Dharmakīrti's Philosophy of Mind Among Other Things

Interview by Richard Marshall

'Dharmakīrti was surely the most influential Buddhist philosopher in India, having decisively changed the course not only of Buddhist philosophy, but of Indian philosophy more generally.'

'He was an unusually thoroughgoing reductionist whose basically psychologistic account of mental content has real affinities with the projects of some contemporary physicalists – but he was also a philosophical idealist whose influential arguments for rebirth amount to a concerted critique of physicalism.'

'Systematically working to maintain consistency with his orienting commitments as a Buddhist, Dharmakīrti thus argued that no conceptual constructs carve nature at its joints; everything that Wilfrid Sellars would include in the logical space of reasons thus reflects, for Dharmakīrti, merely our psychic capacities and explanatory interests, which are more salient for the ways in which they mislead us than for their facilitating our flourishing.'

'I have found philosophers like Dennett and Fodor helpful in characterizing Dharmakīrti’s project, while philosophers like Sellars and McDowell have seemed to me particularly helpful in critiquing that project.'

'Jerry Fodor has aptly said that the availability of the computer metaphor represents “the only respect in which contemporary Cognitive Science represents a major advance” over the representational theories of mind upheld by its eighteenth- and nineteenth-century predecessors. I’d like to think that if Fodor had known Dharmakīrti’s philosophy, he might just as well have said that the availability of the computer metaphor represents the only really significant difference between his program and that of the 7th-century Buddhist Dharmakīrti.'

Dan Arnold is interested in philosophy of mind and Buddhism. Here he discusses the Indian Buddhist philosophy of Dharmakīrti through the lens of contemporary philosophy, its link to computational theories of mind, its views on physicalism and nihilism, its links to Dennett and Fodor, its links to McDowell and Kant, Madhyamaka critiques of Dharmakīrti, whether Buddhist foundationalism survives Madhyamaka transcendental arguments, and Buddhist doctrines of the middle way and selflessness.

3:AM:  What made you become a philosopher?

Dan Arnold: Notwithstanding the bewilderment I sometimes feel at finding myself where I am (for I wouldn’t say I’ve experienced myself as having directed the course of my life in such a way as to become a philosopher), I am a strong enough proponent of the view that human freedom consists in our responsiveness to reasonsthat I am inclined to begin by saying that nothing, of course, mademe become a philosopher.

Having gotten that out of the way, I would say both that I was slow to discover that my main interests were philosophical, and that once having discovered as much, it was possible to see that they had clearly always been so. As an undergraduate, I majored in History, focusing on Medieval Europe for no other reason than that I particularly appreciated the professor whose area that was (Carleton College’s Philip Niles, who still looms in my imagination as one of the most exemplary teachers I have had). As a senior, I did a semester-long Tibetan studies program in India and Nepal, on the basis of which I went on to complete a senior thesis project that was not really historical so much as historiographical. The thesis was a strange exercise in thinking through conceptual issues in the treatment of Medieval European history as relevant to understanding why it was as I ought to have expected that my experience among Tibetans would not readily yield to any of the preconceptions with which I had embarked on that experience. The point is, my thesis was not at all an exercise in historical research, but rather an entirely conceptual reflection on what it means, in general, to study history.

That meta-level sort of orientation is among the things I would subsequently come to recognize as reflecting my basically philosophical predilections, though I did not at the time recognize it as such. So, I first began graduate school (at Columbia University in 1989) in an area studies program (“Indic Languages and Cultural History”) in which I aimed to continue as a historian now oriented toward the study of Indo-Tibetan civilization. Instead, I spent three years desultorily wondering what the hell I was doing, and more interested in reading novels than anything in my supposed curriculum. Nevertheless, I got three very important things out of my years in New York City: I met the woman who would become my wife; I got three years of training in Sanskrit; and, in the course of what turned out to be my last year at Columbia, I finally discovered that it was actually philosophy that interested me. That discovery resulted from my chancing to read together two books that dove-tailed for me in all sorts of productive ways: Daniel Dennett’s Consciousness Explained, which had just been published and widely reviewed (and which I hoped might afford some sort of explanation for my desultory search for direction), and Steven Collins’s Selfless Persons, which has for a few decades been recognized as the standard account of the Indian Buddhist tradition’s orienting doctrine, which is anātmavāda– the “doctrine” (vāda) that persons are “without selves” (an-ātma). In reading these two very different engagements with broadly reductionist accounts of persons and experience – oneinformed by state-of-the-art cognitive-scientific research, the other focused (as its subtitle says) on “Imagery and Thought in Theravāda Buddhism” – I first came under the spell of philosophical reflection on mind, consciousness, and being a person.

My excited discovery of those two books came during my third year as a graduate student at Columbia, which turned out to be my last; in the event, I bailed on grad school in New York and returned to my native Colorado, where I got a job at an extraordinary bookstore (Denver’s Tattered Cover), where I would work for five years. Rather to my surprise, though, I found that despite my earlier preference for novels over anything academic, I suddenly found that I wanted to read only philosophy and Buddhist studies. Deciding that my discovery of philosophy warranted some intellectual retooling, I enrolled part-time at Denver’s Iliff School of Theology, which not only had a rich tradition of process theology but was also at the time the academic home of eminent Buddhist studies scholar José Cabezón. The commencement of my philosophical education thus involved a Master’s degree in Theology and Philosophy of Religion along with the self-guided explorations my work as a bookseller facilitated, and from the beginning my philosophical studies went hand-in-hand with the continued study of Buddhist thought. Some important books for me at the time included Heidegger’s Being and Time and Whitehead’s Process and Reality, as well as lots of Kant and William James. My experience of these, however, was all along inextricably woven together with my discovery of things like Jay Garfield’s Fundamental Wisdom of the Middle Way– a translation of the magnum opus of the great second-century Indian Buddhist Nāgārjuna, including Garfield’s own philosophically probing account thereof – and Paul Griffiths’s On Being Mindless: Buddhist Meditation and the Mind-Body Problem, which incisively and critically engages some central Buddhist views that complicate the no-self doctrine.

As it happened, Paul Griffiths himself figured importantly in the philosophical education I began in Denver, as my studies at Iliff actually began with a summer course Griffiths taught as a visitor. While I resisted many things about Griffiths’s approach – his widely respected work in Buddhist studies began with his basically apologetic motives as a theologically conservative Anglican, though it was chiefly his arch rationalism that rankled me – I loved him as a teacher and was captivated by his intellectually rigorous critical engagement with Buddhist thinkers. Chief among the many things he taught me is that the culturally relativist way in which I had earlier been inclined to approach Indo-Tibetan Buddhism does not, in fact, represent the most charitable (or even ethical) way to work at understanding the culturally alien. Hermeneutically charitable engagement involves, rather, taking Buddhist thinkers seriously as defending the kinds of claims they have taken themselves to make (i.e., claims proposed as really true, and as supremely important) – which means disagreeing with them where they seem off target, working to further their claims where they seem promising, and otherwise taking them seriously as philosophers, every bit as worthy of our serious and disciplined attention as philosophers still think Aristotle and Kant are.

During the years I thus spent as a bookseller and seminarian in Colorado, I thought all along that I either wanted to get the whole philosophy thing out of my system, or else resolve to pursue my philosophical interests in the comparative program in Philosophy of Religions that Paul Griffiths had started at the University of Chicago Divinity School. Leery of the likelihood that professional training in the field would distort the philosophical interests I was finding so exciting, I all along sort of rooted for the former outcome, hoping a few years of concerted study would sufficiently scratch my new intellectual itch that I could then get on with something else. In the event, though, I couldn’t manage to put my philosophical questions to rest, and so, I did indeed go on to do a Ph.D. at the University of Chicago Divinity School, where I was privileged to be part of a remarkable cohort of doctoral students doing philosophically constructive work in conversation with Buddhist and other traditions of Indian philosophy.

And now it happens that I have myself been a member of the faculty of the Divinity School’s distinctive program for (can it really be...?) over fifteen years. I am happy to say that I continue to find at the University of Chicago an environment uniquely hospitable to philosophically serious engagement with the fathomlessly rich traditions of Indian Buddhist thought, though I do sometimes find myself looking back on my Denver years (1992-97) as the time of my most existentially committed philosophical inquiries. Though I surely count it a blessing to be gainfully employed in the field, there is no doubt that one’s relationship to philosophy is changed when it becomes inflected by professional conferences and academic politics and grant applications and the like. I can still become nostalgic for the time when my studies were motivated only by a passionate conviction that there could be no better way to spend my time than in philosophically reflective encounters with history’s great traditions of spiritual transformation.


3:16: You’ve looked at the Indian Buddhist philosophy of Dharmakīrti through the lens of contemporary philosophy of mind. So to begin with can you say something by way of an introduction to this thinker? What salient claims does he make that make him of interest to contemporary philosophy of mind?

DA: Dharmakīrti, a Buddhist who lived and wrote some time in the seventh century c.e., was among the most influential of all Indian philosophers. While there were other Indian Buddhists whose works were more consequential elsewhere in Asia – for example, Nāgārjuna (fl.ca.150 c.e.) and Vasubandhu (fl.ca. 360 c.e.) – Dharmakīrti was surely the most influential Buddhist philosopher in India, having decisively changed the course not only of Buddhist philosophy, but of Indian philosophy more generally. When Brahmanical philosophers writing after him engaged “the Buddhist” position, they typically meant Dharmakīrti’s position, and to this day his works figure centrally in the philosophical curricula of Tibetan monastic education.

While Dharmakīrti’s influence extended to many fields of philosophy (but especially logic, epistemology, and philosophy of language), what I find perhaps most interesting about him is how he could uphold commitments that are eminently resonant with influential strands of contemporary philosophy of mind, while at the same time being committed to views that most contemporary thinkers would think diametrically opposed to those. In particular, he was an unusually thoroughgoing reductionist whose basically psychologistic account of mental content has real affinities with the projects of some contemporary physicalists – but he was also a philosophical idealist whose influential arguments for rebirth amount to a concerted critique of physicalism. He was one of the most influential contributors to India’s sophisticated history of logic and epistemology, and his famously difficult works make him seem to many modern interpreters like a real philosopher’sphilosopher; at the same time, he was every bit a Buddhist philosopher, and he was thus shaped by a tradition of reasoning for which the possibility of liberative self-transformation was axiomatic. Insofar as he thus held together a range of commitments that will seem to many contemporary thinkers to be in tension with one another, Dharmakīrti productively challenges what are too often taken to be settled views on a wide range of philosophical issues.

With regard to contemporary resonance, Dharmakīrti represents the apotheosis of mainstream Indian Buddhist analyses of self and personal identity, which were broadly reductionist in character. The orienting doctrine for Indian Buddhist philosophers was the “no-self” (anātma) doctrine – the doctrine that persons do not have enduring, unitary “selves,” instead consisting simply in causally continuous series of momentary events. In systematically elaborating his views in epistemology and philosophy of language, Dharmakīrti added to this broadly reductionist orientation a particular emphasis on causal efficacyas the criterion of the real; only the kinds of things that can enter into causal relations with one another count, for Dharmakīrti, as finally real. On his typically Buddhist view, the self– a fictitious “whole” that supposedly exists over and above the parts (embodied sensations and cognitions that are essentially momentary) that alone constitute the career of a “person” – is the paradigm case of something that cannotenter into causal relations with anything. To that extent, the kind of unitary and enduring “selves” we habitually take ourselves to be cannot, in fact, explain anything at all about our experience; the very idea of a self is, for Dharmakīrti as for all Indian Buddhists, an incoherent construct that badly misleads us.

That we are, in general, habitually mistaken about what is real and what we’re like is, for Buddhists, compellingly evident in the existential suffering that Buddhists take as endemic to the condition of living beings; we suffer, that is, because the actions we understand as advancing our interests mostly function only to further enmesh us in vicious cycles of frustrated desires. Among the intellectual virtues exemplified by Dharmakīrti is an eminently systematic concern to understand what consistency with the no-self doctrine demands when it comes to all sorts of other philosophical issues. To the extent, for example, that conceptually constructed entities are epitomized by the selves that Buddhists are in the business of denying, it stands to reason that other instances of conceptual unity and continuity would come in for the same treatment, and so Dharmakīrti’s commitments led him to develop, e.g., a distinctive and sophisticated sort of nominalism that was meant to explain how abstractions like conceptsand linguistic referentscan be useful even though they, like selves, are finally unreal. Systematically working to maintain consistency with his orienting commitments as a Buddhist, Dharmakīrti thus argued that no conceptual constructs carve nature at its joints; everything that Wilfrid Sellars would include in the logical space of reasons thus reflects, for Dharmakīrti, merely our psychic capacities and explanatory interests, which are more salient for the ways in which they mislead us than for their facilitating our flourishing.

Given his remarkably uncompromising reductionism and his emphasis on causal efficacy, not to mention his broadly empiricist orientation – perception represents, for Dharmakīrti, a privileged epistemic faculty just insofar as perception, which he takes to be constitutively non-conceptual, affords immediate acquaintance with real particulars – Dharmakīrti can seem like an eminently contemporary thinker. Indeed, I would say he epitomizes the considerations that have led many modern interpreters to consider Buddhist philosophers as intellectual fellow travelers – veritable “mind scientists,” in one popular idiom, who, like today’s cognitive scientists, will have no truck with anything like the supposedly Cartesian idea that persons are individuated by the “thinking substances” that are souls. While it would be wrong to suppose that Dharmakīrti held his views as a result of his application of anything like scientific method – he did philosophy in the religiously scholastic context of a tradition that took it as axiomatic that the Buddha had already discovered everything we need to know – it’s certainly right to say that he made sophisticated arguments for views that many contemporary thinkers might find attractive.

And yet: For all that he might thus make common cause with the anti-essentialists of today, Dharmakīrti was very much a dualist. It’s important to recognize, in this regard, that it’s possible to be a reductionist (and thus to hold that everything consists not in substances but in events), and yet to hold that there are, nevertheless, essentially different kinds of events – that mental events, in particular, are ontologically distinct from physicalevents. Moreover, Dharmakīrti wouldn’t likely be much troubled by the problems that beset dualism (such as how to explain why mental events, if they’re essentially distinct from physical ones, nevertheless occur only as embodied); for he was, at the end of the day, almost certainly an idealist (though Dharmakīrti was famously cagey about his final views, always preferring to maintain maximal room for philosophical maneuver). His point in arguing that mental events cannot coherently be thought to depend on the body was finally to show simply that mentalevents are explanatorily basic – if, at the end of the day, it turns out to be hard to reconcile such events with the physical world, well, so much the worse for the latter.

It seems to me quite interesting that while contemporary philosophers are mostly apt to suppose that reductionism, ipso facto, vindicates physicalism, Dharmakīrti was a dyed-in-the-wool reductionist who thought some sort of idealism was finally the right view to hold. He thought as much, moreover, because he was first and foremost a Buddhist philosopher, for whom it was axiomatic that thought, which creates the world we experience, is the most “real” sort of thing there is. To be sure, the reality of thought does not, for a Buddhist, consist in its being any kind of substance; nonetheless, one of Dharmakīrti’s orienting intuitions is the basically idealist notion that nothing at all can be known or experienced except insofar as it shows up for consciousness, which is what therefore must be transformed if we are to overcome the existential suffering that Buddhists take as the problem to be addressed by the practice of the Buddhist path.

I think, then, that contemporary philosophers, particularly insofar as they are apt to find much that is philosophically familiar in Dharmakīrti’s works, stand to learn something when they find themselves brought up short by what can be discovered by more deeply probing his project. It can be quite eye-opening to see that prima facie familiar arguments and positions might not, in fact, entail just the conclusions we habitually suppose they do – that they might, indeed, be deployed in support of views that fundamentally challenge the very conclusions we might suppose they clearly entail.

[Dan Dennett]

3:16: Why do you discuss the disagreement within contemporary philosophy of mind between, on the one hand, Dennett and Fodor, and, on the other, McDowell? How does this help to clarify why the Buddhists can be taken to have something to contribute?

DA: In thinking through the projects of first-millennium Indian Buddhists, I am generally inclined to study them in conversation with contemporary philosophers for a number of reasons. Perhaps chief among these is a basic hermeneutical consideration: Understanding the arguments of any Sanskrit philosophical text is not just a matter of determining what the sentences say; it also requires understanding what sense it makes that they say that– and there is no other way to get at that sort of interpretive question than to think philosophically along with the texts. I often tell students, in this regard, that to really understand any of these thinkers is to become capable of giving one’s own accountof the sense it makes to think their claims true – where giving one’s own such account means, among other things, imagining how the author might respond to philosophical objections not explicitly entertained in the text. One way to begin to think one’s way into that is by first considering some of the arguments against Buddhist views leveled by other contemporaneous Indian philosophers, whose critiques can bring their adversaries’ views more sharply into relief, albeit, as in the history of Western philosophy, sometimes owing to the “productive misreadings” that so often figure in the unfolding of traditions of reasoning. Appreciating that Buddhist philosophical views thus make sense as part of a larger economy of Sanskritic discourse, and that one is only hearing part of that conversation if the Buddhist views are considered alone, it can then come to seem natural to search for other, non-contemporaneous interlocutors whose thought might also afford some purchase on the views at issue.

Given, then, that understanding any of these texts necessarily means thinking philosophically along with them – and given, as well, that there are ample resources in contemporary literature for doing so – I frankly do not see how else to engage these texts but in conversation with contemporary philosophers. But of course, it’s necessary to remain alert to the possibility of having been misled by one’s initial sense of the right conversation partners; one must continually tack back and forth between the questions and inspirations one takes from contemporary readings, on one hand, and, on the other, premodern Indian texts that might just as well resist our questions as yield to them. It is this dialectic that is often called the hermeneutic circle, and while the achievement of understanding this way is more art than science, I have continued reading Dharmakīrti and other Buddhist philosophers in conversation with modern thinkers just insofar as the latter have helped illuminate the first-millennium arguments I’m trying to understand. Thus, I have found philosophers like Dennett and Fodor helpful in characterizing Dharmakīrti’s project, while philosophers like Sellars and McDowell have seemed to me particularly helpful in critiquing that project. In the course of staging these anachronistic philosophical debates, all kinds of things can be learned – that, for example, McDowell’s arguments, as having considerable purchase against Fodor, may be more metaphysically contentful than McDowell allows, or that one’s contemporary interlocutors cannot have exhausted the possible options insofar as the project of a Buddhist philosopher turns out to resist characterization in any of the terms they suggest, or that the character of the Buddhist contribution itself becomes more clear in virtue of the attempt to understand why the latter is so.

In proceeding this way, I have come to think that despite the stake Dharmakīrti has in resisting physicalism, what may finally be most salient about his project is the extent of his affinities with cognitive-scientifically inclined philosophers like Dennett and Fodor, with whom he shares not only a basically empiricist orientation but also an overriding commitment to the idea of causal explanation as the gold standard when it comes to understanding anything. Like these contemporary thinkers, Dharmakīrti holds, in effect, that the basic questions in philosophy of mind are all empirical questions, and that the way to answer them is thus to identify the causal mechanisms that explain the empirical occurrence of psychological processes. As against this, scholars in Buddhist studies have appreciated for a couple of decades that Wilfrid Sellars’s critique of the “myth of the given” may be fruitful in characterizing the kinds of philosophical problems raised by Dharmakīrti’s way of proceeding (though there are others who reasonably reject the idea that Dharmakīrti presupposed anything like Sellars’s “given”). 

[John McDowell]

In my view, John McDowell’s more thoroughgoingly Kantian elaboration of Sellarsian arguments has been especially helpful in this regard. Given the extent to which Dharmakīrti valorizes supposedly non-conceptual awareness – and given the problems that result from his so sharply sundering supposedly non-conceptual perceptionfrom all other ways of knowing – it stands to reason that McDowell’s strongly conceptualist bent would be helpful in identifying some of the problems Dharmakīrti faces. In particular, McDowell’s arguments to the effect that, (1), our conceptual capacities are constitutively normative (which is among other things to say that they essentially resist any reduction to causal relations), and, (2), such capacities are always already in play in any epistemically significant experience – these arguments cut directly against some of Dharmakīrti’s most significant guiding commitments. Whether or not one credits McDowell’s arguments, then, entertaining his arguments as affording critical purchase on some of Dharmakīrti’s characteristic arguments can bring the logic of the latter more clearly into focus.

I would also say, finally, that there’s something profoundly and instructively humbling in the realization that contemporary thought is not as far advanced as we are often inclined to suppose. Part of what I want my readers to appreciate is the extent to which some first-millennium debates among Indian philosophers – debates between Buddhists and their Hindu critics, for example, about the ontological status of universals – represent not just quaint and historically interesting moments in the history of thought, but debates that centrally concern issues still very much at issue among contemporary philosophers. This should serve as a corrective to some regrettably but widely skewed preconceptions about what philosophy is, showing that pre-modern Indian thinkers are every bit as deserving of a place in the history of philosophy as their Western counterparts. But there is also, I think, something just quintessentially philosophical in the humbling realization that what some Indian philosophers wrote around 500 c.e.might just as effectively characterize the state of a philosophical problem as anything by, say, Daniel Dennett or Jerry Fodor.

That is not, of course, to say there are no consequential differences in terms of the state of our empirical understanding; obviously there’s vastly more known about, e.g., the brain and central nervous system now than there was even fifty years ago, never mind 1,500 years ago. Thinking philosophically across traditions can, however, help us get clear on just what kinds of questions are really at issue. If, for example, it makes sense for a Buddhist philosopher to be a reductionist without being a physicalist, that may tell us that there are, when it comes to enduring issues in philosophy of mind, more conceptually basic issues at stake than what kind of stuff mental events are made of. Jerry Fodor has aptly said that the availability of the computer metaphor represents “the only respect in which contemporary Cognitive Science represents a major advance” over the representational theories of mind upheld by its eighteenth- and nineteenth-century predecessors. I’d like to think that if Fodor had known Dharmakīrti’s philosophy, he might just as well have said that the availability of the computer metaphor represents the only really significant difference between his program and that of the 7th-century Buddhist Dharmakīrti. It seems to me that awe is the most appropriate response to the recognition that this could be so.

3:16: Do you think that Dharmakīrti’s account of the mind is somehow computational?

DA: As I understand it, computationalism essentially involves the idea of rules (algorithms, instructions, etc.); to that extent, Dharmakīrti’s account of the mind and mental content may not be computational, as he’s really committed to explaining everything about the mental with reference only to unique particulars – specifically, with reference to particular, self-intimating mental states or episodes, which (as I read him) he takes to be not just explanatorily basic, but indeed finally to be the only real existents there are. In terms of the Buddhist tradition’s crucial thought that there are two basically different kinds of “true” descriptions – conventionally true descriptions of our common-sense intuitions, and the revisionary descriptions thereof that are taken by Buddhist tradition to be ultimately true– this is the claim that an ultimately true account can make reference only to momentary mental events, and that everything about common-sense experience must therefore be explicable in terms of these.

To be sure, Dharmakīrti’s apoha (“exclusion”) doctrine – on which, more below in answer to the question about Dharmakīrti’s account of universals – can be said to elaborate what is aptly called a rule with regard to the construction of mental content: stable conceptual content is generated merely by excluding all those particulars that do not produce whatever effect is desired on the occasion of concept-use. Moreover, Dharmakīrti’s elaboration of the “exclusion” mechanism in particularly causalterms surely brings him still closer to computational approaches to cognition. But it seems to me that the exclusion mechanism by itself is insufficient to account for mental content, as we are still owed an account of rules for determining what effects, exactly, ought to be preferred on any occasion, as well as for determining how we are in the first place to know what to exclude as not producing those. Perhaps, though, that is only to say that I’m sympathetic to contemporaneous Indian critiques of the exclusion doctrine, which typically argued that Dharmakīrti’s account finally presupposes just what it was supposed to explain – that the doctrine itself is intelligible only insofar as we already know, e.g., what kinds of things do and do not produce the effects that motivate the initial process of exclusion, where it’s the very idea of kinds that was supposed to be explained.

In any case, I am most inclined to characterize Dharmakīrti’s basic approach not as computational but as paradigmatically psychologistic, in precisely the sense targeted by the likes of Frege and Husserl – everything about the nature of thought, on his account, must be explicable with reference only to empirically describable occurrences of particular mental states. I’ll say more on all this below, in answer to the question about Dharmakīrti’s account of universals. By way of advance press for that discussion, though, I’ll here add that one of the most intensely debated issues in classical Indian philosophy, as in much Medieval Western philosophy, was that of the ontological status of universals – and among my aims has been to show how that concern, far from being an arcane preoccupation of merely historical interest, represents a viable way to characterize basic issues in philosophy of mind.

3:16:  Don’t Buddhists think that physicalism is tantamount to nihilism in that it eliminates everything of value?  In his ‘Critical Commentary on Epistemic Criteria’ doesn’t Dharmakīrti explicitly take on and refute physicalism? Are his reflections of Buddhist proofs of rebirth particularly important for this?

DA: As I briefly noted above, Dharmakīrti, along with pretty much the entirety of the mainstream Indian Buddhist philosophical tradition, exemplified a combination of commitments that will strike many contemporary readers as counter-intuitive: He was an uncompromising reductionist, but at the same time strongly critical of physicalism. (As noted above, there is no inconsistency in holding both that all the macro-objects of the common-sense world are reducible to atomic moments, and at the same time that there are essentially different kinds of moments.) The reason this is so important for Buddhists is that the idea of “rebirth” makes sense only given their dualism – and Buddhists were all along worried by the idea that absent “rebirth,” the no-self doctrine would amount to nihilism.

My scare quotes here are to acknowledge the fact that rebirth is not a very apt term in a Buddhist context (though it’s used widely enough that we’re probably stuck with it); for the prefix re-suggests there is one thing recurrently subject to the same process, which of course is just what the Buddhist no-self doctrine denies. What Indian Buddhist philosophers affirm, then, is just the post-mortem continuation of mental continua that comprise only momentary particulars. Indeed, continuum (Sanskrit santāna) is an important term of art in Buddhist philosophy, typically adduced as explaining what a “person” really consists in – “persons” are just continua that consist in causally related psycho-physical events. The Buddhist idea of “rebirth,” then, is just that insofar as mental events are essentially distinct from physical events, the continuity of a series of mental events is not impacted by the death and decomposition of a physical body.

This raises, of course, a number of problems, beyond the familiar problem of how, given the essentially distinct character of mental and physical events, cognition can be an invariably embodied phenomenon. (I’ve already said that insofar as Dharmakīrti was finally an idealist, he would happily conclude that any problems in this regard count simply against the viability of the physical; what cannot be doubted, on his view, is the explanatory basicness of consciousness.) There was, for example, a Buddhist school of “Personalist” (pudgalavādin) philosophers who raised the problem that the idea of a “continuum” seems to presuppose (and therefore cannot explain) personal identity, and that the Buddhist no-self doctrine must therefore be compatible with a recognition of the ineliminable basicness of the category of persons. Likewise, typically Buddhist emphasis on the ethical significance of rebirth seems to require that there be a robust sense in which it is the same mental continua that are taken up at the start of new lifetimes, which might be thought at odds with the no-self doctrine. Regardless of the various difficulties with the idea of rebirth, though, Buddhist philosophers were ever at pains to argue that the no-self doctrine does not entail that our actions are ethically insignificant – that even though there is a basic sense in which future moments of the causal series that is “me” are no more closely related to what I am now than are contemporary “others,” it nevertheless makes sense that I care about the consequences of my actions, particularly insofar as the consequences vastly transcend any individual lifetime.

The Buddhist tradition well recognized, then, that its cardinal doctrine would be supposed by many to fly in the face of ordinary ethical intuitions, and Buddhists were concerned from the start to argue not only that nihilist conclusions are not entailed by the no-self doctrine, but that any such conclusion us badly misguided. Pretty much all Buddhist philosophers thus shared the conviction that their characteristic refutation of selves steered clear of ethical nihilism, and were confident that they could account for the intelligibility of ordinary ethical intuitions with reference only to causal continuity. While it is true, they would thus argue, that there is a sense in which the person who committed some heinous crime on one occasion is not identical to the person who later stands trial for that, it is also true, Buddhist philosophers emphasized, that these two person-stages are not essentially distinct, either. In thus embracing the view that (according to a venerable Buddhist trope) that any two person-stages are “neither the same as nor different from” one another, Buddhists argued that the degree of relatedness that obtains owing merely to causal continuity is sufficient for maintaining something like the basically forensic idea that John Locke took to be most salient for the category “persons.”

But it was not just in order to retain ordinary ethical intuitions that Buddhists argued against physicalist understandings of the no-self doctrine; indeed, more basically important in this regard is what Buddhists thought nirvāṇa consists in. On the Buddhist diagnosis of the human predicament, the problem to be overcome by practicing the Buddhist path is that life is essentially characterized by suffering; we are finite beings whose existence in time necessarily means that everything is finally lost,as most poignantly evident in the fact of our mortality, in the face of which we desperately cling to sources of meaning and significance that are ultimately fleeting and unsatisfying. But whatever else it means (and Buddhists were plenty willing to allow that we can’t imagine what it would be like), nirvāṇa was thought to effect the elimination of that suffering insofar as it consists in the cessation of the endless cycle of death-and-rebirth: to achieve nirvāṇa just is to stop being reborn.

So here’s the thing: without the idea of “rebirth” – without, that is, the idea that suffering is not eliminated simply by dying, and this because mental continua are not, in fact, brought to an end by bodily death – it seems that nirvāṇa, understood as the cessation of rebirth, becomes meaningless; release from suffering would then come for all beings just because they die, completely regardless of how they had lived. Indian Buddhists found this to be a deeply unacceptable consequence, and were much at pains to show that it was not, in fact, entailed by the no-self doctrine. Indeed, the familiar idea of Buddhism as a “middle way” has precisely to do with circumventing that worry; the two extremes eschewed by the middle way are those of eternalism, on the one hand– the view that there is something essentially unchanging that represents what any person “really” is, which of course is what the no-self doctrine refutes – and, on the other, the extreme of nihilism, which we might also call (with a nod towards the kind of physicalism avowedly held by some contemporary philosophers) eliminativism.

To be sure, first-millennium Indian Buddhist philosophers didn’t spend much time engaging physicalism, as that was not a sort of view widely entertained by contemporaneous Indian philosophers (and it should be noted that much of this looks rather different in the context of Buddhism in, say, China, where rebirth was not a widely prevailing idea). Nonetheless, physicalist views in philosophy of mind would paradigmatically exemplify the intuitions that Buddhists thus resisted: the intuition that consciousnessis explicable in terms of physical existents that are taken as more certainly known (which cuts against the Buddhist tradition’s guiding thought that consciousness, though not a substance, is at least as real as anything else), and the intuition that suffering is eliminated simply by death (which cuts against the tradition’s thought that suffering is a uniquely challenging problem precisely because it is notended by death). Dharmakīrti’s views in this regard were, then, quite typical of mainstream Indian Buddhists, although he was rare among Indian Buddhists in having lengthily argued for the reality of rebirth; it is clear he considered India’s token physicalist school, however marginal at the time, to represent a serious challenge to Buddhist ideas.

Dharmakīrti’s arguments for rebirth are really just to the effect that mentalgoings-on cannot be explained with reference to physical events; his arguments for “rebirth,” in other words, were mostly just arguments for dualism. Given, however, some Buddhist views about the nature of causation, the fact that mental events cannot have physical causes means that every mental event must be caused by preceding events of the same kind – which, he argued, means that what we think of as the first moment of a newborn’s conscious experience cannot, in fact, really be first, and must instead represent the effect of the last mental events of a previous “lifetime.” While there are perhaps few modern readers who are likely to find Dharmakīrti’s arguments to that effect convincing, I will say a bit more below  about some other cogent reasons that some of Dharmakīrti’s rival Buddhists would adduce as decisively countering physicalism.

3:16: If Buddhism isn’t a physicalist doctrine then how can it be thought to be compatible with the cognitive-scientific accounts of the likes of Dennett and Fodor? Even if it’s not dualist it ends up looking like Idealism doesn’t it? So in what way is Dharmakīrti’s account of the mind vulnerable to arguments that make physicalism vulnerable?

DA: I would begin by emphasizing that it is not “Buddhism” that has affinities with physicalism, but rather Dharmakīrti’s philosophical elaboration of Buddhist insights that does – though that is an interesting point particularly given that Dharmakīrti was, as noted, rare among Indian Buddhists in having lengthily argued against physicalism. What his thought, in particular, can help us see is that reductionism does not by itself entail physicalism. Dharmakīrti’s philosophy thus challenges us to recognize that when it comes to philosophy of mind, the question of what kind of “stuff” mind consists of may be rather a red herring; there may, to that extent, be more conceptually basic issues at stake in debates between physicalists and defenders of consciousness.

Such is my thought in trying to show that notwithstanding his own commitment to refuting physicalism – his own commitment, indeed, to a kind of philosophical idealism – Dharmakīrti nonetheless shares with contemporary physicalists a basic commitment that is arguably more significant than their divergent views about what kind of “stuff” there is. What Dharmakīrti most importantly shares with most contemporary reductionists, I’ve argued, is the view that causal efficacy is the criterion of the real, and that the only things that belong in a final ontology are therefore causally efficacious particulars. It is, then, because of his commitment to a causal and atomistic account of the mental that Dharmakīrti’s position turns out to be vulnerable to some of the same arguments typically directed against physicalist accounts in philosophy of mind – contemporary arguments, in particular, to the effect that normativity essentially characterizes mental content, and that mental phenomena therefore will not admit of exhaustively causal explanation just insofar as normativitycannot be so explained.

Among the relevant arguments in this regard are all those directed against psychologistic accounts of mental content. What cannot be accounted for by a characteristically empiricist attention only to particular mental occurrences, according to this line of argument, is the conceptual (logical, semantic, etc.) content that innumerable mental events can commonly be about. Frege’s argument, e.g., against a view paradigmatically exemplified by Locke – Frege’s argument, i.e., that the content of a thought (Gedanke) must be distinguished from (and cannot be gotten out of) any of the empirically occurrent “Ideas” that Locke, always capitalizing the word, took to have thoughts as their content – cut just as surely against Dharmakīrti as they do against the classical empiricists.

[Jerry Fodor]

Now, contemporary thinkers like Fodor well understand the problems with psychologistic accounts like Locke’s; indeed, Fodor’s whole program largely concerns the importance of restoring semantic content to mental events that can also be characterized in terms of their causal properties. (Recognizing that predecessor approaches to the problem of mental causation failed chiefly insofar as they explained how mental states could be causally efficacious only at the cost of making the content of these states epiphenomenal, Fodor has said that “Cognitive Science is the art of getting the cat back in.”) For his part, Dharmakīrti, too, appreciated the problems here; his “exclusion” (apoha) doctrine was meant precisely to explain how repeatable, intersubjectively available mental content (“universals”) could be explained given an ontology in which only unique particulars are finally real – and he recognized that that requires explanation because he understood that otherwise there could be no making sense of language. Among the questions that Fodor and Dharmakīrti commonly face, though, is whether their attempts to explain semantic content can really work. I would argue that both accounts face intractable problems, fundamentally centering on the question of whether they must finally presuppose the very sorts of things they claim to explain.

In this regard, Dharmakīrti’s contemporaneous Indian critics argued that his “exclusion” doctrine – on which, more shortly – was undermined by a basic circularity, insofar as his attempt to explain universals was itself intelligible only if one already has universals at one’s disposal. There are, I think, variations on the same problem with Fodor’s account, which can help us appreciate the purchase of a line of critique that I take as commonly cutting against him and Dharmakīrti – a critique, which I take to be epitomized by Kant’s Critique of Practical Reason, based on the ineliminable priority of “practical reason” (or as I would say, of first-personally understood reasoning). The problem that Fodor has a hard time escaping, as I see it, is that even if it’s granted that he does manage to get semantic content back into the picture, he is nevertheless committed to the view that it’s only in terms of the causal properties of mental states that any explaining is done. So, even if Fodor is able to make sense of the semantic content of mental states, that content remains, on his account, a gear that doesn’t turn any other wheels. To that extent, his is a view according to which it is (as Kant put it) “only as empirically conditioned” that our responsiveness to reasons makes any difference – it is, in other words, only the things that “have” mental content (occurrent mental states) that do the explaining, not the content itself. That is, however, to say that we are not, after all, responsive to reasons as such, but only to “reasons” under an altogether different description – and that, in Kant’s terms, is to deny that reason really is “practical.” And the Kantian argument that I take to cut decisively against any such view is to the effect that it cannot coherently be denied that reason is practical (that we are responsive to reasons as such) just insofar as it is only by reasoning that it is possible to make sense of the denial itself.

As noted below, a key part of my case for thinking Dharmakīrti vulnerable to the same critiques leveled at physicalists is that some of Dharmakīrti’s contemporaneous Indian critics made arguments conceptually similar to Kant’s argument from practical reason. Thus, for example, some proponents of Madhyamaka, a rival Buddhist school of thought, would argue against Dharmakīrti that the very idea of ultimately real existents is incoherent just insofar as that idea is itself intelligible only relative to conventionally real persons – an argument that has affinities to Kant’s pragmatic-transcendental approach. In both cases, the argument is basically that there is a performative self-contradiction involved in any explanation of conscious experience as “really” being something else; for no such explanation can be entertained without reference to conscious experience on a common-sense understanding thereof.

3:16: How does Dharmakīrti seek to explain universals with reference only to particulars using his ‘exclusion’ doctrine of apoha? You say this concern is also Jerry Fodor’s – does Fodor help us get what Dharmakīrti was concerned with here, and does Fodor bring additional resources to the table even though presumably Fodor’s physicalist approach is rejected in Buddhism?

DA: Dharmakīrti’s “exclusion” (apoha) theory is elusive and sophisticated, but the basic idea behind it is that it’s possible to construct semantic mental content in terms only of negations or “exclusions” – which is, against the many contemporaneous Brahmanical schools of thought that held variously realist views of universals, to say that the content of any concept can be accounted for without reference to any really existent universals.

The exclusion theory was first developed by Dharmakīrti’s predecessor Dignāga (who flourished in the sixth century c.e.), and it’s important to understand that these two Buddhist philosophers elaborated the idea in different (if complementary) ways. For Dignāga, the exclusion theory had to do with the relative determinacy of conceptual content, with respect to which exclusion represents an elegant way to make sense of the inferential asymmetry that characterizes relations among concepts, any of which invariably has its place in a branching hierarchy of increasing determinacy. In his Foundations of Arithmetic, Frege expressed something much like Dignāga’s orienting thought when he (Frege) affirmed that “the content of a concept diminishes as its extension increases; if its extension becomes all-embracing, its content must vanish altogether.” Conversely expressed in terms of exclusion, the point is that concepts become more precisely determinate just insofar as they exclude more; a concept like being existent has relatively indeterminate content just because it excludes very little, whereas the narrower scope of a concept like being a cow is precisely a function of its additionally excluding all existents that are not cows. What’s more, this way of characterizing conceptual content beautifully explains why inferences to different levels of abstraction work in only one direction. Thus, the concept being an aspen is more determinate than the concept being a tree just insofar as aspen excludes not only everything in the world that is not a tree, but also all trees that are not aspens – and that is precisely why it’s correct to infer that if something is an aspenit is therefore a tree, but incorrect to infer that that if it is a tree it is therefore an aspen (for of course not all trees are aspens).

Again, Dignāga’s approach here was chiefly motivated by his characteristically Buddhist resistance to the idea, typical of most Brahmanical schools of thought, that universals like being a cow (a stock example in Indian philosophy) are real, carving nature at its joints rather than (as Buddhists affirmed) merely representing our conceptual interests. Dignāga did not, however, seek to explain how any of the concepts in a branching hierarchy can relate to real particulars in the world; his version of the exclusion doctrine is not, that is, an account of reference, which is precisely what Dharmakīrti’s further development of the doctrine was concerned to build in. Having brought Dignāga’s idea on board, Dharmakīrti further recruited the “exclusion” mechanism as also explaining how it is that we come to bring real particulars under any concept – as explaining, that is, how we can in the first instance individuate real particulars as the referents of concepts, which is precisely to take them no longer as particulars but rather as tokens of the typesthat concepts represent. The challenge here is to explain how it is that linguistic usage can be pragmatically effective, even though the referents of words are, as essentially abstract, not real existents.

Now, Dharmakīrti held, in this regard, that there are only two kinds of relations: the conceptual relations theorized by Dignāga (the relation, e.g., between being a treeand being an aspen), and the causal relations that obtain only among real particulars. Dharmakīrti, here again exhibiting his affinities with contemporary physicalists, addressed the problem of reference by applying Dignāga’s exclusion mechanism in a way dictated by causal relations. His thought was that when we take any real particular as referred to by a concept, we arrive at the “description” under which we then experience that particular just by excluding everything in the world that does not produce the effect desired on that occasion of concept-use. To the extent that reference is thus rooted, for Dharmakīrti, in causal relations, it surely has affinities with Fodor’s attempt to show how it can be that reference, in general, is grounded in causal relations, even though Fodor allows that many (if not most) particular instances of reference may not admit of causal explanation. The fact that many particular occasions of reference do not readily admit of causal explanation is, Fodor allows, integral to the idea of meaning, which typically exceeds whatever information is conveyed by occasions of language-use (where Fodor’s point is precisely that “information is tied to etiology in a way that meaning isn’t”). Fodor’s basic thought is that even though meaning thus resists causal explanation, it must nevertheless be asymmetrically dependent on instances of causation that are foundational for it. Thus, while it is not the case that all particular utterances of any word will demonstrably be causally relatable to some token of the type denoted, it is nevertheless the case that no particular occasion of reference could make sense as the kind of thing it is unless some such tokens are so relatable. For Fodor, then, in order that there be any cases of rightly calling particular bovine critters cows, there must be some such cases that will admit of causal description (in terms, e.g., of ostensive indication of a particular critter).

While Fodor thus allows that many occasions of reference float free of causally describable encounters (even though he takes the latter as foundational for meaning), Dharmakīrti instead argues that there is an importance sense in which his causally describable exclusion mechanism does, in fact, figure in every instance of language-use. We can see why he says that by noting a possible objection to Dharmakīrti’s thought that reference is enabled just by our excluding whatever does not produce the effect at issue on any occasion of language use. As his contemporaneous Indian critics argued, that may seem to beg the question; after all, if we can experience a particular ruminant as being a cow just insofar as we exclude everything in the world that doesn’t produce, say, the “effect” which is milk, it would seem we are taking the capacity to produce milk as itself a universal – how, then, can this be a viable alternative to any other intensional account of universals, on which, say, “having horns and a dewlap” might be taken as the content of the universal being a cow? Doesn’t “having the capacity to produce milk” work just the same way?

Dharmakīrti’s answer to this worry really shows his psychologistic stripes. He argues that when, on any occasion of bringing something under a concept, we exclude everything other than what produces the effect we desire, the effect he has in mind is itself a particular (and hence, real): specifically, a momentary instance of cognition whose phenomenal content is the kind we expect when we’re concerned (to stay with the example) with cows. To that extent, Dharmakīrti effectively concedes that there is, after all, a kind of “sameness” across instances of experiencing cows – it’s just that the sameness is merely phenomenal, and is thus a function only of our own psychic dispositions, expectations, etc. If asked for an explanation of that– for an explanation, that is, of our disposition to take any number of unique moments of awareness as all having the same phenomenal content – Dharmakīrti basically punts, simply saying (as Buddhists typically do) that this is owing to the beginningless habituation of our conceptual capacities. (As atheists, Indian Buddhists typically held that there is no first beginning of existence. Given, moreover, the principal Buddhist presupposition in play in arguments for rebirth – viz., the presupposition that every existent necessarily has a prior cause of essentially the same kind – this means that every mental continuum is, like the world itself, without beginning. According, then, to the cosmology of rebirth, our dispositions for erroneously experiencing things – and particularly our disposition to take ourselves as enduring selves– are very deeply habituated, indeed, having been reinforced over the course of innumerable lifetimes.)

To the extent that Dharmakīrti must thus appeal to a Buddhist axiom in order to account for the merely phenomenal sameness that he allows, it may still be thought that his account is basically circular; indeed, I am myself sympathetic to that argument. Even so, Dharmakīrti’s account makes sense as undermining the kind of realism about universals that was characteristic of most Brahmanical schools. Far from carving nature at its joints, the concepts we construct through exclusion are fundamentally generated just by our own interests, and what can be brought under them will vary accordingly; what we mean by cow may differ, for example, depending on whether the desired effect is milk or, say, cow dung (often used as fuel in India) – even if what is really excluded is only cognitions whose content isn’t dairy cows (or cognitions whose content isn’t defecating cattle, or whatever).

3:16: Do you think that Dharmakīrti and in fact the Buddhist approach in general to mind is in trouble because of its attempt to naturalise intentionality and not take seriously the normativity of meaning and what McDowell characterizes as the sui generis nature of the space of reasons and Kant as ‘spontaneity’? And do you think Buddhist thinkers criticizing Dharmakīrti were actually on the same kind of track as modern philosophers like McDowell and Sellars? Can you sketch for us what critics like Nagarjuna and Candrakīrti were arguing and whether you think their criticism of Dharmakīrti should be taken more seriously?

DA: As noted above , it is important to appreciate that my claim is not that Buddhist thought, in general, is vulnerable to the same kinds of arguments that some modern philosophers have leveled at physicalism; my claim, rather, is that a particular trajectory of Buddhist philosophy (the one epitomized by Dharmakīrti) is vulnerable to these – and in fact, a key part of my case for thinking so is my argument that some of Dharmakīrti’s contemporaneous Indian critics (including some Buddhist critics) made arguments conceptually similar to the pragmatic-transcendental argument I sketched above ). 

My thoughts, in this regard, are that, (1), McDowell can be read as building on the Kantian version of that line of argument, and, (2), the Madhyamaka tradition of Buddhist thought, epitomized by Nāgārjuna and Candrakīrti, cogently pressed a similar kind of argument against those of their fellow Buddhists who understood Buddhist commitments as Dharmakīrti did. Indeed, in both of the books I have published to date, I have represented the Madhyamaka school of thought – which originates with the work of Nāgārjuna, who wrote centuries before Dharmakīrti (Nāgārjuna likely flourished around 150 ce) – as more philosophically promising than Dharmakīrti’s project. (In fact, most Tibetan Buddhists agree, generally taking Madhyamaka, particularly as elaborated by a rough contemporary of Dharmakīrti by the name of Candrakīrti, as representing the pinnacle of Buddhist thought.)

In sketching the Madhyamaka arguments as I understand them, let me begin by noting that while Dharmakīrti’s arguments for rebirth are unlikely to be found cogent by many modern readers, Madhyamaka represents a radically alternative trajectory of Indian Buddhist thought that may leave room for less ambitious but still significant conclusions regarding the ineliminable character of consciousness.In wrapping up with a sketch of my understanding of Madhyamaka, I will thus emphasize why that school of thought seems to me to offer resources for a more nuanced understanding of Buddhist ideas of consciousness than may be evident in Dharmakīrti’s more ambitious attempt at demonstrating the reality of rebirth.

The Madhyamaka tradition of thought most basically resisted the understanding of the two levels of truth typical of the Buddhist philosophical mainstream that Dharmakīrti epitomized. On Dharmakīrti’s more mainstream view, the difference between conventional and ultimate truth basically tracks a distinction between essentially different kinds of existents: a conventionally true statement is one that, although pragmatically useful in the proper context, makes reference to things that are not ultimately real, while an ultimately true statement makes reference only to irreducibly basic existents. This is the idea, in other words, that nothing that is reducible to more basic constituents belongs in a final ontology, but that it is possible to specify the kinds of irreducibly basic things that dobelong in a final ontology. To that extent, Dharmakīrti’s eminently revisionary account of ordinary experience, which may be thought roughly analogous to a scientific description, is taken as superseding common-sense perspectives on the mental: as against the illusory existents that populate common-sense experience, an ultimately true description succeeds in picking out ontologically basic existents that should be recognized as ultimately real.

Proponents of Madhyamaka, however, thought this a fundamentally flawed – indeed, an utterly incoherent – way of elaborating the Buddhist tradition’s no-self doctrine. To be sure, Madhyamaka holds that the exercise of redescribing experience in wholly impersonal terms does facilitate the discovery of something true: viz., that we do not exist in the way we habitually think we do, and that that fact is evident in the extent to which ordinary experience will indeed admit of the kind of impersonal re-description typically cultivated by Buddhist practice. What Madhyamaka rejects is just the claim that any such re-description could make sense as picking out anything essentially more real. The real insight of the Buddhist tradition, according to Madhyamaka, is precisely that there are no ultimately real existents. According to Madhyamaka, nothing at all can show up for us except under some description, which means that the very idea that we could, from within the common-sense world we inhabit, discover that that world is “really” something else cannot be made coherent. Madhyamaka arguments for that conclusion are generally to the effect that anything proposed as ultimately real – any category taken as a logical terminus of explanation, or as providing an exhaustively explanatory grip on whatever phenomena it would explain – will itself turn out to be intelligible only relative to other existents, including the very ones that were supposed to be explained.

Among the reasons why that is so is that any existents that could figure in an explanation must themselves depend on various causes and conditions, and thus cannot make sense as terminating explanation in the way that, e.g., many theists have supposed a self-existent God can. To that extent, Madhyamaka claims to offer the only thoroughly consistent account of (what is another cardinal Buddhist doctrine) dependent origination. The doctrine of dependent origination (pratītyasamutpāda) represents another way to state the no-self doctrine: the reason there are no selvesis just that every instance of experience originates in dependence upon innumerable further factors, none of which makes sense as being what that instance “really” is. In terms of this doctrine, Madhyamaka’s insight is that dependently originated existents would really be explained only by existents that do not themselves admit of the same kind of analysis – and insofar as the Buddhist tradition’s main insight is that there are no existents that are not dependently originated, nothing could make sense as being at the bottom of the pile.

More subtly, though, Madhyamaka’s insight is that any existent that can be adduced as having explanatory significance is intelligible as such only relative to some explanatory interests. Consider, for example, that causation is a relation that obtains between events. Events, however, do not individuate themselves; indeed, any event typically comprises all manner of further events (obtaining, e.g., at different spatio-temporal scales), and so the identification of anything as an event’s causemakes sense only insofar as one has in the first instance takenthat as the event at issue. Now consider that what is always of paramount interest to Buddhists is how to understand particularly those events that constitute the lives of suffering persons. What is centrally at issue, in that case, is precisely the kinds of things (viz., “persons”) that have explanatory interests, and that therefore make sense as in the first instance taking this or that event as requiring explanation. That being so, what could entitle us to have more confidence in an essentially impersonal causal explanation of that taking– more confidence, e.g., in a neurobiological account of everything involved in the occurrence of any conscious experience of, say, wanting to explain consciousness– what could entitle us to have more confidence in the explanation we entertain than in the fact that we entertain it? How, in other words, could we ever think an exhaustively impersonal explanation of conscious experience represents the ultimate truth of the matter, when it is only as conscious persons that it is so much as possible to entertain the explanation?

Thus, when Madhyamaka philosophers argued for what they took to be the right understanding of the Buddhist no-self doctrine, their guiding thought was that that doctrine cannot coherently be taken to require the specification of what “really” exists instead of the persons we think we are. The kind of impersonal analysis of experience canonically advanced in mainstream Buddhist philosophy is, then, useful in undermining our confidence that we really understand ourselves – but insofar as any such impersonal analysis is itself intelligible only relative to persons, it cannot coherently be thought that the analysis identifies anything essentially more real than the conventionally real persons who first set out to undertake the analysis. To be sure, Madhyamaka’s point is not, against basic Buddhist doctrine, that there are selves after all; it remains the point to realize that persons are just conventionally (contingently, dependently, temporally) existent. The point, rather, is just that the Buddhist tradition’s impersonal analyses cannot be thought essentially more real than these; the supposed ontological primitives posited by mainstream Buddhist philosophy, too, are just conventionally real, which is, indeed, all the more “real” that anything can be.

In terms of Buddhism’s two truths, this line of thought is sometimes expressed as the claim that the ultimate truth, for Madhyamaka, is that there is no ultimate truth– which, despite the whiff of self-contradiction, can be non-paradoxically expressed as the thought that when it comes to the nature of conscious experience, the most important truth is just that there can be no saying what that “really” is. Thus, the various Indian proponents of self-theories can be understood as proposing that what we really are transcends the vicissitudes of temporal existence; while our bodies grow old and die, our selves remain unimpacted. Against such views, mainstream Buddhist philosophers cogently argued that that is incoherent because nothing essentially unchanging can be related to manifestly temporal affairs; these Buddhists went on to affirm, however, that we must therefore recognize that what really exists is just causally continuous series of psycho-physical events. On a Madhyamaka view, however, that affirmation just reproduces the original error in a different guise; the real soteriological problem just is the very idea that there must be something that makes sense as what being a person “really” is. Liberative transformation can come, then, not from adopting an exhaustively impersonal analysis as the ultimate truth of a human life, but only from giving up the very idea that being a person consists in being any kind of thing (or any series of things) at all.

Now consider, in light of Madhyamaka’s radically alternative reading of basic Buddhist commitments, how one might understand Buddhist convictions regarding the ineliminable significance of conscious experience. Contra Dharmakīrti’s idealism, conscious experience is not, for Madhyamaka, to be affirmed as explanatorily basic, much less as essentially more real than anything else; like everything that occurs, conscious experience is likewise dependent upon all manner of causes and conditions, including, no doubt, many physical ones. (The relevant causes and conditions also include, proponents of Madhyamaka can say, all sorts of causes and conditions that are not inside the skull; there is plenty of room, on a Madhyamaka account, to allow, with a nod to Hilary Putnam, that when it comes to mental phenomena, they ain’t just in the head.) But thus to allow that conscious experience is as contingent and dependent as everything else is not, on a Madhyamaka view, to say that it can therefore be explained away; the point is just that nothing adduced as explaining conscious experience could be any more real than that. Conventionally real is all the more real that anything can be, and what must be given up, therefore, is the very idea that things count as “real” only insofar as they are metaphysically substantiated.

What Madhyamaka rejects by refuting all “ultimately real existents,” then, is not existents, per se, but only a particular criterion of the reality thereof. And – to return, finally, to physicalism and the question of rebirth – Madhyamaka arguments against that cut against physicalist versions of reductionism just as much as they do against mainstream Buddhist versions thereof. My account of Madhyamaka thus represents an important part of my case for thinking that Dharmakīrti – despite his clearly idealist predilections and his celebrated Buddhist critique of physicalism – is himself vulnerable to some of the same kinds of arguments some contemporary philosophers have leveled at physicalism. That assessment thus involves showing not only how Dharmakīrti’s psychologistic orientation has affinities with the projects of Dennett and Fodor, but also that critiques of Dharmakīrti by some of his contemporaneous critics exhibit the same logic as some modern critiques of physicalism. In particular, I take proponents of Madhyamaka to have made basically transcendental arguments against the intelligibility of reductionism – arguments, that is, to the effect that ineliminable reference to a “personal” or (as I have also put it) “intentional” level of description is a condition of the intelligibility even of the reductionist’s own account.

That Madhyamaka rejects Dharmakīrti’s intuitions about the uniquely basic character of consciousness does not mean, then, that Madhyamaka plumps instead for physicalism, which Madhyamaka would take to be just as problematic (and for just the same reasons) as any self-theory – that is just what it means for Madhyamaka to avoid both“eternalist” and “eliminativist” extremes. Thus, for a proponent of Madhyamaka to affirm that conscious experience is just conventionally real is not for her to say it is less real than something else; it is, indeed, precisely to say that it is as realas anything can be. To affirm the conventional reality of conscious experience is thus to affirm, however, that while that is not essentially different from physical existents, it is also not essentially the same asthese. To that extent, Madhyamaka will certainly resist Dharmakīrti’s proof of rebirth, with its claim to demonstrate that conscious experience is essentially distinct from anything physical – but that is not to give up on the idea that conscious experience is nonetheless conventionally distinct from the conditions of embodiment. Indeed, that conscious experience is relatively distinct is just as we must allow, since the idea that we could coherently think that consciousness is ultimately something else – whether an eternal soul or events in the brain and central nervous system – is just what Madhyamaka denies.

With all that in mind, consider what the 14th Dalai Lama has to say about the status of conscious experience and rebirth. I’m thinking, in particular, of his book The Universe in a Single Atom(2005), which, among the many books published under his name, is (as I’m reliably informed) one on which the Dalai Lama himself did considerable work. While there is an extensive literature, eminently characteristic of Buddhist modernism, concerned with the supposedly distinctive compatibility of Buddhism and science, this book differs from many examples of the genre in that it not only explores points of compatibility, but also ventures a sustained critique (not of science but) of scientism– a critique, that is, of the kind of overweening confidence in science that finds expression in the thought that there are no meaningful questions to which scientific inquiry cannot provide the answers. While many such critiques rightly argue that, e.g., axiological and ethical questions – questions about what we ought to want and how we ought to live together – will not admit of resolution by scientific research, the Dalai Lama is chiefly motivated here by a typically Buddhist concern with the reality and transformative potential of conscious experience.

It is clear throughout his book that a driving reason for this is the Dalai Lama’s reluctance to give up the idea of rebirth. The book gives many instances of Buddhist doctrines (chiefly having to do with the cosmology and natural philosophy of premodern India) that Buddhists can and should concede are superseded by scientific understanding. Rebirth, however, recurrently figures as one point on which the Dalai Lama is reluctant to yield. In this regard, the Dalai Lama concisely summarizes Dharmakīrti’s arguments for rebirth (at pp.131-32), and maintains, as well, that while the reality of rebirth has not been substantiated by science, it has not been falsified thereby, either.

While the latter point might be thought to represent a rearguard defense no more promising than many modern readers will find Dharmakīrti’s arguments to be, it seems to me the Dalai Lama is unassailably right about a closely related point that is, I think, finally more significant than any understanding of rebirth: physicalism is not a conclusion demanded by scientific research – it is, rather “a philosophical, in fact a metaphysical, position.” (p.12) That is, it may be that only physical entities can figure in any (or even all) scientific explanations; the claim that only such entities are therefore real is, however, an additional claim that is not itself warranted by scientific method. As the Dalai Lama says, he therefore does not resist reductionism, per se (which stands to reason given his own indebtedness to Dharmakīrti’s thought); problems arise, rather, “when reductionism, which is essentially a method, is turned into a metaphysical standpoint.” (p.207)

The point I want to make here is that that conclusion is very much consistent with the Madhyamaka tradition’s understanding of Buddhist commitments – consistent, that is, with an approach according to which Dharmakīrti’s confidence in causal efficacy as the criterion of the real, and his deployment of that intuition in his demonstration of rebirth, can be recognized as being just as problematic as physicalism itself is. Indeed, it is basically by way of Madhyamaka arguments that the Dalai Lama advances his own argument to that effect (which is to be expected, given that his scholastic tradition, like most in Tibet, takes Madhyamaka as the definitive expression of Buddhist insights). Just, then, as Madhyamaka argues that it cannot coherently be thought that there are ultimately real existents that exhaustively explain the conventional truth of ordinary experience, so, too, the Dalai Lama’s point is that conscious experience cannot be explained away by science just insofar as scientific inquiry is itself unintelligible without reference to the conscious experience of an observer. And again, his point is not to affirm (with Dharmakīrti) that consciousnesstherefore has a privileged status; “the reality of the external world is not denied,” he says – it’s just that that, like all existents, “is understood to be relative. It is contingent upon our language, social conventions, and shared concepts. The notion of a pre-given, observer-independent reality is untenable.” (p.63) And, as Madhyamaka argues, that notion is untenable just because it fails to account for the very perspective from which any claims about reality must be entertained.

The only reasonable conclusion, then, is, the Dalai Lama says, that “this world of empirical experience is not an illusion, nor is it unreal. It is real in that we experience it.” (p.67) I’m here reminded of the guiding impulse behind William James’s “radical empiricism”: the atomism of the British empiricists shows them not to have been “empiricist” enough, since their approach leaves out of account the continuity that, phenomenologically speaking, really is experienced. James thus held that “empiricism must neither admit into its constructions any element that is not directly experienced, nor exclude from them any element that is directly experienced” – and, James said with emphatic italics, “the relations that connect experiences must themselves be experienced relations, and any kind of relation experienced must be accounted as ‘real’ as anything else in the system.” Likewise, the Dalai Lama can affirm, with Madhyamaka, that while Dharmakīrti wrongly held that conscious experience alone is ultimately real, that is just because nothing at all exists that way – and particularly insofar as nothing at all, therefore, can be more “real” than ordinary experience, the idea that consciousness cannot count as such becomes downright absurd.

I have here sketched my understanding of Madhyamaka in terms of some Buddhist views of rebirth and consciousness not only by way of emphasizing (what many modern readers may find counter-intuitive) that most Buddhists were reductionists without being physicalists, but also to emphasize that Buddhist views particularly of the status of consciousness should, I think, be recognized as cogently striking what can indeed lay claim to being a “middle way” – which, in fact, is just what Madhyamakameans – between extremes. I am suggesting, in other words, that notwithstanding the convictions of doctrinaire physicalists, to argue that conscious experience is as real as anything else is not to make a radical or extreme point; that there is conscious experienceseems to me, indeed, the incontrovertible starting point of philosophy (a point that Descartes got right), and becomes a problematic claim only insofar as it is taken to warrant the conclusion that there must therefore be some kind of conscious substance(which is where Descartes went off the rails). Madhyamaka can thus help us appreciate that physicalism represents just the same kind of error that Descartes embraced.

3:16:  Does Buddhist foundationalism survive Madhyamaka transcendental arguments?

DA: My own view is that Madhyamaka critiques of Dharmakīrti’s approach are both philosophically cogent, and more thoroughly consistent with basic Buddhist commitments. There is, however, a crucial sense in which Madhyamaka itself makes sense only relative to the reductionist approach typical of the Buddhist philosophical mainstream (and epitomized by Dharmakīrti). In this regard, one basic way of representing Madhyamaka is as radically extending the Buddhist tradition’s no-self doctrine: the Buddhist philosophical mainstream argued that there are no selves, but that it is possible to specify the irreducibly basic existents that explain why we think there are – and Madhyamaka rejoins that these supposedly basic existents, too, are without “selves,” or, in a Madhyamaka idiom, all existents are “empty” of any intrinsic nature.

But here’s the thing: it only makes sense to emphasize that the supposedly basic existents of mainstream Buddhist analyses are empty if one has first entertained the project of reducing experience to them. That this is so is, in fact, integral to the sense it makes that Madhyamaka is not just ratifying our common-sense intuitions about ourselves and thus reneging on the no-self doctrine. Madhyamaka makes sense as a Buddhist school of thought – as, that is, developing the no-self doctrine – just insofar as Madhyamaka can itself be rightly understood only if one has first understood, pacemainstream Buddhist philosophical analyses, that being a person can indeed be recognized as dependent upon all manner of impersonal causes. When Madhyamaka is understood that way, it can be seen that its claim is not so much that mainstream Buddhist analyses stand refuted, as that those analyses, although still integral to Buddhist practice, are not, at the end of the day, themselves ultimately true, either. In that case, one can retain from mainstream Buddhist analyses the realization that we do not really exist in the way we think we do, even as one further realizes that that does not mean there is any other way that we really do exist.

While proponents of Madhyamaka such as Nāgārjuna and Candrakīrti did not themselves explicitly put the matter quite that way, it is typical of the Tibetan scholastic curricula still in use today to emphasize just that point. According, then, to Tibetan appropriations of India’s Buddhist traditions of thought, Madhyamaka can be rightly understood only if it is studied after having mastered the traditions of Buddhist thought epitomized by Dharmakīrti, who thus retains a central place in Tibetan scholastic curricula even as most Tibetan scholars hold that his philosophy, when subjected to Madhyamaka’s critiques, cannot finally withstand scrutiny. This reflects a crucially important distinction between Buddhist philosophy, and the contemporary academic discipline of philosophy: it must always be kept in mind that Buddhist philosophy finally makes sense only as integral to carefully articulated systems of transformative practices meant to effect the elimination of existential suffering. Particularly in light of that fact, it makes sense for Buddhist thinkers to recognize a wide range of philosophical analyses – even those that are finally thought to be superseded – as significant, and indeed necessary, for realizing the religious project of Buddhist practice.

3:16: As a take home can you say where this leaves Buddhist doctrines like the middle way and selflessness and what the mind is? How do you think Buddhist doctrines are best understood as here?

DA: The fact that often divergent Buddhist philosophical schools can, as in Tibet’s scholastic traditions, all be taken as integral to a larger project of radical self-transformation surely indicates an important difference between these traditions, and “philosophy” as practiced in university departments thereof. Buddhist philosophical discourse, as many scholars have in recent years been apt to say, is better understood, with Pierre Hadot, as reflecting philosophy as a way of life– as continuous, that is, with possibly numerous techniques for cultivating persons as certain kinds of moral agents, rather than as an academic specialization whose disciplinary and canonical boundaries are to be guarded from any and all encroachments by disciplinary or canonical “others.”

Nevertheless, Buddhist traditions of philosophy are every bit as sophisticated as any in the world, and they surely comprise all manner of novel philosophical moves that would repay the close attention of today’s professional philosophers. I have, for example, emphasized the important point that Buddhist philosophy attests to the possibility of being at once a reductionist and an idealist. What’s more, the whole world of Buddhist ethical thought and practice represents an ethics that, as against the predominant views of many other traditions, is not intelligible in spite of the fact that there are ultimately no “selves,” but precisely because that is so. Relatedly, the Madhyamaka tradition, in particular, surely represents one of the world’s most thoroughgoingly consistent traditions of critique, as proponents of this school did not shrink from applying their arguments even to their own analyses

Among the really difficult but profound points, then, to be worked out in connection with this school of thought is that Madhyamaka analysis itself is taken to be no exception to Madhyamaka’s claim that there are no ultimately real existents. To that extent, Madhyamaka seems to me one of history’s most philosophically promising attempts to argue that the things that matter most to us are not deprived of their significance because they are, as contemporary historicists of all stripes would urge, recognized as wholly contingent functions of situated creaturely endeavor; indeed, Madhyamaka can be understood as arguing that it is just becauseit cannot coherently be thought that there is any other world than that of ordinary experience that it matters what we do. As Nāgārjuna famously affirmed in this regard, there is, ultimately, no difference at all between nirvāṇa and saṃsāra– a thought, surely, with possibly profound implications for the understanding of all religions.

But while I would thus argue that the regrettable parochialism of academic philosophy departments has meant, among other things, that a vast tradition of great philosophical sophistication has been unduly neglected, and that philosophers would thus do well to engage Buddhist thinkers, too, as philosophers, I would conclude by returning to the point that “philosophy,” in classical Buddhist contexts, is integral to a way of life. Thinking, among other things, of my own misgivings regarding the extent to which the existentially compelling significance I originally found in philosophy sometimes seems diminished by the fact that it is now something I’m paid to pursue, I wonder if what Buddhists might most importantly have to teach contemporary philosophers may not chiefly have to do with this or that distinctive argument or position. Perhaps, then, it would most behoove us to learn from Buddhist thinkers that philosophy ought to be most valued for its significance in transforming the perspectives out of which we live.

3:16:  And finally, are there five books you could recommend other than your own that will take us further into your philosophical world? 


 Nāgārjuna, Fundamental Verses on the Middle Way (as one might translate the title Mūlamadhyamakakārikā). I am most inclined to engage Nāgārjuna’s Sanskrit text, but would particularly recommend two of the many translations thereof: Mark Siderits and Shoryu Katsura, Nāgārjuna’s Middle Way (Wisdom Publications, 2013), which is probably the most reliable translation of Nāgārjuna’s Sanskrit; and Jay Garfield, Fundamental Wisdom of the Middle Way (Oxford, 1995), a translation that is rather less reliable insofar as it is done from the Tibetan translation, but that is greatly enriched by Garfield’s own philosophically incisive commentary.

Steve Collins, Selfless Persons (Cambridge University Press, 1987). One of the first books I typically recommend to people interested in the basics of Buddhist thought.

Paul Griffiths, On Being Mindless(Open Court Press, 1994). A philologically responsible and philosophically incisive book that first taught me that the best way to take Buddhist thought seriously is to engage it critically.

William James, Essays in Radical Empiricism (1912; available in many editions). Notwithstanding James’s own contention that his pragmatism can be evaluated independently of his “radical empiricism,” I would contend that these essays are absolutely integral to an understanding of his overall orientation – in addition to which, they crystallize a distinctively American tradition of naturalist, historicist metaphysics that seems to me very promising.

J. J. Valberg, Dream, Death, and the Self (Princeton University Press, 2007). My favorite of any philosophical work I’ve read in the last decade or so, and not just because it is such a fine development of the idea of transcendental subjectivity, but especially because Valberg so sensitively brings out the existential significance of that idea.


Richard Marshall is biding his time.

Buy his new book here or his first book here to keep him biding!

End Times Series: the index of interviewees

This site was built using