Perceiving Reality: Consciousness, Intentionality, and Cognition in Buddhist Philosophy

OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN Perceiving Reality Consciousness, Intentionality, and Cognition in Buddhist Philosophy !"! Christian Coseru 1 00_Coseru_Prelims.indd iii 7/4/2012 7:15:57 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN C H A P T ER 1 ! Introduction: Taking the Structure of Awareness Seriously If one is impressed by the irreducible uniqueness of mental life, and yet happens to be a natu- ralist, or even a physicalist, one would want to carve out a niche within the heart of one’s naturalism in order to find a place secure enough for the intentional. (Mohanty 1986: 505) “T he mind,” Hume once argued, “can never exert itself in any action, which we may not comprehend under the term of perception.”1 Hume was right, and in that sense he comes very close to a position that Buddhist philosophers have advocated for two and a half millennia: perceptual aware- ness in its multifaceted forms is the beginning and end of our conscious lives. This book is about the structure of that perceptual awareness, its contents and character, and about what we stand to learn when we realize that the world we inhabit is inseparable from our perception of it. A distinctive and influential philosophy of perception emerges in the Buddhist tradition from the analyses of consciousness and cognition asso- ciated with that system of thought and method of descriptive analysis known as the Abhidharma.2 With Dignāga and Dharmakīrti, the initia- tors of a Buddhist school of epistemology, this systematic inquiry into the 1. Hume (2000: 3.1.1.2). 2. Lit. “concerning” (abhi) “the teachings” (dharma), usually translated as “higher doctrine”—the systematic scholastic analysis of the Buddha’s teachings as con- tained in the eponymous genre of philosophical literature. See Frauwallner (1995) and Willemen, Dessein, and Cox (1998) for detailed discussions of the origins and scope of this literature, which develops over a period of several centuries beginning around 300 B.C.E. (1) 01_Coseru_CH 1.indd 1 7/4/2012 6:25:08 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN (2) Perceiving Reality philosophical foundations of empirical knowledge plays a formative role in shaping what would eventually become the dominant approach to Buddhist philosophy of mind at such well-established universities as Nālandā and Vikramaśīla. According to this approach, a philosopher’s views on percep- tion are central to his or her epistemological and metaphysical commit- ments. Thus, questions about what there is and how we come to know it, become questions about the nature of awareness itself, its modes of disclo- sure, and its contents. And here the Buddhist comes surprisingly close to a position that is widespread among Western phenomenologists: to say about something that it exists is not primarily to make an ontological assertion but to provide a descriptive account of experience. On this view of perception, empirical awareness is intrinsically perspectival: it does not simply manifest a given object, a particular, but it is also in some sense self-manifesting, self- given. To perceive is to occupy a specific vantage point. This dual aspect view of awareness, whose origins may be traced to early Abhidharma accounts of cognition in terms of luminosity (viz., the lamp that makes itself manifest while illuminating others),3 becomes an axiomatic principle of Buddhist epis- temology and a subject of debate between the Buddhists4 and their principal opponents, the Naiyāyikas (“philosophical logicians”) and the Mīmām  sakas (“hermeneuticians”). The central concern of this book is a range of arguments advanced by two prominent philosophers at the university of Nālandā, Śāntaraks ita and Kamalaśīla,5 in support of the role that a particular understanding of the structure of awareness must play in settling epistemological disputes. What is significant about these arguments is that they provide a model for integrating the phenomenological and cognitive psychological concerns of Abhidharma traditions within the dialogical-disputational context of Buddhist epistemology. In unpacking the central arguments of Buddhist epistemology I am not simply pursuing a project in the history of philoso- phy, much less in the history of Buddhist thought alone. Rather, I am com- mitted to the view that both the specific style of these broadly Sanskritic 3. The Mahāsām  ghika (“Great[er] Community”) view of self-cognition found in Mahāvibhās ā (“Great Commentary”) articulates precisely such an account. See Yao (2005: 15). 4. The “Buddhists” stand here for those who adopt the epistemological concerns of Dignāga and Dharmakīrti. These authors’ views on mind and cognition differ signifi- cantly from those advanced by Mādhyamika philosophers who follow Candrakīrti’s (fl. 640) critique of reflexive awareness (for which, see chapter 3.3). Unless other- wise noted, all unspecified uses of “Buddhist” (or its plural form) refer to the former group. 5. The dates for Śāntaraks ita (c. 725–788) and Kamalaśīla (740‒795) are those given by Frauwallner (1961: 141–143), based on textual references and historical records, including their visits to Tibet at the invitation of king Khri srong lde btsan (c. 740–798). 01_Coseru_CH 1.indd 2 7/4/2012 6:25:08 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN I N T R O D U C T I O N : T A K I N G T H E S T R U C T U R E O F AWA R E N E S S S E R I O U S LY 3 argumentative strategies and the universality of the metaphysical and epis- temological theses under dispute are better showcased (and understood) if made continuous with contemporary philosophical concerns. The principal methodological reason for emphasizing continuity over comparison reflects a specific intuition about the scope of philosophical inquiry: one which says that its problems, though often couched in historically and culturally contin- gent terms, are nonetheless grounded in all aspects of conscious experience for a person at any given time.6 To the extent that contemporary philosophi- cal debates—and the theoretical advances and empirical findings that inform them—provide clear accounts of a wider range of conscious experience, they can be profitably used in probing the Buddhist epistemological project such that the implications of its theses (as well as their strengths and limitations) are fully borne out. A central argument of this book is that epistemological inquiries in India, particularly with regard to examining the sources of reliable cognition, have never displayed the sort of non-naturalism characteristic of the Cartesian and Kantian traditions in Western philosophy. Thus, with the return to naturalism in epistemology and phenomenology, hence to understanding cognition in embodied and causal terms,7 we are now in a better position to appreciate the contributions of Buddhist philosophers to epistemology. Let me clarify from the outset that I take naturalism to be a commit- ment to considering the empirical evidence from the sciences of cognition in settling questions about the acquisition of beliefs. More broadly, natural- ism refers to the notion that reality is exhausted by nature, though there is no agreement among contemporary philosophers on exactly what counts as “nature.” Indeed, philosophers with weak commitments to naturalism oper- ate with a rather unrestricted notion of nature, whereas stronger adherents to naturalism define it more stringently.8 Eliminativist positions, which seek to reduce mental content to, say, neurophysiological processes do not, I think, do justice to the phenomenal or qualitative aspects of experience, and provide little or no support for the framework of a phenomenological epistemology. My position on naturalism, which I shall henceforth refer to as phenomenological naturalism, closely aligns with the enactive and embodied 6. A similar approach is at work in Ganeri (2001). Note that Ganeri does not label his analysis of the philosophical literature of classical India, which he considers nei- ther comparative nor historical in scope. 7. For one of the earliest accounts of causal theories of knowledge, see Goldman (1967). A causal account of knowledge simply states that a person’s belief that p, say that it is raining, is true iff it is the case that p (that is, if it is actually raining). For a review of various causal theories of mental content, principally those advanced by Fred Dretske, Jerry Fodor, Ruth Garrett Millikan, David Papineau, Dennis Stampe, and Dan Ryder, see Rupert (2008). 8. See Papineau (2007) for an extensive survey of various positions on naturalism. 01_Coseru_CH 1.indd 3 7/4/2012 6:25:08 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN (4) Perceiving Reality approach to cognition: cognitive awareness is to be thought not as an inter- nal state of mind or brain locked into linear causal chains of sensory input and behavioral output. Rather, it is to be understood as a structure of com- portment, an intentional orientation and attunement to a world of actions, objects, and meaning.9 The enactive approach to cognition, as championed by its proponents, is non-reductionist because it relies primarily on dynamic systems theory, whose working premise is that in order to understand com- plex systems such as sentient beings, we must pay close attention not only to their constitutive elements but also to their organization.10 As such, it allows for cognition to be understood in causal terms without reducing the contents of awareness to noncognitive elements. A focus on causal accounts of cognition is shared by all Indian epistemo- logical theories, Buddhist or otherwise. It is with Dharmakīrti, however, that an examination of the underlying processes of cognition becomes instru- mental in determining which epistemic practices are conducive to effec- tive action, captured by the well-known theory of the pragmatic efficacy of cognitions. By contrast, modern Western philosophers, beginning with Descartes, argue that justification, specifically the justification of reasons for why certain beliefs ought to be classified as knowledge, is the main focus of epistemic inquiry. What sets the two traditions apart seems to be the fact that, as Jitendranath Mohanty pointed out some time ago, “the distinction, common in Western thought, between the causal question and the question of justification was not made by the Indian theories.”11 While Mohanty is right about the absence of this distinction in Indian and indeed Buddhist epistemology, this should not necessarily be seen as an unfor- tunate oversight. Rather, the absence of a distinction between the causal ques- tion and the question of justification is indicative of the fact that epistemic inquiry in India is primarily driven by pragmatic rather than normative con- cerns (that is, by concerns about how we come to believe something rather than why might we be justified in believing it). Indeed, if Indian epistemolo- gies treat as warranted only that cognition which corresponds to its object and 9. I follow Thompson’s (2007b: 38–43) account of the relevance of dynamic sys- tems theory in bridging human experience and cognitive science. This conception of enactive and embodied cognition finds its roots in Husserl’s notion of the life-world (Lebenswelt). The paradigm of embodied and enactive cognition is explored at length in Dreyfus (1979), Hutchins (1995), Clark (1997), Hurley (1998), Lakoff and Johnson (1999), Noë (2004), Gallagher (2005), and Thompson (2007b). 10. Th is account of the relevance of dynamic systems theory to understanding cognition in causal terms follows closely Varela (1999) and Thompson (2007a). As Thompson notes, “because dynamic systems theory is concerned with geometrical and topographical forms of activity, it possesses an ideality that makes it neutral with respect to the distinction between the physical and the phenomenal, but also applicable to both” (2007a: 357). 11. Mohanty (2000: 149). 01_Coseru_CH 1.indd 4 7/4/2012 6:25:08 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN I N T R O D U C T I O N : T A K I N G T H E S T R U C T U R E O F AWA R E N E S S S E R I O U S LY 5 is produced in the right way, it seems to me, they have a way of explaining epistemic dispositions as resulting from our embodied condition (rather than as attitudes of a disembodied cogito serving the justification of its beliefs).12 This book should be of interest both to philosophers looking for non- Western approaches to consciousness and cognition, and to scholars of Indian and Buddhist philosophy primarily interested in various aspects of the Buddhist epistemological tradition. For the benefit of readers with little or no background in Indian and Buddhist philosophy I have adopted throughout English equivalents of technical philosophical terms, providing the original Sanskrit term in parentheses where necessary. In order to cap- ture the polysemy of Sanskrit philosophical concepts and show sensitivity to context, occasionally a key technical term would have more than one English equivalent. For instance, vijñāna is translated both as “consciousness” and as “cognitive awareness,” since it designates both a basic form of sentience and a discerning type of awareness. Likewise, pramān a,13 which I will translate as “source of knowledge” or “reliable cognition,” could be rendered as “epistemic warrant” (since it also refers to the specific aspect or quality of a cognition that makes it an instance of knowledge14). Brief explanations for the choice of English equivalents of Sanskrit philosophical concepts are provided in 12. As it turns out, recent research programs in cognitive psychology have not validated the distinction between causal and normative account of our epistemic intuitions. Indeed, traditional normative accounts of epistemic intuitions have been challenged in recent years by a substantive body of empirical research, which shows that there are significant variances across cultural subgroups in the way the objects of experience are described and categorized. These studies appear to confirm Alvin Goldman’s (1992: 160) point that the perceived uniformity of our epistemic intu- itions is most likely attributable to the “fairly homogeneous subculture” to which philosophers belong. See Nisbett and Ross (1980), Haidt, Koller, and Dias (1993), Weinberg, Nichols, and Stitch (2001), and especially Nichols, Stich, and Weinberg (2001) (the last two studies for an interesting take on epistemology as ethnography). I thank my colleague Sheridan Hough for bringing to my attention the relevant work of Nichols to the topic of the empirical investigation of our epistemic intuitions. 13. Th is key term is variously translated as “true cognition” or even “truth,” and more literally as “means of knowledge,” though the sense of the term varies signifi- cantly depending on the context. For instance, Matilal (1985: 203) takes pramātva, usually translated as “truth,” to be just one of meanings of prāmān ya, the theory of the means for the apprehension of truth. Following his brief survey of various translations in French and German for pramān a, Ruegg (1994b: 403‒404) observes that although we do not yet have a satisfactory translation, “means” or “instrument of knowledge (or cognition)” captures the sense of the term in its technical use. At least in the Buddhist context, a pramān a is primarily concerned with, and oriented toward, achieving pragmatic aims, and thus bears on objects capable of bringing about intended results. See also Hugon (2011). 14. The notion of “epistemic warrant” I have in mind here is as developed by Alvin Plantinga who extends the traditional notion of epistemic justification (of true beliefs) also to include as a necessary condition for the warrant the optimal functioning of one’s cognitive systems in the production of the respective belief. See Plantinga (1986, 1993). 01_Coseru_CH 1.indd 5 7/4/2012 6:25:08 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN (6) Perceiving Reality the footnotes, where the reader will also come across modern philosophical glosses on different concepts and theories. Now, let me briefly explain why I am situating my project at the intersec- tion of phenomenology and analytic philosophy of mind. First, in the giving and defending of “reasons” Nyāya and Buddhist epistemologists operate on similar principles to those found in the tradition of analytic epistemology. Drawing from contemporary debates in epistemology between, for instance, internalists and externalists, or foundationalists and coherentists, becomes an essential and indispensable step in assessing the positions of the Buddhist epistemologists on such topics as the nature of evidence.15 But it also guaran- tees an innovative treatment of these South Asian philosophical materials: the goal is to go beyond the task of historical reconstruction and endeavor to propose novel solutions to enduring and genuinely universal philosophical problems.16 Second, the Abhidharma traditions with their phenomenological approach to investigating the elements of existence and/or experience pro- vide the basis on which Dignāga, Dharmakīrti, and their followers deliberate on such topics as the ontological status of external objects and the epistemic import of perceptual and intentional states of cognitive awareness. Finally, approaching historical authors that belong to a different cultural and philo- sophical horizon than that of the modern West also demands that herme- neutical considerations come into play. In framing the approach of this book in philosophical terms I do not mean to downplay alternative methodologies, principally those concerned with text-critical (that is, philological) issues pertaining to editing and translat- ing Buddhist philosophical texts.17 Nor do I want to imply that such studies fail in some significant way to capture the scope of the Buddhist epistemo- logical enterprise. My contention is simply that our understanding of, and engagement with, the Buddhist epistemological project is better served by the methodologies and conceptual resources of philosophy. The real chal- lenge, as I see it, is devising the best possible approach to integrating such complex and diverse epistemological canons as those of classical India and the West.18 15. For an Indian philosophical account of evidence, see Matilal (2002b). For a contemporary philosophical account of evidence, see Achinstein (2001). 16. For an encouraging recognition of the universality of some of our philosophi- cal problems, and the ways in which comparative philosophy can make a compelling case for integrative solutions that bridge the cultural and historical divides between Western and non-Western philosophy, see Strawson (1998: 327). 17. Detailed discussions pertaining to the translation and interpretation of Buddhist philosophical literature are found in Lopez (1988), Powers (1993), Ruegg (1995), and Tillemans (1997). I address some of these interpretive issues in chapter 2. 18. Significant contributions in this direction are found in Matilal (1971, 1986), Mohanty (1992b), Dreyfus (1997), Ganeri (1999a, 2001), Siderits (2003), and Arnold (2005a). 01_Coseru_CH 1.indd 6 7/4/2012 6:25:08 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN I N T R O D U C T I O N : T A K I N G T H E S T R U C T U R E O F AWA R E N E S S S E R I O U S LY 7 The working assumption of this book is that the tradition of episte- mological inquiry within Buddhism could be seen as advocating a form of naturalism that links it strongly to the Sautrāntika and Yogācāra19 formu- lations of Abhidharma reductionism. Although epistemological theories in Buddhism ultimately attempt to justify a core set of Buddhist principles, and thus reflect specific doctrinal concerns, questions regarding the founda- tion of these principles are addressed on both metaphysical and empirical grounds. I will argue that naturalized epistemology (and phenomenology) and the cognitive sciences which inform it share with the Buddhist episte- mological tradition of Dignāga and Dharmakīrti a common concern: devel- oping a theory of knowledge that does not divorce logical arguments from descriptive accounts of cognition. For the Buddhist, the ultimate source of these descriptive accounts is the Abhidharma, in its specifically Sautrāntika- Yogācāra synthesis of thinkers like Vasubandhu (c. 350 C.E.).20 We must be careful, however, not to read too much into a philosophical program whose roots are historically and culturally different that those of the modern West. At the same time we can (and indeed should) profit- ably engage the thought of historical Buddhist thinkers when such thought addresses perennial philosophical problems. Such investigations, of course, must be mindful of Hans-Georg Gadamer’s caveat about the objectivist claim that it is possible to interpret the thought of a historical author in such a way as to suggest that the interpreter does not enter in the event.21 One of the most enduring themes of the Sanskritic philosophical tradi- tion is the debate over the number and nature of reliable means of belief formation (although, this tradition is principally concerned with cognitive “events” or states of “cognition” (jñāna) that are epistemically warranted 19. Sautrāntika and Yogācāra stand respectively for “Follower of [Canonical] Discourse School” and “Practice of Yoga School.” 20. Vasubandhu is generally thought to have written his earlier works, in particu- lar the Abhidharmakośa, from a Sautrāntika perspective, while most of his later trea- tises are expositions of various aspects of Yogācāra philosophy. Th is claim is based on the assumption that there is only one Vasubandhu. On the hypothesis of two Vasubandhus, proposed because of discrepancies in matters of style and doctrine in the works that are attributed to him, see Frauwallner (1951). 21. Gadamer (1975) argues for the well-known “pluralist” or “anti-objectivist” view, according to which understanding a text, historical event, or cultural phe- nomenon is a complex process in which there is a “fusion of horizons,” such that the object of interpretation and the interpreter’s perspective are not easily distin- guished. Gadamer’s insistence on there being a plurality of views does not mean that he endorses a type of relativism (he does acknowledge that there are criteria for distinguishing between right and wrong interpretations). Rather, he simply insists that an interpreter’s prejudices and prejudgments (Vorurteile) are not only inescap- able but also indispensable, for no interpreter stands outside the horizon of history. For a critical defense of Gadamer’s view that objectivism is not possible because the object of understanding is always constituted anew in each act of understanding, see Weberman (2000). 01_Coseru_CH 1.indd 7 7/4/2012 6:25:08 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN (8) Perceiving Reality (pramā), rather than with “beliefs” proper22). Much of this debate centers on an examination of our cognitive capacities and on issues pertaining to the structure and function of Sanskrit grammatical thought. While controver- sies are at the heart of any tradition of epistemological inquiry, in India they most often reflect (and are enforcing of) scholastic and doctrinal affi liation: philosophical views are framed within a specific scholastic paradigm which is in turn expanded within a genre of commentarial literature that seeks to explain and facilitate access to the original insight. When innovation occurs, as it often does, it is presented as statements or modes of reasoning that make explicit what might have only been alluded to in a canonical or founda- tional text of that particular school of thought. Dignāga’s Pramān asamuccaya (“Collection on the Sources of Knowledge,” hereinafter the Collection) is generally recognized as the first systematic treatment of the sources of knowledge from a Buddhist perspective. Indeed, Dignāga (c. 480‒540) is rightly credited with having inaugurated a new era in Indian thought with his synthesis of epistemological, grammatical, and psychological theories.23 This new model of epistemological inquiry, which is expanded in great detail less than a century later by Dharmakīrti (c. 600‒660)24 (and significantly altered in certain respects) rests on two sets of premises: first, it adopts a specific view of language as a means of reason- ing and deliberation first articulated in the Sanskrit grammatical tradition but also formalized by the early Naiyāyikas; second, it incorporates insights from the Abhidharma traditions concerning the phenomenology of percep- tion and conceptual cognition. The specifics of these two sets of premises, their formative influence, and the manner in which they contributed to the development of a Buddhist epis- temology of perception, in particular as reflected in the encyclopedic work of Śāntaraks ita and Kamalaśīla—the Tattvasam  graha (“Compendium of True Principles”) and its Pañjikā (“Commentary”),25 hereinafter the Compendium and its Commentary—form the major coordinates of this book. These two sets of premises were instrumental in the adoption by the Buddhist episte- mologists of a phenomenological perspective in describing the role of percep- tion for knowledge. Key to this phenomenological account of perception is the notion that self-awareness is constitutive of perception. 22. For discussions of the difference between cognitive “events” and states of “belief,” see Matilal (1986: 101ff.), Mohanty (1992b: 134‒135), and Patil (2009: 42‒43). For an account of “belief” that captures both its dispositional (or occurrent) and phenomenal (or conscious experience) aspects, see Schwitzgebel (2002). 23. Dignāga’s contribution to the development of Indian logic and epistemology is treated at length in Vidyābhus an a (1921), Mookerjee (1935), Frauwallner (1959), Hattori (1968), Hayes (1988), Matilal (1998), and Pind (2009). 24. For a review of the debate about Dharmakīrti’s dates, see Lindtner (1980). 25. For recent reviews of the scholarly literature on the TS/P, see Funayama (1992), Steinkellner and Much (1995), and McClintock (2010). 01_Coseru_CH 1.indd 8 7/4/2012 6:25:08 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN I N T R O D U C T I O N : T A K I N G T H E S T R U C T U R E O F AWA R E N E S S S E R I O U S LY 9 I will argue that—in conceiving of self-awareness as constitutive of per- ception— Śāntaraks ita and Kamalaśīla, like other followers of Dignāga and Dharmakīrti, share a common ground with phenomenologists in the tradi- tion of Edmund Husserl and Maurice Merleau-Ponty,26 and with analytic phi- losophers of mind interested in phenomenal consciousness,27 all of whom contend that perception is best understood as bearing intentional content. I refer here primarily to the notion of “intentionality” as initially developed by Brentano and (following him) by Husserl and Merleau-Ponty. On this phe- nomenological account of intentionality, to perceive an object (or to have a perceptual experience) is to apprehend an intentional relation: whether the object intended in perception (the one the perception is of ) is real is less important than how it is intended or that it is intended. Indeed, one of the features of intentionality is that it reveals the co-constitutive nature of perception and that which is perceived; as such, it discloses the world rather than attempting to establish a relationship to a discrete, “external” world. I shall argue that the Buddhist epistemologists’ treatment of self-awareness (svasam  vedana) displays similar features. Can knowledge be established on a foundation of phenomenal or conscious experience, specifically on a type of cognitive awareness that is nonconcep- tual and inerrant, as the Buddhist epistemologists claim? Isn’t this type of foundationalism, like the sort one associates with the sense data theorists, a discredited philosophical program (as Wilfrid Sellars and all critics of the “Myth of Given” have argued)? Can the Buddhist epistemologists’ definition of perception be justified in terms of our understanding of the phenomenol- ogy of perception? Is it correct to assume that what turns the continuous flow of experience into perceptually distinct objects are the conceptual and categorizing tendencies of an embodied mind? Finally, how might we reason on the basis of such empirical testimony? These questions are at the heart of a lengthy philosophical dispute between the Buddhist epistemologists and their opponents, chiefly the Naiyāyikas and the Mīmām  sakas. Insofar as this dispute can be integrated into contemporary philosophical debates, it ought to invite the same sort of scrutiny one would expect of all enduring philosophical problems, regardless of their historical origin. If philosophi- cal issues raised by the Buddhist epistemologists can be shown to have any 26. Note that philosophical phenomenology as currently understood is a het- erogeneous discipline, including, but not limited to, such subfields as transcenden- tal phenomenology (concerned with how objects are constituted in pure awareness), naturalistic phenomenology (concerned with how consciousness presents the natu- ral things), existential phenomenology (concerned with the experience of choice and action in concrete situations), and hermeneutical phenomenology (concerned with the interpretive structure of experience). For a detailed survey of the various subfields of phenomenology, see Embree et al. (1997). 27. Including such representative accounts as one finds in Siewert (1998), Kriegel (2009), Chalmers (2010), Bayne (2010), and Schwitzgebel (2011). 01_Coseru_CH 1.indd 9 7/4/2012 6:25:09 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN ( 10 ) Perceiving Reality relevance beyond the confines of the history of Indian and Buddhist thought, the need to engage them from the perspective of our expanded understand- ing of the nature of consciousness and cognition (as exhibited, for instance, in the sciences of the mind) becomes imperative. I spell out the reasons for such an engagement, while addressing a broad set of methodological and metatheoretical issues, in chapter 2. The Buddhist epistemologists advance what might initially look like a rep- resentationalist theory of knowledge. On closer scrutiny, this apparent “rep- resentationalism” in effect masks a complex phenomenalism that ascribes nonconceptual cognitive content to direct experience, as well as a causal theory of cognition that aims to explain the relation between varying modes of awareness and their corresponding constitutive elements. In order to fully unpack this Buddhist theory of knowledge I shall draw from the internalism vs. externalism debate in analytic epistemology,28 and the critique of that debate from the standpoint of philosophical phenomenology.29 The guid- ing methodological insight at work in this book is that a phenomenological account of perception on models first provided by Husserl and Merleau-Ponty best serves to translate the intuitions of the Buddhist epistemologists about the cognitive function of perception. Given the immediacy and directness of sense experience, as defined by Dignāga, Dharmakīrti, and their follow- ers, the sense-object relation is an issue of continuous concern (just as for Husserl perception and that which is perceived is ultimately constituted by intentional content). The Buddhist epistemologist thus insists on treating each cognitive event as a new introduction to an object. Furthermore—as I will demonstrate—this continuing concern for the sense-object relation also explains why the Buddhist epistemologist treats cognition as bearing the characteristic marks of embodiment: it is the dynamic of the five aggre- gates30 that ultimately gives the cognitive event its expression. The theory of knowledge advanced by Śāntaraks ita and Kamalaśīla resem- bles Western variants of representationalism as proposed by, among others, Descartes and Locke. These similarities, however, extend only as far as the 28. Following Kornblith (2001: 4ff.), I understand this debate as an outcome of the dilemmas faced by the Cartesian epistemologist, whose internalist perspective on the justification of belief must confront the possibility that such internal belief could be dubitable, and thus be constrained to fi nd a way to coordinate internal belief with external reality. 29. Such a critique is advanced by, among others, Carman (1999, 2007), Siewert (2005), Thompson (2007a), and Zahavi (2004a, 2009). 30. The Buddhist answer to the problem of personal identity stands in contradis- tinction to most classical philosophical attempts to reduce the individual, or his or her mind, to some metaphysical core such as a “soul” or “self.” Instead, the individual personality is regarded as the dynamic product of the five aggregates of grasping: form or body, sensations, apperception, volition or dispositional formations, and consciousness. See chapter 3 for a detailed analysis of the five aggregates. 01_Coseru_CH 1.indd 10 7/4/2012 6:25:09 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN I N T R O D U C T I O N : T A K I N G T H E S T R U C T U R E O F AWA R E N E S S S E R I O U S LY  11  general view of representationalists who admit that while perception may provide immediate access to the domain of phenomenal experience, we come to apprehend it as a realm of discrete entities only as a result of categori- cal and conceptual discriminations. These conceptual discriminations (or representations) are in fact the cognitive counterparts of empirical objects, although the precise ontological status of empirical objects is a matter of dispute among representationalists. Indeed, it is hard to find a general defi- nition of representationalism in Western (or for that matter South Asian) philosophy of mind that adequately covers the various theories subsumed under it. Although most proponents of representationalism agree on the existence of different “mediums” of representation (e.g., ideas, images, lan- guage, symbols, etc.), there is significant debate concerning the content and nature of these representations and the causal relations that obtain between that content and its referent (viz., the empirical reality).31 This explains in part why representationalists can hold both internalist and externalist posi- tions.32 Seeking to reconcile nominalist views of concepts (which regard con- cepts as abstract objects) with realist views of mental content, some have proposed that we treat mental content as a mode of presentation that views concepts as psychological properties of mental representations.33 For Śāntaraks ita and Kamalaśīla, as for all Buddhist epistemologists, the apprehension of a resemblance (sādr śya) between different objects, which marks the transition from an indistinct perceptual experience to a distinct cognitive event, is in itself a form of conceptual apprehension. It should be noted that unlike, say, Locke’s attempt to solve the problem of represen- tationalism by dissociating between two types of ideas (viz., sensible and intelligible), Śāntaraks ita and Kamalaśīla regard any cognitive event that involves discerning or discrimination as bearing the characteristic of con- ceptuality. While this position implies that resemblance relations should be taken as unreal—and here Śāntaraks ita and Kamalaśīla are in agreement 31. The dominant view, known as the representational theory of mind, is that the internal systems of representations have clear structure and a language-like syntax. Th is view has also come to be known, after Fodor (1975), as the language of thought hypothesis. 32. An example of a realist form of representationalism is articulated by the psy- chologist Max Velmans who argues for a distinction between descriptions of phe- nomena (which represent the phenomena themselves) and theories about phenomena (which represent their causes and several other inferred patterns). For Velmans, even appeal to such categories as “universals” to describe the phenomena or their causal relations suggest that “there is a ‘reality’ which is like something” (Velmans 2000: 163). For a general overview of representationalism in Western philosophy from Plato and Descartes to contemporary cognitive philosophers, see Watson (1995). A detailed survey of representational theories of consciousness is found in Lycan (2006). 33. Th is is particularly the case with the unified theory of concepts advanced by Laurence and Margolis (2003, 2007). 01_Coseru_CH 1.indd 11 7/4/2012 6:25:09 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN ( 12 ) Perceiving Reality with Dharmakīrti’s critique of relations—this does not necessarily invali- date representationalism. For Śāntaraks ita and Kamalaśīla, representation- alism reflects an understanding of the nature of epistemic practices rather than commitment to a certain ontology. What we perceive is mediated by our internal expectations, propensities, as well as the range of our sensory sys- tems.34 How we identify and conceive of what is perceptually apprehended, however, depends on our linguistic and conceptual practices.35 This approach to cognition represents an important aspect of Buddhist epistemology and demands a closer and more detailed scrutiny than this introduction can provide. For the moment, let me clearly emphasize its sig- nificance for one of the premises outlined above, namely, that conceptual analysis is a reliable source of knowledge only to the extent that it is grounded in nondeceptive and inerrant experiences. As I shall argue at length in chap- ter 6, the observation of similarity between objects is not the direct cause of internally apprehended relations of resemblance. Instead, the represen- tational content is generated following a process of aspectual rendering, which is in keeping with the reflexive nature of cognitive events. The cogni- tive model which Śāntaraks ita and Kamalaśīla develop—primarily in their discussion of the characteristics of external objects—centers on the theory of cognitive aspects (ākāravāda), more specifically on the notion that in cog- nizing we are directly acquainted both with the phenomenal content (vis aya- ākāra) and the phenomenal character (jñāna-ākāra) of experience. While this model finds parallels in the sense data theories advanced by Bertrand Russell and George E. Moore, I will argue that Husserl’s phenomenological account of intentional objects offers a more viable alternative. Like many of their predecessors, Śāntaraks ita and Kamalaśīla are preoc- cupied with explaining what exactly lies at the juncture of perception and conception. As in many other instances, Dharmakīrti provides the standard answer. For him, conceptual cognitions arise from grasping an object by means of an act of recognition (pratyabhijñāna) that brings it under a certain concept. However, the concept by means of which an object is grasped as such, say as a “blue lotus,” does not entirely correspond to the perceptual aspect of the object as phenomenally given. Rather, how the object is conceived also depends upon the evocative capacity of language, as the medium of concep- tual apprehension, to represent the object.36 For Śāntaraks ita and Kamalaśīla, 34. Whether the appearance of an object in cognition is due to the apprehension of a sensory object (viśaya) or merely to the presence of an internal object (ālambana) is a complex issue that calls into question the ontological positions of the Buddhist epistemologists. Th is issue is addressed below (chapter 4.4). 35. See, for example, TS 1214–1217 and TSP, loc. cit. 36. Dharmakīrti defines a concept as “a cognition with a phenomenal appearance that is capable of being conjoined with linguistic expression” (NB I, 5: abhilāpasam  sargayogyapratibhāsā pratītih kalpanā). 01_Coseru_CH 1.indd 12 7/4/2012 6:25:09 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN I N T R O D U C T I O N : T A K I N G T H E S T R U C T U R E O F AWA R E N E S S S E R I O U S LY  13  this argument provides sufficient ground for claiming that conceptions have the capacity to capture the contents of experience only to the extent that we bracket common assumptions about the function of linguistic reference, a position that finds full expression in the semantic theory of exclusion (apoha). For example, the category of distinction (vyavaccheda), which can be univer- sally applied to all objects of experience, is not an intrinsic feature of the object, but of the cognitive process that entails the recognition of contrasting features in the apprehension of objects. It is perhaps safe to assume that the origins and development of Buddhist epistemology as a distinct type of discourse mark the gradual acceptance of certain canons of logic and argumentation by those Buddhist philosophers who regarded polemical engagement with their Brahmanical opponents as vital to influencing their standing in the wider philosophical community. A good reason for this engagement might have been an eagerness on the part of the Buddhists to guarantee that their mode of argumentation is com- mensurable with the methods of reasoning formulated by the Naiyāyikas. Perhaps the most important concern of the Buddhist epistemologists, one that finds a clear articulation in the Compendium and its Commentary, is the need to withstand the criticism that core doctrinal principles such as those of momentariness and dependent arising cannot be defended on rational grounds (or, as the case may be, lack empirical support). As Buddhist philosophers would argue, our cognitive propensities are beginningless, each thought being merely the continuation of an endless series of previous thoughts, which constantly inform, influence, and direct our intentional acts.37 These cognitive propensities manifest most vividly as traces of memory and conceptual elaborations. The Buddhist epistemologists came to reject memory as a reliable source of knowledge and to regard conception as completely dissociated from perception. While an exploration of the his- torical context in which this dissociation occurred is beyond the scope of this book, in chapter 3 I offer a speculative reconstruction of the sort of empirical reasons that might warrant limiting perception (as a source of knowledge) to nonconceptual states of cognitive awareness. In this regard, it ought to be noted that—for the Buddhist epistemologists— perception is not only an epistemic modality for establishing which cognitive event (or what aspect of it, and under which circumstances) counts as knowl- edge but also a cognitive process to be understood within the framework of classical Abhidharma psychology. Indeed, Nyāya and Buddhist philosophers did not make a radical distinction between epistemological and psychologi- cal accounts of cognition (at least not in the way that dominant currents in modern Western philosophy drifted away from naturalist explanations 37. See AKBh ad AK III, 19d. 01_Coseru_CH 1.indd 13 7/4/2012 6:25:09 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN ( 14 ) Perceiving Reality after Kant).38 It is precisely the practice of translating logical arguments back to their perceptual source that resulted in Indian theories of inference being branded as forms of psychologism, a derogatory label for the seem- ing conflation of logical reasoning with the psychology of perception mainly associated with Gottlob Frege and the tradition of logical positivism. That exclusionary branding no longer holds true. As Francis Pelletier, Renée Elio, and Philip Hanson39 argue in their examination of what Willard Van Orman Quine called “the old anti-psychologistic days,”40 the newly expanded under- standing of our cognitive architecture provided by the sciences of the mind makes it possible to be psychologistic about logic, albeit in a novel way. I shall explore the implication of this resurgent psychologism for our evaluation of the Indian theories of inference in my brief treatment of the semantic theory of exclusion in chapter 4. In this book I am primarily concerned with the analysis of perception, although an account of the relation between perception and conception will also be provided. The analysis of perception provides the empirical founda- tion that gives Buddhist epistemology its pragmatic anchorage. The precise definition of perceptual knowledge (an important subject of dispute among the Indian epistemologists) led to several interpretive solutions, the details of which form the primary contents of my analysis. Dharmakīrti’s overarch- ing influence on subsequent generations of Buddhist and Brahmanical phi- losophers meant that his ideas took on the appearance of a standard account of the Buddhist epistemological standpoint. While no one disputes the para- mount importance of Dharmakīrti, both Śāntaraks ita and Kamalaśīla dis- play a sufficient degree of originality in espousing their views to warrant an independent consideration.41 Our main source, the Compendium and its Commentary, is not only a vast collection of Buddhist doctrines recorded in the second half of the eighth century C.E. but also a highly polemical work bear- ing testimony to the sustained disputes between Buddhist and Brahmanical philosophers during what is perhaps the golden era of Indian philosophy. The polemical nature of the Compendium and its Commentary is further evinced by the importance that Śāntaraks ita and Kamalaśīla accord to issues of metaphysics and epistemology in their attempt to refute a multitude of views and establish their own perspective on the nature of reality. This Buddhist perspective on the nature of reality is anchored in a thorough defense of the principle of dependent arising. The aim of chapter 5 is to examine the 38. See Mohanty (1992b: 130). 39. Pelletier, Elio, and Hanson (2008: 9ff.). 40. Quine (1969: 84). 41. Unless, of course, one is specifically concerned with the various ramifications of Dharmakīrti’s epistemological innovations as reflected in the work of his imme- diate commentators, as is the case in Dunne (2004), or in the works of his Tibetan interpreters, as is the case in Dreyfus (1997). 01_Coseru_CH 1.indd 14 7/4/2012 6:25:09 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN I N T R O D U C T I O N : T A K I N G T H E S T R U C T U R E O F AWA R E N E S S S E R I O U S LY  15  context and scope of Śāntaraks ita and Kamalaśīla’s encyclopedic work and map out its epistemological and ontological concerns, with a focus on the tension between epistemic dispositions and altruistic concerns. More immediately, the three principal points of debate that form the sub- ject of the present investigation concern: (1) whether a perception that is “devoid of conception” (kalpanāpod ha) can be said to be cognitive, and if so in what sense; (2) what is the precise relation between language and con- ceptual analysis, and what sort of causal conditions, if any, are at work in the generation of meaning from verbal content; and (3) in what ways do the Abhidharma analyses of consciousness and cognition constrain the Buddhist epistemologist’s understanding of the epistemic role of perception. Chapter 6 and parts of chapter 4 deal with these three principal points of debate. Adequately addressing these issues demands that the views of Śāntaraks ita and Kamalaśīla are separately treated, taking into account their individual styles of argumentation and their specific audiences. However, while seeking to locate differences in their approaches, one should not overlook the fact that in the Compendium and its Commentary, Śāntaraks ita and Kamalaśīla con- cern themselves with answering questions that form an integral part of the Buddhist epistemological enterprise, and their solutions to these questions are to a large extent contingent on its conceptual and theoretical resourc- es.42 This enterprise is concerned, inter alia, with answering the following questions: What forms the proper basis of a knowledge event? What sort of objects do perceptual cognitions ultimately intend? Is perceptual knowledge representational in character or should it rather be understood as a type of embodied action? What sort of phenomenological account of perception does the Buddhist epistemological tradition advance? These questions are taken up in chapter 7, which addresses the wider philosophical implications of the Buddhist analysis of particulars in the context of contemporary debates about foundationalism in epistemology and phenomenology. In the end, no discussion about perception and its mode of presentation can take place outside the horizon of consciousness. It is consciousness that ultimately provides the evidential ground for all modes of inquiry. For the Buddhist epistemologists consciousness, as reflexive awareness, is not just another event in the chain of dependently arisen phenomena but its dis- closing medium. The precise nature of this reflexive awareness is the sub- ject of chapter 8, in which I argue for the possibility of an intentional but 42. For several arguments about why the TS/P might be regarded as an epistemo- logical (that is, as a pramān a) work, primarily on account of its structure and scope, see Blumenthal (2004: 30ff.) and McClintock (2010: 61‒62). Indeed, the emphasis on reasoning (yukti), and on a specific audience composed primarily of so-called “propo- nents of reasoning” (nyāyavādins), is a clear indication to this effect. It is generally assumed that Śāntaraks ita and Kamalaśīla adopt mainly a Madhyamaka position in their later works. 01_Coseru_CH 1.indd 15 7/4/2012 6:25:09 PM OUP UNCORRECTED PROOF – REVISES, 07/04/12, NEWGEN ( 16 ) Perceiving Reality nonrepresentational form of consciousness. I conclude in chapter 9 with some reflections on the promise of cross-cultural approaches to Buddhist philosophy of mind, with hopes of tackling, among other issues, the old dilemma of the distinction between seeing and seeing as from the perspec- tive of an enactive approach to cognition. For the Buddhist, thus, perceiving reality ultimately marks an intentional orientation in a world that is expe- rientially constituted. Examining the nature of this distinctive perceptual orientation in the arena of current philosophical debates is the main concern of this book. 01_Coseru_CH 1.indd 16 7/4/2012 6:25:09 PM