Introduction

THE ESSAYS COLLECTED here are on the Madhyamaka, the “philosophy of the middle” that begins with Nāgārjuna and Āryadeva in the second century CE and evolves throughout Indian, Tibetan, and Chinese Buddhism.¹ Many have appeared previously and are reprinted with a few revisions; some (chapters 5, 7, and 12) are new. As in my previous work on other Buddhist philosophers, notably Dharmakīrti, these essays zigzag between a historico-philological approach and philosophical analysis. It is a delicate balance. Madhyamaka needs to be understood charitably with the best philosophical reading that textual data and intellectual contexts permit.

The book is loosely organized in terms of four topics: Madhyamaka’s philosophical promise (chapters 1 and 2), features of its philosophy of logic and language (chapters 3–7), its ethics and spiritual path to Buddhist enlightenment (chapter 8–10), and its potential contributions to contemporary philosophical controversies (chapters 11 and 12).

This philosophy has had a mixed reception, both in the past and nowadays. It has been taken as the pinnacle of subtlety by its promoters but also vilified, and even simply ignored, by many Buddhists and non-Buddhists alike, one of the frequent objections being that it distorted opposing realist positions willfully and thus refuted straw men. The first essay, “Trying to Be Fair,” attempts an initial view of the lay of the land, in a charitable light. The second, “How Far Can a Mādhyamika Refute Customary Truth?” takes up issues of relativism, truth, and ontology. Chapters 3 and 4 deal with the question of whether the Mādhyamika respects universally the law of noncontradiction; the argument is that while a weak, nonadjunctive type of inconsistency may well fit some sūtras and possibly Nāgārjuna, inconsistency of any sort is considered as a fault by the time of the sixth century and is thus not endorsed by major Mādhyamikas such as Candrakīrti, Bhāviveka, and their successors. Chapter 5, “Prasaṅga and Proof by Contradiction in Bhāviveka, Candrakīrti, and Dharmakīrti,” takes up a crucial application of the law of noncontradiction: reductio ad absurdum, or prasaṅga, which has long been considered the preferred strategy of the Mādhyamikas who follow Candrakīrti (commonly known as prāsaṅgikas) to expose the internal flaws of philosophical positions while not taking a position of their own. The use of proof by contradiction in the Mādhyamika thinkers Candrakīrti and Bhāviveka is contrasted with that of Dharmakīrti, a sixth/seventh century non-Mādhyamika metaphysical realist, who along with the founder of the Epistemological School, Dignāga (fifth century), is sometimes the direct adversary, and often the éminence grise, in so many later Indo-Tibetan Madhyamaka debates.² Chapter 6, “Apoha Semantics: What Did Bhāviveka Have to Do with It?” is the most historical and philological chapter of the book. It shows how close the relationship between Epistemologists and Mādhyamikas was historically in India, as the dominant semantic theory adopted by the leading Indo-Tibetan Buddhist thinkers seems to have been quite significantly influenced by the sixth-century Mādhyamika Bhāviveka. Chapter 7, “What Happened to the Third and Fourth Lemmas in the Tibetan Madhyamaka?” takes up problems that arise in Tibetan interpretations of the third and fourth lemmas in the famous tetralemma, or catuṣkoṭi, whose use Mādhyamikas claimed as essential to their abstention from all theses (pakṣa, pratijñā) or philosophical positions (abhyupagama). The question arises whether Mādhyamikas can and should accept a law of double-negation elimination (pratiṣedhadvayena prakṛtagamana; dgag pa gnyis kyis rnal ma go ba “understanding the main [proposition] by means of two negations”) and whether they respect De Morgan’s Laws. The issues are real, even if the formulation of the problem was skewered in thinkers like Gorampa (Go rams pa), Khedrup Jé (Mkhas grub rje), and others due to syntax-provoked ambiguities in the Tibetan language that led them to widely confuse statements of the form “neither φ nor ψ” with those of the form “not both φ and ψ.”

From logic and language we turn to ethics and the Buddhist spiritual path. Chapter 8, “Madhyamaka Buddhist Ethics,” examines, inter alia, whether a Mādhyamika’s reliance on scriptural authority to settle ethical issues can be justified, as Candrakīrti claims, on the basis of what the world itself recognizes (lokaprasiddha). The Candrakīrtian argument is, however, an interesting failure. The upshot is that traditional Buddhist ethics founded on the law of karma involves a type of epistemic protectionism that, if left unchallenged, prices Buddhism out of secular ethical reflection. Chapter 9, “Reason, Irrationality, and Akrasia (Weakness of the Will) in Buddhism,” looks at recurrent arguments from the eighth-century Mādhyamika Śāntideva that suggest a recognition of incontinence, or weakness of the will, as a significant factor in Buddhist ethics. “Yogic Perception, Meditation, and Enlightenment,” chapter 10, is a reexamination of the philosophical themes in the key eighth-century debate between the Chan monk Heshang Mohoyen and the Indian Mādhyamika paṇḍit Kamalaśīla.

The final two chapters are applications of Madhyamaka to contemporary problems in the philosophy of mind and ontology. “On Minds, Dharmakīrti, and Madhyamaka” argues that the usual Buddhist defenses of mind will fare badly against modern eliminative materialism; a Madhyamaka approach would hold more promise. Chapter 12, “Serious, Lightweight, or Neither: Should Madhyamaka Go to Canberra?” looks at how Madhyamaka might find a place in current metaontological debates on the legitimacy and merits of pursuing philosophical questions of what there really is. “Canberra” refers not only to the political capital of Australia but also to a philosophy department that is especially associated with the revival of interest in metaphysics in analytic philosophy. Should a modern Madhyamaka philosophy also tread the Canberran path? Perhaps the Svātantrika could. Prāsaṅgika-Madhyamaka, however, is best seen as a quietist philosophy that is fundamentally out of step with most of the new analytic metaphysics as practiced in Australia and elsewhere.

* * *

The theme that comes back repeatedly in the essays is indeed quietism, or the Mādhyamika’s reasoned disengagement from all philosophical theses, and hence debates (vivāda) about them. Here is how Nāgārjuna put it in verse 50 of his Yuktiṣaṣṭikā:

           Superior individuals have no theses (pakṣa, phyogs) and no philosophical debates; how could there be any opposing theses for those who have no theses [themselves]?³

That quietism is not just prudential advice — like what we find in the Pāli canon — to avoid destructive and fruitless discussions with hostile adversaries. Nor is it, in my opinion, a dishonest refusal to show one’s own hand, as the non-Buddhists regularly caricatured Mādhyamikas in calling them vaitaṇḍika: “cavilers,” “sophists.” At its best, it is a defendable truth: that one cannot reasonably have philosophical theses and take sides in debates on how things really are in themselves, or how they are in an ultimate fashion. Finally, quietism is not only a stance for philosophers who argue. It figures in meditation and practice too, as Mahāyānists generally agree that there comes a point on the Buddhist spiritual path where views (dṛṣṭi, darśana), and indeed all conceptual thinking (vikalpa, kalpanā), are to be left behind. Where that point is — either at the beginning or the end of the path — constitutes the subject of the well-documented historical confrontation between Chan and Madhyamaka whose philosophical issues are my focus in chapter 10. Suffice it to say here that both Chan and the Madhyamaka recognized that — sooner or later — Buddhist practice is quietistic.

There is, however, an obvious tension here. As the articles in the present book show, there are such ample Madhyamaka contributions to typical philosophical controversies in logic, language, causality, epistemology, and ethics, to name a few, that the question naturally arises as to how such inquiries could be pursued if Mādhyamikas supposedly remained unengaged, advocating instead the āryatūṣṇīṃbhāva, the “silence of the noble ones.” What room, if any, do quietistic Mādhyamikas leave for truth claims? Can they engage in philosophical debate in some limited way? The answers to these questions are complex and will vary considerably from one thinker to the next and from one period to the next. Let us look at some of the varieties of quietism in more detail.

One possible — admittedly speculative — way to understand Madhyamaka quietism would be to see it as resulting from a certain tolerance of contradiction. As developed in chapter 3 and 4, the author(s) of the Prajñāpāramitā and other sūtra literature, and perhaps Nāgārjuna, could be read as regularly asserting and denying the same statements p, q, r, and so on, so that p, q, and r are affirmed for the worldly reasons that the common man would give but are also denied from an ultimate point of view, as there is nothing that has an ultimately existing “intrinsic nature,” or svabhāva, or in other words, really is what it is anyway — that is, independently of all relations it might enter into and whatever one might say or think about it. Affirmations and denials of the same statements cancel each other out, leaving disengagement from both and suspension of belief. Nāgārjuna, read in this way, would be quietistic in not himself making truth claims.

Interesting as it may be, such reliance on paradox, however, cannot easily be read into post-third-century writers like Candrakīrti who respect strictly the law of noncontradiction. Their quietism turns on other considerations. A common traditional Indian and Tibetan interpretation of Candrakīrti (what I term in chapter 2 “typical Prāsaṅgika Madhyamaka”) holds that things lack real intrinsic natures and — what is the same — that things are therefore false appearances (ābhāsa, snang ba) that merely, in an oft-repeated stock phrase, “exist from the point of view of mistaken minds” (blo ’khrul ba’i ngor yod pa) or, less literally, “just seem to be thus and so to the mistaken.” This generally unpacks as the idea that for Mādhyamikas there are no genuine “sources of knowledge” (pramāṇa, tshad ma), because what people term “right understanding” is nothing more than a type of “seeming to be right” that meets with widespread acceptance by the world, which in any case invariably labors under illusions. This is a strongly antirealist way of reading Madhyamaka: there is nothing “out there” that really is what it is anyway and can be represented correctly by our thoughts and language; minds and conventions make the world’s entities, which work as they do because people believe in them.⁴ Mādhyamikas are then quietists because they shun any and all deeper questions about what there actually is and content themselves with what the world acknowledges (lokaprasiddha); they thus use reductio ad absurdum of others’ positions to demolish them but don’t affirm anything that they themselves hold to be true. In this book, unless otherwise indicated, I’ll generally read Candrakīrti in this way — as a “typical Prāsaṅgika.” Such is the book’s default setting: Candrakīrti read in a natural Indian fashion. In some chapters nothing much turns on it, but sometimes it is important. Chapter 5 on prasaṅga, for example, turns on this “typical Prāsaṅgika” reading.

Other Mādhyamikas, such as Tsongkhapa (Tsong kha pa, 1357–1419) and his followers — “atypical Prāsaṅgikas” — think quite differently. They do accept that there are sources of knowledge (pramāṇa) and thus have a much greater place for beliefs and truth claims — there are true/right answers, where “true” is not just “seeming to be true,” “generally thought to be true,” “believed in,” et cetera. In their logic, although they adopt the moniker Prāsaṅgika, they believe and prove truths in a minimalist fashion not involving a metaphysics of intrinsic natures. Crucially, then, they insist upon a split between how things are confusedly grasped in the fashion of a metaphysical realist committed to intrinsic natures and how they are understood innocently and rightly without such commitments. The daunting task for these quietists, then, is to disentangle the innocent from the confused, and stay unengaged on positions that involve the latter.

A thought such as this in which there is a richer place for truth and reality is, in my opinion, considerably more sophisticated and promising than the global error theory or fictionalism of the typical Prāsaṅgika. I fully recognize that it is going to be disturbing to some to read that a fourteenth-century Tibetan philosophy may have been, in certain significant respects, clearer and even much better philosophy than that of the Indian thinkers on which it was based. That evolution toward sophistication should be unsurprising to the historian who recognizes that traditional religious thinkers are regularly obliged to disguise their innovations and creativity. Nonetheless, it means that much of the effort to read Tsongkhapa and other later Mādhyamika thinkers back onto Candrakīrti or Nāgārjuna is strained. Let’s be clear: Madhyamaka changes significantly over time, sometimes for the better.

But could this type of Madhyamaka philosophy work? Or would its acceptance of sources of knowledge, right answers, truth, and reality entail that the Mādhyamikas came full circle and had to accept fundamental entities with fully fledged intrinsic natures? In chapter 2, the limited realism and truth that these atypical Prāsaṅgika Mādhyamikas conserved is explained in terms of a form of deflationism, where there is nothing more to be said about truth and real facts other than platitudes like the so-called T-sentences of a semantic theory of truth: <p> is true if and only if p, “‘Snow is white’ is true in English if and only if snow is white.”⁵ Of course, a bottom line of deflationary truths and facts is thin fare for those who expect more from an account of truth and reality or from a Buddhist religious philosophy of liberation. Deflationism does nonetheless retain the crucial minimal features of the concepts of truth, objectivity, and reality, namely that truth involves an all-important contrast between being right and just seeming to be right and that any viable concept of reality thus needs a difference between what is the case (or what has been the case all along) and what just seems to be the case or is widely believed to be so. This is quite a lot, and it comes without metaphysical strings attached. It brings no commitments to one who rejects svabhāva and is thus already disinclined to ontology. The radical point is thus that such deflationary truths and facts may be all the Mādhyamika Buddhist needs for his idea that true and false statements, existence and nonexistence, realities and unrealities are established customarily (vyavahāratas), by genuine sources of knowledge, and never ultimately (paramārthatas).⁶

What then is the place for philosophy? Svātantrika-Mādhyamikas like Bhāviveka, Śāntarakṣita, and Kamalaśīla not only make truth claims but have a much greater place for positive, constructive, philosophical theorizing than their Prāsaṅgika counterparts. Indeed, I have argued that one of the most important differences between the two is that Svātantrikas reinstate much of the ontology, philosophy of language, and epistemology of Dignāga and Dharmakīrti as customary (rather than ultimate) truths.⁷ The Prāsaṅgikas, by contrast, have no interest in such constructive philosophizing, and thus their quietism also ranges over ontology and epistemology even when transposed to customary truth: they generally have nothing constructive to offer as a theory of universals, particulars, negative facts, perception, memory, reflexive awareness, and the like.

In chapter 12, I try to make sense of this wide-ranging quietism by making a distinction, following Kit Fine, between purely quantificational and ontological questions — roughly those that ask, in an innocent, noncommittal way, whether there is an x such that x is F and those that ask whether the x that is F really exists; the latter questions do, thus, involve ontological commitment. Much of science and common sense’s interest in what there is arguably involves the first sort of question. For example, when a mathematician has found a number that has the property F, it is trivial and of no metaphysical import to infer “There is an x such that x has F.” Philosophical debates on the existence of universals, absences, nonnatural properties, numbers, and the like have, of course, been generally framed in terms of questions of the second sort. If, however, we were also to regard the existence of universals, absences, and so forth in the purely quantificational way, then it might well become harmless and even obvious that we should accept them. And doing so would not be a philosophical thesis at all but just an acceptance of uncontroversially true statements coupled with a dismissal of the philosopher’s sense of the strangeness of the entities concerned. I would contend that something along those lines is a plausible rational reconstruction of the Prāsaṅgika quietist’s move when he grants that there are universals and so on but stays clear of all philosophical/metaphysical debates about them.

Not surprisingly, many modern writers would not agree with a purely negative, quietistic, characterization of Prāsaṅgika-Madhyamaka and insist that it also has, or perhaps just should have, a minimal positive philosophical explanation of why the things of the world are precisely the ones they are. That positive account is presented as a corollary of the Mādhyamika’s idea of emptiness, and it goes something like this: because objects have no (i.e., are empty of) genuinely real intrinsic natures and thus have no real way of being as they are anyway, then it must be minds, human society, and language (and possibly other factors) that somehow bring about the things of our world. The Mādhyamikas, so it is claimed, are after all providing a positive account of the genesis of objects, even if it is not a full-fledged ontology of real entities.

The most plausible of such positive accounts are formulated in terms of pragmatism: we adopt, or even choose, certain objects that are “convenient fictions,” useful to us in meeting our needs and furthering happiness, and we reject or ignore those that are not. Jay Garfield, in the introduction to his translation of Nāgārjuna’s Mūlamadhyamakakārikā, put his version of that positive Madhyamaka account in terms of choices of useful mereological sums. Others, like Mark Siderits, but in a broadly similar vein, speak of us adopting “conceptual fictions” — tables, chairs, personal identity — because of our interests and needs. We can, for our purposes, take the two positions together. Here is the version in Garfield 1995, 89–90:

           To say that it lacks essence [i.e., svabhāva], the Mādhyamika philosopher will explain, is to say, as the Tibetans like to put it, that it does not exist “from its own side” — that its existence as the object that it is — as a table [Garfield’s italics] — depends not on it, nor on any purely nonrelational characteristics, but depends on us as well. That is, if our culture had not evolved this manner of furniture, what appears to us to be an obviously unitary object might instead be correctly described as five objects: four quite useful sticks absurdly surmounted by a pointless slab of stick-wood waiting to be carved. Or we would have no reason to indicate this particular temporary arrangement of this matter as an object at all, as opposed to a brief intersection of the histories of some trees…. The table, we might say, is a purely arbitrary slice of space-time chosen by us as the referent of a single name and not an entity demanding, on its own, recognition and a philosophical analysis to reveal its essence.

What is being maintained here, if I have understood correctly, is in part the recognizable antirealist position that entities of the world being what they are depends not only on things but “depends on us as well.” We carve out our worlds for our own reasons.⁸ Thus, the components of our world — tables, chairs, chariots, people — are what they are because of a set of largely collective decisions; for the world as we know it, and not another, was somehow found useful to our interests (especially interests conditioned by preexisting social institutions) or was even chosen because of its utility.⁹ And this is not a blind causal process, like an evolutionary acquisition of thumbs or digestive enzymes. Instead, some brute, unstructured objects, like pieces of wood and chunks of metal, were interpreted as belonging together and thus structured as tables and chairs for strategic reasons. We could have taken them otherwise — it was arbitrary — but we did not because we somehow saw that it was not in our advantage as rational agents. The parallel might be with a group of rational agents’ economic decisions aiming for prospective profits.

Any philosopher who deliberately stays close to ordinary usage should feel uneasy here. After all, we commonly understand what it is ordinarily to choose rationally some thing or person or to choose an alternative course of action, but what would it mean to stretch that further and say that people also choose the entities of their worlds? Of course, people ordinarily do have interesting and sophisticated ways to answer legitimate questions about the origins of specific things — for instance, “Why are there planets, human beings, wars, and so forth?” These can be answered with complicated reasons from science or simpler ones from common sense. But a story about how we interpret things to make up useful mereological sums or adopt “conceptual fictions” like tables, people, and planets for reasons of convenience is never that type of answer. (We know what it means for vanadium to be used in making steel alloys or in flow batteries but not what it means for vanadium to exist because of our uses.) For a Mādhyamika who places store on what the world acknowledges, something would seem to be going badly wrong. Instead of the usual talk about how various things come about, why they exist, how they are used, and so on, we are attempting to extend that talk and say, in philosophical terms, how acceptance of anything at all is grounded in virtue of a further type of fact — namely, decisions about utility. And that step beyond is suspect.

It is not only suspect, but such an account of why things exist could not be given on the wide scale Garfield’s and Siderits’s position would demand. Indeed very large scale or exclusive appeals to usefulness and human ends to explain the existence of objects and people would seem to involve a vicious circularity, for in order to determine ends and usefulness to them in human enterprises, we already need to have a world in its broad outlines, with people and many macroscopic objects too. In short, usefulness of carts, tables, and the like to people presupposes a context in which there are people, their environments, and their complex interactions with a lot of quite different sorts of objects. If strategies to further human ends were themselves responsible for the genesis of all these entities, their genesis would seem to become unintelligible.¹⁰

Actually, the Mādhyamikas themselves do not seem to have used their own part-whole argumentation to come up with anything like Garfield’s positive account of how thought, language, or societal transactions lead us to choose mereological sums strategically. True, reasons of utility do explain why buddhas or bodhisattvas supposedly describe the world in such and such fashion to guide disciples along the path — what is commonly known as “skill in means” (upāyakauśalya), where teachers make pedagogically useful statements that should not be taken at face value but need exegesis (neyārtha) to get their message across. But that is not what is going on here. The question facing us is whether ordinary beings themselves rely on reasons of utility to choose the world they have and whether the world thus “depends on us as well” in this fashion. The Buddha’s pedagogy and pragmatic skill in means just concern his methods for teaching others.

We do find numerous statements to the effect that everything is a mere designation (prajñaptimātra), dependently arisen, designated in dependence on something else (upādāya prajñapti), empty of intrinsic natures, and that intrinsic natures are superimposed (samāropita), and so on and so forth. It is well known that Nāgārjuna (in the Mūlamadhyamakakārikā) and Nāgasena (in the Questions of King Milinda) held that a chariot or the self is prajñapti/paññati. The temptation may indeed be great to gloss these terms, as does Siderits, as meaning “convenient designator”¹¹ or a “useful fiction.”¹² But let’s be clear: “convenient” and “useful” are additions. Siderits admits as much in his account of customary/conventional truth in Vasubandhu and the Abhidharma (an account that he seems to apply to Madhyamaka too):

           A sentence is said to be conventionally true if and only if it is assertible by the conventions of common sense, where these are understood as standards based on utility.¹³…

                These accounts are extrapolations from what is actually said about conventional truth (saṃvṛtisatya) and ultimate truth (paramārthasatya) in the Abhidharma literature…. Vasubandhu says that the conventions for the terms “pot” and “water” having been made, the statement “Pots and water exist” is conventionally true. And since he has made clear that the relevant conventions involve the aggregation of either spatial parts or atomic properties, it seems fair to say that these conventions, and thus conventional truth, reflect the pragmatic standards of common sense.¹⁴

Is this actually fair to say? It is one thing to give a negative account (as Mādhyamikas or Ābhidharmikas do) of why objects like pots and water are unfindable under analysis and are mere designations (prajñapti, Pāli: paññati) or are customary truths/realities (saṃvṛti, sammuti) that are commonly accepted; it is another to move to justificatory talk about the convenience and usefulness of certain designations and thus introduce pragmatism and talk of strategies. Although the second move is easy to make,¹⁵ it does not seem to be there, explicitly or implicitly, in the Abhidharma/Abhidhamma accounts of paññati (designation, concepts), sammuti (convention, consensus),¹⁶ nor in the well-known Madhyamaka etymological explanations of three uses of the term saṃvṛti as “what is real for the obscured,” “what arises dependently,” and “what is governed by agreement.”¹⁷ Indeed, significantly enough, when Buddhists do provide accounts — over and above the world’s own explanations — as to why we experience and accept the customary realities we do, the account is often framed in terms of a moral causation: karma accumulated in past lives and ripening in the present in unfathomable ways. Or sometimes it is said that we have “beginningless tendencies” (anādivāsanā) to experience things in a certain way.¹⁸ Or it is simply said that such are our customs and practices, our vyavahāra. In any case, our making choices for reasons of utility doesn’t seem to play any role in the explanation. It is not said that we interpret otherwise unstructured brute data in certain ways because we aim at useful fictions or sum individuals. We could perhaps go further: arguably, the Buddhist account of how all things come to be in terms of humanly unfathomable karma, “beginningless tendencies,” and custom is itself a backhanded admission that constructive philosophical accounts of the origin of everything cannot be given and should not even be attempted.¹⁹

So much for the texts. It will probably be replied that the point is more about what the Mādhyamikas should say philosophically than what they did say. In other words, even if they did not fully, or properly, develop their own philosophy of emptiness (śūnyavāda) in that constructive pragmatic way, they would nevertheless need the pragmatic complement to avoid an otherwise incomplete picture. They would need some such account to develop their own ideas of how vyavahāra — customs, accords, societal transactions and practices — determine our acceptance of things. Many are, for example, the attempts to explain the evolution of cultural phenomena in broad accordance with the same evolutionary game theories initially developed in biology.²⁰ Why then not also our strategic choices of mereological sums?

Let’s expand the debate a bit to see how a modern Mādhyamika could resist that seemingly constructive pragmatic explanation of how our world as a whole comes to be. Indeed, whether self-professed quietistic philosophies should nonetheless admit such a constructive counterpart is not just a Madhyamaka matter but in many respects an East-West problem. It is well known that Wittgenstein, in his later philosophy, repeatedly invoked forms of life and linguistic/social practices when confronted with philosophically inspired questions about how understanding and meaning are possible. This has led many commentators to see him as providing the makings of actual philosophical answers to those questions in terms of societal practices, or even “social pragmatism.” Such is, in effect, the positive, constructive reading of the Philosophical Investigations and other works of Wittgenstein that we find in Crispin Wright and others, and which is contested by philosophers like John McDowell, for whom it is a serious, albeit very seductive, mistake.²¹ On a strongly quietist reading like that of McDowell, Wittgenstein’s point is that our most basic understandings do follow customs in which we have been trained but without the overintellectualization that comes from seeking interpretation and choice “all the way down.”

It is not controversial to say that we have customary practices within which the useful is valued. No doubt, too, our practices generally have little or no place for mereological sums when the components would be so discontinuous as to make the sum impossible to use. And a good number of entities — clearly so in the case of tables, carts, borders, and stock markets — are obviously unintelligible taken independently of customary practices in which they figure. Indeed, if the Mādhyamika is right, all are. The controversial philosophical step is to say that we therefore somehow interpret otherwise neutral data to pick out what sums and kinds are useful to us in those practices and that this explains why we have the world we do and not another.

Much of the Philosophical Investigations and Remarks on the Foundations of Mathematics is consecrated to the “paradoxes of rule-following” and seeks to show in detail how no one could ever understand basic facts — like someone’s linguistic behavior being in keeping with a particular rule — if such an understanding were itself an interpretation of data that could always have been taken otherwise.²² We thus arrive at Wittgenstein’s given: customs, forms of life, are the ground or bedrock beyond which the philosopher’s justificatory quest is misguided, as it is a search for one set of reasons and one interpretation too many.²³ Something analogous may well be going on in the Madhyamaka idea of vyavahāra admitting of no analysis (avicāra), or “being fine as it is unanalyzed” (avicāraramaṇīya).²⁴ To say that designated things depend on vyavahāra is thus to arrive as far as one can go: justificatory reasons stop there and can go no further in spite of the philosopher’s constant temptation to uncover ever more interpretations and choices. Why should the modern Mādhyamika stop there? Well, for one thing, those Wittgensteinian paradoxes apply to rule following across the board: in arithmetic, in meaning such and such by words, and interestingly enough, they would also apply to rules of how to group things together under kinds or combine things to make mereological sums. If rule following was an interpretation, we would never know which rule we, or anyone else, were following, and which kinds or sums people (ourselves included) intend.

How then would the Madhyamaka idea of an unanalyzable vyavahāra, “fine as it is,” fit into such an updated picture? The Madhyamaka — indeed the Mahāyāna in general — consecrates considerable effort to show that it is impossible that our ideas of entities are copies of, or somehow dictated by, entities as they are intrinsically. They are unintelligible without rule-governed contexts — customs and practices. But that would not mean that they owe their being to human strategic choices and interpretations. Following rules and customs could better be seen as no more a strategic choice or interpretation than would be a drowning person’s understanding of the word “Help!”²⁵ It would be a natural fact in need of no further justification and analysis, “as much part of our natural history as walking, eating, drinking, playing” (Philosophical Investigations §25). The point in saying that the customary is unanalyzable would then be this: there is no deeper account to explain entities, and indeed our lives, being what they are, whether in terms of intrinsic natures or in terms of strategic choices. The felt necessity to ground forms of life on something else (how reality is intrinsically, our reasoned choices, interpretations, what have you) is the mistake. Quietism, be it Madhyamaka or Wittgensteinian, is a type of vigilance.

* * *

It remains for me to acknowledge people who have helped me along the madhyamā pratipad over the years, especially Jacques May, a consummate scholar of Buddhist Studies and uncompromising Madhyamaka quietist, and the late Geshé Tamdin Rabten (Dge bshes Rta mgrin rab brtan), with whom I and a number of other Westerners of my generation studied Dbu ma/Madhyamaka and Tshad ma/Pramāṇa in the 1970s and early 1980s. The ideas in this book would not have taken the form they did without the backdrop of exchange with Mark Siderits, Jay Garfield, and Georges Dreyfus, who all hold quite different positions from mine. The debt to them is obvious throughout, and I thank all three of them for years of debate on Madhyamaka. Shōryū Katsura, David Seyfort Ruegg, Graham Priest, Yasuo Deguchi, Arindam Chakrabarti, Eli Franco, Dan Arnold, Sara McClintock, John Dunne, Malcolm Eckel, Johannes Bronkhorst, José Cabezón, Jiří Holba, Jan Westerhoff, Koji Tanaka, Bronwyn Finnigan, Helmut Krasser, Birgit Kellner, David Higgins, Pascale Hugon, Vincent Eltschinger, Ken Liberman, and Stephen Batchelor all deserve my heartfelt thanks for hearing me out, sometimes correcting unclarities, and often helping me formulate ideas in one way or another. Sincere thanks to David Kittelstrom and the team at Wisdom Publications for their editorial rigor. There are many others who have helped.

Some of the papers here came out of my lectures, seminars, and conversations at the University of Lausanne in the former Department of Oriental Languages and Cultures. I was blessed with excellent doctoral students over the years. I also have a great debt to the Center of Buddhist Studies at the University of Kathmandu, where I presented ideas in annual conferences and in intensive courses, and to my students at the University of Vienna for a delightful class on Indo-Tibetan Buddhism as philosophy that stimulated me every week throughout the summer semester of 2012.

Finally, a book such as this one, a collection of articles written over more than a decade, will inevitably have some repetitions, changes of perspectives, and other inelegances. With the exception of chapter 1, the original articles have been left mostly as they were, as it was practically not feasible to do otherwise. Some translations and terms might also merit rethinking, and on occasion the philosophical analyses were too cautious, perhaps because Buddhist studies was, for a long time, so cautious about the methods of analytic philosophy in general. The more serious problems, however, are the philosophical implications that I did not see or only dimly understood. I can only hope that others will see them more clearly than I did.

Notes to the Introduction

  1.   The translation of Madhyamaka as “philosophy of the middle” is in keeping with Seyfort Ruegg 2010. Should the term be Madhyamaka or Mādhyamika? The attempt to see the former as the word for the school of thought and the latter as the word for the adherents to this thought is, as Jacques May pointed out quite some time ago, not grounded in rigorous data from Sanskrit texts. Even if Madhyamaka might be more of a designation of the system, Mādhyamika is attested for both, viz., the system and its adherents. The situation in Sanskrit is not parallel to that in Tibetan where dbu ma and dbu ma pa do respectively designate the philosophical system and its adherents. See May 1979, 472. In what follows, we have adopted this new convention of distinguishing between Madhyamaka (the school, the philosophy) and Mādhyamika (the adherents, the philosophers). Avoiding orthographic chaos is probably enough of a justification.

  2.   For a summary presentation of my views on Dharmakīrti’s philosophy, see Tillemans 2011a — that is, the Stanford Encyclopedia of Philosophy entry, plato.stanford.edu/entries/dharmakiirti.

  3.   Che ba’i bdag nyid can de dag // rnams la phyogs med rtsod pa med // gang rnams la ni phyogs med pa // de la gzhan phyogs ga la yod //. Tibetan text in Lindtner 1990, 114.

  4.   For antirealist interpretations of Madhyamaka, see Siderits 1988 and 1989 and Westerhoff 2011.

  5.   On the problems and advantages of a deflationary theory of truth and its T-sentences for Madhyamaka, see Priest, Siderits, and Tillemans 2011.

  6.   Deflationism might capture philosophically important features of Madhyamaka, but it would be wildly bizarre to imagine that simply thinking about deflationism or concentrating on it is somehow what Buddhists promote as a path to enlightenment. Our claim is just that it might describe in philosophical terms what the quietistic Prāsaṅgika perspective on truth would have to be like. Buddhist paths — sūtra or tantra — involve many quite sophisticated psychological techniques and ethical practices that induce personal transformations in ways that philosophizing about truth theories cannot. Also, without additional understanding of how intrinsic natures are somehow incoherent, a deflationism would be little different from the commonsense realism that Madhyamaka diagnoses as confusedly grasping at reifications. Cf. MacKenzie 2009. Many of these same considerations would apply, mutatis mutandis, to quietism and the abstention from philosophical positions.

  7.   See Tillemans 2003. The terms Svātantrika and Prāsaṅgika are, as is relatively well known by now, Sanskritizations of the Tibetan terms rang rgyud pa and thal ’gyur ba. This should be borne in mind. Jay Garfield has finally convinced me that it is pedantic to put an asterisk in front of these Sanskritizations. For more on these and related Tibetan terms for subschools of the Madhyamaka, see chapter 2, note 7.

  8.   Cf., for example, Hilary Putnam’s “internal realism,” according to which the world is not independently as it is because the various “carvings” into kinds and individuals are brought about by our cognizing minds.

  9.   Cf. Siderits 2003, 8: “[O]ur ontological intuitions are being guided by our institutionally arranged interests,” by which he decribes how we assemble various parts to make a cart because we have an institutionalized use for such an entity — transportation — but not for cart parts scattered all over a battlefied.

10.   Amber Carpenter (2015, 14–15) makes the same point about people being clearly presupposed in explanations that turn on human ends.

11.   See Siderits 2007, 49. “‘I’ is what they call a ‘convenient designator,’ a word that refers to something that is just a useful fiction. The person is that useful fiction. The person is a whole made of parts. And wholes are not themselves real things, only the parts are. I think that ‘I’ must refer to one and the same thing every time I use it because I have forgotten that the person is a useful fiction. I have forgotten that ‘I’ is just a useful way to talk about all the parts taken together.”

12.   See, e.g., Siderits and Katsura 2013, 23.

13.   Siderits 2003, 7.

14.   Siderits 2003, 14.

15.   Indeed Guy Newland and I made that conceptual slide in an earlier publication entitled “An Introduction to Conventional Truth.” See the Cowherds 2011, 12: “Which wholes are created and focused upon and which are not is largely a matter of tacit common agreements about what is useful to the human form of life.”

16.   Y. Karunadasa has done admirable studies on the Theravāda Abhidhamma notions in which he shows in detail that common agreement or consensus is essential to sammuti and paññati but never suggests that the Buddhist claims that people use reasons of interest or usefulness to settle upon one particular agreement rather than another. See, e.g., Karunadasa 1996.

17.   See Newland and Tillemans 2011.

18.   The Vaibhāṣika schools have a bewildering variety of causal accounts of how karma works to create the world. See, e.g., Bareau 1955. In the Mahāyāna, Yogācāra’s causal account of the world’s genesis typically turns on the ālayavijñāna (storehouse consciousness), in which karmic seeds and tendencies are “planted” and come to maturity — this is part and parcel of their idealism. In Madhyamakāvatāra 6.89 and Bhāṣya, Candrakīrti interprets the sūtra passages that state that all is mind alone (cittamātra) to be expressing the Abhidharma principle that everything in the world, be it animate or inanimate, is caused by karma, which is itself dependent upon mental factors. What is important for our purposes is that in such accounts, considerations of utility do not figure at all to explain why we have the world and the sentient beings we do. They are purely causal. Note, finally, that the Buddhist Epistemologists’ theory of apoha, especially in its Dharmakīrtian version, is more complicated in its emphasis on human interests in the construction of common properties and kinds (see Dunne 2011a). Dharmakīrti’s Yogācāra maintains that various inherited imprints/tendencies (vāsanā) and other causal processes lead us to understand kinds as we do. Does his apohavāda then also maintain that we somehow choose how to understand them or interpret them in the ways we do for various reasons, and that in principle those choices could always have been radically different, perhaps yielding weird sum individuals and funny composite kinds like GRUE (i.e., things observed to be green before time t or not so observed before t and blue. See N. Goodman 1983)? Functions are no doubt important in defining kinds for Dharmakīrti and avoiding realism about universals; it is not clear to me how much liberty he leaves us to choose for our reasons.

19.   To get an idea of the philosophical work typically done by the term anādivāsanā from Dharmakīrti to Klong chen pa, see p. 123–24n12, and p. 198n27.

20.   There are a considerable number of attempts to apply evolutionary games theory to cultural phenomena, from altruism, senses of fairness, to social contracts. See Alexander 2009.

21.   See, e.g., McDowell 1998, 277: “Readers of Wittgenstein often suppose that when he mentions customs, forms of life, and the like, he is making programmatic gestures towards a certain style of positive philosophy: one that purports to make room for talk of meaning and understanding, in the face of supposedly genuine obstacles, by locating such talk in a context of human interactions conceived as describable otherwise than in terms of meaning or understanding. But there is no reason to credit Wittgenstein with any sympathy for this style of philosophy.”

22.   Cf. Philosophical Investigations §201: “It can be seen that there is a misunderstanding here from the mere fact that in the course of our argument we give one interpretation after another; as if each one contented us at least for a moment, until we thought of yet another standing behind it. What this shews is that there is a way of grasping a rule which is not an interpretation, but which is exhibited in what we call ‘obeying the rule’ and ‘going against it’ in actual cases.” See also chapter 1 on Wittgenstein’s “paradoxes” that arise when we interpret behavior as in conformity with such and such a rule.

23.   Cf. Philosophical Investigations §226: “What has to be accepted, the given, is — so one could say — forms of life.” Ibid. §217. “I have exhausted the justifications … my spade is turned.” Finally, cf. Remarks on the Foundations of Mathematics VI–31 (cited and discussed in McDowell 1998, 242): “The difficult thing here is not to dig down to the ground; no, it is to recognize the ground that lies before us as the ground.”

24.   See chapter 1, page 23. A quietistic interpretation of the Madhyamaka stance about the world’s things existing customarily (vyavahārasat) is not that thought, language, and human transactions actually create them but that one cannot intelligibly come up with any undergirding for them that would be radically outside, or beyond, thought, language, and transactions. Those two interpretations are not the same. To take a provocative Tibetan formulation: there is no other way of establishing things (’jog byed gzhan med pa) besides, or beyond, our vyavahāra. See chapter 11, page 209–10 on the differences between an antirealist and a quietistic interpretation of the Madhyamaka idea that things are what they are because of, or in a context of, thought, language, and human transactions.

25.   Cf. Remarks on the Foundations of Mathematics VI–35 (cited in McDowell 1998, 240): “How do I know that the colour that I am now seeing is called ‘green’? … If I am drowning and I shout ‘Help!,’ how do I know what the word Help means? Well, that’s how I react in this situation. — Now that is how I know what ‘green’ means as well and also know how I have to follow the rule in the particular case.”