contextual definition is a device for ­elimination (and, therefore, for a kind of reduction), for Frege it is not so. In fact, the possibility of giving a contextual definition for a singular term occurring in a sentential context shows, for Frege, that the singular term does have reference after all.6 By contrast, Russell gave his definition of definite descriptions—for example, of ‘the present king of France’—just to show that the definite description does not have any reference.

286

B. K. MATILAL AND P. K. SEN

It is true that Frege’s contextual definition is different from Russell’s in this respect, and that we have not so far laid any emphasis on this fact. (This is on the assumption that we are not making any mistake by ascribing to Frege the use of the idea of contextual definition.) But our critique of the interpretation of the context principle in terms of contextual definition cannot be faulted on that ground.The critique remains valid even if the difference just noted between the two conceptions of contextual definition exists and is otherwise important. On the Fregean conception, the contextual definition would indeed tell us that an expression, typically a singular term, has reference, and, for that matter, sense as well. But telling that an expression has reference (or sense) is certainly different from telling what that reference (or sense) is. If it is said that the reference, and consequently the sense, are given by the expression itself (as Dummett has suggested)—the singular term ‘the morning star’ gives both the planet and a mode of presentation of the planet, say—then what does the definition do? The definition itself sets up equivalence between sentences; but it does so without specifying the ground of this equivalence: the meaning, that is, the sense or the reference. The weak interpretation of the context principle has the merit of being quite uncontroversial. Who will deny that the best way, and, in the case of some expressions, the only way, in which we can ascertain the meaning of sub-­sentential expressions is by observing how they are used in sentences to say something? Consider how the archaeologists have done their job of deciphering inscriptions in long lost and dead languages. Their only method has been that of observing patiently, sometimes over the years, how the same character occurs in different constructions. But since the context principle, on this interpretation, becomes so unexceptionable, we may doubt whether this is all that Frege had in his mind. One of the things which Frege certainly meant us to take seriously—this becomes evident when we consider his actual use of the principle in the development of his theory of number—is that it does not matter whether we are able to identify in each case something to be the meaning of a given sub-­sentential expression and form a self-­subsistent idea to represent this meaning; it is sufficient if we can say what the sentence in which it occurs means as a whole, and tell how the expression in question is used in it. So what Frege had in his mind when he advocated the context principle is not just a matter of discovering the meaning of an individual expression, it is also something about the individual expression’s having the meaning it has.The weak interpretation therefore, is too weak to bring out the full significance of the Fregean principle. Thus, it seems that we shall have to reject both the strong interpretation given in terms of the idea of contextual definition and the weak interpretation which treats the principle just as methodological advice. We are then left with what we have called the intermediate interpretation. To a consideration of this we may turn now. III

What the intermediate interpretation tells us is that the meaning of a sub-­sentential expression is nothing but its contribution to the meaning of the sentence in which it occurs. Dummett, who apparently takes the context principle in this form, also gives what seems to us to be the best possible defence of the principle if it is so taken. The

287

THE CONTEXT PRINCIPLE

defence goes as follows: the main, if not the sole, function of language is that we use it for ­saying things, to perform linguistic (speech) acts. But we cannot say anything, in the strict sense of the word ‘say’, without the use of whole sentences. It is only by use of whole sentences that we can make ‘moves’ in language, to put it in Dummett’s words. It follows from this that the significance of sub-­sentential expressions also lies in that they enable us to say whatever we want to say. But how do these expressions help us in the matter of saying whatever we want to say? Surely by forming parts of sentences and contributing to the significance of sentences of which they form parts. Thus the meaning of the sub-­ sentential expressions, we can now say, consists in the contribution which they make to the meaning of the sentences. The exact significance of the context principle is not really so easy to bring out even in this intermediate or moderate form. In being clear about the meanings of the individual expressions, in terms of contributions made to the meanings of sentences, we shall have to keep in mind the distinction between sense and reference drawn by Frege, although this is a distinction which is absent from Frege’s early writings, including The Foundations of Arithmetic where the context principle is broached. The context principle says that the meaning of a sub-­sentential expression is nothing but what it contributes to the meaning of the sentence. But what is the meaning of a sentence? Granting that the meaning of the sentence is what Frege calls its ‘sense’, we shall have to say that the meaning of the sentence is its truth-­conditions (at least that is the usual understanding of the idea of sense in Frege). What then is the contribution of a sub-­sentential expression to the meaning of the sentences in which they occur? It would surely be the contribution which it makes to the determination of the truth-­value of these sentences. If we now ask the further question regarding what, or how, the individual expression contributes to the truth-­conditions of the sentences, say, of the atomic sentence ‘Socrates is wise’, we find that the name ‘Socrates’ contributes to its truth-­conditions by introducing the individual Socrates, and the predicate, ‘wise’ or ‘is wise’, or ‘(…) is wise’, or whatever it is—does it by introducing the concept wise. But we have to realize that the individual and the concept are both referents of the respective expressions, or what are called by Dummett their semantic values. But this realization would actually make us less confident about the correctness of this approach to the idea of the meaning of a sub-­sentential expression being what it contributes to the truth-­conditions; at the same time, it would also cast doubt on the idea that the meaning (or sense) of the sentence is nothing but its truth-­conditions. The general principle which seems simultaneously to undermine both is that it is the reference of the complex expression which is determined by the reference of the constituents. We can, of course, think of ways out of this difficulty, but that will have to be worked out only in the light of a more detailed consideration of the sense/reference distinction. IV

Some scholars writing on the context principle have suggested that in putting forward the context principle Frege was actually trying to find his own solution to what is basically a Kantian problem, namely, the problem of explaining the unity of thought. The unity of

288

B. K. MATILAL AND P. K. SEN

thought, for Kant, is a judgement. A judgement involves synthesis, and it is this synthesis which brings about the unity of the judgement, the fact that a judgement is one in spite of its essential complexity (which complexity is in the long run due to the complexity of the manifold of sense). The synthesis, bringing about the unity in a judgement, is guided by a rule (and it is this which makes the synthesis intelligent, as opposed to blind, and confers whatever objectivity a judgement has despite the fact that the synthesis, and to that extent the judgement itself, is a subjective act operating on a subjective manifold). This rule, for Kant, is always some concept. In fact two kinds of concepts are involved in every judgement, the concepts functioning as the predicate(s) and the concepts functioning as the categories of understanding. (Thus ‘Socrates is wise’ involves the concept wise as well as the concept (category) of substance-­and-­accidens.) So it is true to say that the unity of the judgement, and so of the thought, is explained by Kant in terms of the concepts. We may now identify the concepts to which this function is ascribed by Kant with the concepts, or, more generally, functions, Frege is talking about. If we do so, we can say that it is the same role which is ascribed to the concepts by both Kant and Frege; for certainly it is the concept (or function) which gives unity to the thoughts according to Frege: being the unsaturated component in our thought it is the concept (function) which makes its unity possible. It is a very good question to ask here whether this identification of the Kantian concept with the Fregean is at all correct. There are some obvious differences, for sure, b­ etween the two. Concepts, and functions generally, form an important item in Frege’s ontology, but they are not accorded any ontological status by Kant. Concepts confer unity to our thoughts, judgements in Kant’s view, because they are the rules of synthesis, which is a mental act; but there is no question of any synthesis being involved in bringing about the unity for Frege. In fact, the unity of thought is not a thing which is brought about by an act of the mind, in Frege’s theory, it is something which is out there, albeit in the third realm, independently of us and of our activities. Thoughts are objective entities, their unity is also something objective. One may, however, say that these differences do not really matter for the limited perspective from which the analogy is drawn. From that perspective, the only thing which is important is that concepts, for both philosophers, are what explain the unity of thought, whatever may be the details of the explanation offered by any of them. Let us grant this. But even so, we are not entitled to conclude that the context principle is put forward by Frege to explain the unity of thought. The unity of thought is explained by Frege by invoking the idea of an unsaturated part of the thought, which unites itself with the other part, without requiring a tertium quid. To say now that it is here that the context principle is brought into play is to make exactly the mistake of supposing that the context principle is to be understood in terms of the saturated/unsaturated distinction.7 V

A few words are in order here concerning the relationship of the context principle and what is now called ‘the composition principle’. The composition principle says that the meaning (sense) of a whole sentence is determined by (composed of) the meaning (sense) of the words out of which the sentence is composed. Apparently, at least, this principle is

THE CONTEXT PRINCIPLE

289

inconsistent with the context principle. For while the context principle seems to take the meaning (sense) of the sentence to be primary, the composition principle seems to take the meaning (sense) of the word to be primary. In his first book on Frege, Frege: Philosophy of Language, Michael Dummett discusses the question whether it is possible to reconcile the context principle with the composition principle, to both of which Frege seems to have been committed. His answer to the question is that they can be reconciled, and he suggests that they can be reconciled in the following way: ‘In the order of explanation the sense of a sentence is primary, but in the order of recognition the sense of a word is primary.’8 In so far as the knowledge of what a particular significant stretch of discourse means is concerned, knowledge of the sense of individual words must precede any knowledge of the sentence as a whole; the reason for this being that the sense of the sentence is itself determined, and determined in the very strong sense of being actually made up of, the senses of the words which constitute the sentence. (This is the Composition Principle.) Since the meaning of the sentence is what is determined by the meaning of the words, we cannot know what the meaning of the whole sentence is unless we know what the words which make up the sentence mean. We thus derive our knowledge of the sense of any given sentence from our previous knowledge of the senses of the words that compose it, together with our observation of the way in which they are combined in that sentence. (p. 4)

But, on the other hand, it is the sense of the sentence which must be regarded as primary in the order of explanation of the sense of any significant stretch of discourse; this explanation of the sense of the words, as well as sentences, being understood as a general explanation of what it is for sentences and words to have a sense, ‘that is, of what it is for us to grasp their sense’. Dummett continues, For Frege the sense of a word or of any expression not a sentence can be understood only as consisting in the contribution which it makes to determining the sense of any sentence in which it may occur. Since it is only by means of a sentence that we may perform a linguistic act—that we can say anything—the possession of sense by a word or complex expression short of a sentence cannot consist in anything else but its being governed by a general rule which partially specifies the sense of sentences containing it.

Dummett is aware that this way of understanding the relationship of the sense of sentences and the sense of words can be defended only if we could give an account of what it is for a sentence to have sense without any reference to the sentence being formed out of meaningful words. (That is, we should not say that for the sentence to have sense is to be composed of words which have sense.) So Dummett adds further: If this is so, then, on pain of circularity, the general notion of the sense possessed by a sentence must be capable of being explained without reference to the notion of the senses of constituent words or expressions. This is possible via the conception of truth-­conditions: to grasp the sense of a sentence is, in general, to know the conditions under which that sentence is true and the conditions under which it is false. (pp. 4–5).

290

B. K. MATILAL AND P. K. SEN

VI

We now propose to examine the issues before the Indian philosophers, issues which provided a context in which it has been thought that some sort of a ‘context’ principle was enunciated against an atomistic theory where the sentence-­meaning as a whole is supposed to be constituted by the atoms of word-­meaning. One issue was about the significant units of language, that is, the proper locus of meaning. (Frege might or might not have been concerned with this issue in Grundlagen.) The task before us now is to provide an exegesis on certain arguments of the Indian philosophers to illuminate the central point in the controversy. But at the very outset, one word of caution to avoid possible misunderstanding. In what follows we will not make the usual distinction between sense and reference. Although a somewhat similar distinction between the ‘mode of presentation’ and the ‘reference’ was known to the Naiyāyikas, for the present purpose we would use the ambiguous word ‘meaning’, and accept some version of the reference theory of meaning. In Vākyapadīya, ch. 2, Bhartṛhari notes that regarding the notion of the sentence and sentence-­meaning there are two principal philosophical theses: one is called the ‘indivisibility’ thesis (a-­khaṇḍa-­pakṣa) and the other is the ‘divisibility’ thesis (khaṇḍa-­pakṣa). The first thesis is what Bhartṛhari himself maintains while the second is held by his opponents, the Mīmāṃsakas. For our purpose we propose to call the first ‘sentence-­holism’ and the second ‘atomism’. They had two main questions before them: what is a sentence, and what constitutes the sentence-­meaning? More specifically, how is a sentence constituted, and how is the meaning of a whole sentence cognized by the hearer after the utterance is made? According to sentence-­holism, sentences are wholes and they are the unanalysable units of meaningful discourse. Similarly, the meanings of sentences themselves are wholes. In fact they are also timeless. For destruction is usually believed to be dissolution into parts. We reach words as parts of the sentence, and word-­meanings as parts of the sentence-­ meaning through ‘analysis, synthesis and abstraction’ (a method that is called apoddhāra). This method is only instrumental in facilitating our language-­learning, a convenient way of making explicit our implicit linguistic competence. The words are no less abstractions than the letters are. The meaning of a word in isolation is an imaginary construct. In fact words are as much devoid of meaning as the letters or some syllables in a word, like ‘rat’ in ‘Socrates’. The meaning of a complete sentence is given to us as a whole block of reality. We chip this whole and correlate such abstracted (extracted) bits and pieces of meaning with words and particles which are also reached by such a process of breaking apart the whole sentence. On this theory, a sentence cannot be a composite entity with words as constituent elements, and the meaning of a sentence likewise cannot be given by the allocation or computation of word-­meanings individually considered. This view is very similar to that of W.V. Quine’s, as noted already, and hence Quine’s reference to it was justified. The whole meaning though expressed by a sentence can share a common structure, and have common ‘parts’ but such parts would not be capable of existing in isolation from the rest. In this sense, they could be just our own ‘abstractions’. A weaker implication may be that in ontological terms, the wholes (whether sentences or other wholes) may have parts but such parts lose their significance (perhaps ontic significance) as

291

THE CONTEXT PRINCIPLE

soon as they lose their contextuality in the whole. The opponent would have to say that there may be wholes which have parts but the latter will not lose any ontic significance if they lose contextuality. This holistic solution of Bhartṛhari was seriously challenged by the Mīmāṃsakas in the tradition. VII

We shall leave aside Bhartṛhari and discuss the so-­called atomistic views of the two Mīmāṃsā sub-­schools.9 Both atomistic views recognized that the sentence is a composite entity composed of elements which we call words, particles, etc. These elements are meaningful units of expression. The sentence-­meaning must be connected with these units. The hearer grasps the meaning of a sentence or what is spoken, provided he has what we may call linguistic competence, that is, knowledge of the meanings of words and particles as well as of how that particular language works. On this view, it will be unreasonable to take sentences as the smallest meaningful units, for sentences are virtually countless and we certainly cannot learn a language by learning those countless sentences and their meanings. It is only by learning a few (a finite number of) words and seeing how that language works that we gain the linguistic competence described above. What has been stated in the last paragraph is commonly held by the two sub-­schools of Mīmāṃsā, the Bhāṭṭa and the Prābhākara. They both reject Bhartṛhari’s view. But internal differences between these two sub-­schools led to great controversies for several centuries. This becomes clear as soon as we ask the following question: how does the competent hearer recognize sentence-­meaning as a whole from hearing, in bits and pieces, the constituent words in sequence? Does he first cognize or recognize the meaning of each constituent word and then join these bits and pieces of meaning together to cognize a connected whole—the sentence-­meaning? If our answer is yes, then we are talking from the Bhāṭṭa point of view. Designation by words first, then the designata are connected to form a unity (abhihitānvaya). Alternatively, the Prābhākara says that a person recognizes the meaning of the whole sentence by hearing simply the constituent words put together syntactically: ‘connected designation (by any word)’ (anvitābhidhāna). On the first (Bhāṭṭa) view, the hearer recognizes the meaning of the whole sentence by figuring out first the meanings of individual words whereas on the second (­Prābhākara) view, he recognizes the meaning of the sentences directly from the words themselves skipping the intermediate step of grasping singly the individual word-­meanings. Phrased in this way the distinction may seem trivial, but it is not really so. A little reflection shows that on the first view meanings of words are assumed to be independent units, as complete objects. In recognizing the meaning of a sentence (i.e., interpreting a sentence made of several words), we as hearers must first obtain these self-­subsistent building blocks (meanings) and then cement these blocks to obtain the connected meaning of the sentence. This implies that the distinction between word-­meanings and sentence-­meaning is one of building blocks and the building itself. Here we move close to the intuition which prompted the modern ‘composition’ principle. Notice that the words cannot have meanings only in the context of a sentence on this view, and hence it does not seem compatible with what we have called the strong interpretation of the context principle.

292

B. K. MATILAL AND P. K. SEN

Further the word ‘directly’ in the second view, that we recognize or obtain the sentence-­meaning directly from the words themselves, means that there is no intervening event such as that of our getting hold first of the so-­called word-­meanings as building blocks, ­between our knowledge of the words (through hearing) and our knowledge of the meaning of the sentences made of such words.This has the implication that the meanings of the words are not, in some sense, context-­free, independent objects. Whatever a word designates, it is always related or connected (anvita) with the designation of other words in the sentence. Notice that it comes very close to saying that a word gains its proper meaning only in a context, that is, in the context of a sentence. In fact, the second view expressly advocates that we know or learn the meaning of a word only by considering the sentential context in which it occurs. Apart from this, we must note that the main point in the dispute is epistemological. The question is how do we come to know, as we invariably do, the complete and connected meaning of the sentence simply on the basis of our knowledge of the constituent elements, the words? It is clear that we derive our knowledge of such distinct (constituent) elements through hearing (or seeing) the words as well as their interconnections. (It should be noted that knowledge of meaning of the sentence is what is aimed at, not mere understanding of its meaning, but it would be beyond the scope of this paper to argue this point.) VIII

Both sides in the dispute appeal to a general theory about language acquisition. We learn a language invariably by acquiring knowledge of word-­meanings given in the context of sentences whose meanings are also known or given. A child learns his language in this way by observing the linguistic and other behaviours of the adults. The older adult (uttamavṛddha) commands something to the younger adult (madhyamavṛddha) who acts to obey. ‘Bring a horse’, and a horse is brought. ‘Bring a dog and tie the horse’, and so it is done. That is how our acquisition of the meanings of individual words is explained. The Bhāṭṭa says here that there is thus the denotative power in the individual words, to give us isolated objects, actions, qualities, and relations (i.e. meanings). Hence given any newly formed sentence we can derive its meaning following the ‘expected’ syntactic pattern (ākāṃkṣā) by computing and manipulating such individual meanings to construct a whole. But the Prābhākara disagrees. He says that since individual word meanings are derived only in the context of some sentence or other and therefore from words already syntactically connected with other words, we learn such word-­meanings along with their possible (semantic) connections with other word-­meanings. The denotative power of a word gives us not simply the object, or action, or quality or relation but also each item’s possible connection with other items. Hence being presented with a sentence we do not waste time by first computing meanings from words and syntax and then manipulating such meanings into a whole, but straight away: we derive the connective meaning whole, objects with action, quality with the qualified, and a relation with a relatum. We shall first discuss one of the arguments usually given in support of the Bhāṭṭa view. Sentences are innumerable, but the word-­lexicon has a manageable size. The logic of parsimony demands that it is the word that should be endowed with the designative

THE CONTEXT PRINCIPLE

293

power (śakti), not the sentence. Consider the following four sentences: ‘Bring (a) cow’, ‘Bring (a) horse’, ‘Tie (a) cow’, ‘Tie (a) horse’. The child’s ability to learn the language is facilitated by learning the four words (Sanskrit does not use articles) and their corresponding meanings (real elements of the world) as opposed to learning the four sentences and their corresponding meanings. Add a word ‘black’ to each of the four sentences and see that by learning five words we can interpret eight more combinations. We can also better explain our ability to interpret new combinations which we have never heard before, such as new poetical compositions by a poet. Moreover, if there are several unfamiliar words in a sentence, we cannot cognize the sentential meaning. Using such and other arguments, the Bhāṭṭa repudiates Bhartṛhari’s sentence-­holism. Against the Prābhākara, however, the Bhāṭṭa argues as follows. If isolated, atomic word-­ meanings are like the discrete points of distinct iron pins (ayaḥśalākā), how could they constitute a continuous line representing the unity of the sentential meaning? For obviously the separate elements being independent of each other, cannot naturally merge into each other to form the continuous line. The Bhāṭṭa’s answer is that it is done through ākṣepa, that is, an extrapolative judgement (when word-­meanings are individually cognized), or a sort of suggestive inference (arthāpatti) on the part of the hearer. A few words to explain ākṣepa or arthāpatti may be in order. When what is presented seems incomplete to us we are forced to imagine some suitable additional (unrepresented) element for completion. This is called ākṣepa. By looking at a baby in a cradle one may imagine by ākṣepa that there is a mother who is around. Arthāpattii has a slightly different meaning. It is a proper inference from the given data. If I see that my desk is no longer in this room, I can easily infer that it has been removed (otherwise it would be impossible to explain such absence of the desk from the room under the circumstances). It is argued that our knowledge of the missing connectors between two word-­meanings is suggested by such ‘extrapolation’ or ‘inference’. Each of the words gives some independent object as its complete meaning and then since they are in a sentence together (āsatti) along with syntactic expectancy (ākāṃkṣā) and semantic fitness (yogyatā), we infer these appropriate connections to obtain the connected meaning of the sentence. Notice that the Bhāṭṭa plays down the logical role of ākāṃkṣā or syntactic expectancy. Or, he might have regarded it as simply a psychological factor. The Naiyāyika, who is the third party in this atomistic framework, emphasizes, however that the interconnection between word-­meanings is derived from the ‘syntax’, that is, from ākāṃkṣā, which is defined (non-­psychologistically) as the interconnectedness, order, etc., of the elements of representation. Here the basic issue seems to touch the well-­entrenched disputed area where we sometimes talk about interdependence or one way dependence between different elements or constituents (subject and predicate, proper name and general term, noun and verb) that seems to form a unity in a proposition.The Bhāṭṭa says that the designata of words are unrelated objects and hence to connect them we need a presumptive judgement (inspired by the psychological factor ‘expectancy’). On the second (Prābhākara) view the word’s designative power extends to a designatum plus possible linkages with others. The word means the related item. Here the same underplaying the importance of syntax is noticeable. The unity of the sentential meaning is guaranteed here by the semantic contribution (an object plus possible relation) of the words themselves. There is no need for additional suggestive

294

B. K. MATILAL AND P. K. SEN

inference or extrapolative judgement to cognize this unity. The Bhāṭṭa argues in reply that on the Prābhākara theory we cannot explain satisfactorily our prompt understanding of the meaning of many new sentences which we have never heard during our days of language-­learning.We may note that due to similar misgivings the ‘composition’ principle is thought to have some edge over the ‘context’ principle. IX

It has been suggested that the Prābhākara theory of ‘related designation’ is an extreme form of syncategorematism.10 The question is: does the strong context principle ‘words have meanings only in the context of a sentence’ necessarily entail syncategorematism? Usual examples of syncategorematic words are grammatical particles, adverbs, prepositions, etc., ‘sake’ in ‘for the sake of ’, to take a typical case. Using the older idea of a term, one can say that a syncategorematic word is one which cannot be used as a term independently but only in conjunction with another word or words. Quine11 extended the notion to include certain adjectives, ‘little elephant’, ‘little butterfly’, ‘poor violinist’, and ‘true artist’. In these cases, the qualifying words may have some independent meanings, but they are largely irrelevant, and from the meaning of combined phrases, meanings of such words are not (easily) separable. Recently Paul Gochet12 has argued that Quine on the whole prefers a syncategorematic treatment of predicates or general terms. And even if we deny this generally, we have to admit that some (at least one) predicates are syncategorematic (example ‘α’, & ‘=’). An extreme form of syncategorematism, it appears, would have to be a claim that all words are like this. On a milder interpretation, syncategorematism may be a vague way of underscoring the later Wittgensteinian claim that the meaning of a word is the use it has in language. Does the Prābhākara view come somewhat close to such a position? There is an obvious difficulty here. It is important to see the contrast. Wittgenstein’s motivation was to move away from the idea that our talk of the meaning of a word is a talk of the object it stands for, or the entity with which it is somehow correlated. The Prābhākara’s main concern was to account for how the constituent word-­meanings, if they are given in isolated forms, could be linked up, hooked up, with one another in order to form a unity. In fact, one way to describe the Prābhākara view is to say that for him such a word’s semantic value is ‘an object with a hook (to pick out another object)’ so that two or several of them in a sentence can naturally cling together to form a whole. Besides, if admission of syncategorematic words in language presupposes presence of categorematic words in combination with which they would form meaningful units, then this is not the view of the Prābhākara. For him, each word needs another to form a meaningful unit. In fact, this general point can be used as a criticism of the strong context principle, if it is construed wholly in the syncategorematic way. We must note that while both the Prābhākara view and at least some version of the context principle may tell us that the meaning of a word is the contribution it makes to the meaning of a sentence in which it occurs, the ‘context’ principle requires that a word unconnected with other words cannot have a meaning while the Prābhākara requires that a word cannot have such a meaning as is unconnected with the meanings of other words.

295

THE CONTEXT PRINCIPLE

X

We shall now present the argument of the Prābhākara in defence of his theory.The Bhāṭṭa argues that the Prābhākara by making all word-­meanings ‘context-­sensitive’ faces a dilemma. Consider a two-­word sentence; ‘XY’ meaning a connected unity. If we ask what meaning is conveyed by ‘X’ we have to answer that if it conveys any meaning at all then it conveys the unitary meaning of ‘XY’ itself. And the same is the case with ‘Y’. This is so because the Prābhākara has claimed that the meaning of the word of a sentence contains within itself, though implicitly, the whole sentence-­meaning, that is, the connected meaning. This seems to amount to sentence-­holism. But the Prābhākara maintains that sentences are made of parts which are words and if the meaning of one part contains the meaning of the whole, the other part becomes redundant. Hence the dilemma before the Prābhākara: either redundancy or sentence-­holism. The Prābhākara answers that the word ‘cow’ in the sentence ‘Bring a cow’ or ‘The cow is white’ designates a cow along with the possibility of its linkage with all other possible objects, or a cow with all the possible qualities, modalities, and actions, and the second element in the sentence is necessary only to help us determine which particular linkage to the exclusion of all other possible linkages is to be taken into account. But this is not enough. For one thing, if by the utterance of the word ‘cow’ one becomes aware of the cow linked up with all the possibilities (but no specific linkage is given) one is in effect aware of specifically nothing at all. It is an incomplete, and hence a very vague, awareness of meaning. (For a thirsty person a salty ocean is no better than a dry dreary desert. This analogy is from Jayanta. Knowledge of the object with all possible linkages is equal to no knowledge at all!) A dilemma arises again: (a) if the second word is necessary, is it so by its mere presence? If so, we again embrace sentence-­holism. (b) Is it necessary because it contributes its own meaning to the whole? If so, it resolves into the Bhāṭṭa view: words give their meanings first and then the sentential meaning is derived from them. To escape between the horns of this dilemma, the Prābhākara proceeds as follows: first he concedes that the expression ‘a cow’ means the object cow with infinite possibilities of linkage and the function of the other phrase ‘bring’ or ‘is white’ is to exclude other possibilities except the particularly intended one. The second phrase performs this function by its mere presence, not by contributing its own meaning. This is not sentence-­holism. For holism demands that the combination as a whole has the combined (whole) meaning where contributions of individual elements are not recognized at all, save through an artificial analysis. But the Prābhākara theory recognizes that the second element’s contribution lies only in excluding all other possible combinations save one that is intended by the sentence. Jayanta explains the point with the help of several analogies. Cooking is the result of many factors: burning of wood, a pan holding water, etc. They all individually contribute to the combined effect, cooking, by performing their own functions which can be individually recognized. Similarly the unitary knowledge of the meaning of a sentence is the result of the interrelated but separately recognizable functions of its constituent words. A wagon moves and each part is also moving. We can only recognize separately the mutually connected function of each part but all such parts jointly produce the motion

296

B. K. MATILAL AND P. K. SEN

of the wagon as a whole. The functions of such parts (or their motion) are not separable in reality. Similarly our knowledge of the meaning of a sentence is produced by the interrelated function of the constituent words. We can of course recognize the unconnected ‘own’ meaning of each constituent word just as we can observe the individual functions of all parts of the moving wagon, which cause it to move. But such functions do not have separate existence outside the context. We may be reminded here of an already noted comment by Dummett: in the order of recognition, the sense of the word is primary. But the point here straddles between both epistemic and ontological concern. Unconnected word-­meanings can be recognized but it would be wrong to construe them as separate entities. Each of them can play a role only in combination with others. The Bhāṭṭa makes a mistake of construing them as separate reals and identifies them with the (material and immaterial) objects such as a pot, blue colour, and action. More generally, combination of factors produces a combined effect and each factor in combination produces its own effect, which is discernible only in that combined effect. The designative power of a word becomes manifest only in combination with other words or only when it is placed in its natural home, a sentence (one word sentences being allowed). A word may remind us of an isolated independent object, but to contribute to the sentence-­meaning it must mean directly an object with a linkage. To put it bluntly, on the view we are considering, a word does not mean an isolated object, although it may remind us of such an object through associative psychological connection, while what it means is what it contributes to the sentence-­meaning, that is, an object with linkage with others. In Jayanta’s language:13 A word does not mean a pure object. For we cannot find pure (isolated) words which are not functioning in combination with others. These words are not employed separately to give their ‘own meanings’ and then the meaning they have in combination with others. They are always used to give their meaning in combination with others. But when they are used in this way it is not that we cannot recognize their own function (or own meaning). Therefore the sentence cannot be a partless whole (nor can its meaning be so either), for the individual functions of the parts are recognized.

Thus, the Prābhākara believes that perils of holism can be avoided and drift towards extreme atomism can be stopped. XI

Let us see whether we can tie up some of these different issues together. The context principle may be taken to be a very general thesis about meaning, and as a general thesis it would oppose what has been called epistemological atomism.14 This is the view that at least some objects are ‘given’ to us in sense perception or intuition—and hence our knowledge is in the first instance knowledge of isolated objects (and their properties). This view would then construct the meaning (sense or reference) of complex expressions (sentences etc.) in terms of those sense-­perceptible givens or the isolated objects. In Der Gedanke, Frege seems to have rejected this view impressionistically. Perception of objects, he said, involved grasping of thoughts. It is not to be confused with pure sensory reaction.

THE CONTEXT PRINCIPLE

297

Knowing is always knowing that.The context principle, viewed as a general thesis of meaning, would also oppose such atomism. This undoubtedly offers some insight into our discussion of the Prābhākara theory which opposes a similar sort of epistemological atomism of the Bhāṭṭa school. Kumārila Bhaṭṭa15 cited the case of a perceptual judgement constructed out of the bits and pieces of the sense-­g iven. We can take this to be another version of epistemological atomism, which is then extended to the philosophy of language to explain our knowledge of the connected meaning of sentences. Here is the example. Seeing a white flash moving swiftly and hearing the noise of the hoofs and neighing, one judges perceptually ‘a white horse runs’. Here the bits of the sense-­g iven are white flash, hoof-­noise, and neighing, but a judgement that connects these bits together is reached through the operation of the mind. Similarly there is the operation of the mind which connects the bits and pieces of isolated meanings, to obtain the connected sentential meaning. The Bhāṭṭa’s point seems to be this. The bits and pieces of the objective world, i.e. the isolated objects themselves, possess the capacity (power) in themselves to stimulate the observer enough first to grasp the isolated objects and then to formulate a judgement that connects them together. There are presumably three constituent items in the resulting judgement. They are separable as (1) the white flash presented visually, (2) the notion of horsehood presented by the instant inference from hearing of the neighing, and (3) the notion of running presented by the inference from the noise of the hoofs.These three are presented in three different ways (through three different avenues of knowledge, pramāṇas) and hence presumably they are grasped as unconnected bits of objects. Having been grasped they can by themselves evoke a judgement which unites them. This shows that a judgemental knowledge is possible simply on the basis of the presentation of the isolated object-­atoms themselves. Similarly let each word in the sentence present the individual unconnected meaning (objects, properties, actions, etc.).When such isolated meanings are grasped, there will automatically arise the judgement of the connected sentential meaning. The Prābhākara disagrees. The example, he says, is wrongly construed. There is no doubt that the three bits of object are separately presented in the given example, their sources being different in each case. But a connected judgement automatically arises in the person, as soon as he can locate all these three bits into one spatial location or in one particular substance, the horse. Śālikanātha following the Prābhākara argues that if the person is simply unaware or uncertain of the connectedness of the three bits, he would have three disconnected awarenesses: ‘There is a horse there, something is white, and somebody is running.’ But from the bits and pieces of the given, the required judgement, ‘A white horse is running’ arises since he can recognize both neighing and the action of running as belonging to the (same) substratum where the white flash belonged. Or, if he is unaware of the lack of connection of these objects he would have the required judgement. The judgement may finally be based upon knowledge of connected facts or, even lack of knowledge of the dis-­connection of isolated bits. In the case of a sentence, the words themselves as constituents provide, by way of presenting their meanings, such connected facts, but since such connected facts and the sentential meaning are not different from each other, we know the sentential meaning directly from the knowledge of the words and need not go through the collection of (word-­) meaning-­atoms.

298

B. K. MATILAL AND P. K. SEN

Sālikanātha was concerned with the epistemological question: how do we as hearers know the sentence-­meaning with our usual linguistic competence (vyutpatti)? But it is by no means clear whether an ontology of connected facts is conceded here by the ­Prābhākara. Perhaps not. What is asserted is rather that word-­meanings properly understood are connected facts, not isolated, unconnected bits of object. For otherwise it would be impossible to derive knowledge of the connected sentential meaning from unconnected bits.To imagine any additional device as the Bhāṭṭa does, for providing the required connection between isolated objects would be going against the principle of parsimony. The Prābhākara points out that the Bhāṭṭa may be violating the principle of parsimony in more than one way. The Bhāṭṭa imagines first that the words have one kind of dispositional property, that of being able to produce in the hearer the cognition of their ‘own’ (individual) meanings. Second the word-­meanings themselves (objects, properties, actions) should then have the disposition to generate the hearer’s cognition of the linkages among themselves. Moreover, the words must have another dispositional property for producing in the word-­meanings such a (second) dispositional property as would be capable of generating the cognition of the linkages. So the Bhāṭṭa view implies that there should be in all three dispositional properties, two in the word itself and one in the word-­ meaning, in order that we can account for the verbal cognition of the whole sentence meaning satisfactorily. If, however, we agree, along with the Prābhākara, that words themselves have a dispositional property—that of producing in the hearer a cognition of their ‘own’ meanings plus linkages—then we can practice the desired economy of dispositional properties (one instead of three). Besides, when we deal with language we can never find a word that is only a word being completely isolated from other words, for at the end some sort of word such as ‘is’ or ‘exists’ will be understood always when one word is uttered or heard. Hence a cognition of its meaning will necessarily bring in the linkage, its connection with the meaning of the other word through association. The dispute between the two groups rolled on for several centuries. Their arguments and counter-­arguments became increasingly subtle and technical. At some point, the exponent of the Prābhākara conceded that the isolated meaning of the word without the linkage can very well be recollected by the hearer as soon as the word is heard. And this facilitates our language learning procedure.The Prābhākara confirmed: the isolated meaning, the object cow, from the word ‘cow’, is quickly recollected because of intensity and recurrence. But this recollection simply facilitates our awareness of the proper meaning of the word in the context; we become aware of the object cow plus its possible relation, and the second awareness yields the knowledge of the connected sentence-­meaning.With this concession, it was claimed that it became a more defensible theory about meaning, which avoided the problem of sentence-­holism as well as that of ‘unconnectedness’ which the extreme forms of atomism might imply. If connected facts are not admitted as real entities ‘out there’ but at the same time it is claimed that words in a sentence designate connected items, that is, objects with linkage, and not objects as such, then this ‘designation’ relation of words is supposed to pick out what we may call epistemic objects, not the actual items or objects in the domain of the reals. It is our knowledge of such a ‘designation’ relation of words that gives us the knowledge of the sentence-­meaning as a whole. This seems to call for a tentative distinction to

299

THE CONTEXT PRINCIPLE

be made between the isolated objects, the ontological domain, and what we call a semantic domain which will include designata of words, such connected facts, the epistemic objects, that is, objects with linkage. We are not however sure whether the Prābhākara would be accepting this consequence, but this seems to follow from his view. Dummett repeatedly says that the context principle as applied to reference (as applied to sense, the principle seems to him unproblematic) created a tension in the kind of realism that informs Frege’s whole philosophy. The Prābhākara view is an attempt to avoid a crude theory of meaning, which demands that the meanings of our words be construed as independent and isolated pieces of reality. We have shown that the Prābhākara can avoid this construal and still maintain realism in his ontology by confining his doctrine of meaning to epistemological level; it becomes an epistemological thesis about the origin of our knowledge of sentence-­meaning. The moral seems to be this. If we flout the context principle, as well as the Prābhākara view of connected facts as word-­meanings, we are hard put to explain how we recognize the connectedness of these individual atoms in our judgements. In fact, this will be a general problem for any form of atomism, epistemological or ontological. To conclude: it is obvious that the context principle was formulated to answer presumably a different set of questions, but some of the philosophical issues raised by it were not entirely different from the issues raised by the age-­old controversy between the Bhāṭṭa and the Prābhākara about how we grasp the (whole) meaning of the sentence. Among other things, Frege was concerned with the dismissal of the psychologistic interpretation of number-­entity. That was apparently not the concern of the Indian philosophers we have talked about. It is however not clear whether Frege was making an epistemological point about how we grasp thoughts, or an ontological point about the parts in the context of the whole. He did seem to worry at times about how the whole manages to hold together. The Prābhākara, on the other hand, explicitly makes an epistemological point about how we grasp the sentence-­meaning. By positing such semantic or epistemic objects as things, properties or actions with possible linkages, constituting the domain of the meaning of words in a sentence, he steers clear of the two extremes: the Scylla of crude realism implicit in the extreme atomism of the Bhāṭṭa and the Charybdis of a sort of idealism implied by Bhartṛhari’s holism. APPENDIX

1 On page X of The Foundation of Arithmetic (J. L. Austin’s translation) Frege writes: ‘Never to ask for the meaning of a word in isolation, but only in the context of a proposition.’The principle is here laid down as one of the three principles Frege says he has kept to in his enquiry in the book. So it does seem that the principle is laid down as a methodological principle (or a heuristic device). He explains, after laying down the principle, that a violation of this principle forces one to take mental pictures or acts as the meanings of words, and thus to a violation also of what is now known as anti-­psychologism. (The three principles which are laid down by Frege in the Introduction are anti-­psychologism, the context principle, and the object-­concept distinction.)

300

B. K. MATILAL AND P. K. SEN

2 In Section 60, p. 71, Frege says: ‘Only in a proposition have the words really a meaning.’ Here, the principle is invoked to argue that from the fact that we cannot form a (separate) idea (i.e. mental picture) for a number word it does not follow that it does not have any meaning. Thus he adds: ‘It is enough if the proposition taken as a whole has a sense; it is this that confers on its parts also their content.’ It is very significant however that in the second part of the sentence Frege speaks like a (sentence-­)holist: it is the sense of the sentence which confers on the words the content which they have. 3 Two pages later, in Section 62, p. 73, Frege talks again of the context principle in the same manner. 4 Frege refers to the context principle again towards the end of the book, while giving a resume of his investigations in the Foundations, in Section 106. On page 116 he writes the following: We next laid down the fundamental principle that we must never try to define the meaning of a word in isolation, but only as it is used in the context of a proposition: only by adhering to this can we, as I believe, avoid a physical view of number without slipping into a psychological view of it.

Reference to the idea of definition in context seems to suggest the case of contextual definition of words and sub-­sentential expressions. Wittgenstein’s own form of the context principle was inspired by the second and third formulations in Frege. In 3.3 of the Tractatus he writes: ‘Only propositions have sense; only in the nexus of a proposition does a name have meaning.’ Wittgenstein uses two terms in German while expressing the context principle—‘Sinn’ and ‘Bedeutung’—and the first in connection with propositions (sentences), while the second in connection with names. If this is taken really seriously, we can take Wittgenstein to have said that the sense is something which belongs to the sentence, and not to the words, not especially to names; and the names have reference, as opposed to sense, but the reference of the name is determined by the sentential context in which it occurs. If there is anything in this interesting idea, what Wittgenstein advocated in the Tractatus must be very different from what the context principle is taken to say in any of its interpretations we have discussed, and for which we have any textual evidence. NOTES * We benefited from conversation with Michael Dummett at a seminar given by us in Oxford in 1984.We wish to thank also the following persons who read the paper at its various stages and made important comments: Peter Strawson, Michael Resnik, and Simon Blackburn. 1. Gottlob Frege, 1980. 2. Frege and the Philosophy of Mathematics, Ithaca, 1980. 3. ‘Frege’s Context Principle’, Mind, 1986, pp. 491–5. 4. We apologize for introducing two rather formidable Sanskrit terms at the outset. We have given literal translations of the two terms here. They are ascribed to the two philosophers of the Mīmāṃsā school, Prābhākara and (Kumārila) Bhāṭṭa. Later we propose to use these two proper names to refer to these two rival views. 5. ‘On Denoting’, Mind, 1905, pp. 479–93.

THE CONTEXT PRINCIPLE

301

6. This aspect of Frege’s use of the contextual definition is brought out extremely well by Crispin Wright in his Frege’s Conception of Numbers as Objects, Aberdeen University Press, 1983. But Peter Milne (Mind, 1986, pp. 491–6) is right in saying that the context principle cannot be exhausted by Crispin Wright’s syntactic priority thesis, that is, by the thesis that if an expression has all the syntactic features of a singular term, and we can give sense to the sentence as a whole, then we can say that the singular term stands for an object, even if there is no independent evidence for the existence of such an object. (For one thing, expressions other than singular terms must be brought within the scope of the context principle; and, for another, the principle should not be assumed to concern reference alone of expressions, it can concern their sense as well.) However, the point which Milne himself makes, namely that the context principle was taken by Frege (in his pre-­sense/reference thinking which was at least partially retained even at the time of writing The Foundations), to give an account of identity statements in terms of a shift of reference from the object to the sign, cannot add anything to a specification of the content of the context principle. It only tells us something about what this principle was taken to explain. Besides, it is both desirable, and possible, to find out some interpretation of the context principle which can survive the sense/reference distinction. 7. We believe that it is a mistake to suppose that the principle applies to unsaturated expressions alone. For, to be sure, the context principle, according to Frege, should apply to all expressions. Resnik has added another reason (private correspondence): unsaturatedness is only needed after sentences are given parts that have separable meanings, while the strong context principle would imply that parts of the sentence may not have separate meanings. 8. M. Dummett, Frege: Philosophy of Language, 2nd edn, p. 4. 9. Arguments for both schools are to be found in many Sanskrit philosophical texts. For convenience, we have used Jayanta’s Nyāyamañjarī of 10–11th century AD. See the edition of 1934, pp. 365–9. 10. F. Staal, ‘Sanskrit Philosophy of Language’ in History of Linguistic Thought and Contemporary Linguistics, ed. Herman Perret, Berlin, 1976, pp. 102–36. 11. W.V. Quine, Word and Object, pp. 132–3. Also Theories and Things, p. 68. 12. P. Gochet, ‘The Syncategorematic Treatment of Predicates’, in Analytical Philosophy in Comparative Perspective, ed. B. K. Matilal and J. L. Shaw, 1985, pp. 61–80. 13. Jayanta (1934 edn), p. 207. 14. See M. Dummett, The Interpretation of Frege’s Philosophy, London, 1981, pp. 345–59.The term is used by Sluga. 15. The discussion here centres around a frequently quoted verse of Kumārila Bhaṭṭa. We follow the explanation by Jayanta and Śālikanātha.

CHAPTER

17

Bhartr.hari’s view of sphot.a Bimal Krishna Matilal

Source: The Word and the World: India’s Contribution to the Study of Language (Delhi: Oxford University Press, 1991), pp. 84–98. I

The sphoṭa doctrine has been most prominently associated with Bhartṛhari. But scholars have held different views about the exact significance of this concept in Bhartṛhari’s thought. The situation is further complicated by the fact that later grammarians attribute a much crystallized and ostensibly different doctrine of sphoṭa not only to Bhartṛhari but also to Patañjali. Early Indologists (Keith, 1928, 387; De, 1925, 180) described sphoṭa as a mysterious or mystical entity, and this was probably due to its association with Bhartṛhari’s notion of śabda-­brahman or the Eternal Verbum (Sastri, 1959). But this was a mistake. In spite of its metaphysical underpinning which M. Biardeau (1964, 268) rightly emphasized, there is a linguistic treatment of the concept well-­documented by Bhartṛhari himself, and other grammarians. Brough (1951, 34) and Kunjunni Raja (1969, 97–148) were right to stress the point that sphoṭa was not a mysterious entity. There is another dispute that is connected with it; whether the sphoṭa in Bhartṛhari is simply the linguistic sign in its aspect of meaning-­bearer (Brough, Kunjunni Raja), or whether it represents an abstract class of sounds sorted out and extracted by the listener from gross matter ( Joshi, 1967, 40). Cardona (1976, 302) thinks that Brough’s thesis should be modified since Bhartṛhari also talks about varṇa-­sphoṭa, which refers to a sound unit of the language system, not to any meaning-­bearing unit. Iyer (1969, 158–9) has, however, refuted Joshi’s rather sweeping comment that Bhartṛhari’s sphoṭa had nothing to do with the meaning-­ bearing speech unit. In Mahābhāṣya-­dīpīkā Bhartṛhari reinterpreted Patañjali’s use of the word śabda. This was noted by Patañjali as the meaning-­bearing element in the Paspaśā section. Bhartṛhari glossed this as sphoṭa and characterized it as eternal. 302

303

BHARTṚHARI’S VIEW OF SPHOṬA

II

In Vākyapadīya Bhartṛhari clearly develops the threefold doctrine of sphoṭa related to letters or phonemes, words and sentences.This is explicitly mentioned in the Vṛtti. Sometimes he uses śabda and sphoṭa interchangeably, which might have been the source of confusion. The sphoṭa is further described as partless and indivisible, and as devoid of internal sequence. A pada-­sphoṭa, i.e. the sphoṭa identified as a word, seems to be a meaning-­bearing unit. But, for Bhartṛhari, the vākya-­sphoṭa, i.e. the sphoṭa in the form of a sentence, is the most important. In the second kāṇḍa of Vākyapadīya, he deals with various definitions of the sentence, and finally concludes that a sentence is a sequenceless, partless whole, a sphoṭa, that gets ‘expressed’ or manifested in a sequential and temporary utterance. This is also the primary meaning-­bearing element. For Bhartṛhari, however, this is a wrong term; ‘meaning-­bearing unit’. Sphoṭa is the real substratum, proper linguistic unit, which is identical also with its meaning. Language is not the vehicle of meaning or the conveyor-­belt of thought. Thought anchors language and language anchors thought, Śabdanā, ‘languageing’, is thinking; and thought ‘vibrates’ through language. In this way of looking at things, there cannot be any essential difference between a linguistic unit and its meaning or the thought it conveys. Sphoṭa refers to this non-­differentiated language-­principle. Thus I believe that it is sometimes even incorrect to ask whether sphoṭa is or is not the meaning-­bearing speech unit in Bhartṛhari’s system. III

Bhartṛhari begins the discussion of his theory by a reference to the distinguishing of the two aspects of language by his predecessors. In verse I.44 he says that the linguists comprehend two types of śabda among the upādāna śabda, ‘linguistic sound’: one is the causal root of its manifestation and the other is applied, being manifested, to convey meaning. Of these two, the second is the linguistic unit properly understood, it is the real language, while the first is what ‘manifests’ or ‘expresses’ it. Bhartṛhari, and following him some later grammarians, related this duality to what I shall call the sphoṭa-­nāda distinction of language. Nāda manifests sphoṭa and sphoṭa conveys meaning.We need to explain such expressions as ‘manifests’ and ‘expresses’.The sphoṭa is an indivisible unit, a partless, sequenceless whole, which is connected with the verbal dispositional ability of the speaker or the hearer. For the sake of communication between language-­users, sphoṭa needs to be made explicit, i.e. potentiality must be actualized, so that the hearer may receive it.This cannot be done without nāda, the sequential utterances of sound-­elements.This is how the nāda becomes the causal factor for making sphoṭa explicit.The speaker cannot but utter nāda in a particular sequence, and nāda therefore reveals sphoṭa in this way, in sequence, and part by part. It is argued that sphoṭa in this way appears (falsely) to have parts and temporal sequences just as the moon reflected in wavy waters appears to be wavy and disintegrated. Since the nāda is also identified with the sphoṭa certain spurious features are superimposed on the sphoṭa. The sounds uttered by the speaker make the real linguistic units, primarily a sentence, explicit, but this is the sphoṭa of the speaker. Sphoṭa is also shared by the hearer, and as a result the hearer’s sphoṭa is ‘awakened’ by the utterance of the speaker.This awakening of the hearer’s sphoṭa is what is called the comprehension by the hearer of the sentence uttered.

304

BIMAL KRISHNA MATILAL

This is what is meant by the claim that the sentence uttered must ‘already be present’ in the hearer. From the point of view of the speaker, however, the sphoṭa ‘already present in him’ will be the causal condition for the nāda or the sequential word utterance, which will convey the sphoṭa to the hearer. The metaphysical view of Bhartṛhari is that whatever is called śabda, ‘language’ and artha, ‘meaning’, ‘thought’ or ‘things meant’, are one and undifferentiated in their pre-­verbal or potential state. Before the utterance, it is argued, the language along with whatever it conveys or means is like the yolk of a peahen’s egg. In that state all the variegated colours of a full-­grown peacock lie dormant in potential form. Later these colours are actualized. Similarly in the self of the speaker or hearer, or whoever is gifted with linguistic capability, all the variety and differentiation of linguistic items and their meanings exist as potentialities, and language and thought are identical at that stage. Bhartṛhari even believes that the nature of the self is nothing but identical with the nature of language-­thought. This state of complete identity of language and thought is called the paśyantī stage of language. Before the proper articulation of the sound-­sequence or utterance, there is another ‘intermediate’ stage (called madhyamā vāk) where the language and the thought it conveys are still one and undifferentiated, but at this ‘pre-­verbal’ stage the speaker sees them as differentiable. In other words, he recognizes the verbal part, which he is about to verbalize either to himself or to another, as distinct and separable from the artha, ‘meaning’ or ‘thought’. This perception impels him to speech which results in the nāda-­sphoṭa differentiation. It may be useful to quote here some relevant verses of Bhartṛhari to underline the sphoṭa-­nāda distinction. Verse I.44: Linguists (śabdavidaḥ) comprehend two types of śabda among linguistic śabda. One is the nimitta of the sound and the other designates the object or meaning. Commentary: In the sequenceless nature of the speech (vāk) both powers, the power to be articulated in sound (audible form) and the power to convey meaning, lie intermixed. Verse I.46: Just as light/fire (jyotiḥ) resides in the arani stick and (being manifested) becomes the cause for manifesting other objects, śabda resides likewise in the mind (inner faculty, buddhi) and (being manifested) becomes separately the cause for manifesting the meaning (as well as itself; added in the commentary). Verse I.48: Since nāda (sound) arises in sequence, sphoṭa, which has neither a former nor a latter stage and which is sequenceless (akrama), is apprehended (through nāda) and appears to be having a sequence as well as parts. Verse I.49: (Thus, properties of nāda are transferred to the sphoṭa.) The reflected image (of the moon, for example) although it resides in a separate location, seems to share the operation (i.e. movement of the waves in water) of objects in a separate location; sphoṭa being manifested in nāda shares the properties of nāda in the same way. Verse I.52: A figure being grasped by a single awareness is painted on a canvas (part by part) into another complete, unitary figure (for the viewer to grasp it in one sweep). Similarly in śabda, too, all these three stages are found. Verse I.53: The speaker apprehends beforehand the entire śabda, with regard to which the hearer’s awareness also arises. Verse I.55: Just as fire has both powers—the power to be the object of manifestation and the agent of manifestation—all śabdas individually have both powers likewise. (Fire in manifesting itself manifests others; śabda, too, in manifesting itself manifests others.)

305

BHARTṚHARI’S VIEW OF SPHOṬA

Verse I.56: (For this reason) śabdas do not convey meaning without themselves being the objects of our awareness. They cannot manifest or reveal the meanings simply by their existence, if they (themselves) remain unapprehended. Verse I.57: Hence, when from indistinct utterance, the form of the śabda is not apprehended, one asks ‘what did he say?’ But when the senses reveal objects, those senses do not need to be apprehended themselves.

IV

How is the sphoṭa-­nāda distinction comprehended? Or, we may rephrase the question: how is ‘language’ comprehended? We have seen that Bhartṛhari has posited three stages of language or speech. The first stage, where there is complete identity of language and thought, is called the paśyantī stage; we can call it ‘non-­verbal’. The ‘intermediate’ stage, where despite the identity of thought and language their difference is discernible, can be called the ‘pre-­verbal’ stage. And the third, the vaikharī stage, can be called the ‘verbal’ stage. This is how the matter stands from the speaker’s point of view. But how does the hearer comprehend it? Bhartṛhari states four different views on this point. According to some, sphoṭa is cognized as identical with the sound or nāda. One who grasps the nāda grasps the sphoṭa at the same time. Since basically the sound or nāda is identical with the sphoṭa, and they are so to say only two sides of the same coin, the grasping of one cannot be distinguished from that of the other. In other words, he who has not grasped one has not grasped the other.The commentator has supplied a beautiful analogy to elucidate the point. When a piece of crystal is placed near a red japā flower, the piece (of crystal) cannot be grasped or perceived without the colour red, for it now certainly appears red because of the proximity of the red flower. The sphoṭa is likewise comprehended along with the nāda that manifests it, one grasps the bits of language as sound or utterance, i.e. the sphoṭa as the nāda. It is not clear whether this analogy can be taken to imply that the nāda is only a superimposed feature of the sphoṭa, the real language. For, the piece of crystal is only apparently red due to the conditional superimposition; in reality it is colourless. If the implicit idea is that the nāda or sound is an inessential, conditionally superimposed, feature of the real language or the sphoṭa, then I do not think this would be in exact accord with the doctrine of Bhartṛhari. Perhaps the analogy is not to be taken too far. It may be that simply the identification of the nāda with the sphoṭa is what is implied. In that case, this may well be Bhartṛhari’s own view.The analogy between a piece of crystal and a bit of language was again used by Bhartṛhari in kāṇḍa 3 of Vākyapadīya. But the purpose of that anology was probably different. It was to illustrate a point of Bhartṛhari’s semantics. Objects meant by bits of language are only created by the language, for language is autonomous just as a piece of crystal is believed to be autonomous (according to the pre-­scientific theory of light) in the sense that it ‘reflects’ objects by according an autonomous status to them (the reflector modifies the object in its own way). This analogy has been used with great ingenuity by R. Herzberger in her exposition of Bhartṛhari’s semantics (Herzberger, 1986, 50–3). However, in the context of our comprehension of the sphoṭa through nāda, the ‘crystal’ analogy might serve a slightly different purpose.We grasp

306

BIMAL KRISHNA MATILAL

the sphoṭa as reflected in the nāda, and as almost identical with the nāda itself, just as we grasp redness as presented by the piece of crystal, not in any other way. Bhartṛhari refers to a second view held by some: that the comprehension of the sphoṭa does not require the comprehension of sounds or nāda as a condition. A tentative argument is given in favour of this view. We know that when we cognize an object, say a pot, through visual perception, we do so through the instrumentality of the faculty of vision, the eye, and it is an established fact that we do not need to know the properties or features of the eye-­organ itself. The fact that we have the eye-­organ is enough, for this only is relevant for the knowledge of the object. Similarly we comprehend the sphoṭa through the instrumentality of nāda, sounds. Patañjali (see ch. 7) has contended that sound is the attribute of the sphoṭa. Now when the sphoṭa is presented through sounds, we comprehend it right away even prior to our cognition of all the sound-­symbols, though the latter is indicative of the former. In other words, in this view the cognition of the sounds themselves is not needed prior to our cognition of the sphoṭa. Bhartṛhari has criticized this view saying that as long as the sounds are uttered they are also directly perceived by our sense of hearing. Hence it is impossible to comprehend the sphoṭa without comprehending the sounds. The view, as it stands, is indeed peculiar. Perhaps the upholders of this view were unconsciously arguing in favour of a distinction between sound-­tokens and sound-­types, and they accordingly wanted to say that we do not need to cognize the sound-­types over and above the sound-­tokens prior to our comprehension of the sphoṭa. Sometimes, it may be pointed out, we comprehend the sphoṭa even when only a part of the relevant sound-­ token is heard. The third and fourth views mentioned by Bhartṛhari were also obscure. There were a few thinkers who apparently believed in the metaphysical existence of the sphoṭa but maintained that it could not be amenable to perception.The reason adduced further confounds the issue, for it is said that the distance that separates the sphoṭa from the cognizing hearer makes the sphoṭa imperceptible. The upholders of the fourth view understood the matter in another way. Distance cannot make the sphoṭa imperceptible but can render it rather indistinguishable. In other words, since distance does not remove the object from the field of perception, it can render the sphoṭa, like an object at a distance, indistinguishable from the environment. Thus, we fail to perceive sphoṭa as a distinct entity. Bhartṛhari obviously rejected these views, and it is difficult to make sense of their notion of sphoṭa. There is an obvious problem when we say that the sequential and atomic nāda-­units in combination reveal or manifest the indivisible, impartite sphoṭa. The problem is similar to the problem of perception: how do the parts present the whole? If they present it partially, then the whole will never be presented in one sweep and this will cast doubt upon the contention about the reality of the whole over and above the constituents out of which it is formed. For the grammarians, the sphoṭa is a whole and it is a metaphysical entity, neither an object of construction, nor an abstraction. So the question arises: how do we perceive it from the utterance of a divisible and sequential sound-­stretch? How is the unity perceived through the presentation of plurality? Thus by stating that the sound-­ stretch manifests the sphoṭa we do not explain much or answer the crucial question: how? For surely separate efforts are required to produce different sounds and they are produced in succession. At which point exactly is the unity that we call sphoṭa comprehended? If the

BHARTṚHARI’S VIEW OF SPHOṬA

307

unity is cognized at every instant from the beginning, then there is the fault of repetition and redundancy. If it is by the last (utterance of the) sound-­unit then all the preceding units (or their utterances) are superfluous, for they have been destroyed at the time of comprehension. Bhartṛhari answers this objection as follows: For the sake of convenience, let us use the illustration of a word-­sphoṭa ‘gauḥ’ (the cow). There is a unity here, the word. But the four sound-­units or letters, g, a, u and ḥ, present the sphoṭa, each individually. Bhartṛhari says that each letter here is the medium of manifestation of the unity, the whole sphoṭa. The problem of repetitiveness or redundancy is avoided by postulating a difference or a distinct property (viśeṣa) each time in the resulting awareness or comprehension. The first letter (or the first sound-­unit) shows the whole, but very indistinctly. It becomes gradually clearer and progressively better understood through successive stages until the last unit is uttered. Although the earlier units disappear when the last unit is reached, the memory-­impressions left behind by those earlier units are in the hearer and each time there is qualitative difference in the memory-­impression (saṃskāra). The sphoṭa itself does not admit of any qualitative or quantitative difference, addition or subtraction, but the impression or image of it may be imperfect and different, due to the imperfect nature of the human intellect. Our memory-­impression may be dim or partial, clear or unclear, on various occasions but the object, sphoṭa, will always shine in its undimmed glory. It may be further argued that Bhartṛhari’s explanation of the comprehension of sphoṭa is unsatisfactory. For certainly the letters or sounds, g, a, u and ḥ, reveal themselves only individually. That is, we perceive each unit as it is produced, through our sense of hearing. If the sphoṭa ‘gauḥ’ is also perceived when we perceive g or a, it must be perceived as g or a. But gauḥ is not g or a.The sphoṭa is not identical with any of the individual sound-­units.To perceive something x as what it is not, is to misperceive x. We call it error. It is somewhat unparsimonious to first postulate an entity like the sphoṭa of gauḥ and then claim that it is only misperceived in the first, second or third letter, g, a, u. Besides, each misperception is based upon recognition of some similarity between the object present and the object superimposed. In the present case, it is difficult to obtain any satisfactory account of such a series of misperceptions. Bhartṛhari gives a bold reply to this criticism which also has the metaphysical underpinnings of his sphoṭa doctrine. He says that just as the cognition of the lower numbers, ‘one’ and ‘two’, is the means for understanding a higher number, say, ‘three’, although they are each distinct and different, similarly, comprehension of the sphoṭa, either a pada-­sphoṭa or a vākya-­sphoṭa, is invariably conditioned by the cognition of the so-­called constituents, either the sounds, g, a, u, ḥ, or the word-­elements ‘Devadatta’ and ‘goes’ (in the sphoṭa as a sentence ‘Devadatta goes’).This presupposes an understanding of the Vaiśeṣika theory of numbers. According to this theory, numbers are distinct from one another and they constitute separate entities, and all numbers higher than one are produced by a sort of ‘connective-­comparative’ cognition called apekṣābuddhi. This is the notion that brings many unities under one number or another.When two things are present, one cognizes both as ‘this is one and that is one’. This is the ‘connective-­comparative’ cognition that gives rise to the awareness of ‘two’ or duality, and similarly with each succeeding number. In this theory understanding of the previous number is the conditioning factor for the awareness of the higher number. Similarly understanding of the distinct

308

BIMAL KRISHNA MATILAL

sounds or distinct word-­elements is the means for the awareness of the combined unity, the sphoṭa. Besides, Bhartṛhari claims that for the sphoṭa, or what he would call the real language, to convey some meaning to the hearer, it is an essential and unavoidable condition that it be made explicit through the sequential and transitory nāda elements. Just as an episode of knowledge cannot be known or talked about without any reference to what it represents or what is known by it, similarly the sphoṭa cannot communicate or convey any meaning (or be known) without its being manifested through the sequential nāda or speech.To have a clear perception of a tree, for example, we must proceed from a distance step by step when the vague and indistinct blur gradually gives way to a distinct shape and identity. Similarly the sphoṭa, through steps or sequences, is distinctly understood and identified. Bhartṛhari claims that a man who has mastered the śabdayoga or obtained the light of the Eternal Verbum (some sages have apparently succeeded in this) can perceive or understand the sphoṭa clearly when the first sound is heard, just as a man with a perfect vision or unlimited power of sight (if such a man exists) can see the tree distinctly even from a distance. Comprehension of the sphoṭa is equivalent to such a distinct vision of reality. V

Bhartṛhari has noted also that there is no unanimity among his predecessors regarding the real nature of the sphoṭa. He refers to several earlier views. According to one, the sphoṭa is the universal manifested by the individuals which are nāda elements or sounds. The sphoṭa is thus the class, of which sounds would be members. The commentary quotes a line from the Mahābhāṣya of Patañjali, where the word sphoṭa is used in the sense of the universal. It is the universal of the word gauḥ, not the universal called word-­ness. We may call it the word-­form, realized through the sequential utterance of the sounds. Some later commentators (cf. Bhaṭṭoji) apparently have taken the class-­sphoṭa theory as Bhartṛhari’s own. But this is a mistake, and the claim of some modern scholars to the effect that sphoṭa is nothing but a postulation of a unitary word-­universal, should be rejected. Another view mentioned by Bhartṛhari regards sphoṭa as an impermanent entity, produced by the initial sounds resulting from the contacts and separations of the vocal organs. The initial sounds themselves constitute the sphoṭa, they do not manifest it. But these sounds, despite being momentary, produce further sounds which thus spread in all directions, gradually decreasing in intensity, and reach the hearer’s organ as nāda. The sound produced initially is the sphoṭa, other sounds produced in reverberation, are ‘sound-­produced’ sounds (dhvani or nāda). Another view modifies this position, in that it understands both sphoṭa and nāda to be produced simultaneously through contacts and disconnection of vocal organs. They are like the flame and the light of a lamp. We produce both the flame and the light at the same time. The light ‘travels’, so does the reverberation. The flame is fixed in one place, and so is the sphoṭa, according to this view. None of these views would be acceptable to Bhartṛhari. His idea of sphoṭa is different, as already described. Bhartṛhari draws another interesting distinction between two types of sounds in this connection. They are called the ‘primary’ sounds (prākṛta) and the ‘derived’ or ‘transformed’ sounds (vaikṛta). The usual way to take the ‘primary’ sounds is to refer to the linguistically relevant sound-­sequences which the speaker intends to produce and the hearer

BHARTṚHARI’S VIEW OF SPHOṬA

309

expects to hear. It is the shared ‘speech’ which manifests sphoṭa (where the sphoṭa can be called prakṛti, the ‘original’, and hence the prākṛta is the manifestor of the ‘original’).These primary sounds are not abstractions, but ideal particulars which have sequences, duration and other qualities—all specified by the particular language system. The long sounds should be long, of required length, the short vowels should be short and so on. But this must be conceived as divested of all personal idiosyncracies or ‘mannerisms’ of the speaker who utters them. It is the norm. The non-­linguistic concomitants of any utterance are to be separated from this notion of ‘primary’ sound. This type of sound is also said to be identified with the sphoṭa, though of course wrongly, for the sphoṭa is conceived as a sequenceless, durationless and partless whole. In other words, one (wrongly) cognizes the sphoṭa as united with this ‘primary’ sound-­series. (See also ch. 7, section III.) The ‘secondary’ or ‘transformed’ sounds may therefore be taken to be the individual instances of utterance that either reverberate or continue to show the individual peculiarities of the speakers, various differences in intonation, tempo, pitch, etc. The description here is a bit obscure, for it is also said that the manifestation of the sphoṭa still continues to happen or take place (after the first manifestation by the ‘primary’ sounds) with the help of the ‘transformed’ or ‘secondary’ sounds. ‘Difference in the speed of utterance’ (vṛttibheda) is also a factor in the ‘transformed’ sounds. They continue to manifest the sphoṭa again and again uninterruptedly for a longer period. They are said to resemble the light of a lamp which travels and continues to reveal the object as long as the lamp is lit. If this analogy is not very helpful, we may try another characterization. It is said that the slow utterance of a short a sound does not turn it into a long ā sound, nor does speedy utterance of the latter turn it into the former. The primary sounds a or ā are unchanged although the speed becomes a factor in the ‘transformed’ sound. In his commentary on the Mahābhāṣya, Bhartṛhari again refers to this distinction, but the matter there is further complicated by a reference to the two kinds of primary sounds (Iyer, 1969, 171). Kunjunni Raja’s description of the ‘primary’ sounds as the acoustic image or the abstract sound pattern is much too influenced by a knowledge of modern linguistics (Kunjunni Raja, 1969, 120). That the ‘transformed’ sounds may be just ‘reverberations’ after the utterance (of the primary sounds) which reveal the sphoṭa is a possible interpretation which receives support from Helārāja in his commentary on the third kāṇḍa of the Vākyapadīya. But by ‘reverberation’ we need not think of the echo or returning sound but extend its sense to include the continuous producing of the sound-­series after the initial sound is produced by the impact of the vocal organs. This may presuppose the Vaiśeṣika theory of sound, according to which the first sound is produced by impact, etc., and is believed to be a momentary entity, which dies at the next instant but nevertheless generates another similar sound-­individual, which in its turn generates another before being destroyed in the next moment, and so on. This process continues for a while and then stops (on reaching the hearer’s organ). How long the process will continue and how far it will ‘travel’, will depend upon the intensity of the originally produced sound. Bhartṛhari does not contribute to the Vaiśeṣika theory of sound-­production, but it seems that he would accept this process of sound-­travel with one crucial change made in the theory—instead of talking about production of a new sound at every moment, we should talk about the new manifestation of the same sound at every moment during its persistence.

310

BIMAL KRISHNA MATILAL

VI

Bhartṛhari’s philosophy of language is ultimately grounded in a monistic and idealistic metaphysical theory. He speaks of a transcendental word-­essence (śabdatattva) as the first principle of the universe. His sphoṭa doctrine is finally aligned with the ultimate reality called śabda-­brahman. A self-­realized person attains unity with the word-­principle—a man of perfect knowledge. There is no thought without language, no knowledge without word in it. Consciousness vibrates through words, and such vibrating consciousness or a particular cognitive mode motivates us to act and obtain results. Hence language offers the substratum upon which human activity is based. Language and meaning are not two separate realities such that one conveys the other. They are in essence the two sides of the same coin. The sphoṭa is this unitary principle where the symbol and what is signified are one.To understand each other’s speech and to communicate, we do separate the inseparable, the sound and its sense. This is only instrumental to our mutual understanding. At the ultimate level, they are one. Bhartṛhari talks about three kinds of sphoṭa: letter-­ or sound-­ sphoṭa, word-­sphoṭa and sentence-­sphoṭa, but his primary interest lies with the sentence-­ sphoṭa. He underlines the importance and primacy of the sentence in language analysis in the second kāṇḍa of Vākyapadīya. The sentence is the unit of communication.According to Bhartṛhari it is a unity, a whole, and not the result of joining together smaller units called words. The sentence-­meaning is likewise a whole, not constituted by word-­meanings put together. The Mīmāṃsakas apparently believed that the sentence and the sentence-­meaning are produced by joining the smaller units, words and word-­meanings together. Bhartṛhari examines five different definitions of the sentence put forward by the Mīmāṃsakas. 1 The sentence is a ‘collection’ (saṃghāta) of words, and each word has a certain meaning of its own. When it is used in a sentence, it conveys the same meaning. But when the words are linked in a sentence, then, through mutual linkage, an additional meaning is given to the whole unit. Thus the sentence-­meaning is the ‘mutual linkage’ (saṃsarga). 2 The sentence is the ‘sequence’ (krama) of words, and the sentence-­meaning belongs to this sequence. But since sequence is a property of time, one has to superimpose it upon the linguistic units called words. Both views uphold an extreme form of atomism and are ascribed, later on in history, to the Bhāṭṭa school of Mīmāṃsakas. This Bhāṭṭa theory has been nicknamed abhihitānvaya, ‘designation followed by (syntactic) connection’, theory. In simple language, it means: words designate their meanings and then the meanings are brought together to give the sentence-­meanings. 3 A sentence must have a finite verb, and we may even say that the finite verb is the sentence, for all the other words would be required in any case to be connected with the verb. Hence if we understand the verb, we understand the sentence, all the other kāraka-­words being understood. The sentence-­meaning, on this view, is of the

BHARTṚHARI’S VIEW OF SPHOṬA

311

nature of an action or process. It is the dependence of the meanings of other words on that of the verb that makes the sentence-­meaning possible. 4 Even the very first word can be called the sentence, for the meaning of the sentence is already contained in, though only vaguely or partially expressed by, the first word. 5 A further modification of the above view is that each word in the sentence contains the whole sentence-­meaning, though it is only partly revealed by it. These last three views are supportive of a sort of contextualism. Words may have isolated meanings, but when they are in isolation they cannot express such separate or isolated meanings or meaning-­atoms. Each word in connection with another word has a designation (meaning), which we should call ‘connected designation’; it designates a connected entity. This is ascribed later on to the Prābhākara school of Mīmāṃsakas, and nicknamed anvitābhidhāna, ‘connected designation’. The idea is that since, in order to talk about a word, we talk about a word used in some sentence or other, its meaning cannot be known in isolation. We must necessarily talk about the word’s contextual meaning or rather its meaning in connection with the meanings of other words in the sentence. Bhartṛhari refutes all these five notions of the sentence, and rejects both atomism and the sort of contextualism that has been defended by the second group of Mīmāṃsakas. He puts forward instead a holistic framework and argues that a sentence might be understood as an indivisible sphoṭa. The sentences as well as their meanings are indivisible units. It is for the sake of convenience as well as for facilitating our learning and understanding of a language, that we break the wholes into parts and smaller units and correlate words and word-­meanings. The indivisible sentence is either internal to the language-­user (any member of the speech-­community) or external to him. It is externalized through its manifestation in speech or nāda. And through such manifestation it appears to have divisions (and this is true of both the sound-­aspect and the meaning-­aspect), although essentially it remains indivisible. In other words, appearance of divisibility of sentences and sentence-­ meanings is deceptive. It is like the ‘cognition of multiplicity’ (citra-­jñāna). Although the cognition in such cases is one and indivisible, one sees plurality in it. A drink from the punch-­bowl will have a plurality of flavours and tastes, but it is one and a unique drink—a whole. Bhartṛhari underlines this indivisibility aspect by offering three more definitions of the notion of the sentence in addition to the five already mentioned. 6 The sentence is the class of the sequences of words, or the universal resident in the sequence of words. Or, 7 It is the whole string of words without any divisible part. Both these definitions refer to what has been called the ‘external’ sphoṭa above.The former regards the sentence as a universal while the latter regards it as a particular whole. Technically the later grammarians called them the jāti-­sphoṭa and the vy-­akti-­sphoṭa, respectively. 8 The last definition refers to what has been called ‘internal’ sphoṭa. The sentence is a whole piece of cognitive awareness.

312

BIMAL KRISHNA MATILAL

These three so-­called definitions may be just recognition of the three different levels of the realization of the same sphoṭa or three different ways of capturing the same reality, sphoṭa, the sentence. The sentence-­meaning in all these cases is declared to be given by pratibhā, i.e. by a ‘flash of understanding’ of what is being communicated.This flash of understanding is also declared to be holistic.We do not obtain it bit by bit.That is why some people are said to understand the meaning of the sentence even before the utterance of the whole sentence (see also ch. 10). Several points are not made clear here. The correlation between the ‘external’ and the ‘internal’ sphoṭa must be a crucial factor here. For, an ordinary sentence with indexicals, be it in the form of a universal of the sound-­sequence or a concrete and particular sound stretch, will have to have a different sphoṭa as the speaker, time and place change. This distinction can be maintained by associating the external sound-­pattern with the ‘internal’ sphoṭa of the speaker, which reveals itself to him at a particular time and place. This question has not been raised in this way and therefore the answer has not been given or discussed. Besides, if the sphoṭa is the real language or a particular language system, as Bhartṛhari seems to imply, we must apply the word ‘language’ in an extended sense, not to a system or string of sound-­units having sense, but to what is only made explicit in a system-­of-­sound-­units, not, that is, to the nāda exclusively, but to what is ‘manifested’ by nāda, the sphoṭa. BIBLIOGRAPHY Bhartṛhari, Vākyapadīya, eds K.V. Abhyankar,V. P. Limaye, Poona University, 1965. ———. Vākyapadīya, Kāṇḍa I, ed. K. A. Subramania Iyer, Deccan College, Poona, 1966. Kāṇḍa III with Helarāja’s Commentary,Vol. I, 1963. Biardeau, M. Theorie de la Connaissance et Philosphie de la Parole le Brahmanisme Classique, Paris, 1964. Brough, John. ‘Theories of General Linguistics in the Sanskrit Grammarians’, Transactions of the Philological Society, 1951, 27–46 (included in Staal, 1972). Cardona, George. Pāṇinī: A Survey of Research, Mouton & Co., The Hague and Paris, 1976. De, S. K. Studies in the History of Sanskrit Poetics, Luzac & Co., London, 1925. Herzberger, R. Bhartṛhari and the Buddhists, Studies of Classical India Series No. 8, D. Reidel Publishing Co., Holland, 1986. Iyer, S. Bhartṛhari, Deccan College, Poona, 1969. Joshi, S. D. The Sphoṭa-­nirṇaya of Kaunḍabhaṭṭa, Poona, 1967. Keith, A. B. A History of Sanskrit Literature, Oxford University Press, Oxford, 1928. Kunjunni Raja, K. Indian Theories of Meaning, Adyar Library, Madras, 1969. Sastri, G. The Philosophy of Word and Meaning, Sanskrit College, Calcutta, 1959. Staal, J. F. (ed.). A Reader on the Sanskrit Grammarians, M.I.T. Press, Cambridge (Mass.) 1972.

CHAPTER

18

– – “Aka s’a” and other names Jonardon Ganeri

Source: Journal of Indian Philosophy, 1996, 24, 339–362. Praśastapāda (c. 550 c.e.), the first major commentator on the Vaiśeṣika-­sūtras, initiated a controversy about meaning whose repercussions would be felt for many centuries. He claimed that the usual model for the meaning of nominal referring expression (samjñā), according to which a term refers to an object on the “basis” (nimitta) of that object’s possession of a certain descriptive feature, does not apply in the case of such terms as “time,” “space,” and “ākāśa.” The main reason he gave for this claim was that the descriptive feature with which such a term would be associated, if the standard model is to apply must be singly instanced, contradicting the doctrine, accepted in Nyāya and Vaiśeṣika, that a “real” or “natural” universal property ( jāti) cannot have just one instance. He therefore significantly asserted that these terms belong to a new semantic category, which he called the class of “pāribhāṣikī ” names: In the absence of a [characteristic] real property, which [absence] is due to the uniqueness of ākāśa, space, and time, the three names [“ākāśa,” “space” and “time”] are [to be called] pāribhāṣikī[terms].1

I would like, in this paper, to trace the historical development of Praśastapāda’s idea in later Nyāya and Vaiśeṣika texts. Though the meaning of the term “pāribhāṣikī” was subject to a great deal of change, the effects of Praśastapāda’s recognition that such terms form a special class can be felt even in a text as late as Gadādhara’s Śaktivāda (c. 1650 c.e.). At the same time, we will see how systems of classification for nominal expressions evolved from relatively ad hoc listings of intuitively distinct nominal types to principled classificatory schemata. 313

314

JONARDON GANERI

Praśastapāda himself offered no positive account of the semantics of this new class of expression, nor did he give any clear criterion for deciding which terms belong to it. The terms cited seem, indeed, to have more than one unusual property. First, unlike ordinary generical nominals like “[the] pot,” whose reference is token-­reflexive in that the reference of such a term depends on the particular context in which it is uttered, the reference of a pāribhāṣikī name is context-­independent. Each such term is semantically associated with a single individual, to which any utterance of that term refers. An apparent corollary of this is that such terms require no feature to delimit the domain of possible referents. Second, the individuals to which they refer are all homogenous, partless substances. In this, they differ from ordinary proper names, the names, for example, of people. Patañjali argued, and in this he was followed by the majority of later philosophers, that ordinary proper names are semantically continuous with common nouns, for each utterance of a name like “Devadatta” refers to a temporal slice of Devadatta, the “basis” being the unifying feature Devadatta-­hood. This was supposed to show that Vyāḍi’s theory of meaning, according to which the meaning of a noun consists entirely in its being associated with an object, could not account for such terms.2 It seems, however, that the application of Vyāḍi’s model to the terms distinguished by Praśastapāda stands some chance of issuing in a successful semantic theory, since the objects to which they apply do not change over time. Third, and perhaps most significantly, the pāribhāṣikī terms are “theoretical” terms, terms introduced into the Vaiśeṣika philosophical discourse via specific reference-­fixing stipulations. Take for example the term “ākāśa.” The Vaiśeṣikas argue that sounds, which are considered to be qualities, must subsist in some substance. Yet no perceptible substance, nor any imperceptible substance whose existence is already acknowledged, can be the substratum of sound, since none displays the properties the substratum of sound must have. There has, then, to be some other substance which is, among other things, the substratum of sounds. A name for this substance is now introduced via the following stipulation:〈Let the term “ākāśa” stand for that substance which is the substratum of sounds〉 . Analogous chains of reasoning presage the introduction of the terms “time” and “space” into the Vaiśeṣika discourse. In the next sections, I will review the discussion of the pāribhāṣikī terms in the later literature, beginning with Udayana’s Pariśuddhi, and also the Kiraṇāvalī, his commentary on Praśastapāda. I will look too at Vardhamāna’s Kiraṇāvalī-­prakāśa, Raghunātha’s Kiranāvalī-­ prakāśa-­dīdhiti, and Jagadīśa’s Sūkti, a commentary directly on Praśastapāda. The views of Gadādhara, whose analysis of the term “ākāśa” in the Śaktivāda is easily the most detailed we have, can then be examined. I believe that we can locate in these discussions a well developed theory of what are now called “descriptive names,” names introduced by a descriptive reference-­fixing stipulation. The discussion of these terms in Nyāya-­Vaiśeṣika is interesting for another reason. It has become a platitude to say that the philosophers of classical India do not distinguish between the analytic and the synthetic, and, in particular, that they do not recognise the existence of a class of statements which are “true in virtue of meaning.” It might seem, however, that the definitional sentence “ākāśa is the substratum of sounds,” which the Nyāya call a “combined application” (sahaprayoga), is as good a candidate for analyticity as any. It is also widely held that, at least partly because of the empiricism of the pramāṇa theory, Indian philosophers fail to recognise the class

“ĀKĀŚA” AND OTHER NAMES

315

of a priori  truths. Again, the “combined” sentence would usually be regarded as true a priori. What the Nyāya has to say about pāribhāṣikī terms and their associated definitional sentences is, therefore, of some relevance to our broader understanding of Indian philosophical thinking. UDAYANA ON THE CLASSIFICATION OF NOMINAL TYPES

In his Pariśuddhi (1911a: 363), Udayana (c. 1050 c.e.) reports “a view of some modern ­Naiyāyikas” who claim that there are five distinct kinds of nominal expression. In one of these kinds the term is said to pick out its referent “directly” śṛṅgagrāhikayā, lit. “by grasping by the horns”), unaided by any indicating features (nimittopalakṣaṇarahite). According to these theorists, proper names like “Caitra” and “Maitra,” and perhaps also the demonstrative pronoun “that” (“asau”),3 belong to this class. In a second kind, the referent of the term is indicated by a feature which, although not an essential property of the referent, serves to individuate it (tatasthopalaksana). We might say that the feature is a “syndrome” of the object.The usual example is the term “earth,” where the mark odiferousness ( gandhavattva) serves to indicate the referent, a piece of earth. The theorists who promote this view say that this category has a subclass, which they call the class of paribhāṣaṇika terms, and of which “ākāśa” is considered to be an instance. The difference between the two classes is that, in the former, the feature functions as the “basis” for referring, whereas, in the later, it does not. The two other kinds of reference mentioned here pertain respectively to terms whose reference is based on etymology (nimitte ‘ntarbhūtopalakṣaṇe nirvacanikayā …), like the verbal noun “[the] cook” (“pācaka”), and terms whose reference is based on a bundle of features, or one feature restricted by another (nimittasaṅkocanikayā …).The example given is “heaven” (“svarga”), which picks out an object on the basis of its being the abode of goodness, as qualified by not being the abode of evil.4 Saṅkara Miśra (c.  1450 c.e.) is the only post-­Udayana Nyāya-­Vaiśeṣika I know of explicitly who adopts and advocates this five-­fold division. Some of the comments in his Kaṇāda-­rahasya help to clarify the intentions behind the classification. He raises the following question (1917: 26–27): how can it be said that the name “Devadatta” grabs its referent “by the horns,” when there is a feature, namely Devadatta-­hood, which can be the basis for the reference? After all, there must be some criterion of distinguishing the referent of “Devadatta” from other objects, like a post or a pot.5 But if this is so, then there is no justification for the claim that names like “Devadatta” belong to a different semantic category than ordinary nominals like “[the] cow”: both refer to objects on the basis of the object’s possession of a certain feature. This is, of course, just the point made by those who accept Patañjali’s thesis about grammatically proper names. Śaṅkara Miśra’s reply is interesting. He notes that this feature Devadatta-­hood is presumably that property which collects together all the different temporal stages of Devadatta, and so at any given time, it is possessed by just one individual. Suppose then that there is someone who does not know who Devadatta is, and to enlighten her, we say “This is Devadatta.” Here, since the utterance occurs at a particular time, the feature Devadatta-­hood cannot be doing any discriminatory work. So it cannot be the basis of the association between the name and the individual Devadatta. Śaṅkara Miśra’s point seems to be that even if we associate a name

316

JONARDON GANERI

with a diachronic “criterion of identity,” it will not help to distinguish the referent of the name at any given time. It is not clear, however, why he thinks this, since presumably at any given time the feature Devadatta-­hood is possessed by at most one individual, and would suffice to discriminate that individual from all others present at that time. Saṅkara Miśra also clarifies the category of pāribhāṣikī terms like “ākāśa.”These are ones for which a “syndrome” (tatasthalaksana) such as “being the substratum of sounds” serves as an indicator (upalakṣaṇa) of the referent.6 The word “indicator” is used here in that sense in which, to repeat a somewhat hackneyed example, the crow sitting on the roof “indicates” a certain house. That is to say, it is an “adventitious qualifier,” a convenient but accidental mark by which a certain house can be distinguished. As used in the context of the term “ākāśa,” the idea seems to be that the feature “being the substratum of sound” has an evidential, heuristic role in the identification of the referent of the term. Although it is as yet far from clear what this role is to be, we will see later how another Naiyāyika, Vardhamāna, tries to make the idea more precise. What is in any case clear is that those who endorse the five-­fold classification refuse to reduce the semantics of the pāribhāṣikī terms to that of ordinary proper names: the former are tied, in one way or another, with a descriptive feature, but the latter refer directly. Although it was Udayana who mentioned the above doctrine, it is not one which he himself accepts. In the Kiraṇāvalī, he distinguishes just three categories of referring expression: naimittikī, aupādhikī, and pāribhāṣikī. Both the first and second kind are ordinary nominals, which refer on the basis of a feature. In the first case, this feature is a generic property or natural kind (jāti), whereas in the second case it is an “imposed” or “nominal” kind (upādhi).7 Examples are the terms “(the) cow” and “(the) beast” respectively, where a beast is defined to be a thing having a tail, hair, and certain other properties. The notion of a “basis” (nimitta) for reference is here thought of as according to the following definition: a word w refers to an object x on the basis of a feature just in case there is a property π, natural or imposed, such that w refers to x (if and) only if x possesses π. Udayana thus understands the distinction between “imposed kind” terms and “natural kind” terms as being derived from an ontological distinction among the properties on which they are based, but it is not clear why he thinks this will lead to any semantic differences between the two. However, in view of the fact that the majority of imposed kinds are composite properties (sakhandopādhi), it might be that the difference is between homophonic and non-­homophonic meaning clauses. Thus, we have: 〈“[the] cow” refers to x only if x is a cow〉 , but〈“[the] beast” refers to x only if x has a tail, hair, etc.〉 . It then remains an open question whether terms having an etymological analysis (e.g., “pācaka,” “dhenu”) and terms having a logical analysis (e.g., “bhūta,” “svarga”) can both be assigned non-­ homophonic reference-­clauses. Udayana now distinguishes a third kind or referring expression, to which he applies Praśastapāda’s label ‘pāribhāṣikī’ term. In his comments on Praśastapāda, he says this (Udayana, 1971: 70; 1956: 333–336): The three pāribhāṣikī terms, “ākāsa,” “time” and “space,” are [terms which] refer directly, without a basis (nimittam antareṇa śṛṅgagrāhikatayā). (Question:) What is the reason [for saying this]? (Answer:)  It is because of the absence of any universal such as ākāśa-­hood, etc.

“ĀKĀŚA” AND OTHER NAMES

317

(Question:)  And why is that? (Answer:)  It is due to the fact that there is no plurality of particulars, since [ākāśa, etc.] are unique [entities]. That is to say, the [Vaiśeṣika] definition of a universal, as that which is eternal, unitary, and occurring in a plurality of particulars, is not satisfied, the obstacle being due to the nature [of ākāśa, etc.] itself. (Opponent:)  Even then, [the term “ākāśa”] cannot be a pāribhāṣikī term, for it is held to be imperceptible. What is said [however] is that “ākāśa is that which is sound’s substratum.” So the property of being sound’s substratum could be a “complex condition” (upādhi). (Reply:)  No! [That is not right] because [the function of] the property of being sound’s substratum is to be an indicative mark (upalakṣaṇatayā taṭastha), just as is “this” when we say “This is Devadatta.” Otherwise, the combined sentence (sahaprayoga) “Ākāśa is the substratum of sound” would not be proper [i.e., non-­tautologous]. (Opponent:) The addition of [the abstraction suffices] -­tva and -­tal to the word “ākāśa” would then [form a] senseless [word]. (Reply:)  No. [The term “ākāśa-­hood”] denotes the nature of ākāśa, either metaphorically or [in the sense of its being] beyond distinctions [i.e., homogenous].

The pāribhāṣikī terms, he stipulates do not refer on the basis of any associated descriptive clause.8 Borrowing instead an earlier idea, he says that they grasp their referents “directly” (śṛṅgagrāhikayā). Unlike the Naiyāyikas who endorse the five-­fold schema, he includes “ākāśa” in this group. Thus Udayana rejects their distinction between P ­ raśastapāda’s ­pārbhāṣikī terms and ordinary proper names, and considers both to refer directly to their referents. For him the meaning clause for “ākāśa” is just MCdirect “ākāśa” refers to ākāśa. Udayana considers the suggestion that a term like “ākāśa” should be classified as an aupādhikī, a term whose reference is based on an “imposed” kind, since the term-­introducing stipulation〈Let “ākāśa” stand for the substratum of sounds〉links “ākāśa” with the composite descriptive clause “the substratum of sound.” This is just to analyse “ākāśa” as a descriptional name, a name whose reference is fixed by description. Its meaning clause would be MCdescription (∀x) (“ākāśa” refers to x iffdefn x is the substratum of sound),9 and it is, as we shall see, explicitly advocated by a later Naiyāyika, Jagadīśa. Udayana’s argument against this view is that it leaves us unable to explain why it is that the ‘combined application’ (sahaprayoga) sentence, “ākāśa is the substratum of sound,” can be informative. If “ākāśa” is tied definitionally with “the substratum of sound,” then this sentence should be equivalent in meaning with the uninformative sentence “The substratum of sound is the substratum of sound.” Udayana claims that the view that “ākāśa” is a descriptional name misconstrues the role of the descriptive clause. This, he says, functions simply as an indicative syndrome (upalakṣaṇayā taṭastha), which he compares with the role of the demonstrative “that” in “That is Devadatta.”The name “Devadatta” is not associated with any descriptive clause, yet its reference can be indicated by means of some other expression like a demonstrative, functioning as an heuristic device. The feature “being the substratum of sounds” is, for Udayana, just one way of many to show others which object is the referent of the term “ākāśa,” and has no conventional tie with this term.

318

JONARDON GANERI

Udayana also notes, as Praśastapāda had already, that “ākāśa” cannot be a naimittikī or term which refers on the basis of a natural kind, because of the doctrine, universally accepted in Nyāya-­Vaiśeṣika circles, that a natural kind (jāti) must be plurally instanced. Thus, even if we can form grammatically the abstract term “ākāśa-­hood” (“ākāśatva”), this term does not denote the “essence” of ākāśa, but simply the property of being distinct from other substances, i.e., the property ‘(λx)(∀y)(x ≠ y iff y ≠ ākāśa).’10 In Udayana’s three-­fold schema, terms like “ākāśa” belong to the same semantic category as proper names like “Devadatta,” both referring directly to their referents. We saw above that Śaṅkara Miśra, though substantially echoing Udayana’s arguments, retains an earlier distinction between these two classes. His proposal is that the term is semantically tied with a descriptive clause, a tie which is not definitional but evidential. Let us now see how later Nyāya authors developed these ideas and, in particular, how they modified Udayana’s schema. – – – LATER NAIYAYIKAS: VARDHAMANA, RAGHUNATHA –’ AND JAGADI SA

The Navya-­Naiyāyika Vardhamāna (c.  1440 c.e.) wrote an influential commentary on Udayana’s Kiraṇāvalī, the Kiraṇāvalī-­prakāśa, in which he sets down, perhaps for the first time, the classic definition of a “basis for reference” (pravṛtti-­nimitta) or meaning invariant for an ordinary nominal. He defines it as a feature which i occurs in all the individuals to which the term can refer, ii is itself a meaning relatum (vācya) of the term, and iii regulates the presentation of the term’s referents (pravṛttinimittaṃ ca vācyaṃ sadvācyavṛttitve sati vācyopasthāpakam, 1933: 34). The first clause states that the ‘basis’ delimits the domain of applicability of the term, while the second clause states that it gives the “descriptive content” of the term. The last clause states that the basis regulates the way the referent is grasped in the hearer’s understanding. Vardhamāna uses this definition to isolate a category of referring expressions called “indicator terms” (upalakṣaṇavatī).The features attached to these terms satisfy only the second and third clauses of the standard definition: they guide the interpretation of the term, but are not themselves ‘contained in its semantic power’ (tacca vācyopasthāpakamātram na tu śaktyantarbhūtam). In other words, the hearer must identify its referent as that which is f, but it is not the case that it simply abbreviates a description, “the f.”Vardhamāna significantly claims that “ākāśa” is an example of an “indicator” term, saying that “ākāśa” refers to that which is indicated by the feature “substratum-­of-­sound-­hood.” Vardhamāna does not believe that the reference of ordinary proper names can be explained in this way, and he therefore reintroduces the category of terms which single out their referents directly, a category into which he places names like “Caitra.” Following Udayana, but contradicting the earlier nomenclative practice, it is to these terms that he applies the label pāribhāṣikī. The stipulation or convention governing such a term concerns a particular object, and such terms lack any link with a descriptive or heuristic “basis” (pravṛttinimittopalakṣaṇa-­yogābhāve saṅketaviṣayasya vyaktiviśeṣasya).Vardhamāna adds

“ĀKĀŚA” AND OTHER NAMES

319

the intriguing remark that the referent of such a term must be presented perceptually.This is in sharp contrast with ākāśa, which is supposed to be imperceptible. Thus Vardhamāna’s pāribhāṣikī terms are something like ostensive names, and his “indicator” or upalakṣaṇavatī terms are rather more like theoretical or technical terms. Vardhamāna, then, arrives at a classification in which there are four, rather than three, nominal types. I have cited his definitions of the types in the Appendix. We have so far gathered the following facts about the Naiyāyikas’ use of the term “pāribhāṣikī.” The general practice was constrained by Praśastapāda’s assertion that “ākāśa” is a pāribhāṣikī term, and this led most of the authors to name accordingly the category into which they thought this term fell. This led to a certain verbal incongruity, at least for those authors who, following Udayana, did not regard this class as in any significant sense comprising defined or technical terms. Vardhamāna retains Udayana’s use of the term “pāribhāṣikī” to label the directly referring terms, but has then to introduce a new term “indicator name” to label the class into which he considers “ākāśa” to fall. The inappropriateness of the label “pāribhāṣikῑ” led later to a new nomenclative practice, namely to use this label to designate terms introduced by an explicit stipulation by some author (ādhunika-­saṅketa), perhaps for theoretical reasons or for the sake of economical expression, in contrast with terms which form part of the natural language, or whose introduction into the language is untraceable (c.f., ājānika-­saṅketa). Those who followed this new practice (especially Gadādhara, see below) could no longer regard the category of pāribhāṣikī terms as comprising only proper names, for there is no reason why general names (jāti-­śabda) cannot be introduced by explicit stipulation. Let me now turn to Raghunātha (c. 1500 c.e.). Raghunātha, in his sub-­commentary Kiraṇāvali-­prakāśa-­dīdhiti (1932: 47–49), discusses Vardhamāna’s theory, and appears to have some sympathy with it. He begins by reporting Udayana’s three-­fold classification, but suggests that it should be drawn in a different way. He defines a pāribhāṣikῑ term simply as an expression which refers to a single object (eka-­vyaktimātra-­bodhikā saṃjñā pāribhāṣikī), an aupādhikī term as an expression which refers (on different occasions) to a plurality of objects with the help of an imposed property (upādhi-­puraskāreṇāneka-­vyakti-­bodhikā ­aupādhikī), and a naimittikī term is an expression which refers (on different occasions) to a plurality of objects with the help of a natural kind property (jāti-­puraskāreṇa bodhikā ­naimittikī). The point for Raghunātha is that these divisions do not represent significant semantic distinctions; ontological differences in the “basis” (nimitta) do not affect semantic properties. Raghunātha’s claim is that a single semantic proposal can account for the behaviour of all nominal expressions.11 Let us see how this proposal applies to the case of the term “ākāśa.” Raghunātha endorses Vardhamāna’s suggestion that the basis need not itself be a meaning relatum (na tu śakyāntarbhūtam iti), but believes that this is true, not just for a restricted class, but for all nouns. He does not think that the basis loses all semantic significance, however. First, it still delimits the class of permissible referents for the noun (aśakyenāpi śakyānugamasaṃbhavāt) (c.f., clause (i) in Vardhamāna’s definition). Moreover, it can still, as Vardhamāna suggests, regulate the hearer’s interpretation of the term, by restricting the way in which the referent is represented in the hearer’s language-­based thought ­(śābdabodha). We are not forced to accept Udayana’s contention, that the referent is grasped in a pure unqualified state. Specifically as regards the term “ākāśa,” Raghunātha mentions a view according to which there is no conventional restriction on the mode under which ākāśa

320

JONARDON GANERI

is to be grasped. Someone who thinks of its referent under description “the substance different from the other eight”12is just as correct as one who thinks of its referent under the description “the substratum of sound.” The only conventional constraint is that they all associate the same object ākāśa with the word “ākāśa.” Raghunātha’s response to this proposal is interesting. He says that we should agree that there are variations in the descriptions to which different hearers attach the word “ākāśa” (c.f., vyutpattivaicitrāt). Yet, for the sake of economy (lāghavāt), we must regard one description as principal. Different hearers can associate the word with different descriptions without thereby excluding themselves from the linguistic practice in question, but one description, the one by means of which the reference of the term is fixed in the linguistic rule, has still a privileged status. Use of a particular description in this rule, the description “being the substratum of sound,” carries an implication, though nothing stronger, that this is the way the referent is to be thought of. Raghunātha’s discussion had a considerable impact, especially on two later Naiyāyikas, Jagadīśa (c. 1600) and Gadādhara (c. 1650). Jagadīśa wrote a commentary entitled Sūkti ( Jagadīśa, 1927), directly upon the Praśastapāda-­bhāṣya, one which is heavily influenced by later developments. He re-­employs, at least formally, Udayana’s three-­fold classification of nominal types, but like Raghunātha believes that the principles on which Udayana draws it are unsound. In particular, he does not accept Udayana’s claim that there can be referring expressions which pick out their referents unaided by any basis or descriptive feature. His reconstruction of the classificatory schema is as follows. A pāribhāṣikī term is now defined as a term the linguistic rule for which is delimited by a singly instanced property (ubhayāvṛtti-­ dharmāvacchinna-­saṅketavattvam eva pāribhāṣika-­saṃjñātvam). An aupādhikī term is defined as a word whose rule is delimited by a plurally instanced “imposed” property (anugatopādhyavacchinna-­). A naimittikī term is, as for Udayana, a word whose rule is delimited by a natural kind or universal.The difference between the first two categories is now purely numerical, for, as we have already seen, according to Nyāya-­Vaiśeṣika, a singly instanced property must be an imposed property. This way of drawing the tripartite distinction differs importantly from Raghunātha’s in that for Jagadīśa every nominal must have a descriptional basis. In fact, Jagadīśa explicitly defends the view that “ākāśa” has as its basis the imposed property “substratum-­of-­sound-­hood,” the descriptional doctrine criticised by Udayana and encoded in MCdescription. Udayana’s objection to this view was that it renders the sentence “ākāśa is the substratum of sound” vacuous. Jagadīśa replies that this is not correct, for the sentence should be construed as “The word “ākāśa” refers to the substratum of sound” (sahaprayogasyākāśapadavācyaḥ śabdavān ityarthakatvenāpyupapatter).That is to say, it should be read as meta-­linguistically asserting the linguistic rule governing the word “ākāśa.” He adds that Udayana’s view is unsatisfactory in its own right, for it is contrary to experience (anubhavavirodhāt; counter-­intuitive?) to suppose that the word “ākāśa” presents ākāśa by itself, i.e., in a pure unqualified form.This move reflects the Naiyāyikas’ growing discomfort with the idea that objects can appear as unconceptualised “givens” in thought.13 Jagadīśa also criticises the ‘indicative syndrome’ theory, mentioned in the five-­fold classification and supported by Śaṅkara Miśra. He says that since this view, which denies that terms like “ākāśa” have a descriptive content (śakyatāvacchedaka), is equally applicable to ordinary nouns like “[the] pot,” it would have the unacceptable consequence that they too do not possess descriptive content. Unfortunately, he does not elaborate on this remark,

“ĀKĀŚA” AND OTHER NAMES

321

and as it stands, there seems no reason why we cannot follow Śaṅkara Miśra in drawing a sharp line between the semantics of common nouns and that of the pāribhāṣikῑ terms. Gadādhara’s contribution to the development of Nyāya thought about pāribhāṣikī terms, contained in his tract Śaktivāda, is best read as a critique of the views of his predecessors. Gadādhara isolates five such views, traceable to some extent in the literature we have just reviewed. In the next section, I will examine Gadadhara’s criticisms of each of these theories, and conclude by considering his own view, which seems to be a modification of that of Jagadīśa. – GADADHARA’S CONTRIBUTION TO THE DEBATE

We have seen how a number of authors categorise the types of nominal expression.When the expression in question is a token-­reflexive common noun, the classifications are based either on the ontology of the invariant descriptive basis, or on the existence of significant etymological or logical structure. A greater part of the discussion, however, has concerned those expressions which are not token-­reflexive, expressions which, in virtue of the conventions which govern their use of interpretation, are semantically associated with one and only one object. Among the authors we have considered, it is Raghunātha who most clearly isolates this class, using the label “pāribhāṣikī” to reserve those referring expressions which designate exactly one object. The standard examples of these expressions are Praśastapāda’s names, “ākāśa,” “space” (“dik”) and “time” (“kāla”). Some authors include grammatically proper names such as “Caitra,” but others deny that these are strictly proper names. Following Gadādhara, we distinguish five primary theories about the semantics of names in this literature, which I will call the ‘direct’ theory, the ‘common core’ theory, the ‘indicative syndrome’ theory, Raghunātha’s ‘implicature’ theory, and Jagadīśa’s ‘descriptional’ theory. The ‘direct’ theory

Early Navya-­Nyāya authors, Udayana, Vardhamāna and Śaṅkara Miśra, recognise the existence of a class of name which “grasps by the horns” (śṛṅgagrāhikayā) its referent. Such a term refers directly – there is no descriptive feature, semantically tied to the term, on the basis of which it picks out an object. The proper names, like “Caitra” and “Devadatta,” as well as the demonstrative pronoun “that” (asau), are supposed to belong to this class, and Udayana, at least, includes even the pāribhāṣikī names, “ākāśa” and so on. The origins of this view in Nyāya-­Vaiśeṣika can be traced back at least to Jayanta (c. 850 c.e.), who remarked that Simply a particular is “meant” (ucyate) by a word in whose referents there is no common property. Thus in the case of a word like “Ḍittha,” understood to be a proper name, the “meaning” is a particular, since we cannot speak of a universal [Ḍitthahood]. We therefore call such a word a substance-­word.14

Associated with this theory is the slogan “even from a word, an object can be presented in a pure, unqualified state” (padād api nirvikalpakaṃ). Although we do not know who

322

JONARDON GANERI

first said this, Gadādhara (1927: 65) claims that it characterised the view of the “older” Naiyāyikas. In its crudest form this view seems to state that to understand such a term is to be in a state of “bare acquaintance” with it. The influence of Vyāḍi’s theory here is unmistakable. There is also a certain similarity between the view of these Naiyāyikas and Russell’s theory of logically proper names, which, though striking, should not be over-­stressed. Russell claimed that to understand a logically proper name, one must be in acquaintance with its referent, and believed that we are acquainted only with sense-­ data, and perhaps also universals, but not with ordinary objects. The Naiyāyikas who propound this view claim that it is not the names of sense-­date but names of ordinary objects which refer in this way. In this, we may note, they differ from those who would endorse Patañjali’s thesis about grammatically proper names.Vardhamāna, however, insists that the referent must be presented perceptually, and in doing so conformed with the general opinion in later Nyāya that if there are states of pure, construction-­free awareness of things, they must be perceptual states. So the most likely candidates for the direct theory are names and demonstrative pronouns referring to objects in the common perceptual field of speaker and hearer. The central drawback to this theory lies in its commitment to a doctrine of ‘pure acquaintance’ with the objects referred to. On behalf of the advocates of this theory, Gadādhara notes that this doctrine has a strong and a weak reading (1927: 65). According to the strong reading, it states that someone who understands a directly referring term grasps the referent as completely unqualified by any feature. The weak reading states that the referent is not grasped as qualified by any feature conventionally tied with the word, “ākāśa” say. However, in any sentence in which the uninflected nominal stem “ākāśa” occurs, there will be other words, nominal affixes, etc. and someone who understands the sentence will grasp ākāśa as qualified by their referents.15 For example, in the sentence “ākāśa is a substance,” the word “ākāśa” is construed with the masculine singular affix from the first triplet (i.e., nominative). So ākāśa will be qualified by singularity, etc. Gadādhara notes two problems even with this weaker version of the direct theory. He points out first that the direct theory is technically inconsistent with well-­ established Nyāya doctrines. The Nyāya claim that in the specification of the content of a thought, a qualifier of something may only be itself unqualified if it is either a universal (jāti) or a non-­composite imposed property (akhaṇḍopādhi). Ākāśa, however, is neither of these things, and yet becomes the qualifier in a sentence like “[A bird] is a sky-­dweller” (“ākāśavāsin”), in which it is not itself qualified by further elements in the sentence. Gadādhara’s second, and perhaps better, argument is as follows. The hearer is said to know that “ākāśa” refers to ākāśa, where ākāśa is not singled out by a description. However, in order to acquire such an item of dispositional, memory-­based knowledge, the hearer must have been introduced to ākāśa in some way, and, since ākāśa is imperceptible, this is presumably by means of some description. So the proponent of this view must somehow explain away the natural idea (which Gadādhara ascribes explicitly to Murāri Miśra), that in our dispositional or memory-­based knowledge of things we always grasp those things under the mode in which we were first presented with them.16

“ĀKĀŚA” AND OTHER NAMES

323

Gadādhara appears to think that this argument extends beyond the pāribhāṣikī terms, to ordinary proper names like “Ḍittha,” whose referents are perceptible (c.f.,Vardhamāna). His point, perhaps is that even if one can learn a word by acquainting oneself with its referent, one would only be said to understand that word if the knowledge so acquired grounds a dispositional, memory-­based knowledge of which object is its referent. To reach the conclusion that understanding cannot consist in a “pure,” qualification-­free (nirvikalpaka) grasp of the referent, he must supplement this observation with an argument to the effect that memory-­based thoughts are necessarily descriptive, and this seems improbable. Russell famously claimed that memory is a mode of acquaintance, that “we are immediately aware of what we remember, in spite of the fact that it appears as past” (1912, p.  26). And more recent literature points to the existence of “non-­conceptual memory states.”17 We should note, however, that Gadādhara recognised the difficulty, and even wrote a short tract on the question, the Nirvikalpakasmaraṇavāda, in which he tries to supply arguments to support his contention that memory cannot present a past object in a qualification-­free state. The existence of qualification-­free memory states seems indeed to have been a source of great controversy among later Naiyāyikas, presumably because on this question hinges the feasibility of a direct or realist theory of meaning for ordinary names.

The common core theory

Gadādhara considers a view which like the first view denies that the public conventions governing a name tie it to an associated descriptive clause, but avoids that view’s dependence on a doctrine of bare, unqualified acquaintance. He calls this the “indicator” theory (upalakṣita-­śakti-­vada), defining an “indicator” as a property whose sole function is to delimit the reference of the term, without entering the content of the thought.18 The meaning clause for the term would be MCindicator (∃x: substratum of sound) (“ākāśa” refers to x), where the sole function of the descriptive condition “substratum of sound” is to restrict the permitted values able to be taken by the quantifier, and hence is substitutable for any other co-­extensive predicate.19 To use a modern phrase, it merely “fixes the reference” of the term, but is not part of its meaning. Any other property which picks out the same object as referent will serve the purpose equally well. According to this view, each language user identifies the referent of “ākāśa” under a mode or descriptive feature, but the particular mode may vary from person to person. One hearer might identify ākāśa under the mode “the substance different from the other eight substances,” another under the mode “the substratum of sound.” There is no hard and fast rule.20 The only restriction on the idiosyncratic descriptive features is that they all have the same extension, viz. the object ākāśa. I think Wiggins has something similar in mind when he remarks that

324

JONARDON GANERI

the sense of a proper name simply consists in its having been assigned whatever reference it has been assigned: to know the sense of n is to know to which entity n has been assigned, a single piece of knowledge which may be given in countless different ways by countless different descriptions” (1975, p. 11).

Evans, citing this passage, remarks that “it is difficult to hold, of ordinary proper names, that there is some particular description semantically associated with the name” (1982, p. 48). One might imagine the linguistic community as comprising individuals each of whom identifies the referent of the name “Caitra” in their own private fashion: for one, Caitra is the brother of Devadatta, for another he is the man last seen in the launderette, etc. (c.f., vyutpattivaicitrāt – Raghunātha). But what guarantees that they all understand the same name “Caitra” in a public language is that for each it is true to say that they know that “Caitra” refers to Caitra, and it is the statement of this common core fact which should be the meaning clause for the name.Thus, the function of the domain restrictor in the meaning clause is simply to fix the extension of the term. It is, in Gadādhara’s jargon, an “indicator” or upalakṣaṇa. As an account of what is involved in understanding ordinary proper names, this theory has much to recommend it.21 But as applied to the pāribhāṣikī term “ākāśa,” it is less plausible. The common core theory now says that “ākāśa” refers to a certain object, which is indicated by the description “the substratum of sound” (śabdāśrayatvopalakṣita evākāśapada-­ śaktir). In this view, the only distinction which accrues to the description “the substratum of sound” is that it is the description associated by the name-­giver with the name; it has no other semantic salience. Both the direct and the common-­core theories are motivated by a desire to explain the apparent informativeness of the sentence “ākāśa is the substratum of sounds” by analysing it as having a simple name-­predicate form. Gadādhara, however, rejects this view, precisely because it does not accord a sufficiently distinctive status to the description “the substratum of sound” in the semantics of “ākāśa.”The sentence “ākāśa is the substratum of sounds” seems to be a priori; on the current proposal, however, the truth expressed by it is a posteriori for anyone who does not identify ākāśa under the description “the substratum of sounds,” and vacuous for anyone who does.22 The syndrome theory

The traditional form of the “indicator” theory seems rather different from Gadādhara’s reconstruction. The advocates of the five-­fold classification say that “the substratum of sound” is a “syndrome” (taṭastha) for the referent of “ākāśa.” This suggests that it has a more significant role in the recognition of the referent.Vardhamāna adds a remark which may give a clue to the nature of this role. He suggests that the feature is a sort of “interpretational heuristic” (vacyopasthāpaka). Perhaps his idea comes to this. Someone who understands the name “ākāśa” knows that there is associated with it a descriptive feature “the substratum of sound” which she is able to draw upon as an aid in identifying the referent of the name. However, it is not required for understanding that the person should think of

“ĀKĀŚA” AND OTHER NAMES

325

the referent under the description “the substratum of sound,” nor should this description matter to the hearer’s interpretation of the statement in which the name occurs, or to the truth-­value assigned to it. The description is simply a community-­wide “dossier of information” associated with the name.23 If this is right, then the view bears some resemblance to Donnellan’s analysis of the “referential use” of definite descriptions, according to which the descriptive clause is only a heuristic device for making a reference.24 Matilal seems to think that this is the way to understand the doctrine: a distinguisher or qualifier may identify an object in a totally irrelevant way, or it may be very pertinent to the identification of the object in the context of the further attribution of properties of it. I can identify an object as the man with the whisky-­glass in hand, and go on to say that he is a logician, where the distinguisher, viz. having a whisky-­glass in hand, is totally irrelevant to his being a logician. But if I go on to say that he is going to get drunk very soon, then at least the audience may understand that the identifying qualifier is relevant. Nyāya underlines the distinction by calling the relevant distinguishers viśeṣaṇa, and the irrelevant ones upalakṣaṇa. One can in this way identify an object by an irrelevant qualifier, and assign a proper name to it (1985: 387).

So it may be that the doctrine that a descriptive feature is a “syndrome” (taṭastha) or “interpretational heuristic” (vācyopasthāpaka) is a defensible one in some cases, such as the referential use of nominal descriptions. But once again, it does not seem satisfactory as an account of the relation between “ākāśa” and the description “the substratum of sound.” The reason is that this description is our only access to the referent of “ākāśa,” an entity which is by nature imperceptible. Someone who used the term “ākāśa” has no other means than the associated description to identify the referent, so the role of that description cannot be merely evidential. – tha’s theory Raghuna

Raghunātha has a quite different idea. His view is roughly as follows. The linguistic rule MCindicator (∃x: substratum of sound) (“ākāśa” refers to x) just states that “ākāśa” refers to ākāśa. However, the use of the descriptive phrase “the substratum of sound” is not simply an arbitrary way of fixing the reference of the name: the linguistic rule non-­semantically implicates that ākāśa should be identified under the mode “the substratum of sound.” In other words, this descriptive feature, as the mode under which ākāśa is presented in the public linguistic rule (saṅketīyaviśeṣyatāvacchedaka), regulates by implication the way hearers identify the reference of “ākāśa.” Raghunātha nevertheless agrees that different hearers identify ākāśa in different ways, and are not thereby linguistically incompetent. The point is that each such hearer is in accordance with what is literally said by the rule, even if they do not respect its non-­literal implication. Of the theories considered, Raghunātha’s is the first to assign to the descriptive clause a distinctive semantic, as opposed to evidential, role.This becomes clear when we consider

326

JONARDON GANERI

the sentence “ākāśa is the substratum of sound.” What this sentence literally states, on the current proposal, is that a certain object is the substratum of sound, and as such, has the same status as any contingent name-­predicate assertion. But anyone who fully understands the word “ākāśa” will know that the conventional implication of the linguistic rules guarantees that the sentence is true. So there would seem to be a clear sense in which the truth expressed by the sentence is a priori. This imaginative way of dealing with the peculiar semantics of descriptive names deserves further attention.25 Gadādhara criticises Raghunātha’s view in a terse and rather enigmatic passage (1927: 62). He suggests that there are certain terms whose linguistic conventions must be regarded as explicitly stating, and not merely implicating, the description or feature under which their reference is to be picked out. Gadādhara notes that Kaṇāda, the author of the Vaiśeṣika-­sūtra, gives the word “substance” (dravya) a technical meaning by introducing it via the rule “Let the word ‘substance’ be construed as about an entity under the mode possessor-­of-­qualities.”26 The point, presumably, is that such a rule fixes, not just the reference, but the descriptive content of the term as well. According to Gadādhara, however, there is no significant semantic difference between such terms, whose introduction can be traced to some explicit definition (c.f., ādhunịka-­saṃketa), and those which cannot, since the same account is given of what it is to understand them: to understand such a term is to know, in any given context, which object is its reference and how the reference is to be identified. Gadādhara believes that his own proposal not only adequately explains the semantics of these noun phrases and singular terms, but can even be extended into an account of pronominal reference. If this is correct, then his account is to be regarded as preferable on the grounds of having greater explanatory power. The descriptional theory

The last doctrine to be considered is the theory that names are distinguished from ordinary nouns only in that their associated descriptive feature is singly instanced. This view is most closely associated with Jagadīśa.27 On this view, it is a definitional truth that a name w refers to an object x iff x possesses a certain unique feature (or satisfies a definite descriptive condition) π.Views such as this are commonly called “description theories of names,” and names (if there are any) for which this theory is true are called descriptional names.The associated descriptive clause in the case of the name “ākāśa” is “the substratum of sound,” and its reference clause is: MCdescription (∀x) (“ākāśa” refers to x iffdefn x is the substratum of sound). According to Jagadīśa, in the case of ordinary proper names such as “Devadatta,” there is the feature Devadatta-­hood or the clause “the thing x such that x = Devadatta,” ­Jagadīśa apparently goes so far as to say that for every individual, there is a peculiar property unique to it, the property of being that individual (tadvyaktitva; compare the medieval notion of haecceity). One need not, however, agree with him about this in order to accept a descriptional theory of names. It is a consequence of this view that the link sentence

“ĀKĀŚA” AND OTHER NAMES

327

“ākāśa is the substratum of sound” is a tautology, but as we have already seen, Jagadīśa explicitly recommends a meta-­linguistic reading of this sentence, as asserting “the term ‘ākāśa’ refers to the substratum of sound.” This view sees names like “ākāśa” as theoretical terms introduced into the discourse to abbreviate a longer description. It should be born in mind, however, that the longer description is itself used to make a reference, and the description theory does not commit one to the Russellian thesis that such names are disguised quantifier expressions. The Nyāya, as already noted, claim that to understand such a name is to entertain a descriptive singular thought about its referent.28 There are, of course, special problems with the notion of having singular thoughts about imperceptible objects like ākāśa, or abstract objects like numbers, but if the notion of descriptive singular thoughts, thoughts about objects under descriptive “guises” (dharmitāvacchedaka), is at all defensible, then it might well be extended to what is not perceivable. Gadādhara himself tentatively accepts Jagadīśa’s view on the matter, but notes that it requires us to attach a non-­literal interpretation (lakṣaṇā) to the term “ākāśa” in the ‘combined’ sentence “ākāśa is the substratum of sound” (1927: 73), for we have there to construe it as synonymous with the phrase “the referent of ‘ākāśa.’” He therefore suggests, as an alternative, a way in which this term can be regarded as occurring literally. In what seems to be a statement of the modified version of the indicator theory, he says that: in order to explain [the possibility of] an interpretation under the mode “the substance different from the other eight substances” from the name “ākāśa,” used literally, one should admit that the hearer knows the reference under a mode [m, viz. the substance different from the other eight substances] which is not a semantic value, [and that this mode] regulates an interpretation [of the reference] under that mode [m].29

When a term behaves in this way, he calls it a “viśiṣṭāvācaka” name, a name for which the basis of application is not a meaning relatum (c.f., Vardhamāna’s definition of an uplakṣaṇavatī name). The idea is that anyone who understands the term must think about its referent in a certain way, but that there is no one such way fixed by the linguistic rules. That Gadādhara is not comfortable with this solution is clear from his remark that it “has to be accepted because there is nothing better,” and his unease might be due to the fact that the distinctive semantic role of the description “the substratum of sound” has once again dropped out of the picture. Gadādhara, in the end, leaves the matter open to further research. In the Śaktivāda, Udayana’s three-­fold distinction is dismantled, and replaced with a case-­by-­case examination of individual words. He does retain the term “pāribhāṣikī,” but defines it now as a noun whose convention is “novel” (ādhunika, 1927: 5). The pāribhāṣikī terms, for Gadādhara, are those whose introduction can be traced back to some explicit naming act. His discussion of the meaning of “ākāśa” reveals a fascinating debate that raged within Nyāya circles, and the importance they attached to the topic. We have been able to trace some of the participants in this debate, and the views they espoused, and in doing so hope also to have revealed something about the process by which schools like the Nyāya developed, a process which involved internal criticism and disagreement just as much as debate with philosophical adversaries.

328

JONARDON GANERI

APPENDIX: TABLE OF CLASSIFICATORY SYSTEMS The five-­fold schema mentioned by Udayana: Name

Definition

Example

1 2 3 4 5

. nimittopalaks.an.a-­rahite s’.rnga-­gra–hikaya– nimitte ‘ntarbhu–topalaks.an.e nirvacanikaya– nirnimitta eva tat.asthopalaks.an.e paribha–.san.ikaya– – – sanimitte tat.asthopalaks.an.a evam . praka rikaya . vis’es.avannimitte nimittasankocanikaya–

“Caitra” “pa–caka” “a–ka–s’a” “pr.thivi-­” “svarga”

‘direct’ ‘etymological’ ‘pa–ribha–s. iki-­’ ‘syndrome-­based’ ‘bundle-­based’

’ S’am . kara Mis ra’s five-­fold schema: Name

Definition

Example

1 2 3 4 5

s’.rn˙ga-­gra–hikaya– tat.asthopalaks.an.a antarbhu–topalaks.an.a sanimitta . nimittasankocaniya

“asau,” “Devadatta” “a–ka–s’a” “pa–caka” “go” “svarga”

‘direct’ ‘pa–ribha–s.iki- ­’ ‘etymological’ ‘basis-­possessing’ ‘bundle-­based’

Udayana’s three-­fold schema: Name

Definition

Example

1 ‘pa–ribha–s. iki- ­’

. nimittam antaren.a s’.rngagra–hikaya–

2 ‘aupa–dhiki- ­’ 3 ‘naimittiki- ­’

upa–dhi-­nimittaka ja–ti-­nimittaka

“a–ka–s’a,” “ka–la,” “dik” “pas’u” “ru–pa,” “gandha”

–na’s four-­fold schema: Vardhama Name

Definition

Example

1 2 ‘upalaks. an.avatı¯’ 3 ‘pa–ribha–s. ikı¯’

ja–tiru–papravr.ttinimittavatı¯ va–cyopastha–pakama–tra pravrttinimittopalaks.an.ayoga–bha–ve sam . keta-­vis. ’ ayasya vyaktivises. asya pratyaks. en.-­aupastha–paka, . s’.rngagra–hikaya– pacikriya–ntarbha–ven.a pa–ka–nuku–lakr.tiyogyata–ma–tre

“go” “a–ka–s’a” “Caitra”

4 ‘aupa–dhikı¯’

“pa–caka”

329

“ĀKĀŚA” AND OTHER NAMES

– tha’s three-­fold schema: Raghuna Name

Definition

1 ‘pa–ribha–s. iki-­’ 2 ‘aupa–dhiki-’­

ekavyaktima–trabodhika upa–dhi-­puraska–ren.a anekavyaktibodhika ja–ti-­puraska–ren.a bodhika

3 ‘naimittiki-’­

Example

Jagadı-­s’a’s three-­fold schema: Name

Definition

Example

1 ‘pa–ribha–s. iki-­’ 2 ‘aupa–dhiki-’­ 3 ‘naimittiki-’­

. ubhaya–vr.ttidharma–vacchinna-­sanketavat – anugatopa–dhyavacchinna-­sam . jña – (anugataja–tyavacchinna-­sam . jña )

“a–ka–s’a”

NOTES 1. ākāśakāladiśām ekaikatvād aparajātyabhāve sati pāribhāṣikyas tisraḥ saṅjñā bhavanti. Praśastapāda (1927: 308). The term “pāribhāṣikī” is usually translated as “technical term,” but in view of the wide range of theories about these terms, such a translation may easily become misleading. I prefer, then, to leave it untranslated. It would be interesting to investigate whether ­Paśastapāda’s use of this term is in any way influenced by the Pāṅinian term “paribhāsa” or “meta-­rule.” 2. I have discussed Vyāḍi’s theory in more detail in Ganeri (1995). 3. This is claimed by Śaṅkara Miśra, for whom see below. 4. A comparative list of this and the other nominal classificatory schema discussed here is given in the Appendix. 5. satyam acetena stambha-­kumbhādau devadatta-­saṃjñāyās tādrūpyenābhimatatvāt (1917: 26–7). 6. pāribhāṣikī yathākāśam iti iyam eva taṭasthalakṣaṇety ucyate śabdāśrayatvasyo-­palakṣaṇasya taṭasthatvāt (1917: 27). 7. For this distinction, see Matilal (1968: 50) and (1986: 417–8). 8. C.f., Jagadīśa (1930: 309): niravacchinna-­saṅketa-­śālitvam eva pāribhāṣikatvam … ityācāryamatam. 9. The intention behind the use of “… iffdefn …” in such a clause is that no other co-­extensive description can be substituted on the right hand side. This might be achieved better, though with greater theoretical commitment, by preceding the clause with a modal operator. An alternative (here I am indebted to Eberhard Guhe) would be to introduce some means by which we can name intentionally individuated properties, such as that of Bealer (1982). 10. C.f., Jagadīśa (1930: 309): na caivam ākāśatvam ityādiprayogo na syāt praktyar-­ thatāvacchedaka-­dharmasyaiva tvatalādyarthatvād iti vācyaṃ vyāvartaka-­dharmasyaiva tvatalādinā bhāva-­pratyayenābhidhānāt. 11. It should be noted, however, that he does not explicitly mention ordinary proper names here. 12. There are nine categories of substance in the standard Vaiśeṣika ontological scheme. See for example Annaṃbhaṭṭa (1983). 13. For more on this point, see Matilal (1988). 14. yeṣām artheṣu sāmānỵaṃ na sambhavati taiḥ punaḥ ucyate kevalā vyaktir … evam ḍitthādiśabdānāṃ saṃjñātvaviditātmanām abhidheyasya sāmānyaśūnyatvād vyaktivācitā ata eva hi dravyaśabda ity ­ucyate (Jayanta, 1936: 298).

330

JONARDON GANERI

15. nirvikalpakatvaṃ ca tadaṃśe tatpadopasthāpyāprakāratvaṃ, na tu tadaṃśe sāmānyato niṣprakāratvam (1927: 65). 16. tatra nirvikalpakarūpasya smaraṇasyānupapattiḥ, anubhavasya svasamāna-­prakāraka-­mātra-­smṛti-­ janakatvādīti (1927: 67). 17. See Evans (1982: esp. 239). 18. C.f., Gadādhara’s commentator, Harinātha: upalakṣaṇatvaṃ pada-janya-bodhīya-viṣayatāvacchedakatvāvacchinna-śaktiviśayatā-śunyatvam (1929: 91). This technical use of the term ‘upalakṣaṇa’ in the Śaktivāda has to be distinguished from the use of the term in earlier literature as meaning a heuristic device, a use dating back at least to Bhartṛhari. 19. tatprakāraka-­śābdabodhe tadaṃśe ‘natiriktaprasaktatvarūpa-śakyatāvacchedakat-vāvagāhitvenopalakṣte śaktijñānasya hetutvād (1927: 71). For details about restricted quantifiers, as denoted by such formulations as “(∃x: F) (Gx),” see Neale (1990: 41–43). 20. śaktyākāśapadād aṣṭa-­dravyātirikta-­dravyatvādinā ‘py ākāśa-­bodho bhavati na tu niyamataḥ śabdāśrayatvenaiva (1927: 70). 21. It is partially defended, for example, by Evans (1982: 399–400). However, the independence he alleges to exist between using a name to refer in a given practice and understanding the name does not seem to be acknowledged by the Naiyāyikas. 22. See also Viśvabandhu (1994). 23. C.f., Evans (1982: 399), who notes that even if everyone now associates the name “Homer” with the description “the author of the Iliad and the Odyssey,” it is still a proper name for Homer. 24. Donnellan (1966). Kaplan (1989) also recognises the existence of a heuristic element in reference. Discussing Donnellan’s usual example of a referential use of a definite description (“Who is the man drinking martini?”), Kaplan remarks that “according to my new view of what determines the referent of a demonstrative, the demonstration (here, the description) is there only to help convey an intention [to refer to the perceived object] and plays no semantical role at all.” He calls the description an “off-­the-­record element in language.” 25. The notion of non-­literal implicature is developed by Grice; see his (1989). 26. dravyapadād guṇavattvādinā dravyaṃ boddhavyam ityākāraka ādhunikaḥ kasya cit saṅketo dravyatvādyavacchinna-­viśeṣyatā-­sambandhena gṛhītas tatra tādṛśa-­saṅketa-­grahād dravyatvena dravya-­bodhāpatter guṇatvādi-­prakāraka-­bodhānupapatteś ca durvāratvād (1927: 62). 27. ubhayāvṛtti-­dharmāvacchinna-­saṅketavattvaṃ pāribhāṣika-­saṃjñātvam (1930: 309). 28. McCullock (1989: 291–2) is one modern author who entertains such a theory of descriptive names. 29. ākāśapadāc chaktyāṣṭadravyātirikta-­dravyatvādinā bodha-­nirvāhāya viśiṣṭāvācakā-­kāśādipadasya prakārāṃśe vācyatvānavagāhita-­śaktigrahasyaiva tad-­dharmāvacchinna-­viśeṣyakasya tatprakāranvatyabodhe hetutvam agatyāṅgīkāryam (1927: 73). Compare also the comment in the Mañjūṣā: tadrṣa-­ [aṣṭadravyātiriktadravyatva]-­dharma-­prakāraka-­śabdabodhe taddharmaṃ dharmitāvacchedakīkṛtya pada-­prakāraka-­janya-­śābdabodha-­viṣayatva-­niṣṭha-­prakāratā-­nirūpita-­viśeṣyatā-­saṃsargaka-­śakti-­ jñānaṃ hetur (ibid.). REFERENCES Annaṃbhaṭṭa. (1983). Tarkasaṃgrahadīpikā on Tarkasaṃgraha. Ed. and trans. G. Bhattacharya. Calcutta: ­Progressive Publishers. Bealer, G. (1989). Quality and Concept. Oxford: Oxford University Press. Donnellan, K. (1966). ‘Reference and Definite Descriptions’, Philosophical Review 75. Evans, G. (1982). The Varieties of Reference. Edited by J. McDowell. Oxford: Clarendon Press.

“ĀKĀŚA” AND OTHER NAMES

331

Gadādhara Bhaṭṭācārya. (1892). Śaktivāda.With Sudarśanācārya’s Ādarśa. Ed. Pt. Sudarśanācārya. Calcutta: Own Press. Gadādhara Bhaṭṭācārya. (1927). Śaktivāda.With Kṛṣṇa Bhaṭṭa’s Mañjūṣa, Mādhava Bhaṭṭācārya’s Vivṛtti and Sāhitya Darśanācārya’s Vinodini. Ed. Gosvami Damodara Sastri. Benares: Kashi Sanskrit Series no. 57. Gadādhara Bhaṭṭācārya. (1929). Śaktivāda. With Harinātha Tarkasiddhānta’s Vivṛtti. Ed. Gosvami ­Damodara Sastri. Benares: Kashi Sanskrit Series no. 77. Gadādhara Bhaṭṭācārya. (1933). Nirvikalpakasmaraṇavāda. In Vādavāridhi. Ed. B. Misra and Dh. Sastri. Benares: Chowkhamba Sanskrit Series nos. 421, 446. Ganeri, J. (1995). ‘Vyāḍi and the Realist Theory of Meaning’, Journal of Indian Philosophy 23: 403–408. Gaṅgeśa Upādhyāya. (1901). Tattvacintāmaṇi, śabdakhaṇḍa. With Mathuranātha Tarkavāgīśa’s Rahasyam and Jayadeva Miśra’s Āloka. Ed. Kamakhyanatha Tarkavagisa. Calcutta: Bibliotheca Indica. Grice, P. (1989). Studies in the Way of Words. Cambridge, Mass.: Harvard University Press. Jagadīśa Tarkālaṅkāra. (1930). Sūkti. In Praśastapāda (1930). Jayanta Bhaṭṭa. (1936). Nyāyamañjarī. Ed. S. N. Sukla. Benares: Kashi Sanskrit Series no. 106. Kaplan, D. (1989). ‘Demonstratives: An Essay on the Semantics, Logic, Metaphysics and Epistemology of Demonstratives and Other Indexicals’. In J. Almog, J. Perry, and H. Wettstein, eds., Themes From Kaplan. Oxford: Oxford University Press. McCullock, G. (1989). The Game of the Name. Oxford: Clarendon Press. Matilal, B.K. (1968). The Navyanyāya Doctrine of Negation. Cambridge, Mass.: Harvard University Press. Matilal, B.K. (1982). Logical and Ethical Issues in Religious Beliefs. Calcutta: University. Matilal, B.K. and Shaw, J.L. eds. (1985). Analytical Philosophy in Comparative Perspective. Dordrecht: Reidel. Matilal, B.K. (1986). Perception: An Essay on Classical Indian Theories of Knowledge. Oxford: Oxford University Press. Matilal, B.K. (1988). ‘Some Issues in Nyāya Realism’, addressed to the Oxford Radhakrishnan Centenary conference Realism and Non-­Realism. Stephen Neale. (1990). Descriptions. Cambridge, Mass.: MIT Press. Praśastapāda. (1927). Praśastapādabhāṣya. With Jagadīśa’s Sūkti, Padmanābha Miśra’s Setu and ­Vyomaśivācārya’s Vyomavatī. Ed. G. Kaviraja and Dh. Shastri. Benares: Chowkhamba Sanskrit Series no. 374. Praśastapāda. (1971). Praśastapādabhāṣya. With Udayana’s Kiraṇāvalī. Ed. J. S. Jetly. Baroda: Gaekwad ­Oriental Series 154. Raghunātha Śiromaṇi. (1930). Pratyakṣa-­maṇi-­dīdhiti (prāmāṇyavāda). With Gadādhara’s Gādādhari. Ed. Prativādibhayaṅkara Anantācārya. Raghunātha Śiromaṇi. (1932). Kiraṇāvalī-­prakāśa-­dīdhitti. Ed. Gopinatha Kaviraja. Benares: Princess of Wales Saraswati Bhavana Texts no. 38. Śaṅkara Miśra. (1917). Praśastapādabhāṣyatīkāsaṃgraha, including Śaṅkara Miśra’s Kaṇāḍarahasyam. Ed. V.P. Dvivedin. Benares: Chowkhamba Sanskrit Series 231 & 255. Udayana. (1911a). Nyāyavārtikka-­tātparya-­pariśuddhi. With Vardhamāna’s Nyāya-­nibandha-­prakāśa. Ed. V. P. Dvivedin and L. S. Dravida. Calcutta: Bibliotheca Indica, no. 192. Udayana. (1911b). Kiraṇāvalī. With Vardhamāna’s Prakāśa and Rucidatta Miśra’s Vivṛtti. Eds. S. ­Sarvabhauma (Fasc. I–III) and N.C. Vedantatirtha (Fasc. IV, 1956). Calcutta: Bibliotheque Indica 190. Reprinted in 1989 by the Asiatic Society, Calcutta. Udayana. (1967). Nyāyavārtikka-­tātparya-­pariśuddhi. See Gautama (1967). Udayana. (1971). Kiraṇāvalī. See Praśastapāda (1971). Vardhamāna Upadhyaya. (1911). Kiraṇāvalī-­prakāśa. See Udayana (1911b). Vardhamāna Upadhyaya. (1933). Kiraṇāvalī-­prakāśa. Ed. G. Kaviraja. Benares: Princess of Wales Saraswati Bhavana Texts no. 45. Viśvabandhu Tarkatīrtha. (1994). ‘Proper Names and Individuals’, in B. K. Matilal and A. Chakrabarti, eds., Knowing From Words. Dordrecht: Kluwer Academic Publishers.

P A R T

VIII LOGIC

CHAPTER

19

Semiotic conceptions in the Indian theory of argumentation* Bimal Krishna Matilal

Source: The Word and the World: India’s Contribution to the Study of Language (Delhi: Oxford University Press, 1991), pp. 156–166. DEBATE AND LOGIC

Logic developed in ancient India from the tradition of vādavidyā, a discipline dealing with the categories of debate over various religious, philosophical, moral and doctrinal issues. There were several vāda manuals available around the beginning of the Christian era.They taught students how to conduct debate successfully, what tricks to learn, how to find loopholes in the opponent’s position, and what pitfalls to be wary of. Of these manuals, the one found in the Nyāyasūtras of Akṣapāda Gautama (c. ad 150) was perhaps the most systematic. We shall follow it in this exposition. Debates in Akṣapaāda’s view, can be of three types; (i) an honest debate (called vāda) where both sides, proponent and opponent, are seeking truth, i.e., wanting to establish the right view; (ii) a tricky debate (called jalpa) where the goal is to win by fair means or foul; and (iii) a destructive debate (called vitaṇḍā) where the goal is to defeat or demolish the opponent, no matter how. This almost corresponds to the cliché in English: the good, the bad and the ugly. The first kind signals the employment of logical arguments, use of rational means and proper evidence to establish a thesis. It is said that the participants in this kind of debate were the teacher and the student, or the students themselves, belonging to the same school. The second was, in fact, a ‘winner-­takes-­all’ situation. The name of the game was wit or intelligence. Tricks, false moves, and unfair means were allowed according to the rule of the game. But if both the debaters were equally clever and competent, this could be kept within the bounds of logic and reasoning. Usually two teachers of different schools 335

336

BIMAL KRISHNA MATILAL

would be participants.This used to take place before a board of jury called the ‘madhyastha’ (the mediators or adjudicators) and a chairman, usually a king or a man with power and money who would organize the debate. The winner would be declared at the end by the consensus of the adjudicators. The third type was a variety of the second type, where the winner was not supposed to establish his own position (he may not even have had a position) but only to defeat the opponent using logical arguments, or as the case was, tricks or clever devices. It was explicitly destructive and negative, hence philosophers like Vātsyāyana (c. ad 350) denounced this form of debate in unambiguous language. Again, a clever and competent opponent might force the other side into admitting a counter-­position (‘If you refute my thesis p, then you must admit the thesis not-­p, therefore, please establish your thesis’), and if the other side yielded, the debate was decided in favour of the former, or it would turn into the second form of debate. The notoriety of the third type was universal, although some philosophers argued (cf. Nāgārjuna, Śrīharṣa) that if the refutations of the opponent were done on the basis of good reason and evidence (in other words, if it followed the model of the first type, and the second type) then lack of a counter-­thesis, or non-­establishment of a counter-­thesis would not be a great drawback. In fact, it could be made acceptable and even philosophically respectable. That is why Gauḍa Sānātani (quoted by Udayana, see Matilal, 1986, 87) divided the debates into four types; (i) the honest type (vāda), (ii) the tricky one ( jalpa), (iii) the one modelled after the tricky one but where only refutation is needed, and (iv) the one modelled after the honest one where only the refutation of a thesis is needed. Even the mystics would prefer this last kind, which would end with a negative result (see Appendix I). Apart from developing a theory of evidence (pramāṇa) and argument (tarka) needed for the first type of debate, the manuals go on to list a number of cases or situation-­types where the debate will be concluded and one side will be declared as ‘defeated’ (or nigraha-­ sthāna, the defeat situation or the clinchers). The Nyāyasūtra lists 22 of them. For example, (a) if the opponent cannot understand the proponent’s argument, or (b) if he is confused, or (c) if he cannot reply within a reasonable time limit—all these will be cases of defeat. Besides, these manuals identify several standard ‘false’ rejoinders or jāti (24 of them are listed in the Nyāyasūtra), as well as some underhanded tricks (chala) like equivocation and confusion of the metaphor for the literal. – ya model A. The Nya

Akṣapāda defines a method of philosophical argumentation, called the nyāya method or the nyāya model. This is the symbol for an ideally organized philosophical disputation. Seven categories are identified as constituting the ‘prior’ stage of the nyāya. It starts with an initial doubt whether p or not-­p, and ends with a decision p (or not-­p as the case may be). The seven categories are: Doubt, Purpose, Example, Basic Tenets, the ‘limbs’ of the formulated reasoning, Supportive Argument, and Decision. The first two categories are self-­explanatory. The ‘example’ is needed to ensure that the arguments would not be just ‘empty’ talk. Some of the basic tenets supply the ‘ground rules’ for the argumentation.

SEMIOTIC CONCEPTIONS

337

The ‘limbs’ were the most important formulation of the structure of  a  logical reasoning. This was a landmark in the history of Indian logic. According to the Nyāyasūtras, there are five ‘limbs’ or ‘steps’ in structured reasoning. They should all be articulated into language. The first step is the ‘statement of the thesis’, the second the statement of reason or evidence, the third would be citation of an example (a particular case, well recognized and acceptable to both sides) which will illustrate the underlying (general) principle and thereby support the reason or evidence. The fourth is the showing of the present thesis as a case that belongs to the general case, for reason or evidence is essentially similar to the example cited. The fifth is the assertion of the thesis again as proven or established. Here is the time-­honoured illustration: Step 1  Step 2  Step 3  Step 4  Step 5 

There is fire on the hill. For there is smoke. (Wherever there is smoke, there is fire), as in the kitchen. This is such a case (smoke on the hill). Therefore there is fire on the hill.

The Buddhists and others argued that this was too elaborate for the essential structure. All we need would be the first two or the first three. The rest would be redundant. But the Nyāya school asserted all along that this nyāya method is used by the arguer to convince the others, and hence to completely satisfy the ‘expectation’ (ākāṃkṣā) of the other, you need all the five ‘limbs’ or steps. This is in fact a full-­fledged articulation of an inference schema. The ‘supportive argument’ is needed when doubts are raised about the implication of the middle part of the above inference schema. Is the example right? Does it support the evidence? Is the general principle right? Is it adequate? The ‘supportive arguments’ would examine the alternative possibilities, and try to resolve all these questions. After the ‘­supportive argument’ comes the ‘decision’ one way or another. Another seven categories were identified as constituting the ‘posterior stage’ of the Nyāya method. They consist of three types of debate (already mentioned): the group of trickery, the false rejoinders and the clinchers (also noted already), and the other important logical category, that of pseudo-­reason or pseudo-­evidence. Pseudo-­evidence is similar to evidence or reason, but it lacks adequacy or the logical force to prove the thesis adduced. It is in fact an ‘imposter’.The Nyāyasūtra notes five such varieties. Although these five varieties were mentioned throughout the history of the Nyāya tradition (with occasional disagreement, e.g., Bhāsarvajña, who had six), they were constantly redefined to fit the developing logical theories of individual authors. The five were: the deviating, the contradictory, the unestablished or unproven, the counter-­balanced, and the untimely. Since there can be fire without smoke (as in a red-­hot iron-­r ing), if somebody wants to infer presence of smoke in the kitchen on the basis of the presence of fire there, his evidence would be pseudo-­evidence called the ‘deviating’. Where the evidence (say, a pool of water) is usually the sign for the absence of fire, rather than its presence, it is called the ‘contradictory’. An evidence-­reason must itself be established or proven to exist, if it has to establish something else. Hence, an ‘unestablished’ evidence-­reason is pseudo-­evidence or  a  pseudo-­sign.  A  purported evidence-­reason may be countered by  a  purported

338

BIMAL KRISHNA MATILAL

counter-­evidence showing the opposite possibility. This will be a case of the ‘counter-­ balanced’. An ‘untimely’ is one where the thesis itself precludes the possibility of adducing some sign as being the evidence-­reason by virtue of its incompatibility with the thesis in question. It is called ‘untimely’ because as soon as the thesis is stated, the evidence will no longer be an evidence.

B. The sign and the signified

All this implicitly spells out a theory of adequate sign. What we have been calling ‘evidence’, ‘reason’ and sometimes ‘evidence-­reason’ may just be taken to be an adequate or ‘logical’ sign.The Sanskrit word for it is liṅga, ‘a sign’ or ‘a mark’, and what it is a sign for is called liṅgin, the ‘signified’, the ‘marked’ entity. This is finally tied to their theory of sound inference, that is, inference of the signified from the observation of the logical sign.This is the pre-­theoretical notion of ‘sign-­signified’ connection, as explained here. Note that this notion of ‘sign-­signified’ relation is different from the ‘signifier-­signified’ relation that we talked about in chapter 11. A sign is adequate or ‘logical’ if it is not a pseudo-­evidence, that is, a pseudo-­sign. And the five types of pseudo-­sign have already been identified. This is a negative formulation of the adequacy of the sign.  A  little later on in the tradition the positive formulation was found. The fully articulated formulation is found in the writings of the well-­known ­Buddhist logician, Diṅnāga (in his theory of the ‘triple-­character’ reason, see below). In fact, an adequate sign is what should be non-­deviating, that is, it should not be present somewhere when the signified is absent. If it is present in that location, it would be ‘deviating’. Thus, the identification of the first pseudo-­sign captured this intuition, although it took a long time to get this fully articulated in the tradition. This may be called logical in this sense for it ensures the correctness of the resulting inference. Thus, we have to say: if the sign is there, can the signified be far behind? C. The triple nature of the sign

Diṅnāga (c. ad 500) formulated the following three conditions which a logical sign must fulfil: 1 It should be present in the case under consideration. 2 It should be present in a similar case or a homologue. 3 It should not be present in any dissimilar case, any heterologue. Three interrelated technical terms are used here. The ‘case under consideration’ is called a pakṣa, the ‘subject-­locus’. The ‘similar case’ is called a sapakṣa, the ‘homologue’. The ‘dissimilar case’ is called a vipakṣa, the ‘heterologue’. These three concepts are also defined by the theory. The context is that of inferring a property A (the signified in our new vocabulary) from the property B (the sign) in a location S. Here the S is the pakṣa, the subject-­locus. The sapakṣa is one which already possesses A, and is known to be so.

339

SEMIOTIC CONCEPTIONS

And the vipakṣa is one which does not possess A. The ‘similarity’ between the pakṣa and the sapakṣa is variously explained. One explanation is that they would share tentatively the signified A by sharing the sign B. An example would make it clear. Smoke is a sign of fire on a hill, because it is present on that hill, and it is also present in a kitchen which is a locus of fire, and it is absent from any non-­locus of fire. The third condition is easily explained. The sign must not be present when the signified is not present. For otherwise, as we have already noted, the sign will be deviating, and would be a ‘pseudo-­sign’. Why the second condition? Did Diṅnāga overshoot his mark? Is not the second condition redundant (for the first and the third seem to be sufficient to guarantee adequacy)? These questions were raised in the tradition by both the Naiyāyikas like Uddyotakara (c. ad 550), and the Buddhists like Dharmakīrti (c. ad 650). Some, such as Dharmakirti, maintained that it was slightly repetitious, but not exactly redundant.The second condition states positively what the third, for the sake of emphasis, states negatively.The second is here rephrased as: the sign should be present in all or some sapakṣas.The contraposed version can then be formulated with a little ingenuity as: the sign should be absent from all vipakṣas. For sapakṣa and vipakṣa, along with the pakṣa, exhaust the universe of discourse. Other interpreters try to find additional justification for the second condition to argue against the ‘redundancy’ charge. The interpretation becomes complicated, and we are not going into the details here (see various papers and the introduction in Matilal and Evans, 1986). Logically speaking, it seems that the second is redundant, but epistemologically speaking, this may be needed, i.e. a case of the co-­presence of A and B, to suggest the possibility, at least, that one may be the sign for the other. Perhaps, Diṅnāga’s concern here was epistemological. . – D. The wheel of reason-­sign of Dinna ga

When a sign is identified, it has three possibilities. It may be present in all, some or none of the sapakṣas. And likewise, it may be present in all, some or none of the vipakṣas. To identify a sign, we have to assume that it is present in the pakṣa, however (i.e., the first condition is already satisfied). Combining these, Diṅnāga constructed his ‘wheel of reason’ with nine distinct possibilities, which may be tabulated as follows: 1 + vipakṣa + sapakṣa

2 − vipakṣa + sapakṣa

3 ± vipakṣa + sapakṣa

4 + vipakṣa − sapakṣa

5 −vipakṣa − sapakṣa

6 ± vipakṣa − sapakṣa

7 + vipakṣa ± sapakṣa

8 − vipakṣa ± sapakṣa

9 ± vipakṣa ± sapakṣa

all = +

some = ±

none = −

340

BIMAL KRISHNA MATILAL

Of these nine possibilities, Diṅnāga asserted that only two are illustrative of sound inference for they meet all the three conditions. They are nos. 2 and 8. Notice that either (− vipakṣa and + sapakṣa), or (− vipakṣa and ± sapakṣa) would fulfil the required condition. Diṅnāga is insistent that at least one sapakṣa must have the positive sign. No. 5 is not a case of sound inference. This sign is a pseudo-­sign. For although it satisfies the two conditions 1 and 3 in C above, it does not satisfy condition 2. So one can argue that as far as Diṅnāga was concerned all the three were necessary conditions. The second row does not satisfy condition 2 and hence none of nos. 4, 5 and 6 are logical signs. They are pseudo-­signs. Nos. 4 and 6 are called ‘contradictory’ pseudo-­signs—an improvement upon the old Nyāyasūtra definition of contradictory. The middle one, no. 5, is called ‘uniquely deviating’ (asādhāraṇa), perhaps for the reason that this sign becomes an unique sign of the pakṣa itself, and is not found anywhere else. In Diṅnāga’s system, this sign cannot be a sign for anything else, it can only point to itself reflexively or to its own locus. Nos. 1, 3, 7 and 9 are also pseudo-­signs. They are called the ‘deviating’ signs, for in each case the sign occurs in some vipakṣa or other, although each fulfils the second condition.

E. Development of the wheel by Uddyotakara

Diṅnāga’s system of nine reason-­types or sign-­types was criticized by Uddyotakara, the Naiyāyika, who argued that it was incomplete because he did not consider at least two further alternatives: (a) a situation-­type where there is no sapakṣa, and (b) a situation-­type where there is no vipakṣa. The sign’s absence from all sapakṣas (or all vipakṣas) should be distinguished from these two situations. Let us use ‘0’ for the situation-­type which lacks any sapakṣa or vipakṣa and ‘−’ for the situation-­type where the sign is present in no ­sapakṣa or no vipakṣa (as before). Hence combining the four possibilities,  +  sapakṣa,  ± sapakṣa, − sapakṣa, 0 sapakṣa (no sapakṣa) with the other four (+, ±, −, 0) vipakṣa, we get sixteen in our wheel of sign, and the new wheel contains more sound inferences, i.e., adequate signs. For example, This is nameable, because this is knowable.

Here ‘knowability’ is the sign, adequate and logical, for showing the nameability of an entity, for (in the Nyāya system) whatever is knowable is also nameable (i.e., expressible in language). Now we cannot have a heterologue or vipakṣa here, for (again, according to the Nyāya system) there is nothing that cannot be named (or expressed in language). Within the Buddhist system, another example of the same argument-­type would be: This is impermanent because it is a product. For in Buddhism, everything is impermanent and a product. Later Naiyāyikas called this type of sign ‘kevalānvayin’, the universal-­ positive-­sign, that is, it is the characteristic of every entity. Uddyotakara captured another adequate reason or logical sign, but he formulated the reasoning or inference negatively, i.e., in terms of a counterfactual.This was done probably to avoid the doctrinal quandary of the Nyāya school (to which he belonged) in which

SEMIOTIC CONCEPTIONS

341

explanation of analytical or  a  priori knowledge always presents  a  problem. His typical example was: The living body cannot be without a soul, for if it were it would have been without life.

This is the generalized inference called ‘universal negative’—‘kevalavyatirekin’ in the tradition. The subject S which has a unique property B cannot be without A, for then it would have been without B.

Since B is a unique property of S, and since the presence of A and B mutually imply each other, there is no sapakṣa. But it is a correct inference. Bhāsarvajña (ad 950) did not like the rather roundabout way of formulating the inference-­type. He said, The living body has a soul, for it has life.

But this would verge on unorthodoxy in Nyāya, for: (a) the statement of the thesis includes the sign already, and (b) there seems to be a necessary connection between having life and having a soul. The later Nyāya went back to the negative formulation but got rid of the reflex of the counterfactual that Uddyotakara had. If A and B are two properties mutually implying each other such that B can be the definiens (lakṣaṇa) and the class of those possessing A can be the definiendum, then the following inference is correct: The subject S differs from those that are without A, for it has B. (And A is defined in terms of B.)

This seems to be equivalent to. S has A, for it has B.

The verbal statement ‘S has A because it has B, however, does not expose fully the structure of this type of inference. For one thing, in this version it becomes indistinguishable from any other type of correct inference discussed before. In fact, the special feature of this type of inference is that the inferable property A is uniquely present in S only, and nowhere else, and hence our knowledge of the concomitance or pervasion between  A  and  B  cannot be derived from an example (where their co-­presence will be instantiated) which will be a different case from the S, the case under consideration. In fact,  S  here is  a  generic term and it will be proper to say: all S’s have A, for they have B, and a supporting example will have to be an S, i.e., an instance of S. To avoid this anomaly, a negative example is cited to cover these cases. Thus we can say, a non-­S is a case where neither A nor B is present. This will allow one to infer, for example, absence of B from absence of A and also (since A and B are co-­present in all cases) absence of A from absence of B. But the evidence here is B. Hence by seeing B in all S’s we can infer absence of absence of A. Such a roundabout formulation was dictated by the peculiar nature of the Diṅnāga-­Uddyotakara theory of inference.

342

BIMAL KRISHNA MATILAL

To explain: In this theory, what legitimizes the inference of  A  from the sign  B  is the knowledge that B is a logical sign of A, and to have that knowledge we must have another knowledge that B has concomitance, i.e., invariable connection with A, and the second knowledge must be derived empirically, i.e., from an example where it is certain that A as well as B is present. Without such an example, we would not recognize B to be a logical sign of A. This limitation precluded the possibility of inferring A from B, where the case is such that all that have A are included in pakṣa, the subject-­locus of the inference. For the convention is that the said example cannot be chosen from the members of the pakṣa, i.e. of the set of S’s. Hence the difficulty. Uddyotakara saw this problem and extended the scope of the theory by saying that in these cases, a negative example, a non-­S having neither A nor B, and absence of any counter-­example (the sign’s absence from all vipakṣas) will be enough to legitimize the inference. Udayana (c. ad 1050) later on defended this type of inference as legitimate. For, he said, if we do not admit such inferences as valid then our search for a defining property of some concepts could not be justified. Suppose we wish to define cow-­hood: what is the unique property of a cow? Now, suppose having a dewlap is a unique property of a cow; it exists in all and only cows. What is the purpose of such a ‘definition’, if we can call it a definition (lakṣaṇa)? We can differentiate all cows from non-­cows. How? We do it by the inference: Cows are distinct from non-­cows, for cows have dewlaps. Of course, ‘cows are distinct from non-­cows’ is equivalent to ‘cows are cows’, but when it is negatively put, the purpose of such inference becomes clearer. F. Concomitance or invariable relation

Diṅnāga defined the invariable relation or concomitance of B with A, which legitimizes the inference of the ‘signified’ from the sign B, as follows (in Pramāṇasamuccaya): liṅge liṅgi bhavaty eva liṅginy eva itarat punaḥ When the sign (liṅga) occurs there, the signified, that of which it is a sign, has to occur there also. And if the sign has to occur somewhere, this has to occur only where the signified is occurrent.

This has been quoted frequently by the Naiyāyikas, Jainas and other logicians. This actually amounts to saying that all cases of B are cases of A, and only cases of A could be cases of B. Dharmakīrti (c. ad 630) described the invariable connection in two ways. A sign could be the ‘own-­nature’ or essential mark of B. That amounts to saying that B is either an invariable or a necessary sign of A. Thus, we infer that something is a tree from the fact that it is a beech tree, for a beech tree cannot be a beech tree without being a tree. This only defines invariability or necessary connection. The second type of sign is one when we infer the ‘natural’ causal factor from the effect, as we do infer fire from smoke. It is also the nature or the essence of smoke that it cannot originate without originating from fire. Hence invariable relation means: (i) an essential or necessary property of the class, and (ii) a causally necessary relation between an effect and its invariable cause.

343

SEMIOTIC CONCEPTIONS

The later Naīyāyikas said that the absence of a counter-­example is what is ultimately needed to legitimize the inference-­g iving relation between  A  and B. If  B  is the sign then B would be the logical sign if and only if there is no case where B occurs but A does not occur. If A occurs where B does not, that would be a counter-­example to the tacitly assumed rule of inference, ‘If B then A’. As we know from the truth-­table of the propositional logic, ‘If B then A’ is falsified only under one condition, when there is not-­B and A. Thus Gaṅgeśa (14th century ad) defines this relation (one of the fourteen definitions): B’s non-­occurrence in any location characterized by absence of A.

Alternatively, another definition is given: B’s co-­occurrence with such an A as is never absent from the location of B.

The first is a rephrasing of the first definition of vyāpti in the Vyāptipañcaka of Gaṅgeśa. The second is an abbreviation of what is called his siddhāntalakṡaṇa. NOTE * Semiotics is  a  modern academic discipline. The conception of sign and the signified is however one of very ancient origin. Modern semioticians are interested in extending their notion of sign and thus semiotics has become an all-­inclusive discipline.The old connection between philosophy of language and semiotics is obvious.The material in this Appendix was written as part of a larger article (in collaboration with Dr Jogesh C. Panda) under the title ‘Sign Conception of India’. The article is due to appear in an Encyclopedia of Semiotics. BIBLIOGRAPHY Diṅnāga. In (1) M. Hattori, Diṅnāga, On Perception, Cambridge (Mass.), 1968; and (2) Muni Sri ­Jambuvijayaji, Dvādaśāranayacakra, Parts I & II, Bhavanagar, 1966 & 1976. Matilal, B. K. Perception, Oxford, Clarendon Press, 1986. Matilal, B. K. and R. Evans (eds). Buddhist Logic and Epistemology, D. Reidel Publishing Co., Holland, 1986. Nyāyasūtras of Akṣapāda Gautama, Nydyādarśana, ed. T. Tarkatīrtha and A. Tarkatīrtha, Kyoto, Rinsen edn. 1982. Uddyotakara. Nyāya-­vārttika, see Nyāyasūtra.

CHAPTER

20

Jaina logic and the philosophical basis of pluralism Jonardon Ganeri

Source: History and Philosophy of Logic, 2002, 23:4, 267–281. There was in classical India a great deal of philosophical activity. Over the years, certain questions came to be seen as fundamental, and were hotly contested. Are there universals? Do objects endure or perdure? Are there souls, and, if so, are they eternal or non-eternal entities? Do there exist wholes over and above collections of parts? Different groups of philosophers offered different answers to these and many other such questions, and each, moreover, was able to supply plausible arguments in favour of their position, or to offer a world-view from which their particular answers seemed true. The body of philosophical discourse collectively contained, therefore, a mass of assertions and contradictory counter-­ assertions, behind each of which there lay a battery of plausible arguments. Such a situation is by no means unique to philosophical discourse. Consider, for instance, the current status of physical theory, which comprises two sub-theories, relativity and quantum mechanics, each of which is extremely well supported, and yet which are mutually inconsistent. The same problem is met with in computer science, where a central notion, that of putting a query to a data-base, runs into trouble when the data-base contains data which is inconsistent because it is coming in from many different sources. For another example of the general phenomenon under discussion, consider the situation faced by an investigator using multiple-choice questionnaires, when the answers supplied in one context are in conflict with those supplied in another. Has the interrogee said ‘yes’ or ‘no’ to a given question, when they said ‘yes’ under one set of conditions but ‘no’ under another? Do their answers have any value at all, or should we simply discard the whole lot on account of its inconsistency? Perhaps the most apposite example of all is the case of a jury being presented with the evidence from a series of witnesses. Each witness, we might suppose, tells a consistent story, but the total evidence presented to the jury might itself well be inconsistent. 344

THE PHILOSOPHICAL BASIS OF PLURALISM

345

The situation the Jainas have in mind is one in which a globally inconsistent set of propositions, the totality of philosophical discourse, is divided into sub-sets, each of which is internally consistent. Any proposition might be supported by others from within the same sub-set. At the same time, the negation of that proposition might occur in a distinct, though possibly overlapping subset, and be supported by other propositions within it. Each such consistent sub-set of a globally inconsistent discourse, is what the Jainas call a ‘standpoint’ (naya). A standpoint corresponds to a particular philosophical perspective. Let us say that a proposition is arguable if it is assertible within some standpoint, i.e. if it is a member of a mutually supporting consistent set of propositions.The original problem posed was this: what is the rational reaction to a class of propositions, each of which is, in this sense, arguable, yet which is globally inconsistent? It seems that there are three broad types of response.The first, which I will dub doctrinalism, is to say that it will always be possible, in principle, to discover which of two inconsistent propositions is true, and which is false. Hence our reaction should be to reduce the inconsistent set to a consistent subset, by rejecting propositions which, on close examination, we find to be unwarranted. This is, of course, the ideal in philosophical debate, but it is a situation we are rarely if ever in. The problem was stipulated to be one such that we cannot decide, as impartial observers, which of the available standpoints, if any, is correct. If doctrinalism were the only option, then we would have no choice but to come down in favour of one or other of the standpoints, basing our selection, perhaps on historical, cultural or sociological considerations, but not on logical ones. A second response is that of scepticism. Here the idea is that the existence both of a reason to assert and a reason to reject a proposition itself constitutes a reason to deny that we can justifiably either assert or deny the proposition. A justification of a proposition can be defeated by an equally plausible justification of its negation.This sceptical reaction is at the same time a natural and philosophically interesting one, and indeed has been adopted by some philosophers, notably Nāgārjuna in India and the Pyrrhonic sceptics as reported by Sextus Empiricus. Sextus, indeed states as the first of five arguments for scepticism, that philosophers have never been able to agree with one another, not even about the criteria we should use to settle controversies. The third response is that of pluralism, and this is the response favoured by the Jainas. The pluralist finds some way conditionally to assent to each of the propositions, and she does so by recognising that the justification of a proposition is internal to a standpoint. In this way, the Jainas try “to establish a rapprochement between seemingly disagreeing philosophical schools (Matilal 1977: 61)”, thereby avoiding the dogmatism or “one-­sidedness” from which such disagreements flow. Hence another name for their theory was anekāntavāda, the doctrine of “non-one-sidedness” (for a good outline of these aspects of Jaina philosophical theory, see Matilal 1977 and Dundas 1992). In spite of appearances to the contrary, the sceptic and the pluralist have much in common. For although the sceptic rejects all the propositions while the pluralist endorses all of them, they both deny that we can solve the problem by privileging just one position, i.e. by adopting the position of the doctrinalist. (It seems, indeed, that scepticism and pluralism developed in tandem in India, both as critical reactions to the system-based philosophical institutions.) Note too that both are under pressure to revise classical logic. For the sceptic, the problem is with the law of excluded middle, the principle that for all p, either p or ¬p.

346

JONARDON GANERI

The reason this is a problem for the sceptic is that she wishes to reject each proposition p without being forced to assent to its negation ¬p. The pluralist, on the other hand, has trouble with a different classical law, the law of non-contradiction, that for all p, it is not the case both that p and that ¬p, for she wishes to assent both to the proposition p and to its negation. While a comparative study of the two responses, sceptical and pluralist, would be of interest, I will here confine myself to developing the version of pluralism developed by the Jainas, and discussing the extent to which their system becomes paraconsistent. It is very often claimed that the Jainas ‘embrace’ inconsistency, but I will be arguing that this is not so, that we can understand their system by giving it a less strongly paraconsistent reading. 1.  JAINA SEVEN-VALUED LOGIC

The Jaina philosophers support their pluralism by constructing a logic in which there are seven distinct semantic predicates (bhaṅgī), which, since they attach to sentences, we might think of as truth-values (for a rather different interpretation, see Ganeri 2001, chapter 5). I will first set out the system following the mode of description employed by the Jainas themselves, before attempting to reconstruct it in a modern idiom. I will follow here the twelfth century author Vādideva Sūri (1086–1169 ad), but similar descriptions are given by many others, including Prabhācandra, Malliṣena and Samantabhadra.This is what ­Vādideva Sūri says (Pramāṇa-naya-tattvālokālaṅkāraḥ, chapter 4, verses 15–21): The seven predicate theory consists in the use of seven claims about sentences, each preceded by “arguably” or “conditionally” (syāt), [all] concerning a single object and its particular properties, composed of assertions and denials, either simultaneously or successively, and without contradiction. They are as follows: 1 Arguably, it (i.e. some object) exists (syād asty eva). The first predicate pertains to an assertion. 2 Arguably, it does not exist (syān nāsty eva). The second predicate pertains to a denial. 3 Arguably, it exists; arguably, it doesn’t exist (syād asty eva syān nāsty eva). The third predicate pertains to successive assertion and denial. 4 Arguably, it is ‘non-assertible’ (syād avaktavyam eva). The fourth predicate pertains to a simultaneous assertion and denial. 5 Arguably, it exists; arguably it is non-assertible (syād asty eva syād avaktavyam eva).The fifth predicate pertains to an assertion and a simultaneous assertion and denial. 6 Arguably, it doesn’t exist; arguably it is non-assertible (syān nāsty eva syād avaktavyam eva). The sixth predicate pertains to a denial and a simultaneous assertion and denial. 7 Arguably, it exists; arguably it doesn’t exist; arguably it is non-assertible (syād asty eva syān nāsty eva syād avaktavyam eva). The seventh predicate pertains to a successive assertion and denial and a simultaneous assertion and denial.

The structure here is simple enough. There are three basic truth-values, true (t), false (f ) and non-assertible (u).There is also some means of combining basic truth-values, to form four further compound values, which we can designate tf, tu, fu and tfu. There is a hint too that the third basic value is itself somehow a product of the first two, although by some other means of combination—hence the talk of simultaneous and successive assertion and denial. Thus, in Jaina seven valued logic, all the truth-values are thought to be combinations in some way or another of the two classical values.

THE PHILOSOPHICAL BASIS OF PLURALISM

347

There is, however, a clear risk that the seven values in this system will collapse trivially into three. For if the fifth value, tu, means simply “true and true-and-false”, how is it distinct from the fourth value, u, “true-and-false”? No reconstruction of the Jaina system can be correct if it does not show how each of the seven values is distinct. The way through is to pay due attention to the role of the conditionalising operator “arguably” (syāt). The literal meaning of “syāt” is “perhaps it is”, the optative form of the verb “to be”. The Jaina logicians do not, however, use it in quite its literal sense, which would imply that no assertion is not made categorically, but only as a possibility-claim. Instead, they use it to mean “from a certain standpoint” or “within a particular philosophical perspective”. This is the Jaina pluralism: assertions are made categorically, but only from within a particular framework of supporting assertions. If we let the symbol “∇” represent “syāt”, then the Jaina logic is a logic of sentences of the form “∇p”, a logic of conditionally justified assertions. As we will see, it resembles other logics of assertion, especially the ones developed by Jaśkowski (1948) and Rescher (1968). The first three of the seven predications now read as follows: /p/ = t iff ∇p

(1)

In other words, p is true iff it is arguable that p.We are to interpret this as saying that there is some standpoint within which p is justifiably asserted. We can thus write it as /p/ = t iff ∃σ σ :p,

(1)

where “σ :p” means that p is arguable from the standpoint σ. For the second value we may similarly write, /p/ = f iff ∇ ¬p.

(2)

That is, /p/ = f iff ∃σ σ :¬p. The third value is taken by those propositions whose status is controversial, in the sense that they can be asserted from some standpoints but their negations from others.These are the propositions which the Jainas are most concerned to accommodate. Thus /p/ = tf iff /p/ = t & /p/ = f. I.e. /p/ = tf iff ∇p & ∇ ¬p, or again /p/ = tf iff ∃σ σ :p & ∃σ σ :¬p.

(3)

348

JONARDON GANERI

This way of introducing a new truth-value, by combining two others, may seem a little odd. I think, however, that we can see the idea behind it if we approach matters from another direction. Let us suppose that every standpoint is such that for any given proposition, either the proposition or its negation is assertible from within that standpoint. Later, I will argue that the Jainas did not want to make this assumption, and that this is what lies behind their introduction of the new truth-value “non-assertible”. But for the moment let us make the assumption, which is tantamount to supposing that every standpoint is “optimal”, in the sense that for any arbitrary proposition, it either supplies grounds for accepting it, or else grounds for denying it. There are no propositions about which an optimal standpoint is simply indifferent. Now, with respect to the totality of actual optimal standpoints, a proposition can be in just one of three states: either it is a member of every optimal standpoint, or its negation is a member of every such standpoint, or else it is a member of some, and its negation of the rest. If we number these three states, 1, 2 and 3, and call the totality of all actual standpoints, Σ, then the value of any proposition with respect to Σ is either 1, 2 or 3.The values 1, 2 and 3 are in fact the values of a three-­valued logic, which we can designate M3. I will argue below that there is a correspondence between this logic and the system introduced by the Jainas ( J3, say).The idea, roughly is that a proposition has the value ‘true’ iff it either has the value 1 or 3, it has the value ‘false’ iff it either has the value 2 or 3, and it has the value ‘tf ’ iff it has the value 3. Hence the three values introduced by the Jainas represent, albeit indirectly, the three possible values a proposition may take with respect to the totality of optimal standpoints. Before elaborating this point further, we must find an interpretation for the Jainas’ fourth value “non-assertible”. Bharucha and Kamat offer the following analysis of the fourth value: The fourth predication consists of affirmative and negative statements made simultaneously. Since an object X is incapable of being expressed in terms of existence and non-existence at the same time, even allowing for Syād, it is termed ‘indescribable’. Hence we assign to the fourth predication … the indeterminate truth-value I and denote the statement corresponding to the fourth predication as (p & ¬p). (1984: 183)

Bharucha and Kamat’s interpretation is equivalent to: /p/ = u iff ∇ (p & ¬p),

(4)

that is /p/ = u iff ∃σ σ :(p & ¬p). Thus, for Bharucha and Kamat, the Jaina system is paraconsistent because it allows for standpoints in which contradictions are justifiably assertible. This seems to me to identify the paraconsistent element in the Jaina theory in quite the wrong place. For while there may be certain sentences, such as the Liar, which can justifiably be both asserted and denied, this cannot be the case for the wide variety of sentences which the Jainas have in mind, sentences like “There exist universals” and so on. Even aside from such worries,

THE PHILOSOPHICAL BASIS OF PLURALISM

349

the current proposal has a technical defect. For what now is the fifth truth-value, tu? If Bharucha and Kamat are right then it means that there is some standpoint from which ‘p’ can be asserted, and some from which ‘p & ¬p’ can be asserted. But this is logically equivalent to u itself. The Bharucha and Kamat formulation fails to show how we get to a seven-valued logic. Another proposed interpretation is due to Matilal.Taking at face-value the Jainas’ elaboration of the fourth value as meaning “simultaneously both true and false”, he says: the direct and unequivocal challenge to the notion of contradiction in standard logic comes when it is claimed that the same proposition is both true and false at the same time in the same sense. This is exactly accomplished by the introduction of the [fourth] value—“Inexpressible”, which can also be rendered as paradoxical. (1991: 10–11).

Matilal’s intended interpretation seems thus to be: /p/ = u iff ∇ (p, ¬p), i.e. /p/ = u iff ∃σ (σ :p & σ :¬p).

(4)

Matilal’s interpretation is a little weaker than Bharucha and Kamat, for he does not explicitly state that the conjunction ‘p&¬p’ is asserted, only that both conjuncts are. Admittedly, the difference between Matilal and Bharucha and Kamat is very slight, and indeed only exists if we can somehow make out the claim that both a proposition and its negation are assertible without it being the case that their conjunction is. For example, we might think that the standpoint of physical theory can be consistently extended by including the assertion that gods exists, and also by including the assertion that gods do not exist. It would not follow that one could from any standpoint assert the conjunction of these claims.Yet whether there is such a difference between Matilal’s position and that of Bharucha and Kamat is rather immaterial, since Matilal’s proposal clearly suffers from precisely the same technical defect as theirs, namely the lack of distinctness between the fourth and fifth values. I will now offer my own interpretation, which gives an intuitive sense to the truthvalue “non-assertible”, sustains the distinctness of each of the seven values, but does not require us to abandon the assumption that standpoints are internally consistent. Recall that we earlier introduced the idea of an optimal standpoint, by means of the assumption that for every proposition, either it or its negation is justifiably assertible from within the standpoint. Suppose we now retract that assumption, and allow for the existence of standpoints which are just neutral about the truth or falsity of some propositions. We can then introduce a new value as follows: /p/ = u iff ∃σ¬(σ :p) & ¬(σ :¬p).

(4)

Neither the proposition nor its negation is assertible from the standpoint. For example, neither the proposition that happiness is a virtue nor its negation receives any justification from the standpoint of physical theory. We have, in effect, rejected a commutativity rule, that if it is not the case that ‘p’ is assertible from a standpoint σ then ‘¬p’ is assertible

350

JONARDON GANERI

from σ, and vice versa [¬(σ :p) iff (σ :¬p)]. Our new truth-value, u, is quite naturally called “non-assertible”, and it is clear that the fifth value, tu, the conjunction of t with u, is not equivalent simply with u.The degree to which the Jaina system is paraconsistent is, on this interpretation, restricted to the sense in which a proposition can be tf, i.e. both true and false because assertible from one standpoint but deniable from another. It does not follow that there are standpoints from which contradictions can be asserted. Why have so many writers on Jaina logic felt that Jaina logic is paraconsistent precisely in the stronger sense? The reason for this belief is the account which some of the Jainas themselves give of the meaning of their third basic truth-value, “non-assertible”. As we saw in the passage from Vādideva Sūri, some of them say that a proposition is non-­ assertible iff it is arguably both true and false simultaneously, as distinct from the truth value tf, which is successively arguably true and arguably false. We are interpreting the Jaina distinction between successive and simultaneous combination of truth-values in terms of a scope distinction with the operator “arguably”. One reads “arguably (t & f)”, the other “(arguably t) & (arguably f)”. If this were the correct analysis of the fourth truth-value, then Jaina logic would indeed be strongly paraconsistent, for it would be committed to the assumption that there are philosophical positions in which contradictions are rationally assertible. Yet while such an interpretation is, on the face of it, the most natural way of reading Vādideva Sūri’s elaboration of the distinction between the third and fourth values, it is far from clear that the Jaina pluralism really commits them to paraconsistency in this strong form. Their goal is, to be sure, to reconcile or synthesise mutually opposing philosophical positions, but they have no reason to suppose that a single philosophical standpoint can itself be inconsistent. Internal consistency was, in classical India, the essential attribute of a philosophical theory, and a universally acknowledged way to undermine the position of one’s philosophical opponent was to show that their theory contradicted itself. The Jainas were as sensitive as anyone else to allegations that they were inconsistent, and strenuously denied such allegations when made. I have shown that it is possible to reconstruct Jaina seven-valued logic in a way which does not commit them to a strongly paraconsistent position. The interpretation I give to the value “non-assertible” is quite intuitive, although it does not mean “both true and false simultaneously”. My interpretation, moreover, is supported by at least one Jaina logician, Prabhācandra. Prabhācandra, who belongs to the first part of the ninth century c.e., is one of the few Jainas directly to address the question of why there should be just seven values. What he has to say is very interesting (­Prameyakamalamārtaṇḍa, p. 683, line 7 ff.): (Opponent:)  Just as the values ‘true’ and ‘false’, taken successively, form a new truth-value ‘true-false’, so do the values ‘true’ and ‘true-false’.Therefore, the claim that there are seven truth-values is wrong. (Reply:)  No: the successive combination of ‘true’ and ‘true-false’ does not form a new truthvalue, because it is impossible to have ‘true’ twice. … In the same way, the successive combination of ‘false’ and ‘true-false’ does not form a new truth-value. (Opponent:)  How then does the combination of the first and the fourth, or the second and the fourth, or the third and the fourth, form a new value? (Reply:)  It is because, in the fourth value “non-assertible”, there is no grasp of truth or falsity. In fact, the word “non-assertible” does not denote the simultaneous combination of

THE PHILOSOPHICAL BASIS OF PLURALISM

351

truth and falsity. What then? What is meant by the truth-value “non-assertible” is that it is impossible to say which of ‘true’ and ‘false’ it is.

This passage seems to support the interpretation offered above. When talking about the “law of non-contradiction” in a deductive system, we must distinguish between two quite different theses: (a) the thesis that “¬(p & ¬p)” is a theorem in the system, and (b) the thesis that it is not the case that both ‘p’ and ‘¬p’ are theorems. The Jainas are committed to the first of these theses, but reject the second. This is the sense in which it is correct to say that the Jainas reject the “law of non-contradiction”. I showed earlier that when we restrict ourselves to optimal standpoints, the total discourse falls into just one of three possible states with respect to each system. The Jainas have a seven-valued logic because, if we allow for the existence of non-optimal standpoints, standpoints which are just neutral with respect to some propositions, then, for each proposition, p say, the total discourse has exactly seven possible states. They are as follows: 1 p is a member of every standpoint in Σ. 2 ¬p is a member of every standpoint in Σ. 3 p is a member of some standpoints, and ¬p is a member of the rest. 4 p is a member of some standpoints, the rest being neutral. 5 ¬p is a member of some standpoints, the rest being neutral. 6 p is neutral with respect to every standpoint. 7 p is a member of some standpoints, ¬p is a member of some other standpoints, and the rest are neutral. Although Jainas do not define the states in this way, but rather via the possible combinations of the three primitive values, t, f and u, it is not difficult to see that the two sets map onto one another, just as they did before. Thus t = (1, 3, 4, 7), f = (2, 3, 5, 7), tf = (3, 7), and so on. Using many-valued logics in this way, it should be noted, does not involve any radical departure from classical logic. The Jainas stress their commitment to bivalence, when they try to show, as Vādideva Sūri did above, that the seven values in their system are all products of combining two basic values. This reflects, I think, a commitment to bivalence concerning the truth-values of propositions themselves. The underlying logic within each standpoint is classical, and it is further assumed that each standpoint or participant is internally consistent. The sometimes-made suggestion that sense can be made of many-­ valued logics if we interpret the assignment of non-classical values to propositions via the assignment of classical values to related items (cf. Haack 1974: 64) is reflected here in the fact that the truth-value of any proposition p (i.e. /p/) has two values, the status of p with respect to standpoint σ (‘/p/σ’) derivatively has three values, and the status of p with respect to a discourse Σ (‘/p/Σ’), as we have just seen, has seven. Consider again the earlier example of a jury faced with conflicting evidence from a variety of witnesses. The Jainas would not here tell us ‘who dun it’, for they don’t tell us the truth-value of any given proposition. What they give us is the means to discover patterns in the evidence, and how to reason from them. For example, if one proposition is agreed on by all the witnesses, and another is agreed on by some but not others, use of the

352

JONARDON GANERI

Jaina system will assign different values to the two propositions.The Jainas, as pluralists, do not try to judge which of the witnesses is lying and which is telling the truth; their role is more like that of the court recorder, to present the totality of evidence in a maximally perspicuous form, one which still permits deduction from the totality of evidence. So far so good. But there is another worry now, one which strikes at the very idea of using a many-valued logic as the basis for a logic of discourse. For, when we come to try and construct truth-tables for the logical constants in such a logic, we discover that the logic is not truth-functional. That is to say, the truth-value of a complex proposition such as ‘p & q’ is not a function solely of the truth-values of the constituent propositions ‘p’ and ‘q’. To see this, and to begin to find a solution, I would like briefly to describe the work of the Polish logician, Jaśkowski, who was the founder of discursive logics in the West, and whose work, in motivation at least, provides the nearest contemporary parallel to the Jaina theory. 2. JAS´KOWSKI AND THE JAINAS

Philosophical discourse is globally inconsistent, since there are many propositions to which some philosophers assent while others dissent. The Jainas therefore develop a logic of assertions-made-from-within-a-particular-standpoint, and note that an assertion can be both arguably true, i.e. justified by being a member of a consistent philosophical position, and at the same time be arguably false, if its negation is a member of some other consistent philosophical standpoint. This move is quite similar to that of the founder of inconsistent logics, Jaśkowski, who developed a “discussive logic” in which a proposition is said to be ‘discussively true’ iff it is asserted by some member of the discourse. Jaśkowski motivates his paper “Propositional Calculus for Contradictory Deductive Systems” (1948; trans. 1969), by making two observations. The first is that: any vagueness of the term a can result in a contradiction of sentences, because with reference to the same object X we may say that “X is a” and also “X is not a”, according to the meanings of the term a adopted for the moment’.

The second is that: the evolution of the empirical sciences is marked by periods in which the theorists are unable to explain the results of experiments by a homogeneous and consistent theory, but use different hypotheses, which are not always consistent with one another, to explain the various groups of phenomena. (1969: 144)

He then introduces an important distinction between two properties of deductive systems. A deductive system is said to be contradictory if it includes pairs of theorems A and ¬A which contradict each other. It is over-complete, on the other hand, if every wellformed formula is a theorem of the system. In classical logic, these two properties are conflated; hence the slogan “anything follows from a contradiction”. The problem to which Jaśkowski addresses himself, therefore, is that of constructing a non-classical system which is contradictory but not over-complete. In classical logic, given two contradictory

THE PHILOSOPHICAL BASIS OF PLURALISM

353

theses A, ¬A, we may deduce first that A & ¬A, using the &-introduction or Adjunction Rule, A, B → A&B. Then, since A & ¬A iff B & ¬B for any arbitrary A and B, and since B & ¬B → B from &-elimination or Simplification, A & B → A, it follows that B. More clearly: 1 A, ¬A 2 A & ¬A, from 1 by Adjunction. 3 A & ¬A iff B & ¬B, for any arbitrary A and B. 4 B & ¬B → B, by Simplification. 5 A & ¬A → B, from 3 and 4. 6 B, from 2 and 5 by Modus Ponens. To get an inconsistent (contradictory but not over-complete) system, at least one step in this sequence must be broken. In Jaśkowski’s new system, ‘discursive logic’, it is the Adjunction Rule which no longer holds. Jaśkowski considers the system in which many different participants make assertions, each thereby contributing information to a single discourse. The best example, perhaps, is one already given, the evidence presented to a jury by witnesses at a trial. Jaśkowski then introduces the notion of discursive assertion, such that a sentence is discursively asserted if it is asserted by one of the participants in the discourse, and he notes that the operator “it is asserted by someone that …” is a modal operator for the semantics of which it should be possible to use an existing modal logic. Thus: A is a theorem of D2 iff ◊A, where D2 is Jaśkowski’s two-valued discursive logic, and “◊” is the operator “someone asserts that …”. For some reason, Jaśkowski chooses a strong modal system, S5, to give the semantics of this operator, but this is surely a mistake. The reason is that the S5 modal principle ‘A → ◊A’ does not seem to hold for a discursive system, since there will be truths which no-one asserts. It would not be difficult, however, to use a weaker modal system than S5, for example S20 or S30, which lack the above principle, as the basis for D2. (The characteristic axiom of S40, ‘◊◊A → ◊A’, does not seem to hold in a discursive system: it can be assertible from some standpoint that there is another standpoint in which p is assertible without there being such a standpoint.) The point to note is that, in most modal systems, the Adjunction Rule fails, since it does not follow that the conjunction A&B is possible, even if A is possible and B is separately possible. And this too, is what we would expect from the discursive operator, for one participant may assert A, and another B, without there being anyone who asserts the conjunction. Jaśkowski therefore arrives at a system which is contradictory, since both A and ¬A can be theses, but, because it is non-adjunctive, is not over-complete. 3.  THE LOGICAL STRUCTURE OF THE JAINA SYSTEM

The parallels in motivation between Jaśkowski’s discursive logic, and the Jaina system are unmistakable. There is, however, an important difference, to which I alluded earlier.

354

JONARDON GANERI

Modal logics are not truth-functional; one cannot, for example, deduce the truth-value of ‘◊(A&B)’ from the truth-values of ‘◊A’ and ‘◊B’. And it seems for the same reason that a discursive logic cannot be truth-functional either. Suppose, for example, that we have two propositions A, and B, both of which are assertible from (possibly distinct) standpoints, and hence both true in the Jaina system. What is the truth-value of A&B? It seems that this proposition could be either true, false, or both.To find a solution to this problem, we must explore a little the relations between many-valued and modal systems. I would like to offer a defence of the Jaina position here. For simplicity, let us restrict ourselves to the Jaina system with only optimal standpoints and just three truth-values. If my suggested defence works here, its extension to the full Jaina system J7, would not be especially problematic. Consider again the three-valued logic, M3, whose values were defined as follows: /p/= 1 iff ∀σ σ :p. /p/= 2 iff ∀σ σ :¬p. /p/= 3 iff ∃σ σ :p & ∃σ σ :¬p. These correspond to the three possible states of a totality of optimal standpoints. When we try to construct the truth-table for conjunction in such a system, we find that it is non-truth-functional. Thus, consider the truth-value of ‘p&q’ when /p/ = /q/ = 3. Here, /p&q/ might itself be 3, but it might also be 2. Thus, the truth-value of the conjunction is not uniquely determined by those of its conjuncts. What is uniquely determined, however, is that the truth-value belongs to the class (2, 3). To proceed, we can appeal to an idea first introduced by N. Rescher in his paper “Quasi-truth-functional systems of propositional logic” (1962). A quasi-truth-functional logic is defined there as one in which “some connectives are governed by many-valued functions of the truth-­values of their variables”. The entries in the truth-table of such a logic are typically not single truth-values but sets of values. It is clear that the system set up just now is, in this sense, quasi-truth-functional. Now, as Rescher himself points out, a quasi-truth-­functional logic will always be equivalent to a multi-valued strictly truth-functional system. The idea, roughly, is that we can treat a class of truth-values as constituting a new truth-value. Typically, if the quasi-truth-functional system has n truth-values, its strictly truth-­functional equivalent will have 2n–1 values. (Rescher notes that “in the case of a three-valued (T, F, I) quasi-truth-functional system we would need seven truth-values, to represent: T, F, I, (T, F), (T, I), (F, I), (T, F, I)” but argues that there are special reasons entailing that for a two-valued quasi-truth-functional system we need four rather than three values.) The seven-valued system which results in this way from the three-valued logic sketched above has, in fact, been studied notably by Moffat (see Moffat and Ritchie 1990, see also Priest 1984). I will therefore call it M7. An initially tempting idea is to identify the Jaina system J7 with M7. This, however, will only work if the fourth value, u, is defined thus: /p/ = u iff ∀σ σ :p ∨∀σ σ :¬p. For then ‘tu’ in the Jaina system will be identical with ‘1’ in the Moffat system, etc. This is, however, not an interpretation which receives any textual support.

THE PHILOSOPHICAL BASIS OF PLURALISM

355

Instead, let us observe that there is a close connection between M7 and the restricted Jaina system, J3. For note that the value (1, 3) in M7 is such that /p/ = (1, 3) iff /p/ = 1∨ /p/ = 3 iff ∀σ σ :p ∨ (∃σ σ :p & ∃σ σ :¬p) iff ∃σ σ :p. Thus (1, 3) in M7 is just the value ‘true’ in J3. Similarly, (1, 2) in M7 is just the value ‘false’ in J3. Thus, although J3 is not strictly truth-functional, its truth-tables are embedded in those of the Moffat logic, M7. I have given some of the truth-tables which illustrate this fact in the Appendix. It is presumably possible to find a quasi-truth-functional system whose truth-tables embed those of J7, the full Jaina system, in an entirely analogous way. Thus, although the loss of Adjunction means that the Jaina logic J7, is not truth-functional, its truth-table is embedded in a suitable quasi-functional system. The lack of truth-functionality is not, after all, a fatal flaw in the Jaina approach. 4.  AXIOMATISATION OF THE JAINA SYSTEM

We have shown that it is possible to use many-valued truth-tables to formalise the Jaina system. This was, in effect, the approach of the Jaina logicians themselves. Yet it would surely be much better to proceed by axiomatising the modal standpoint operator, ∇. Once again we look to Rescher (Topics in Philosophical Logic (1968), chapter xiv). His work on what he calls “assertion logics” is an extension of the work of Jaśkowski. Rescher introduces a system Al, with the following axiomatic basis: (A1)  (∃p)σ :p [Nonvacuousness] (A2)  (σ :p & σ :q) ⊃ σ :(p&q) [Conjunction] (A3)  ¬σ :(p & ¬p) [Consistency] (R)  if p ˫ q, then σ :p ˫ σ :q. [Commitment] Note that one effect of the rule (R) is to ensure that the notion captured is not merely explicit assertion but ‘commitment to assert’, for (R) states that from a standpoint one may assert anything entailed by another of the assertions. I believe that the Jainas would accept each of the axioms (A1) to (A3). Bharucha and Kamat, it may be noted, would reject (A3), while Matilal, as I have represented him, would reject (A2). I have already argued that these claims are mistaken. In particular, with regard to (A2), although it is true that the Jainas reject Adjunction, what this means is that assertions made from within different standpoints cannot be conjoined, not that assertions made within the same standpoint cannot be conjoined. We now introduce the modal standpoint operator, ∇ “arguably”, via the definition: ∇p iff (∃σ)σ :p, and add the axioms of S30 or some other suitable modal system.

356

JONARDON GANERI

Rescher defines some further systems by adding further axioms, none of which, I think, the Jainas would accept. For example, he defines A2 by adding to Al the axiom that anything asserted by everyone is true [(∀σ) σ :p ⊃ p].There is no reason to suppose the Jainas commit themselves to this.The system J3, however, is distinguished by the new axiom (A4): (A4) ¬(∃σ) (¬σ :p & ¬σ :¬p)

[Optimality]

Rescher too proposes a “three-valued approach” to assertion logic, via the notion of ‘the truth status of the assertion p with respect to an assertor’, written ‘/p/σ’, and the definitions: /p/σ = T iff σ :p, = F iff σ :(¬p), and = I iff ¬σ :p & ¬σ :¬p, and he shows that using the axioms of A1, we can derive a quasi-truth-functional logic for this system. These are not quite the Jaina values, as introduced earlier, for they do not quantify over standpoints or assertors. It is clear, however, that the Jaina system is of the same type as a modalized Rescher assertion logic. Their innovation is to introduce three truth-values via the definitions given before (/p/Σ = t iff (∃σ)(σ :p); /p/Σ = f iff (∃σ) (σ:¬p); and /p/Σ = u iff (∃σ)(¬σ :p & ¬σ :¬p), where ‘/p/Σ’ stands for ‘the status of the assertion p with respect to the total discourse Σ’). It is this attempt to take a many-valued approach to the modalised, rather than the unmodalized, version of assertion logic which generates the extra complexity of the Jaina system. I have already noted that, since the axiom “p ⊃ ∇p” is lacking, the modal structure of the system will be no stronger than that of S30. Yet in principle there seems no reason to think that the Jaina system cannot in this way be given an axiomatic basis. 5.  PLURALISM, SYNCRETISM, AND THE MANY-FACETED VIEW OF REALITY

The Jainas avoid dogmatism and a one-sided view of the world simply by noting that assertions are only justified in the background of certain presuppositions or conditions. It is perfectly possible for an assertion to be justified given one set of presuppositions, and for its negation to be justified given another different set. The Jainas’ ingenuity lies in the skill with which they developed a logic of discourse to make more precise this natural idea. However, they also went beyond this, for they added that every standpoint reveals a facet of reality, and that, to get a full description of the world, what we need to do is to synthesise the various standpoints. As Matilal puts it, “The Jainas contend that one should try to understand the particular point of view of each disputing party if one wishes to grasp completely the truth of the situation. The total truth … may be derived from the integration of all different viewpoints (1977)”. But is this further step, the step from pluralism to syncretism, a coherent step to take? In particular, how is it possible to integrate inconsistent points of view? The point is made by Priest and Routley, who, commenting on the Jaina theory, state that “… such a theory risks trivialization unless some (cogent) restrictions are imposed on the parties admitted as having obtained partial

357

THE PHILOSOPHICAL BASIS OF PLURALISM

truth—restrictions of a type that might well be applied to block amalgamations leading to violations of Non-Contradiction” (Priest et al., p. 17). Perhaps we can understand the Jaina position as follows. The so-called ‘integration’ of two points of view, σ1 and σ2, does not mean the creation of some new standpoint, which is the combination of the first two. For this would lead to the formation of inconsistent standpoints unless implausible constraints were placed on what can constitute a standpoint. Instead, what it means is that, if p is assertible from some standpoint σ, then this fact, that p is assertible from σ, can itself be asserted from σ2 and every other standpoint. In this way, each disputant can recognise the element of truth in the other standpoints, by making explicit the presuppositions or conditions under which any given assertion is made. If correct, this idea has an interesting consequence. In moving from pluralism to syncretism, the Jainas commit themselves to the claim that we are led to a complete account of reality by integrating of all the different points of view. It follows from this that every true proposition must be asserted within some standpoint, i.e. “p ⊃ (∃σ)(σ :p)” or “p ⊃ ∇p”. Hence the move from pluralism to syncretism is a move from a logic of assertibility based on S30 or weaker to one based on S3 or stronger. To conclude, we have seen how the Jainas developed a plausible and interesting logic of philosophical discourse, how they did not (or need not) commit themselves to the strongly paraconsistent position normally attributed to them, and how, as they strengthened their position from one of pluralism to one of syncretism, they had also to strengthen correspondingly the modal logic underlying the operator “syāt”’. APPENDIX Some truth-tables for J3 The truth-table for ‘&’ in M7 is as follows: &

1

2

3

(2,3)

(1,3)

(1,2,3)

(1,2)

1 2 3 (2,3) (1,3) (1,2,3) (1,2)

1 2 3 (2,3) (1,3) (1,2,3) (1,2)

2 2 2 2 2 2 2

3 2 (2,3) (2,3) (2,3) (2,3) (2,3)

(2,3) 2 (2,3) (2,3) (2,3) (2,3) (2,3)

(1,3) 2 (2,3) (2,3) (1,2,3) (1,2,3) (1,2,3)

(1,2,3) 2 (2,3) (2,3) (1,2,3) (1,2,3) (1,2,3)

(1,2) 2 (2,3) (2,3) (1,2,3) (1,2,3) (1,2)

The embedded truth-table for ‘&’ in J3 is therefore: &

tf

f

t

tf f t

f f f

f f f

f f (t,f,tf)

358

JONARDON GANERI

The truth-table for ‘¬’ in M7 is:

⊢

1

2

3

(1,3)

(2,3)

(1,2)

(1,2,3)

1

2

3

(2,3)

(1,3)

(1,2)

(1,2,3)

The embedded truth-table for ‘¬’ in J3 is therefore:

⊢

tf

t

f

tf

f

t

REFERENCES Bharucha, F. and Kamat, R.V. 1984. ‘Syādvāda theory of Jainism in terms of deviant logic’, Indian Philosophical Quarterly 9, 181–187. Dundas, P. 1992. The Jains. London: Routledge. Ganeri, J. 2001. ‘Rationality, harmony and perspective’, in Philosophy in Classical India:The Proper Work of Reason. London: Routledge. Gokhale, P. 1991. ‘The logical structure of Syādvāda’, Journal of Indian Council of Philosophical Research 8, 73–81. Haack, S. 1974. Deviant Logic. Cambridge: Cambridge University Press. Jaśkowski, S. 1948.‘Propositional calculus for contradictory deductive systems’, Studia Logica 24, 143–157. Matilal, B. K. 1977. The Central Philosophy of Jainism. Calcutta: Calcutta University Press. Matilal, B. K. 1991. ‘Anekānta: both yes and no?’, Journal of Indian Council of Philosophical Research 8, 1–12. Moffat, D. C. and Ritchie, G. D. 1990. ‘Modal queries about partially-ordered plans’, Journal of Experimental and Theoretical Artificial Intelligence 2, 341–368. Mukerji, R. N. 1977. ‘The Jaina logic of seven-fold predication’, in A. N. Upadhye, N.Tatia, D. Malvania, M. Mehta, N. Shastri, K. Shastri eds., Mahāvīra and His Teachings, Bombay, 225–233. Prabhācandra. 1941. Prameyakamalamārtaṇḍa, M. K. Shastri, ed. Bombay: Nirnayasagar Press. Priest, G. 1984. ‘Hypercontradictions’, Logique et Analyse 107, 237–243. Priest, G., Routley, R. and Norman, J., eds. 1989. Paraconsistent Logic: Essays on the Inconsistent. Munchen: Philosophia Verlag. Rescher, N. 1962. ‘Quasi-truth-functional systems of propositional logic’, Journal of Symbolic Logic 27, 1–10. Rescher, N. 1968. Topics in Philosophical Logic. Dordrecht: Reidel. Shah, J., Nagin ed. 2000, Jaina Theory of Multiple Facets of Reality and Truth. Bhogilal Leherchand Institute of Indology, Delhi. Vādideva Sūri: 1967. Pramāṇa-naya-tattvālokālaṃkāra, H. S. Battacharya, ed. and transl. Bombay: Jain ­Sahitya Vikas Mandal.

I NDEX

Note: Page numbers followed by “n” denote notes. Abhidarmakośābhāṣya 169 Abhidharma 88–101, 104, 110–112, 139n3, 159n17; awareness primary factor 95–96; mental factors and groups (lists) 96–100, 114n12; root and branch afflictions 97–100; tradition and mind view 91–100; and witness consciousness 183, 194n3 Abhinavagupta 33, 145; and Buddhist No-selfism 166–180 abhinna (integer) 223 absence 172–173; in absence 23 abstraction 54, 170, 206, 282, 290, 306, 309 Abu’l Fazl 45 Adamson, P. 3 a-dharma (demerit) 11 adhyāsa (superimposition) 191–192; de-superimposition process 192–193 agent 149–150; self as 149–153, 160n19 ahaṃkāra (ego-sense) 191 Aitareya Āraṇyaka 18 ākāṃkṣā (syntactic expectancy) 293, 337 ākāra (aspect) concept 102 ākāśa (substratum of sounds) 313–329 Akbar, Emperor 36–41, 45–47 Akṣapāda Gautama 34, 335–336; Nyāyasūtra 14, 37–39 ākṣepa (extrapolative judgement) 293 ālaya-vijñāna (store-consciousness) 96, 111–113, 114n10, 130, 139n6, 140n10 Albahari, M. 194n4, 195n21, 197n29 Ambedkar, B. R. 2

Anandavardhana 33 anātman (no-self) 91, 166–186 Anquetil-Duperron, A. H. 46 Anscombe, G. E. M. 137–138 anti-individualism 256–266, 272, 275n6; Burge’s defence 258–259; and disjunctivism 256–264, 272; McDowell’s epistemic and HCF reasoning 259–261; perceptual 261–273, 275n6, 276n9; Twin Earth experiment 258–259, 266–267, 275n5 anugatapratīti (common character sharing) 204 ānvīkṣikī (philosophy) 7–8, 14–15, 17, 21–22, 34; schools and disciplines 12–13 anyāpoha (differentiation) 209 aphorism 9 apoddhāra (analysis, synthesis and abstraction) 290 apoha (exclusion) 167 apperception 103–104, 171 apprehension 174–175 argumentation 335–343; philosophy 8, 37–41 Argumentative Indian (Sen) 36 Aristotle 37, 42, 103, 205–207 Ārtabhāga 63 artha (power/wealth) 9–10, 13, 20 arthāpatti (suggestive inference) 293 Āruṇi 78 Āryabhaṭa 37, 65, 74 Āryadeva 179 Asaṇga 91, 96; and Vasubandhu on selfconsciousness 129–138 āsatti 293

359

360INDEX Ascertainment of Valid Cognition (Vetter) 107 asceticism 54 Ashoka Pandita, Emperor 36–39, 42–44, 209 aspect 102–105; assertion 102; objective 103; subjective 103 āśraya (support) 130, 159n17 asymmetric dependence 256–258, 263–264, 275n3, 275n4 atomism 45–46, 281, 290, 310–311; epistemological 296–298 atoms 211 attention 98–99, 116; training 122–124 Aurangzeb 45 Aureng Zebe (Dryden) 47 Austin, J. L. 259 awareness 29; and cognitive tasks 97, 114n11; and illusion 247–252; mental factors 96–100, 113, 114n12; No-selfism, recognition and perception 168–180; perception and realism 243, 246; primary factor 95–96, 113; of qualities 226; self and witness consciousness 182–197; and selfhood 129–140; and theories of mind 87–112; types 11, 95–96 Ayer, A. J. 259 āyurveda (learning) 13 ayutasiddhavṛtti (inseparable relation) 203 Bagchi, S. 28 Bansat-Boudon, L. 146 Bayle, P. 46 Bernier, F. 45–48 Bhagavad Gītā 19, 74–75 Bhāgavatapurāṇa 48 Bharata 33 Bhartṛhari 22, 26–27, 221, 244–245, 281, 290–291, 299; Mahābhāṣya-dīpīka 302; and sphoṭa 302–312; Vākyapadīya 290, 303–305, 309 Bharucha 348–349, 355 Bhāsarvajña 219, 222–225, 233, 236n6, 341 Bhāskarācārya II 234 Bhattacharya, K. C. 2, 21–22, 33–34, 175 Bhattacharya, S. 27–29 Bhikkhu, T. 126n1 Biardeau, M. 302 Block, N. 120 Bodhi, B. 126n1 Bok, S. 82 bondage 15 Brāhamanas 54–55, 63, 146–147; Śatapatha 62 brahman 25, 55–57, 68, 73–74, 77 Brentano, F. 92–94 Brereton, J. 53–64, 83n16 Brough, J. 281, 302 Buddha (Siddhārtha Gautama) 8, 15, 43, 69, 183, 269, 276n7; see also sūtras (sayings of the Buddha) buddhi (intelligence/awareness) 33, 87–112

Buddhism 1, 2, 9, 15, 39–44, 246; Abhidharma tradition 88, 91–100, 110–111; Dharmakīrti 110–111; epistemology 100–111; Mahāyāna 183; mango concept 148–151, 148, 159n9, 159n12; mental events reflexive nature 102–104; perception theory 104–107; self and stream of consciousness 145–158; and self versus Rāmakaṇṭha and Nyāya 145–165; Theravāda 1, 116, 123–125; thought and language 107–109; Zen and Chan 117, 123 Buescher, H. 140n10 Burge, T. 256–264, 275n5 caitesika (mental factors) 96–100, 113, 114n12 Calvino, I. 79 Campbell, J. 137–138 Candrakīrti 138, 140n13 Caraka 22 Caraka Saṁhita 30–31 Cartesian philosophy 8, 46 Cārvāka (Materialist) school 22, 88, 147 caste divisions 18; high and Brahmins 32 causation 152–153, 170, 175, 204, 210 Chakrabarti, A. 17–35, 166–181 Chakrabarti, K. 28, 203–217 Chatterjee, T. 183, 194n10 Chisholm, R. 83n12 citta (mental activities) 95–96, 112–113, 159n17; bhavaṇga (life-constituent consciousness) 113 cognition (prajñāna) 20, 80–81, 87–88, 100–113, 149–150, 195n14, 195n15, 250; access 120–121; combinative 225–226, 231; connective– comparative 307; and consciousness 186–190; and No-selfism 166–180, 183–186; and Nyāya perceptual theory 255–274; and parasitism 264–267; self as agent (pleasure/seeing/desire) 149–153, 160n19; svasaṃvedana 103–104, 113; tasks 97, 114n11; valid (pramāṇa) 100–101, 107, 112–113, 234, 264, 314, 336 coherence 233, 253 colonialism 40–41 compassion 99 composition principle 281, 288–289 concealment 70–75; cave metaphors 70–71; and self 71–75 concentration 99; practices 122–124 Concept of Rasa (Bhattacharya) 33 conceptions 105–106, 111, 244, 255; formation theory 108–109 concepts 255–258, 275n4; possession and misapplication 257–258; wise 287–288 consciousness 26, 77, 83n13, 172–178, 310; and cognition (prajñāna) 186–190; constancy 155; and the existence-dimension 188; five aggregates 116–126, 129, 138n1, 139n3; momentariness and refutation 145–147, 146, 154–155, 158n5,

INDEX

161n23, 161n24; as nature of self 153–155, 161n20, 161n26; in other bodies (ūhyate) 177; pre-attentive (manas and mineness) 131–132; Rāmakaṇṭha, Nyāya and Buddhism 153–158; relation to objects 156–157, 197n28; and self 153–158, 153, 155, 156, 158, 183–190, 195n17; self as agent of cognition 149–153, 160n19; self, Asaṇga and Vasubandhu 129–131; self as substance 148–151, 159n7, 159n17; states 61, 146–147, 147, 158n6, 175–180; store 96, 111–113, 114n10, 130, 139n6, 140n10; stream 93–94, 114n8, 150, 154; support and transformation 130; temporal extension 156; and theories of mind 87–112; see also witness consciousness context principle 281–301; and the composition principle 281, 288–291; connected designation (anvitābhidhāna) 282, 291, 300n4, 311; and contextual definition 283–286, 301n6; designation before connection (abhihitānvaya) theory 282, 300n4; and interpretation 282, 286, 291; and language acquisition theory 292–294; and meaning controversies 281–300; related designation theory 294–295; sentence and word order sense (divisibility/indivisibility) 289–296; syntactic priority thesis 301n6; and unity of thought 287–288 Cowell, E. B. 48 cross-categoricity 221–222 Dadhyañc Ātharvaṇa 66 Damasio, A. 118; and Parvizi, J. 120–121 Dānishmand Khān 45 Dārā Shukoh 45–47 darśana (philosophical viewpoint) 7; six systems 13 Dasti, M. 256, 264–267, 273 data 250–252; psychological–physical 251; sense 250; theory 252; visual 251 Davidson, D. 24, 249, 253–254 Davis, J. H. 116–128 Daya Krishna 21 death 63–64 debate 44, 335–343; anchor 45; concomitance 342– 343; counter-evidence 338; Diṅnāga’s wheel of reason-sign 339–341; evidence–reason 337; and knowability 340–342; and logic 335–343; Nyāya model 336–338; sign and signified 338; sign triple nature (pakṣa/sapakṣa/vipakṣa) 338–342; structure and steps 337; types and categories 335–337; Uddyotakara’s wheel development 339–342 decision (saṃkalpa-prāyam eva tat) 169 definiendum (lakṣaṇā) 207, 216n12, 341–342 Definition and Induction (Chakrabarti) 28 demarcation 174–175 democracy 36 Democritus 82n7 Der Gedanke (Frege) 296

361 Descartes, R. 45, 88, 103, 259 designation before connection (abhihitānvaya) theory 282, 300n4 desire 73–75, 151–153, 160n19; as experience 184, 194n6; false 73, 81; true 73 de-superimposition process 192–193 determinate application (pravṛttīnimitta) 233 Devadatta-hood 307, 314–318, 321–322, 325–327 devotionalism 145 Dewey, J. 273 dharma (merit/piety/righteousness) 11–13, 19–21, 32–34, 92 Dharmakīrti 88–89, 100–112, 145, 157, 158n1, 161n23, 162n28, 169, 339, 342; inference theory 173 Dharmaśāstra 43–44 Dialogues Concerning Natural Religion (Hume) 46 Dignāga/Dinṅāga 100, 103, 108–109, 132, 185, 245; Pramāṇasamuccaya 15; theory of triple character reason 338–340; wheel of reason-sign 339–341 Diogenes Laertius 83n7 dīpa (lamp light) 10 discernment 98 disjunctivism 256–261, 275n2; and antiindividualism 256–264, 272; asymmetric dependence 256–258, 263–264, 275n3, 275n4; Burge and McDowell 256–264; Dasti and parasitism 256, 264–267; epistemic (ED) 259–273, 275n6, 276n9; and highest common factor view 259–260, 265; misplacement theory of illusion 257, 268–274, 276n9; and perceptual anti-individualism 261–273, 275n6, 276n9; and the Proximity Principle 261–265, 272 dissent 36–49; reason and identity 36–48 Doctrine of Pratibhā in Indian Philosophy (Kaviraj) 25 Donnellan, K. 330 doubt 267, 274; perceptual 269 draṣṭā (seer) 183 Dreyfus, G. 87–115, 140n10 Dryden, J. 47 dualism 91; life–death 147; mind–body 95; non145–146; perception and realism 246–247, 254 duḥkha (undesirable items) 11–14 Dummett, M. 24, 242, 282, 285–289, 296, 299 Dvivedin,V. P. 10 East India Company 48 ego sense (ahaṃkāra) 90, 96, 112, 182, 191–193; and EGOcentricity 133–134, 173; usage 179–180 Elements (Euclid) 83n12 emotions 24, 98, 100, 274, 276n10; and rasa 33–34; valence/tone 118–119 enabling conditions 228 Enlightenment 24, 47 epistemic disjunctivism 259–273, 275n6, 276n9 epistemological atomism 296–298

362INDEX error theory 175–176, 246, 249, 252; asymmetric dependence 256–258; and cognition (true/false) 255–274 essence 78–79, 207 ethics 25, 99 Ethics (Spinoza) 47 Euclid 30, 37, 83n12 Evans, G. 324 experience 104, 109–110, 210; consciousness and first person (I) 186–190; and the existencedimension 188; producer 183; self versus no-self 183–186, 194n8, 194n9, 194n12; world presence 190–192 externalism 252 Faddegon, B. 227 fallacies 28–29 fallibilism 253 Fasching, W. 182–199 fear 66–68 First Person (Anscombe) 138 Fludd, R. 46 foundationalism 15; causal 227 Foundations of Arithmetic (Frege) 287, 299–300 Frege, G. 133, 173, 214, 227, 235; context principle 281–300, 301n6; numbers as properties definition 224, 231–232; and Russell, B. 27 Frege: Philosophy of Language (Dummett) 289 Frege’s Conception of Numbers as Objects (Wright) 301n6 Gadādhara 223, 313–314, 320–327; Śaktivāda 313, 327 Gadamer, H.-G. 24 Galloway, B. 139n3, 139n4, 139n6 Gandhi, Mahatma 2, 39–40 Ganeri, J. 1–3, 36–49, 65–84, 129–141, 218–238, 313–331, 344–358 Gaṅgeśa Upādhyāya 13–14, 31, 248, 343; Anumānakhaṇḍa 14; and non-pervasive location 221; Vyāptipañcaka 343 Gārgī 65 Gassendi, P. 45–47 Gauḍa Sānātani 336 Gautama see Buddha (Siddhārtha Gautama) Geach, P. T. 166 ghana (compact mass) 78–80 Gochet, P. 294 Goekoop, C. 31 Goodman, N. 242 grain measures (āḍhaka/droṇa) 229 Grammarians 25–26 Grundlagen (Frege) 235, 281 guṇa (qualities–pleasure/pain/desire/hate) 23, 89, 151, 161n20, 205–207; concept 148, 159n11;

duality 223–227; number problem 219–222; oneness, minuteness and twoness 220–223, 227, 230; and substance distinction 148 Gunaprabhā 139n3 Hacker, P. 8 Halbfass, W. 25 hallucination 268 Haribhadra 12–13; Ṣaḍdar-śanasamuccaya 13 Harvey, P. 120 Heidegger, M. 245 Helārāja 309 Herzberger, R. 305 hetu (evidence-based rationality) 44 Hinduism 38–45, 88–91; Flood story 48; and Śāmkhya view 89–96, 102, 112; and the Upaniṣads 53–64; see also Nyāya Historical and Critical Dictionary (Bayle) 46 Hitopadeśa 19 holism 281, 299; sentence 282, 290, 295 Hume, D. 32, 46, 173–175, 259, 281 Husserl, E. 92–94, 103, 111, 114n8 ‘I am F’ 129–140; first person psychological judgement 132–135; selfhood and subjectivity 129–140; two uses (object/subject) 135–138 Ibn al Arabi 45–47 idealism 244 identity 36–49, 132–137, 161n20, 170–172; brain-mind 178; knowledge 205, 210; non- and number as relation 219, 222–223; reason and dissent 36–48; reductionist view 184; self- 173, 186, 194n12; versus similarity 208–210 ignorance 99–100 Iliad (Homer) 227 illusion 175, 268; and awareness 247–252; misplacement theory of (MTI) 252, 257, 268–274, 276n9; and perception 247–252 inclusion (saṃbhava) 229 Indian Semantic Analysis (Kahrs) 83n15 individualism see anti-individualism Indra (god) 59–60, 66–69, 73, 76 Inductive Reasoning (Bagchi) 28 inference 100–101, 173, 293, 341 infinite regress 212–213, 216n30 inherence (samavāya) 161n20, 203–205, 213–215; self problem 221, 233 inner sense 91 intellect 90 intentionality 29, 98, 110–111 internalism 252 interpretation 245, 249, 254; mis- 282 Islam 39, 45 Īśvara Pratyabhijñā Vimarśinī 166–172, 181n21; Vivṛti 177, 180

INDEX

Īśvarakṛṣṇa 12–14

Iyer, S. 302

Jacobi, H. 8 Jagadīśa 224, 317–321, 326–329; Sūkti 314, 320 jagat (world) 14 Jahān, Shāh 47 Jaina 39, 42–47, 72, 342; axiomatization 355–356; doctrinalism and one-sidedness 345; and Jaśkowski 352–355; pluralism 345–346, 350–352, 356–357; scepticism 345–346; seven-valued logic and truth tables 346–358; syncretism and reality views 356–357; system structure 353–355 James, H. 84n20 James, W. 92–93, 118, 273 Janaka 17, 59, 65 Jaśkowski, S. 352–355 Jāvāli 36, 40–41 Jayanta Bhaṭṭa 12–13, 160n19, 168–172, 176, 209, 295–296, 301n9, 321; Logic Blossoms 21; Nyāyamañjarī 167–168, 171, 301n9 Jayarāma Nyāyapañcānana 46–47; Padārthamālā 46; Siddhāntamālā 47 Jayarāśi 41 Jones, W. 48 Kabat-Zinn, J. 117 Kahrs, E. 83n15 kāma (pleasure) 13, 20 Kamalaśīla 159n17 Kamat 348–349, 355 Kaṇāda 1, 219, 236n1, 326 Kant, I. 19–20, 40–43, 103, 168, 172, 228, 245, 250, 287–288 Kaplan, D. 330n24 karma (action) 21, 31, 54, 63, 205 Kashmir 166–167; Śaiva 166, 170, 173–176, 180 Kāṭhaka Saṃhitā 236n6 Kathāvatthu 37–39, 44 Kauṭilya 7–8, 13, 20, 37 Kavīndrācārya Sarasvati 47 Kaviraj, M. G. 25–26 Khajjaniya Sutta 118 khandhas/skandhas see skandhas/khandhas (Pāli awareness aggregates) Kierkegaard, S. 69 Klawonn, E. 187 knowability, and debate 340–342 knowledge (pramāṇa) 8–15, 17, 33, 43, 101, 110– 112, 169, 204, 211; correct 21; by identification 33–34; memory-based 322–323; moral 32; non(aprāmā) 28; number 228–230; sacrifice 19; self65–84; and totality of things 58; valid sources 100; words and meaning 297–299 Kramer, J. 139n3

363 Krueger, J. 139n9 kṣaṇikavāda (momentariness) 145–147, 146, 154–155, 158n5, 161n23 Kumārila Bhatta 24–25, 44, 109, 158, 245, 310; context principle and meaning 291–299, 300n4; Tantravārttika 44 Kunjunni Raja 302, 309 lakṣaṇā (derivative meaning) 220 language 1–3, 54–55; acquisition theory 292–294; ākāśa and pāribhāsikī terms (Nyāya/Vaiśeṣika texts) 313–329; apoha (exclusion theory) 108–109; first-order 219, 223; sound symbols 306; sphoṭa doctrine 302–312; stages 304–305; and thought 107–109 Le Vayer, F. de La Mothe 47 Leibniz, G. W. 103 Locke, J. 103, 106–107, 206, 245, 281 logic 3, 101, 284, 292; and debate 335–343; Jaina and pluralism basis 344–358; and reasoning 17–34 Logic of Invariable Concomitance (Goekoop) 31 Logic, Language and Reality (Matilal) 31 Lokāyata view 13 lovingkindness 99, 117 Lutz, A. 123 McDowell, J. 256, 259–264; and HCF characterisation 259–260 MacKenzie, M. D. 140n13, 194n3 McTaggart, J. M. E. 169 Mādhavadeva 46; Nyāyasāra 46 Madhusūdana 48 Mādhyamikas 9, 167, 179, 194n3 Mahābhārata (MB) 13, 19, 32–33, 71; and Kauśika 43; rājadharma origin account 13; Śāntiparva 13; and Yudhiṣṭhira 40 Mahādeva Puntāṃkara 48 Mahāsatipaṭṭhāna Sutta 117–119 Mahasi Sayadaw 124 Mahavibhāṣā 93 Maitrāyanī Samhitā 236n6 Maitreyī 75, 79–80 Malebranche, N. 46 Malliṣena 346 manas (mentation) 90, 96, 112, 129–130; kliṣta (afflictive) 113, 129, 139n3, 139n4, 139n6; and mineness 131–132 manasikāra (making in the mind) 119–121, 124–125 Manu 44 Manu Saṃhita 33 Manusṃrti 44 Marshall, J. 48 Mataṅgavṛtti 158n3, 161n23 mathematics 218–219, 234; see also number, theory of

364INDEX Mathurānātha 14 Matilal, B. K. 7–16, 27–28, 31–33, 109, 234, 241–254, 269, 276n7, 281–301, 302–312, 325, 335–343, 349, 356 Mauni (Tamil story writer) 42 meaning 281–301; and the context principle 281–300; sentence 310–311;Vyādi’s theory 314 medical practice 30–31; factors 30; nurse qualifications 30 meditation 54, 99, 110–111; and mindfulness 116–126; states 98, 114n13 Mehta, J. L. 45, 83n8 memory 170–172, 247–251, 271; -based knowledge 322–323; relations 185; revived 248; traces (saṃskāra) 152, 159n17, 170–172, 176, 250, 307 Merleau-Ponty, M. 111 Mersenne, M. 47 Mill, J. 40 Milne, P. 282, 301n6 Mīmaṃsā 13, 23–25, 40, 44, 168, 177, 208; schools 290–291, 300n4, 301n9, 310–311 mind, theories of 87–114; Abhidharma tradition 88–101, 104, 110–112; awareness primary factor 95; Buddhist epistemology 100–111, 118–122; Dharmakīrti 88–89, 100–112; mental events reflexive nature 102–104; mental factors (lists) 96–100, 111, 114n12; perception theory 104– 107; root and branch afflictions 97–100; Śaṃkhya view of self and mental states 89–91, 113n1, 113n2; thought and language 107–109 mindfulness 99; Based Stress Reduction Program (MBSR) 117; and meditation 116–126; see also skandhas (Pāli awareness aggregates) mineness 130–134, 139n3, 140n10, 195n21 misperception 243, 272, 275n1, 276n7; and hallucination 268; Nyāya theory 268–270 misplacement theory of illusion (MTI) 252, 257, 268–274, 276n9 Moffat, D. C. 354–355 Mohanty, J. N. 21, 27–29 mokṣa (freedom from suffering) 13–15, 20–21, 43 morality 19, 24, 32, 98, 119 More, H. 48 Mukerji, A. 2 mystical empiricism 26 Naciketas 66–70 Nāgārjuna 37, 41 Nagel, T. 186, 242 Naiyāyikas 8, 13, 27, 169, 179, 340–342; disjunctivism or anti-individualism 255–276; and eternal self-substance 145–158; and perception 243–254; versus Rāmakaṇṭha and Buddhism 145–165 Nareśvaraparīkṣāprakāśa 158n3, 161n23 Nature of Indian Logic (Mohanty) 27

Navadvīpa 48 Necessity in Indian Logic (Matilal) 27 Nehru, J. 2 Nemec, J. 146 New Essays Concerning Human Understanding (Leibniz) 103 Nibbedhika Sutta 120 niḥśreyasa (ultimate good) 11 Nikāyas (Pāli texts) 116–126; Majjhima 119 Nirvāṇa 43, 91 non-dualism 26 No-self (Buddhist anātman) 91, 166–186; I to I am my world 179–180; Abhinavagupta’s responses 166–168, 173–180; and cognition 166–180, 183–186; other times and subjects 176–178; recognition 168–176; recognition and perception 168–180; Utpaladeva attack (responses) 169–180 Notebooks (Wittgenstein) 65 number, theory of 218–236; cause, effect and eternal substances 219–220, 236n2; and collection (relation/properties) 223–225; and cross-categoricity 220–222; empirical grounding importance 233–234; epistemology 228–229; ‘many’ meaning 229–233; mathematics and philosophy 218–219, 234; as non-identity relation 222–223; objectivity 219, 226–228, 233; oneness and minuteness 220–221; Praśastapāda’s eight moments theory 225–226; as property of objects 219–222; of qualities problem 219–222; rationality and proof concepts 232–235, 235; and self-inherence 221, 233; twoness and threeness 223, 227, 230 Nyāya 2, 15, 17, 24, 27–34, 39, 43–45, 100–101, 105; debate model 336–338; misperception theory 268–270; perceptual theory 255–274; philosophers 46–48; prasthāna 8; school 15, 167, 171–172; tricks of reasoning 30–31 nyāyavidyā (discipline) 12, 47 nyāyavistara (discipline) 12 object 104–108, 173–176, 190–191, 204, 271; class (upādhi) 216n34; corresponding 249; differentiation 211–214; endure or perdure 344; external and internal 168, 227; fictional 252; giveness 191; identification and representation 104–108; language 305; number as property 219–222; past, present and future 169–172, 177, 189–190; perception and realism 247–253; of properties 219–222, 222–226; self and THAT 168–170, 253 Oldenburg, H. 47 Olivelle, P. 70–71, 83n16 Pāli 116–126, 194n4; texts and terminology 116– 126; see also skandhas/khandhas (Pāli awareness aggregates)

365

INDEX

padārtha (categories) 14 pain 22, 34, 157, 177 Pāṇini 30 pāpa (demerit) 13 Paramokṣanirāsakārikāvrtti 158n3, 161n23 parārdha (manyness) 231, 236n6; meaning 229–233 parasitism 256, 264–267 Parasitism and Disjunctivism in Nyāya Epistemology (Dasti) 256 Parfit, D. 184 pāribhāikī (technical) terms 313–329; and ākāśa 313–329; classificatory systems and tables 313, 328–329; common core (indicator) theory 323–324, 330n21; descriptional theory 326–327; direct theory 321–323; Jagadīśa and Gadādhara 314, 317–329; Nyāya and Vaiśeṣika texts 313–330; Patañjali and Devadetta-hood 314–318, 321–322, 325–327; Praśastapāda and Saṅkara Miśra 313–321, 328; syndrome theory 324–325, 330n24; Udayana and nominal type classifications 314–321, 327–328;Vardhamāna and Raghunātha 314–329 parimāṇa (magnitude) 203–205 pariṇāma (transformation) 130–133 Parmenides (Plato) 208–211 particularities (viśeṣa) 219 Patañjali 26, 30, 90, 302, 306–308, 315, 322; Mahābhāṣya 308–309 Paurāṇikas 229 perception (vijñāna) 75–79, 100–112, 149–151, 206, 306; and brain-in-a-vat situation 252–253; Buddhist theory 104–107; conceptual and non-conceptual 169–172; determinant 105; Gods-eye/No-eye view 242; and the HCF line of reasoning 259–260; identification and categorization 106–107; and illusion 247–252; No-selfism and awareness 168–177; nondoubting and non-verbal qualifiers 244–245; number 228; Nyāya theory 255–276; object/ subject 156–157, 162n28, 168–170, 247–253; realist view 241–254; Sautrāntika 105; theory 104–107; veridical and non-veridical 263–267, 270–271; and waterhood 250–251;Yogācāra 105 perceptual anti-individualism 261–273, 275n6, 276n9 Persian 1, 45–48 pervasion 31–32 Philips, S. 270–272 Philosophy of the Commentators (Sorabji) 83n12 Plato 29, 69, 208–210; Ideal theory 205, 211, 215n10 pluralism 37–38, 47, 147, 151, 154, 273; philosophical basis and Jaina logic 344–358 Politics (Plato) 210 postulation 151 Potter, K. 223–226

Prabhācandra 346, 350–351 Prābhākara 247–252, 311; context and meaning 291–299, 300n4; no-illusion theory and reasoning 248–249 pradhāna (primordial nature) 89, 112 pragmatism 273 Prajāpati (creator deity) 59–61, 68–69, 76–77 prajñā (insight) 26 prakṛti (materiality) 89 prāma 29 Pramāṇavārttika 169 Praśastapāda 215n10, 227, 236n6, 313–318; eight moments theory 225–226 Praśastapāda-bhāṣya 320 pratibhā 25–26 prayatna (will/effort) 149 Price, H. H. 253 Priest, G. 356–357 principle, active 233 Prinz, J. 118 Proclus 83n12 properties (dharmas) 148, 161n20; abstract 212; complex 211; numbers as 224; of objects 219–222, 222–226; simple 213–214 properties, of objects 219–222 property-possessors (dharmins) 148, 157 Propositional Calculus for Contradictory Deductive Systems (Jaśkowski) 352 Proximity Principle (PP) 261–265, 272 puṇya (merit) 13 puruṣa (person) 89 puruṣārtha (goals of human life) 13 pūrvapakṣa (opponent’s position) 12 Putnam, H. 241–243, 252, 258, 275n5 Pythagoras Theorem 234–235, 235 qualifier 29 Quine, W.V. 281–282, 290, 294 Radhakrishnan 25, 215n10 Raghudeva 46–48; Dravyasārasaṃgraha 46 Raghunātha Śiromani 8, 14, 223–225, 233, 318–321, 325–326, 329; Kiranāvalī-prakāśa-dīdhiti 314, 319 rajas (energy/activity) 89 Rāmabhadra Sārvabhauma 47 Rāmakaṇṭha 145–158, 194n6; and the self as dynamic constant 145–158, 161n24, 195n13; works of 158n3 Rāmakṛṣṇa, Dharmarāja 14 Rāmāyaṇa 36 rasa (aesthetic rapture) 33–34 Ratié, I. 146 rationality 17–34, 235; axiomatic grammar 29–30; dharma and self-other comparison 32; emotions and rasa theory 33–34; four worries (ends of

366INDEX man) 18–29; humans as special 18–20; in medical practice 22, 30–31; universal concomitance logical form and confirmation 31–32; way to liberation 34 Rationality in Question (Matilal) 33 Rawls, J. 38 realism 110, 228; causal 250; non-promiscuity theory 243; and perception 241–254; transparency 242–243, 247; truth, falsity and bivalence 242–243, 251 reason 36–49; before identity 37; enriched 42–45; and overlapping consensus 38–39; path of 41; personae 42; public 38–39, 45; reach and Amartya Sen 36–42; and settled principle 39; and seventeenth century globalization 45–48 reasoning 178; and logic 17–34 rebirth 54, 63 Recanati, F. 133 receptivity 37 recognition (pratyabhijñā) 168–176; from nonapprehension to synthesis 173–176; and medieval Buddhist dialectic 168–169; perception and awareness 168–177; transcendental arguments 175–176 reductionism 250 reflexivity 103, 132 reincarnation theory 18 relations 29; circumtaining 224; number of 224 Renou, L. 29–30 representationalism 245–247, 251 Republic (Plato) 210 Rescher, N. 355–356 Resnik, M. 282, 301n7 Ṛgveda 66, 83n8 Roebuck,V. 84n16 Rorty, R. 241–243, 254 Routley, R. 356–357 rūpa (body physical matter) 118, 151 Russell, B. 205–206, 214, 243, 246, 322–323, 327; definite descriptions theory 283–286; and Frege, G. 27; paradox 221, 232–233; Theory of Types 221; uniqueness clause 223 Ryle, G. 253 Śabara 19, 168 sābda (linguistic sound/Eternal Verbum) 302–305,

308–310 sādṛśya (resemblance) 208 Śaiva Siddhānta 145, 158n3 Śaivism 2, 145–146; Kashmiri 2 sākṣin (witness) 183, 188 śakti (word power) 251 śakya (denotation) 204, 214–215 śakyatāvacchedaka (connotation) 204, 215n4 Śālikanātha 208, 297–298 Samantabhadra 346

sameness 152, 161n20 Śaṃkara 90–91, 22–25, 47 Śāṃkhya 12–14, 22, 32–34, 89–96, 102, 112, 139n3, 208; and intellect 90; Kārikā 22; principles 89; self and mental states 89–91, 159n11, 182–183, 194n1, 195n13 saṃyoga (contact between two substance) 203–205 Saṃyutta Nikāya 182–183 Sanatani, Gauḍa 336 Śāṇḍilya 62 Śaṅkara Miśra 54, 83n9, 190–194, 194n9, 194n11, 194n12, 195n16, 196n22, 197n30, 221, 229–230, 233, 315–321, 328; Brahmasūtrabhāṣya 47, 83n9, 192, 195n15, 197n27; Kaṇāda-rahasya 315 saṅkhāra (volitional activities) 119, 183 saññā (Pali khanda) 118–120 Sanskrit 1, 7–10, 17–19, 25–30, 39, 42, 46, 243 Śāntarakșita 159n17 Sartre, J.-P. 103 Sarvāstivādin 169; past and future reality 169 śāstras (treatises) 9–14; Artha 13, 20; Dharma 20, 32, 234; Kāma 20; Nāṭya 20; Tarka 17 ṣaṭ-tarkī 12–13 sati (mindfulness) 117, 123–125, 126n1 sattva (transparency/buoyancy) 89 Sautrāntika school 22, 105, 162n28, 245–246 scepticism 252–253, 261, 266–267, 345–346; Cartesian frame 267, 271–274 Schmithausen, L. 140n10 sectarianism 15 secularism 36, 39, 55–60, 64, 89, 112 self (ātman) 23; as agent 149–153, 160n19; as agent of cognition 149–153, 149, 152, 153, 160n19; cave (guhā) metaphors 70–71; and concealment 71–75; and consciousness 153–158, 153, 155, 156, 158, 183–190, 195n17; as dynamic constant 145–165; minimal 133, 139n9; not a consciousness object 75–77; one and all hidden connection 77–79; philosophical debate 145–146, 158n3; Rāmakaṇṭha versus Nyāya and Buddhism 145–165; reluctance of the sage 65–69; a Śāṃkhya view 89–91; subjectivity and I 129–140; as substance 148–151, 148, 159n7, 159n17; and thinking periphery 79–82; true or false (soul/ desires) 68–69, 73–81, 82n4, 83n15; as unitary essence 146–147, 146, 147; Upaniṣadic 65–84; versus no-self 183–186 self-consciousness 129–138 selfhood 129–140; and I 129–140; and awareness 129–140; criticism 41; giveness 186 self-identity 173, 186, 194n12 self-illuminating 185–187, 194n11, 195n17, 195n18 self-knowledge 65–84 Sellars, W. 106–107 semantics 219, 292–293; epistemologizing 243; value 287

INDEX

semiotics 335–343 Sen, A., and reason 36–42 Sen, P. K. 281–301 sense organ 271 set theory 212–213 Sharma, A. 84n16 Sherma, R. 84n16 Shoemaker, S. 131 skandhas/khandhas (Pāli awareness aggregates) 118–122, 129, 138n1, 139n3, 148, 159n10, 167, 184, 193, 197n29 Sluga, H. 282 Socrates 37, 69, 210–211 Somānanda 145 Sophist (Plato) 210 Sorabji, R. 83n12 Sorites paradox 231–233 soteriology 15 spaceness (diktva) 213–214 sphoṭa doctrine 302–312; and Devadetta-hood 307; external and internal 311–312; and gauḥ 307–308; nāda distinction 303–308; and śabda 302–305, 308–310; sound distinctions (primary/ transformed) 308–309 Spinoza, B. 46–47 Śrīdhara 226–230, 233, 236n5 śruti (scripture) 23–24 Staal, F. 31 Sthiramati 134–135, 139n3 Stoics 37, 42 Strawson, G. 136–137 Strawson, P. F. 131–132, 135–137, 178, 250 subjectivity 129–140, 167; selfhood and I 129–140 substance (dravya) 148–149, 205–206, 228; eternal 219–220, 236n2; self as 148–151 śūdras (social class) 14 suffering 64 sūtras (sayings of the Buddha) 9, 14, 88, 221; Brahma 83n9; Kāma 20; Mīmāṃsā 19, 30; Nyāya 30, 160n19, 241–243, 269, 335–337, 340; ritual 45; Vaiśeṣika 30, 83n10, 219–220, 225, 228, 313, 326; Yoga 30, 245 svasaṃvedana 185 svāsthya (coinciding with oneself) 42–43 Śvetaketu 57–58, 78 symbols 54–56 Syntagma Philosophicum 47 syntax 292–293 synthesis (pratisandhāna) 160n19, 168, 174–178, 288 Taber, J. 24 Tagore, R. 2, 37–40; idea of India 45 Taittirīya Saṃhitā 236n6 tamas (inertia/obstruction) 89 tarka (reasoning) 23–24, 27–28, 39, 44, 234 tarka-ṛṣi (wise seer) 19

367 tat tvam asī 78, 83n16 Thakur, Dr A. 10 Theaetetus (Plato) 29 Theories and Things (Quine) 281 Theravāda Buddhism 1, 116, 123–125 Thirty Verses 134 Thompson, E. 87–115, 116–128, 140n10 thought 107–109; de se 133–134; and language 107–109; tree as idea 107–108 Thought,The (Frege) 133 thought-signs 243 Tibetan 1, 122–123 Torella, R. 146 Tractatus (Wittgenstein) 300 training, attention 122–124 Travels in the Mogul Empire (Bernier) 47 trust 266–267, 274 truth-conditions 284–286 Twin Earth experiment 258–259, 266–267, 275n5 Two Dogmas of Empiricism (Quine) 281 Uḍasta Cākrāyaṇā 65, 75 Udayana 27, 172, 212–215, 216n34, 222, 230, 233, 342; ākāśa and nominal types classification 314– 321, 327–328, 329n9; Kiraṇāvalī 314; Pariśuddhi 314–315 Uddaka Rāmaputta 123 Uddālaka 57–58, 61 Uddyotakara 10–14, 109, 231, 264; Nyāyavārttika 10, 160n19; wheel development 339–342 Unity in Being 45 universals (sāmānya/jāti) 203–216, 246, 308, 313, 344, 348; Being and Becoming 205; classifications 205–206, 215n9; cowness and blueness differentiation 206–214; definite descriptions (son of Yajñadatta) 214–215; and inseparable inherence (samavāya) 203–205, 213–215; Nyāya-Vaiśeṣika theory statement 203–207, 215n1, 215n5, 215n10; other school objections 207–210; positive and negative 340–341; restrictive conditions 210–215; sattā and bhāvatva (existence–subsistence distinction) 206, 216n12; similarity versus identity 208–210 Universals (Staal) 31 upacāra (metaphor of self) 132–138 Upadeśasāhasrī 183, 192–193, 195n16, 195n17, 195n20, 196n22, 196n25 upādhi (class of objects) concept 216n34 upamāna (analogy) 101 Upaniṣads/Upanishads 2, 23, 42, 47, 53–64, 82n1; Īśā 53, 62; Aitareya (AU) 53, 56–58, 81; Bṛhad Āraṇyaka (BU) 17, 53, 56, 59–60, 63, 65–66, 70–71, 74–75, 79–80; Chāndogya (CU) 53, 57–63, 68–73, 76–81, 83n11, 83n16; correlation 55–57; cycles 62–63; emergence and resolution 57–58; hierarchy and secret teachings 58–61, 77; and

368INDEX Hinduism 53–64; integrative vision (totality) 63–64; Kaṭha (KaU) 53, 60, 63, 66–74, 77, 82n5; Kauṣītakī (KsU) 53, 74, 82n1; Maitrī 53; Māṇḍūkya 53, 60; Muṇḍaka (MuU) 53, 70–71, 77; paradox of wealth/poverty 61–62; Persian translation 45–47; and self 65–84; Śvetāśvatara (SU) 53, 61, 72, 77, 81, 82n2; symbols and images 53–57; Taittirīya (TU) 53, 66–67, 71–72 Utpaladeva 145; and Buddhist No-selfism 166–180 utpatti (mathematical demonstration) 234 Vādavinoda 37 Vādideva Sūri 346, 350–351 Vaidya, A. J. 255–277 Vaiśeṣika 2, 30, 46, 135, 149, 183, 195n13; number theory objectivity and proof 218–236; Nyāya 27, 174, 203–216; sound theory 309; universals theory 203–216; view 13 values, Asian/Western 36 Varanasi 46–48 Vardhamāna 314–322, 328; Kiraṇāvalī-prakāśa 314–316 varṇas (social groups/caste) 14 Vasubandhu 91, 95, 159n17, 169; and Asaṇga on self-consciousness 129–138, 139n3; Pañcaskandhaka 139n3 Vātsyāyana, Pakṣilasvāmin 8–10, 13–15, 20, 168, 229, 336, 265; Nyāyabhāṣya 9, 160n19 vedanā (emotion feeling tone/pleasure-pain) 118–119 Vedānta 2, 23, 34, 55, 89, 103, 246; Advaita 2, 25, 54, 91, 117, 145; and witness consciousness 182–193, 194n1, 194n2, 194n4, 194n10 Vedas 13–14, 17–19, 21–26, 30–32; moral authority 43–44; Rig 45; ritual 22, 54, 62–63; Sāma 48 verbal testimony 100 Vetter, T. 107 vidyā (disciplines/learning) 13, 20, 33–34 vidyāsthāna (learning branch) 7, 12

vikalpa (conceptual fabrication) 130–132, 138, 209, 245 vimarśa (self-consciousness) 172 vinaya (monastic discipline) 88 viññāṇa/vijñana (phenomenal consciousness) 120–121, 130–131, 139n3; alaya 130, 140n10; mano- 130–131 Virocana (anti-god) 59, 68–69, 73 Vivekacūḍāmaṇi 190, 197n30 Vṛtti 303 Vrttikāra 160n19 Vyāḍi 314, 322 Ward, W. 17 Watson, A. 145–165, 195n13 witness consciousness 89, 113n3; de-superimposition process 192–193; phenomenological reflections 182–193, 194n2, 196n26; presence and experience 190–192; self versus no-self (interpretation/denial) 183–186, 194n3 Wittgenstein, L. 24–25, 65, 81, 135–137, 178, 294, 300 Word and Object (Quine) 281 Wright, C. 282, 301n6 Yājñavalkya 13, 17, 59–63, 65, 74–76, 79–80 Yama 66–68 Yāska 19; Nirutka 19 Yaśovijaya Gani 47–48 Yoga 54, 90, 178, 182, 194n1; visions 25–26, 30 Yogācāra 22, 105, 139n4, 140n10, 162n28, 169, 173, 179, 185–187, 195n13 Yogācārabhūmi 134 Yogavāśiṣṭha Rāmāyaṇa 19 yogyatā (semantic fitness) 293 yukti 39 Zahavi, D. 132–134, 139n8, 139n9, 195n21, 197n26