Slavery and the Making of Early American Libraries: British Literature, Political Thought, and the Transatlantic Book Trade, 1731-1814 1st Edition Sean D. Moore
Illuminating the Mind: An Introduction to Buddhist Epistemology
Jonathan Stoltz
Buddhist Ethics: A Philosophical Exploration
Jay L. Garfield
How Things Are: An Introduction to Buddhist Metaphysics
Mark Siderits
How Things Are
An Introduction to Buddhist Metaphysics
MARK SIDERITS
3
Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and certain other countries.
Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above.
You must not circulate this work in any other form and you must impose this same condition on any acquirer.
Library of Congress Cataloging-in-Publication Data
Names: Siderits, Mark, 1946– author.
Title: How things are : an introduction to Buddhist metaphysics / Mark Siderits. Description: New York, NY, United States of America : Oxford University Press, 2022. | Series: Buddhist philosophy for philosophers | Includes bibliographical references and index.
Identifiers: LCCN 2021030120 (print) | LCCN 2021030121 (ebook) | ISBN 9780197606919 (paperback) | ISBN 9780197606902 (hardback) | ISBN 9780197606933 (epub)
Subjects: LCSH: Buddhist philosophy. | Buddhism—Relations—Hinduism. | Hinduism—Relations—Buddhism. | Philosophy—Comparative. | Vasubandhu. | Dharmakīrti, active 7th century. | Nāgārjuna, active 2nd century. Classification: LCC B162 .S533 2021 (print) | LCC B162 (ebook) | DDC 181/.043—dc23 LC record available at https://lccn.loc.gov/2021030120 LC ebook record available at https://lccn.loc.gov/2021030121
Paperback printed by LSC Communications, United States of America
Hardback printed by Bridgeport National Bindery, Inc., United States of America
1 Introduction
The Buddhist Metaphysical Landscape
1.1 Buddhism as Philosophy
Not long ago, most philosophers thought it unlikely that there could be much of philosophical interest in what the Buddhist tradition has to say about matters metaphysical. That situation has begun to change. This book tries to address the growing interest, among philosophers trained in the western tradition, in the metaphysical theories of Indian Buddhist thought. The present chapter introduces Buddhist metaphysical theorizing, locating it within the larger projects of Buddhist and classical Indian thought, and it details the approach that will be used in presenting Indian Buddhist philosophy, as well as the motivation behind that approach.
There are still many who are skeptical that Buddhism contains much by way of serious metaphysics. Buddhism is, after all, a religious tradition, and the business of a religion is to help people achieve what is, by its lights, the ideal state of existence. It is now fairly widely known that the doctrine of non-self is a core Buddhist teaching. And the view that there is no self might look like the sort of claim that falls within the scope of ontology. But might the teaching of non-self not instead be a Buddhist precept or faithcommitment, something that one is simply instructed to accept in order to progress toward the state of nirvāna? Is there any reason to think that this or any other Buddhist claim about the fundamental structure of reality is based on argumentation and analysis? Is there reason to think that such claims are not justified merely on the grounds that the Buddha taught them, or on the strictly pragmatic grounds that believing them will have good effects for the believer? For that matter, is there reason to think that such claims are any more than mere speculation, akin to the metaphysical claims found in what we have of pre-Socratic thought?
There are some who would counter this last doubt by insisting that Buddhists eschew metaphysical speculation as mere conceptual wrangling,
something that distracts one from the urgent task of attaining nirvāna. But as we will see soon enough, there is ample evidence against this “Buddhist pragmatist” line of interpretation. Buddhist thinkers did indeed engage in serious and sustained efforts to work out what most fundamentally exists; and their use of the tools of argumentation and conceptual analysis will demonstrate that they did so out of concern to determine how things are anyway, independently of our interests and cognitive limitations. That their projects arose in a soteriological context and with determinately soteric aims should not stand in the way of their efforts being understood as genuinely philosophical. After all, the fact that ancient Greek philosophizing was done in pursuit of finding a path to eudaimonia has not counted against reading the thought of Plato and Aristotle as serious metaphysics and epistemology. The operating assumption here will be that Buddhist thinkers developed and defended their views about the nature of reality out of the conviction that liberation from saṃsāra (the round of rebirths and the suffering that it entails) requires overcoming our ignorance about the fundamental structure of reality, and that philosophical rationality has an important role to play in dispelling that ignorance. Was this the stance of the Buddha himself? Given the state of the textual evidence, we may not be able to answer that question. But it is abundantly clear that this was the stance of the Buddhist thinkers who systematized and developed the Buddha’s teachings. They did not think that philosophical rationality alone would carry one all the way to nirvāna. But the Buddhist thinkers whose views we will examine did take the systematic and rigorous examination of core metaphysical issues to be at least a necessary component of the path to liberation. To see why this was so, we need to look briefly at the core Buddhist project as articulated in the Buddha’s teachings. And this in turn requires us to look (even more briefly) at the Indian context in which those teachings were developed.
The scholarly consensus is that the Buddha (Gautama) lived and taught in the fifth century BCE. This seems to have been a time of some intellectual ferment in India. The Buddha was one of a number of wandering ascetics who sought an alternative to conventional views about the good for humans. One element behind their dissatisfaction with contemporary conceptions of the good life was the newly developed karma-rebirth ideology, according to which at death one undergoes rebirth in some sentient form (human, nonhuman animal, god, or inhabitant of one of the hells). The sort of rebirth an individual will receive is determined in accordance with the moral character of their actions in the life then ending. Moreover, while the prospect
of rebirth as a god, or even as a human of high social status, might seem appealing, it would be just one more step in the beginningless and seemingly endless cycle known as saṃsāra. Rebirth meant perpetual redeath, inspiring at best cosmic ennui. Conventionally prescribed modes of life require performance of various intentional actions, and it is intentional actions that set the stage for rebirth: at the time of death one has typically not experienced all the pleasurable and painful karmic fruits of the good and bad actions performed over the course of this life, so a new life is required to balance the ledger. Only liberation from this cycle could, it was thought, bring about real, lasting satisfaction.
A common response to the dilemma posed by karma was to see conventional views about how best to live one’s life as based on a mistaken theory of what one truly is. Bondage to the wheel of saṃsāra is, on this view, the result of misidentification: identifying with the sort of thing that might be made better off by attainment of one of the three conventionally prescribed goals—sensual pleasure, wealth and power, or honor and virtue. The obvious next step is to suppose that the true self is something quite distinct from the psychophysical elements making up the empirically observable body, senses, and mind. This line of thought led to the variety of egological theories one finds in the later Indian philosophical landscape, including such “orthodox” or Vedic schools as Sāṃkhya, Nyāya, and Advaita Vedānta, as well as the “heterodox” Jain school. The Buddha, though, rejected this approach in favor of a non-egological view. On his analysis, the cause of recurrent rebirth is not identifying with the wrong thing; our mistake lies in identification as such. The ubiquitous “I”-sense—the sense each of us has of being something with a distinctive first-person perspective—is a delusion. Having supposedly confirmed in experience that he had indeed attained liberation from saṃsāra, he proceeded to teach his findings to others.
It is a moot point whether the Buddha saw his teachings as comprising a full-fledged philosophical system. Those teachings do, however, involve both explicit and implicit commitments to a variety of metaphysical positions. There are, for instance, explicit claims to the effect that there is no self, that all existents are impermanent, and that impermanent entities originate in dependence on causes and conditions. And at least implicit in these teachings is the view that person-involving descriptions of states of affairs are always reducible to descriptions involving only thoroughly impersonal entities and events. As these teachings spread across South Asia after the death of the Buddha, efforts were made to clarify and systematize them. The beginnings
of systematic Buddhist philosophizing grew out of these efforts. First one sees attempts to adjudicate disputes over rival interpretations or classificatory schemes developed by different Buddhist communities. But because the interlocutors in these disputes are all Buddhists, their disagreements are often treated as wholly hermeneutical in character—to be settled by appeal to things supposedly said by the Buddha.1 In time the conversation expanded to include non-Buddhist disputants. And in those later debates, neither side could resort to appeal to authoritative texts: Buddhists do not recognize the Vedas as authoritative, and non-Buddhists do not ascribe evidentiary value to the alleged testimony of the Buddha found in Buddhist sūtras. The epistemological systems of classical Indian philosophy grow out of attempts to develop a common set of rules for conducting debates of this sort. And likewise for systematic metaphysics: Buddhists and non-Buddhists alike begin to work out and defend answers to questions that the founders of their systems may not have addressed.
1.2 Two Non-Buddhist Systems
Before sketching the trajectory that systematic metaphysics followed in the Buddhist tradition, I want to first introduce two non-Buddhist systems that played significant roles in that history. These are Sāṃkhya and Nyāya, both of them “orthodox,” that is, schools that accept the Vedas as authoritative texts. But these schools represent very different ways of trying to develop an egological response to the dilemma posed by the karma-rebirth ideology and the pursuit of liberation from saṃsāra. Both seek to show that the self is something that is distinct from the empirically available constituents of the psychophysical complex. This would make the error allegedly behind continued rebirth comprehensible: we take ourselves to be the sort of thing that is observable when in fact the true self is not. But the metaphysical schemes employed to develop this insight are quite different. We start with that of the older of the two schools, Sāṃkhya.
At its most basic, Sāṃkhya metaphysics is a kind of dualism, that of what are called puruṣa and prakṛti. In one respect the dualism here is a familiar one, with puruṣa standing on the side of the experiencing subject and prakṛti on
1 The Buddha’s teachings were not committed to writing until several centuries after his death, and were transmitted in a variety of different regional dialects. This may help account for some of the differences among South Asian Buddhist communities concerning the contents of those teachings.
the object side.2 This is not, however, a substance dualism. First of all, prakṛti is the general term for twenty-four kinds of stuff: one that is completely indeterminate, having no manifest properties whatsoever; and twenty-three more kinds, each of which manifests distinctive properties detectable by the external senses and the inner sense. Second, each of these kinds of stuff is wholly constituted by a combination of three elementary tropes, sattva, rajas, and tamas. These are elementary in the sense that they are not analyzable into combinations of yet simpler properties. And based on the assumption that all coming into and going out of existence involve rearrangement of parts, it is held that the three elementary tropes are eternal. Observable change is the result of rearrangement of triplets of tropes. The unmanifest form of prakṛti occurs when the three elementary tropes are in equilibrium. The nature of any one of the twenty-three manifest forms is explained by relations of dominance and subordination among the three. Change is thus to be understood in terms of the notion of evolution: one sort of stuff emerging from another due to rearrangement of the underlying constituent tropes.
In one respect this scheme looks like the sort of atomism we find in Democritus, or in the theory of elements developed by the Indian heterodox materialist school of Cārvāka. Classical atomism also explains observable change in macrophysical objects in terms of rearrangements of more basic constituents. But in that variety of atomism, the fundamental constituents are substances: atoms of earth, air, water, and fire. (Cārvāka rejects atoms as mere unobservable posits, but also explains change in terms of rearrangement of the elements.) Sāṃkhya metaphysics also has a place for atoms, but their place is to be four of the twenty-three manifest forms of prakṛti; they are not to be found at the level of fundamental constituents. Sāṃkhya thus explains change as evolution or transformation, and this leads to its theory of causation as the “existent effect” theory (satkāryavāda) or inherentism. On their account, the effect exists in unmanifest form in its cause: the fire already exists in the wood that will be its fuel. This account of causation will become a stock target of Buddhist critique. In time, though, one Buddhist school will come to embrace something similar—except that it will employ a levels distinction to answer the objections raised to the Sāṃkhya formulation.
2 The term puruṣa is what Sāṃkhya uses in place of the more common ātman. Both terms are used to denote whatever plays the role of essence of the person. When Buddhists deny that there is an ātman, their denial extends to the Sāṃkhyan puruṣa as well.
The point of Sāṃkhya’s dualism is to try to establish the existence of an experiencing subject as something utterly distinct from the psychophysical complex as given in experience. Their strategy is to argue that while the theory of prakṛti and its evolution accounts for the contents of our experience, certain facts are otherwise inexplicable if we do not posit something extra to serve as a pure witnessing subject. Two of their arguments are of some interest here. These are the argument from experience and the argument from control. Both are reductios on the supposition that nothing exists besides the twenty-four forms of prakṛti. And both employ a principle of irreflexivity, according to which an entity cannot operate on itself: a knife blade cannot cut itself, a fingertip cannot touch itself, etc. The argument from experience begins with the premise that all forms of prakṛti are possible objects of experience.3 But experience always has a subject-object structure: there is the content of the experience, and then there is that to which this content is presented. If every form of prakṛti can be an object of experience, and an entity cannot operate on itself, then no form of prakṛti can be that to which the different forms of prakṛti are presented. It is then said to follow that whatever plays the role of subject of experience is something other than prakṛti.
There are two noteworthy points here. First, what the argument purports to establish is a self not as a substance with experiencing as its essence, but as the enduring occurrence of an experience trope. A substance is necessarily complex, insofar as it consists in a substrate plus essential and contingent properties. And just as Sāṃkhya holds that any existent that is genuinely simple must be eternal, so they maintain that complex entities must be impermanent. Second, puruṣa does nothing, it merely indifferently illuminates whatever happens to occur on the prakṛti side of the puruṣa-prakṛti complex that is a person. The self is no agent but only a witness, on this account. Taking it to be both experiencer and agent is the mistake that fuels the karma-rebirth machinery and keeps us bound to saṃsāra. For it is by thinking of the “I” as being of the nature of prakṛti that one can take things like sensual pleasure or material possessions as suitable candidates for final ends in life. Liberation is achieved by coming to see mere witnessing as the true “me.”
The second, controller argument is like the first in structure. Here the relevant property shared by all forms of prakṛti is not that of being an object of
3 The unmanifest form of prakṛti is never given in sensory experience; Sāṃkhya claims that it is cognized through inference. Its constituent tropes are, however, perceived, namely when they occur in one of the evolved manifest forms.
experience, but rather the property of being subject to control. The premise is that one can evaluate and seek to change any of the forms of prakṛti. And so, once again, by the principle of irreflexivity it is concluded that there must be more to the person than the psychophysical elements. Whatever exercises control over all the forms cannot itself be such a form. There must be more to us than the constituents of the psychophysical complex.
Given what was just said about puruṣa being no more than pure sentience, however, the strategy of the controller argument may seem surprising. A controller is, after all, active, and puruṣa is said to be inactive and changeless. The Sāṃkhyan response is that the self may be said to be a controller in the same sense in which a king may be said to win a battle. The king may merely watch from a neighboring hilltop while the kingdom’s soldiers do the actual fighting. Victory (or defeat) is nonetheless attributed to the king. This is presumably because it is the kingdom’s army that fought: without a king there would be no such army. Attribution of the role of controlling to the self is said to be similarly figurative. While Sāṃkhya’s self does not issue commands or initiate movement, its presence as witness is crucial to the function of control. The prakṛti side of the person is active but insentient, and thus not itself capable of exercising intelligent control.
Peter Strawson’s distinction between descriptive and revisionary metaphysics (Strawson 1959, xiii–xv) can be used to lay out a sort of spectrum. At one end will be metaphysical schemes that are meant to merely systematize the (implicit) commitments of common sense (our “folk ontology”).4 Then there will be schemes that require significant departures from those commitments. The more significant the departures, the less descriptive and the more revisionary the metaphysics. Of course, it is notoriously difficult to pin down just what counts as common sense (which may also be a moving target). And there is unlikely to be consensus on a metric with which to measure the significance of departures from our folk ontology. Still the notion of such a spectrum can be useful in discussing metaphysical theories and schemes. Thus we can judge the scheme of Sāṃkhya to be relatively revisionary as compared with the second system we shall briefly look at, that of Nyāya. One way of reaching this judgment is to count the number
4 It can be questioned whether there actually is such a thing as a folk ontology. It might be said instead that questions of ontology only arise when we begin to seriously reflect on our cognitive and linguistic practices—when we begin to philosophize, which is arguably not a folk practice. This may well be true. If so, then by “folk ontology” is meant the ontology that results when we first set out to regiment our “natural ontological attitude,” and before the many problems that beset such a task have begun to emerge. For a sensitive discussion of this issue see Horgan and Potrč 2006: 7–14.
of categories employed in the two schemes. For Sāṃkhya there seem to be just two: that of elementary trope (the three making up prakṛti and the one an occurrence of which is a puruṣa) and that of compound substance (the 24 forms, the person as psychophysical complex plus puruṣa, etc.). Nyāya has altogether seven categories. Why so many more?
The Nyāya categories are substance, quality, motion, universal, inherence, individuator, and absence.5 Suppose you correctly judge there to be a white cow standing alone in a field swishing its tail. The state of affairs that is the correctness condition for this judgment has a number of constituents, but how many? The descriptive metaphysician takes it that we can answer this question by reading off the structure of the judgment itself. The cow is, of course, a substance, the possessor of various properties. Among those properties is the quality of white, understood as a trope occurring in this cow. Another is the action of swishing, the moving of its tail. Still another is that of being a cow. But this property is one that our cow shares with the many other substances that are cows. Its white color and its action of swishing are particulars whose occurrences are specific to this cow, but its being a cow involves a universal, cowness, that is equally present in all other cows. All three properties are alike, however, in bearing the same relation to this cow, that of inherence or being in. The white color, the swishing of the tail, and cowness all inhere in the cow. (There are also universals for qualities and actions: whiteness inhering in this and other occurrences of white color, downward motion-ness in the swishing, etc.) As for what makes it the case that the cow is alone in the field, this is the absence of other cows, something that is also located in the field. This absence has as counter-positive other cows existing elsewhere. Among those other cows there may be some that are qualitatively identical to the cow present in the field, but our cow is nonetheless distinct from them. The category of individuator is invoked to account for this fact, but what makes each cow be a distinct individual is not an individuator inhering in it. It is rather the individuators inhering in the many atoms that compose each cow. All the atoms of a given element are qualitatively identical; their mutual distinctness is explained by their each being inhered in by its own individuator. It is these that in turn account for
5 Absences were first posited by Nyāya’s sister school Vaiśeṣika, and only later included in Nyāya’s ontology. For our purposes we may ignore the doctrinal/historical differences between the two schools.
the individuation of qualitatively identical macro-substances composed of distinct atoms.
Nyāya distinguishes between two kinds of substance, eternal and noneternal. Non-eternal substances are composite, being composed ultimately of simple (and hence eternal) substances. In the case of macro-physical objects like pots, the ultimate components are atoms. Interestingly, while a given pot can survive qualitative change (as when it darkens due to ageing), it cannot survive replacement of parts; when the handle is removed and replaced by a new one, the original pot goes out of existence and a new pot comes into existence. Sorites difficulties are thereby avoided.
Included among the eternal substances are selves. These are typically associated with the body of a sentient being, but this association may be brought to an end by achieving liberation from saṃsāra. Among the qualities distinctive of selves are cognition, effort (volition), and pleasure and pain. The self thus serves as both subject of experience and agent of actions. But because no quality is deemed essential to the existence of a substance—only the continued inherence of the universal responsible for its being of its kind is considered necessary for the persistence of a particular substance—the self can exist in the absence of experience and agency.
Some Naiyāyikas held that the self is known to exist by means of perception. The idea here seems to have been that one perceives a substance by perceiving qualities or actions inhering in it. But the more common strategy for establishing the existence of a self employs inferences that appeal to its alleged role as synthesizer. Take for instance the case of seeing a mango and then grasping it. There are two separate sensory cognitions here, and they involve different sense modalities operating at distinct times: first vision and then touch. Yet we judge that we now touch what we first saw. Set aside for the moment the question why we take the object seen in the past to persist to the present moment of grasping. (Buddhists will question the assumption that there are persisting substances.) The question needing an answer here is how information from one sense-modality is integrated with information from another if there is no entity to which both report. A similar question is said to arise with respect to the occurrence of desire after seeing something associated with earlier pleasure. In general, Nyāya claims, the facts about experience and agency require that we posit a persisting self that can serve as subject of different experiences at different times, and as agent of actions whose fruits are enjoyed subsequently.
1.3 The Buddhist Soteriological Project
Both Sāṃkhya and Nyāya claim that the ideal state of liberation is attained by ceasing to identify with transitory things like the psychophysical complex or its states and possessions, such as pleasure, repute, and material things; one should instead identify with what one truly is, an eternal self. The Buddha claimed instead that the dissatisfactory state of being in saṃsāra results from identifying with anything at all. It is the “I”-sense, the sense that there is a persisting subject of experience and agent of actions, that leads not only to continued rebirth but also to existential suffering in each life. Our mortality ultimately undermines any quest for eudaimonia organized around the notion of an individual subjectivity. The first-person perspective is, the Buddha claims, illusory. What we think of as an enduring person is in fact nothing more than a causal series of impermanent, impersonal psychophysical elements. And it is ignorance of this fact that is the root cause of existential suffering.
All this is laid out in the Buddha’s core teaching of the Four Nobles’ Truths:6 that there is suffering, that suffering has a cause, that there is the cessation of suffering, and that there is a path whereby the cessation of suffering can be attained. The second of these is spelled out in terms of a chain of twelve causal links beginning with ignorance and ending in old age and death. This is usually understood to represent a sequence of three lives, so that the sequence serves to illustrate how ignorance of the facts concerning sentient existence fuels karma and rebirth. Be that as it may, a key transition in the middle of the sequence has events that culminate in desire serving as cause of something called appropriation, the attitude of considering elements in the causal series as “me” or “mine.” This, plus the fact that the links making up the sequence are described in thoroughly impersonal terms, brings home the point that deployment of a first-person perspective is understood as a remediable error. The fourth of the Four Nobles’ Truths describes the path prescribed to remedy the error and attain the cessation of suffering.
This path is said to consist of eight components, but a broader three-fold division better suits our purpose here: a division into morality, meditation, and wisdom. The first of these specifies the modes of conduct said to be best
6 This doctrine has been more commonly called the Four Noble Truths. But what are said to be noble (ārya) are not the four truths but those who realize them. In Buddhist texts, an ārya is someone who has attained liberation. The doctrine concerns what one must supposedly come to know in order to join the ranks of the āryas.
suited to progress toward cessation. Buddhists should, for instance, refrain from killing sentient beings. The second consists of techniques designed to extirpate those mental habits that perpetuate the “I”-sense: the reactive attitudes of desire and aversion, as well as core mistakes in our everyday conception of the world. Meditational practices in turn include something akin to mindfulness meditation (said to help one observe more clearly the workings of the psychophysical complex), but also exercises designed to foster such virtues as compassion and equanimity. The third of the three, the wisdom component of the path, is devoted to mastery of a correct conception of the nature of reality. This is where the philosophical practices of analysis and argumentation play a role. In this the practitioner emulates the Buddha, who gave arguments for at least some of his central teachings (such as nonself). But it is also understood that to do otherwise—to turn adherence to the central tenets of the Buddha’s teachings into a matter of faith commitment— is to risk making the realization of non-self unachievable. The faith stance of “Here I stand, I can do no other” involves at least implicit affirmation of the “I”-sense. Better that the practitioner arrive at the core convictions of the Buddhist project through means less likely to appeal to the (self-affirming) desire for a life of meaning and significance.
The three parts of the path just described are not seen as occurring in a prescribed sequence. The basic rules of morality are understood as a sort of entry-point for the lay Buddhist, since they tend to counter some of the affective habits sustaining the “defilements” of desire, aversion, and delusion. But deliberate entry onto the path (ordination as a monastic) brings with it a new and more stringent set of moral rules, adherence to which is made easier by various types of meditation. The mastery of philosophical theories also equips the meditator with the conceptual tools necessary to make sense of the altered states of consciousness produced by certain forms of meditation. At the same time, the careful observation of mental states cultivated in some types of meditational practice is said to provide empirical evidence in support of key philosophical theses. The cultivation of wisdom is but one component of Buddhist practice, and it functions in mutual interaction with the other two. It is the philosophical content of this component, though, that shall be our main concern here; in the succeeding chapters we shall have little to say about the soteriological roles of the various Buddhist metaphysical theories we present and assess.
1.4 Key Schools and Figures
Indian Buddhist metaphysics is far from monolithic. Buddhist philosophers engaged in sustained disputes not only with their non-Buddhist rivals, but with one another as well. Since it is not our aim to present a history of Indian Buddhist metaphysics, we shall not discuss these disputes in great detail.7 Our chief focus here will be on the arguments and analyses that we think will be of interest to analytic metaphysicians. Here, though, is a brief outline of the main trends of Indian Buddhist philosophy.
Its history can be divided into three major phases. The first, Early Buddhism, consists of the teachings of the Buddha and his immediate disciples. These teachings are to be found in the sūtras making up the Nikāyas of the Pāli Canon and the Āgamas found in Chinese translation.8 The second phase consists of the many schools making up what we may call the Abhidharma movement. These schools arose out of efforts to classify the many technical terms found in the early Buddhist sūtras, and to address questions of interpretation as these arose. The third phase, Mahāyāna, grows out of the appearance of new sūtras (some ostensibly the utterances of the historical Buddha, others purporting to be the teachings of other buddhas and bodhisattvas), beginning around the first century BCE. These sūtras articulate doctrines not found in the Nikāyas or Āgamas, and two major philosophical movements arose as efforts to explain and defend these novel doctrines: Madhyamaka and Yogācāra. These, it should be noted, did not supplant the older Abhidharma schools, many of which continued to flourish alongside their upstart competitors.9
The proliferation of schools and sub-schools in the Buddhist philosophical tradition grows out of a base of shared commitments. Here are some important areas of agreement among all Indian Buddhist philosophers.
Non-self: all agree that there is no entity that persists over the life (or lives) of a person; none of the empirically given parts of the person endures,
7 For details of this history see Westerhoff 2018.
8 In Pāli, the sūtras are called suttas. Throughout this work, when referring to Pāli materials we shall use the Sanskrit equivalents of any technical terms mentioned.
9 Indeed one Abhidharma school, Theravāda, remains active today in Śri Lanka, parts of Southeast Asia, and the west. (Buddhism died out in the rest of South Asia in the wake of Turkic incursions in the twelfth century.) Because Theravāda was less active in debates with rival Buddhist and nonBuddhist schools, it is more useful as a resource that helps fill in gaps in early Abhidharma history than as a source of novel philosophical theories and arguments.
and there is no sound argument for the existence of a transcendent self; all that is to be found is a causal series of impersonal, impermanent psychophysical elements.
Momentariness: no existent endures for more than an instant; this is a significant strengthening of the Buddha’s teachings, since he claimed only that all existents are impermanent (i.e., non-eternal); while there was disagreement as to whether moments have any temporal thickness, all schools accepted at least one of the arguments developed for the conclusion that everything lasts just one moment.
Mereological nihilism: there are, strictly speaking, no mereological sums; the composite objects of our folk ontology, as well as any composite objects posited by rival metaphysical theories, are at best merely nominally real.
Anti-substance: not only are there no composite objects (which are ruled out by mereological nihilism); there are also, strictly speaking, no simple substances (such as atoms or selves) either; substances, understood as the persisting bearers of properties, are seen as conceptual constructions deployed through a bundling process that serves the interest of ease of communication.
Nominalism: universals are rejected on two grounds: as permanent entities they would be devoid of causal efficacy, and causal efficacy is the hallmark of the real; and there is no plausible account of how the inherence relation might work; in time, what begins as a kind of “ostrich nominalism” gives way to a more radical nominalism driven by the realization that in the absence of universals there can be no real similarities either.
As this list makes clear, the Buddhist metaphysical enterprise stands at the revisionary end of the spectrum. Buddhist philosophers are generally unimpressed by appeals to our intuitions, or to our customary ways of talking. Theory construction is to be guided by “lightness” or parsimony, and should involve evidence available to all parties. As might be expected, the results are ontologies on the decidedly spartan side. In the following chapters we spell out some of the principal arguments for these conclusions and explore their more significant consequences. But first we must introduce some of the major players in the debates that ensued.
We may divide the Buddhist metaphysical landscape into four stances with respect to different varieties of realism: two sorts of dualist realism,
one external-world anti-realism, and one global anti-realism. By “dualist realism” is here meant the view that mental and physical phenomena are of distinct ontological sorts, both equally real. Buddhist dualists will not, though, espouse the more familiar sort of substance dualism. Instead their ontology will consist of two kinds of trope, external or physical, and inner or mental.10 Buddhist dualist realists divide over the question of how sensory cognition of external objects works: directly, or indirectly via representations. The Vaibhāṣika school of Abhidharma espouses a direct realist view, which they support using the doctrine for which the school is best known, its eternalism. Its representationalist rival, Sautrāntika, maintains that a sensory cognition’s veridicality consists in its bearing a form that resembles the object grasped by the sense faculty. Sautrāntika likewise rejects Vaibhāṣika’s eternalism in favor of presentism. And just as in early modern philosophy the representationalism of Descartes and Locke leads to Berkeley’s subjective idealism, so the Sautrāntika representationalist view of perception is followed by the Buddhist idealism (hence “subjectless idealism”) of the Yogācāra school of Mahāyāna. Yogācāra’s chief rival is the other school of the Mahāyāna movement, Madhyamaka. At the heart of the disagreement between Yogācāra and Mādhyamaka is how to interpret the teaching, found in many Mahāyāna sūtras, that all dharmas (ultimately real entities) are “empty” or devoid of intrinsic nature. Yogācāra takes this to mean that reality should be understood as devoid of the subject-object dichotomy that results from acquiescing in the appearance of sensory cognition to represent a distinct external world. Madhyamaka takes the emptiness claim in a far more radical way, as rejecting the very idea of ultimately real entities. To Yogācāra, Madhyamaka’s global anti-realism seems tantamount to a self-refuting metaphysical nihilism. Madhyamaka retorts that Yogācāra’s local external-world anti-realism is a half-hearted halfway measure: if there is nothing external, then the mental realm would seem to be equally dispensable. The result is not (as Yogācāra would claim) that the world consists of ineffable particulars, but rather that the very question of how the world is anyway is of questionable intelligibility. The reader cannot be expected to retain all relevant information from this whirlwind tour of the main schools of Buddhist metaphysics. Hence a
10 This is in line with the list of five kinds of aggregates (skandhas) of psychophysical elements that the Buddha used in his arguments for non-self. The first of these, rūpa, covers the corporeal side of sentient existence, as well as inanimate matter. The other four are collectively referred to as nāma (literally “name”). In secondary literature in English the term rūpa is often translated literally as “form”; this skandha was likely given this name on the grounds that only the physical has shape. Mental elements, by contrast, can only be named.
word of advice: bookmark this page, and return to it as needed. But while we are at it, here are a few more names that will crop up from time to time, in this case not of schools but of individual philosophers. Saṃghabhadra (fifth century CE) is a Vaibhāṣika with important things to say in defense of the school’s eternalism. Vasubandhu (c. 350–430) has a foot in two camps. As a Sautrāntika he develops that school’s definitive responses to any number of Vaibhāṣika views, and his sophisticated formulation and defense of reductionism about persons and personal identity are among the clearest articulations of that position. But he also helps found the Yogācāra school, and his later work gives some of the principal arguments in defense of that school’s idealism. Nāgārjuna (second century CE) is the author of the foundational text of the Madhyamaka school. Dignāga (c. 480–540) is founder of the hybrid Yogācāra-Sautrāntika school of Buddhist epistemology, and Dharmakīrti (c. 550–610) is his principal commentator. Best known for their development of a Buddhist theory of epistemic instruments, Dignāga and Dharmakīrti also make important contributions to the philosophy of mind.
Our aim here is to explore a variety of metaphysical theories developed in the Indian Buddhist tradition. Our interest lies chiefly in the theories themselves, and not in the history of their development. There will consequently be relatively little by way of explication of the texts in which these theories are propounded. There will be the occasional quotation of a passage from a relevant text, but more to illustrate the spirit of the Buddhist philosophical enterprise than to support an interpretation. Where there are reliable English translations of primary sources relevant to the topic at hand, references to these will be provided. There will also be references given to the Sanskrit sources, but these are provided only for the convenience of other scholars of the Buddhist tradition. The use of Sanskrit technical terms will be kept to a minimum. I shall, though, provide the Sanskrit when I first introduce the English expression I treat as its equivalent. There is no scholarly consensus on how to render many of the key terms in the Sanskrit philosophical corpus. (Indeed in the more specialized literature of philologically trained Buddhologists, important terms are often left untranslated.) By providing the Sanskrit original of my translation of a given technical term, I hope to make it possible for the non-specialist reader to continue their exploration of a topic through secondary literature that may translate those terms differently.
This work is intended chiefly for philosophers who lack specialist knowledge of the Indian Buddhist tradition. Such readers may find it somewhat disconcerting to learn that some of the interpretations of that tradition
presented here are controversial among the community of specialist scholars. For the most part I shall here avoid discussing these controversies and defending my own interpretations; that has been done elsewhere.11 Suffice it to say that the readings of the Buddhist tradition presented here are at least defensible, and that they have been arrived at through a decades-long process of trying to develop the sort of stereoscopic vision called for in the enterprise of “fusion” or “confluence” philosophy. There are those who worry that this enterprise serves the interests of a sort of intellectual neo-colonialism.12 This is an important matter. My own thoughts are developed elsewhere.13
There has been much talk in recent years about expanding the canon, incorporating works of non-Western philosophy into the curriculum, and ending the pretense that only those of European ancestry have produced work of sufficient rigor to deserve the name “philosophy.” Proponents of such expansion sometimes appeal to considerations of justice and equity. But there are also those who claim that by ignoring work done outside the European tradition, philosophers are depriving themselves of a treasure trove of intellectual riches from which they could derive enormous benefit. This claim may be exaggerated. It is true that the practice of philosophy seems to have been enriched when historically distinct philosophical traditions came into close contact. The case of medieval Spain, which witnessed a sustained dialogue among Islamic, Jewish, and Christian philosophers, is one corroborative instance. Another is classical India between the fifth and tenth centuries CE. Still it could be argued that the sheer number of well-trained philosophers who are active today virtually guarantees that fresh insights will emerge even if the canon retains its Eurocentric bias. Even if we grant that others outside the European tradition did rigorous work in metaphysics, it could be that any insights we might gain by becoming more familiar with their work would have become available in any event, given the propensity of younger philosophers to challenge orthodoxies, and given just how many such philosophers are active today.
My claim will be more modest. Much of the metaphysical theorizing we examine here will have a familiar cast to those familiar with recent discussions in analytic metaphysics. But the arguments are sometimes couched in a slightly different key, and sometimes that difference can lead to a subtle shift
11 I will, though, provide references to works that present competing views on some of the more important disputes.
12 For a recent and relatively sober articulation of this worry, see Hanner 2018.
13 The interested reader should consult Siderits 2015b.
in the tone of the results. I won’t claim that the reader will find in these pages the solution to the problem of universals, or the key to getting clear about the causal relation, or a promising new approach to the defense of presentism. What I do hope to do is instill in those who worry about such things just a (brief) touch of intellectual queasiness. When I was a child I had a toy stereoscope. Since a stereoscope presents an image slightly differently in the two eyepieces, looking through a stereoscope can at first be somewhat physically unsettling. But once the eyes adjust, one comes to see the scene presented by the image in greater depth. I hope the reader will experience something similar by the end of these investigations.
2
Non-Self I
The Buddhist doctrine of non-self seems, to many philosophers today, simple, straightforward, and clearly correct. Simple it is not, however. This and the next chapter explore its formulation, defense, and some of its consequences. Two chapters are called for because the doctrine of non-self is not merely the denial of a self as conceived by Nyāya, Sāṃkhya, or Descartes. This chapter discusses that denial. But the next chapter examines the ontological status of something else, the person.
The distinction between self and person may be understood in the following way. Buddhists are concerned to discover whether the “I”-sense is veridical or illusory. Their question may be answered by investigating the two basic possibilities concerning the referent of “I”: that it is the self, understood as a simple or impartite persisting substance; and that it is the person, understood as a whole or a composite entity consisting of all or some of the psychophysical elements that occur in the existence of a sentient being. To anticipate the outcome of their investigation: “I” cannot denote a self, since there is no such thing; it does, however, denote a person, understood as something conventionally but not ultimately real. These two claims together make up what may be called Buddhist reductionism about persons. We will return to what such a reductionism amounts to after, in this chapter, examining the denial of a self, and in the next, discussing the relegation of persons to the ontological status of back-benchers.
2.1 Early Buddhist Denial of Self
The Buddha gave two arguments for his denial that there is a self: the argument from impermanence, and the argument from control. Both are based on the assumption that the empirically available constituents of persons consist of no more than what are found in the five groups of psychophysical elements (corporeality and four sorts of mental element). Here the notion of empirical availability may require some explanation. It is essentially what is
also found in classical British empiricism: the notion of amenability to being directly cognized or perceived by means of either the five external senses or the “inner sense” (manas).1 The idea is thus that the corporeal constituents of the person are perceived by means of the external senses, while the mental elements are perceived by the inner sense. The exhaustiveness claim is the claim that the five groups of psychophysical elements constitute a complete classification of the kinds of things one perceives when one perceives a person.
The physicalist will of course question the claim that what one cognizes when one “looks within” is anything other than material phenomena. But we will put that question to one side for now. Still it is true that the notion of an inner sense with its own distinctive objects represents something of a grab-bag in Buddhist psychology, and the ontology that results raises at least as many puzzles as it resolves. First, the inner sense is tasked not only with cognizing mental states but also with deploying concepts. Second, its objects are a varied lot. Some, such as feelings of pleasure, pain, and indifference, fit naturally under the rubric of objects of introspection. Others, though, such as perceptual identifications (saṃjñā), look more like functionally defined events—in this case whatever performs the operation of categorizing a percept (e.g., classifying a visual datum as falling under the concept blue). The third of the four groups of mental elements, mental forces (saṃskāra), exhibits a puzzling process-product ambiguity. And the last of the four, consciousness (vijñāna), will give rise to no end of difficulties in the later tradition. Early texts treat this as a sort of bare awareness or registry, leading to the question of whether there can be such a thing as objectless consciousness. The more pressing issue, though, will be how consciousness could itself be directly cognized. (We discuss the resulting controversy in Chapter 9.) Because the Buddha made crucial use of the doctrine of the five groups of psychophysical elements, it had a privileged status among later Buddhist philosophers. In time, though, it would be quietly set to one side as a merely useful pedagogical device.
The argument for non-self from impermanence (MN 109, Mahāpuṇṇama Sutta) has a very simple structure:
1 As was pointed out in the preceding chapter, some later Buddhist dualists espoused a representationalist view of perception like that of the classical British empiricists. There is, however, no evidence that early Buddhism held anything other than a direct realist view of sense perception.
