Personal identity, minimalism, and Madhyamaka

·期刊原文
Personal identity, minimalism, and Madhyamaka
Roy W Perrett

Philosophy East and West

v52 n.3 (Jul 2002)

p373-385

Copyright 1976 University Press of Hawai'i

--------------------------------------------------------------------------------

Abstract:
The publication of Derek Parfit's "Reasons and Persons" in 1984 revived and reshaped the debate on personal identity in Western philosophy. Not only does Parfit argue forcefully and ingeniously for a revisionary Reductionist theory of persons and their diachronic identity, but he also draws radical normative references from such a theory.

--------------------------------------------------------------------------------

p373
Full Text:

I

The diachronic problem of personal identity is the problem of specifying what it is that makes a person the same person over time. Contemporary Western philosophical theories of diachronic personal identity have tended to be constructed in terms of a quest for the unity relation for persons.1 To specify the unity relation for persons is to specify the relation between person-stages occurring at different times in virtue of which they are all stages of one and the same person. We may think of person-- stages as temporal slices of persons, or, alternatively, as temporal slices of the biographies of persons. Either way, there are quite a number of theories on offer as to what the unity relation for persons consists in. These theories can all be classified, however, as being instances of two general types. The first, and currently most popular, type of theory of personal identity holds that the unity relation between person-- stages can be specified in terms of a relation tha! t does not itself presuppose identity. This is what Parfit calls "Reductionism":

Reductionism: Personal identity just consists in the holding of certain facts that can be described without making reference to personal identity.2

The alternative type of theory is Non-Reductionism, which denies this claim. Non-Reductionists hold that the unity relation for persons is just identity; that is, what makes a set of person-stages occurring at different times all stages of the same person is simply that they all are stages of the same person. Personal identity is simple and unanalyzable: there is no nontrivial and noncircular analysis of the identity conditions for persons, nothing that personal identity "consists in." Currently this is very much the less popular type of theory of personal identity, but it nevertheless has able contemporary defenders.3

Derek Parfit's Reasons and Persons not only defends a Reductionist theory of persons and their identity over time (specifically, a variety of psychological continuity theory) but also seeks to derive normative implications from it. In one sense this is not surprising: Parfit's theory revives and extends various Lockean themes, and Locke had already insisted that personal identity was a "forensic" notion "appropriating actions and their merits." Hence, there is a natural temptation to suppose that a theory of personal identity ought to capture the link between identity and what Marya Schechtman calls "the four features": survival, moral responsibility, self-- interested concern, and compensation.4 Parfit's metaphysical and moral revisionism, however, goes much further than this. First, he argues that personal identity (and

--------------------------------------------------------------------------------

p374

hence survival) is a matter of degree because the psychological relation that constitutes identity (Relation R) itself admits of degrees. Moreover the s! ame relation that imperfectly unifies the self over time also connects us with other people. Second, he infers from the imaginary branching cases that "what matters" in survival is not identity, but the holding of the right sort of psychological relation with some future person who may not be identical with you. The central normative implication of all this, so far as Parfit is concerned, is a drift toward impartiality and impersonality, a lessening of the gap between persons since my relation to others is not so significantly different from my relation to my own past and future.

Parfit claims that if Reductionism is true, then the Self-interest Theory (according to which, as rational agents, we each ought to be supremely concerned about our own futures) is false. However, there remain two other possibilities about what it is rational for an agent to seek:

The Extreme Claim: If Reductionism is true, then we have no reason to be specially concerned about our own futures.

The Moderate Claim: If Reductionism is true, then the holding of Relation R gives us some reason to be specially concerned about our futures.5

In Reasons and Persons Parfit held both of these claims to be rationally defensible, although he personally favored the Moderate Claim.6 More recently he seems to have moved a little bit closer to the Extreme Claim, at least insofar as he is now skeptical about the possibility of justifying our customary practices of desert-based punishment and compensation if Reductionism is true.7

There have been various responses to Parfit's claims about the normative implications of Reductionism. One of these is to reaffirm Non-Reductionism, sometimes also enlisting the Extreme Claim to argue that Reductionism must be false because it cannot explain the importance we attach to identity with respect to the four features.8 Another is to accept Reductionism but curtail Parfit's own permissiveness about the cause of Relation R and instead emphasize the importance of the normal cause of R: namely the physical continuity of a living body. A physically based (but psychologically aimed) Reductionism, it is then argued, would not require us to revise substantially our deepest values, since it can explain why we value whole persons and maintain clear distinctions between them.9

A third type of response, often explicitly influenced by Kant, argues that it is a matter of practical necessity that we conceive of persons as enduring wholes, and this sets limits on the normative implications that can be reasonably drawn from Reductionism. Within this camp we can distinguish a variety of claims of different strengths. Some accept the metaphysics of Reductionism, but argue that there may be baleful ethical consequences if we attempt to excise identity from our practical reasoning.10 Others utilize a Kantian metaphor of two perspectives-that of theoretical reason and that of practical reason-and then insist on the importance for the latter of an agent-centered concern for my own future.11 A more radical, Kantian-- inspired approach is not just to allow the Reductionist metaphysics and try to isolate

--------------------------------------------------------------------------------

p375

it from its apparent ethical consequences, but rather to suggest that considerations of practical reason may themselves place constraints on what is an acceptable metaphysics of persons.12

I shall return later to some of these themes about the relevance of practical reason to personal identity. But for the moment I want to highlight a fourth kind of response to Parfit: Minimalism, as defended in a series of papers by Mark Johnston.13 According to Minimalism, metaphysical pictures of the justificatory undergirdings of our practices do not represent the real conditions of justification of those practices. Any metaphysical view that we may have of persons is not indispensable to the practice of making judgments about personal identity and organizing our practical concerns around these judgments. Pace Parfit, then, the presence or absence of "deep facts" about personal identity is largely irrelevant to justifying our ordinary normative practices because these are founded not on a metaphysics of persons but on our circumstances and needs.

Minimalism implies that the normative significance of the division between Reductionism and Non-Reductionism has been exaggerated. On the one hand, Non-- Reductionism assumes that our normative practices are in need of a "deep" metaphysical justification and posits "superlative selves" to do the job. Indeed, the Extreme Claim says that without such superlative selves we cannnot ground the four features properly in personal identity. Reductionism, on the other hand, denies the existence of such superlative selves. However, when combined with the Moderate Claim, Reductionism instead tries to ground the four features in other metaphysical facts. According to Minimalism, both make much the same mistake. Extremism appeals to a Non-Reductionist metaphysics that is just window dressing so far as the real justification of our normative practices is concerned, and Moderatism appeals to a Reductionist metaphysics that has a similar justificatory status. In other words, both Non-Red! uctionism and Reductionism share a common assumption:

The Grounding Assumption: The justification of our normative practices with respect to the four features requires that they be grounded in facts about personal identity, or in those facts to which personal identity is reducible.

Minimalism rejects the Grounding Assumption.

Minimalism, then, is a deflationary account of the normative significance of ontological Reductionism.14 As such, it rejects Parfit's arguments for the thesis that if Reductionism is true, personal identity is morally and rationally unimportant.15 One of Parfit's arguments to this effect is as follows:

The Argument from Below: If Reductionism is true, then personal identity just consists in certain other facts which do not make mention of identity. But these other facts are not themselves of moral or rational importance. Therefore personal identity, which just consists in these other facts, is also not morally and rationally important.

Mark Johnston rejects this Parfitian argument and replies with

The Argument from Above: If Reductionism is true, then personal identity just consists in certain other facts which do not make mention of identity. But personal identity is itself

--------------------------------------------------------------------------------

p376

morally and rationally important. Therefore those other facts, which personal identity just consists in, are themselves (derivatively) of moral or rational importance.16

Morover, Johnston points out that the Argument from Below is subject to the following reductio. Suppose physicalism is true and hence all facts about persons just consist of microphysical facts. In themselves these microphysical facts would not be morally or rationally important. So we have to conclude that nothing has any importance-which is absurd.17

Parfit has responded to this attempted reductio by distinguishing microphysical reductionism from the kind of reduction that, he holds, personal identity is subject to. He explains:

When I claim that personal identity just consists in certain other facts ... about physical or psychological continuity ..., [I claim that] if we knew the facts about these continuities, and understood the concept of a person, we would thereby know, or be able to work out, the facts about persons.... These claims do not apply to facts about fundamental particles. It is not true for example that, if we knew how the particles moved in some person's body, and understood our concepts, we would thereby know, or be able to work out, all of the relevant facts about this person.18

This reply is obscure because it is obscure just what it is we would not know about persons if physicalism is true and we really did know all the microphysical facts. Perhaps Parfit has in mind something like Frank Jackson's argument:

The Knowledge Argument: Mary is a brilliant scientist raised from birth in a black and white room, denied access to any direct experience of color. Nevertheless she becomes an expert on color and the neurophysiology of color vision, mastering all the physical facts there are about color and color vision. However, there is still something she does not know: what it is like to see color.19

If physicalism is true, then color just consists of the physical facts Mary already knows. But the first time Mary is presented with a ripe tomato she surely comes to know something new that she could not infer from the physical facts she already knew about color: what it is like to see red. So physicalism is false, or at best incomplete.

One way to respond to this argument is to concede that after being presented with the tomato, Mary comes to learn by experience what seeing red is like but still deny that she learns any new facts about color the existence of which would refute physicalism. Her new knowledge is knowledge-how, not knowledge-that. Learning what seeing red is like amounts to gaining new abilities to remember, imagine, and recognize. But if gaining such new abilities involves gaining new information, it is neither any sort of irreducibly nonphysical "phenomenal information" nor new propositional knowledge. (This is the Ability Hypothesis defended by Lawrence Nemerow and David Lewis.)20 If the hypothesis is correct, it seems to imply that knowledge of all the physical facts would indeed exhaust all the propositional knowledge possible. But if that is so, it is hard to see how the Reductionist can avoid admitting that we would indeed know all about persons if it was the case both that

--------------------------------------------------------------------------------

p377

we knew all the microphysical facts and that physicalism was true. (The Ability Hypothesis also usefully reminds us that some knowledge is knowledge-how, not knowledge-that-a point to which I shall want to revert later.)

II

How does all this relate to Indian Buddhist philosophy? Firstly, the Western debate about Reductionist and Non-Reductionist theories of personal identity parallels in many respects the debate in classical Indian philosophy between those who explain diachronic identity by reference to an enduring substantial self (atmavadins) and those who deny the existence of such a self, taking instead a "modal" view of reality (anatmavadins).21 The orthodox Hindu philosophers take the former view: although they disagree on the nature and number of such selves, they are all Non-- Reductionists of some sort.22 Most Indian Buddhist philosophers (including the Theravadins, the Vaibhasikas, the Sautrantikas, the Yogacarins, and the Svatantrika-- Madhyamikas) take the latter view and hence are all plausibly classifiable as Reductionists about personal identity.23

Interestingly, in Reasons and Persons Parfit himself claimed that the Buddha would have accepted Reductionism and even cited a few passages from Indian Buddhist texts in support of this.24 However, more recently he has wanted to distinguish his own Reductionism from what he takes to be the Buddhist view:

I do not accept the Buddhist no self view, since I believe that persons exist. We are persons. But I believe that persons are not entities of a kind that must be recognized in any adequate conceptual scheme.25

But the Indian Buddhist Reductionists I just mentioned would all have agreed with Parfit on this; they were not Eliminativists about persons.26

They also would all have agreed with Parfit that Reductionism has normative significance. It is a familiar claim of the Buddhist tradition that the correct understanding of the self is necessary for liberation from suffering; this is also a claim made by Hindu Non-Reductionists. The normative importance of a correct analysis of personal identity, particularly with respect to the four features, is thus common ground between the Indian Reductionists and Non-Reductionists. But the form of the Grounding Assumption they share is a little different from the one shared by Western Reductionists and Non-Reductionists. Both parties to the Indian dispute about personal identity share the assumption that personal identity needs to be explained in terms of metaphysical simples that have an essential ontological independence, or what Madhyamika Buddhists call "inherent existence" (svabhava). Call this the "Indian Grounding Assumption":

The Indian Grounding Assumption: The justification of our normative practices with respect to the four features requires that they be grounded in inherently existent facts about personal identity, or in those inherently existent facts to which personal identity can be reduced.

--------------------------------------------------------------------------------

p378

It is obvious enough that the Hindu Non-Reductionists' espousal of the ātman theory commits them to such essential haecceities, but it may be less obvious that the Buddhist Reductionists are also so committed. However, the Theravādins, the Vaibhā ṣikas, the Sautrāntikas, and the Yogācārins are all committed to the view that facts about persons can be reduced to facts about the causal continuum, which is in turn made up of phenomena that exist independently with essential natures of their own. True, the Svātantrika-Mādhyamikas at least deny that ultimately the constituents of the continuum exist independently, but they still assert that these constituents possess conventional inherent existence; that is, they possess natures of their own by virtue of appearing to the consciousnesses on which their existence depends.27

A crucial Buddhist exception to this general Indian consensus are the Prāsa ṅgika Mādhyamikas: as Mādhyamikas they deny the inherent existence of phenomena at the ultimate level; as Prāsa ṅgikas they also deny the inherent existence of the phenomena at even the conventional level.28 In other words, they deny the Indian Grounding Assumption common to both Buddhist Reductionists and Hindu Non-- Reductionists. In doing so the Prāsa ṅgikas take themselves to be faithful to the original teachings of Nāgārjuna, the founder of Madhyamaka.

In chapter 18 of his M ūlamadhyamakakārikā Nāgārjuna denies both that the self is the same as the psychophysical constituents to which the Buddhist Reductionists try to reduce it and that it is other than those constituents in the way that the Hindu Non-Reductionists claim.29 He does not mean by this that we do not exist, but that we do not exist inherently in any way. In the Vigrahavyavartani Nāgārjuna rejects the opponent's demand that the doctrine of emptiness (i.e., the doctrine that everything is empty of inherent existence) be grounded in the foundationalist framework of Indian pram.na theory. Trying to meet such a demand, Nāgārjuna argues, will lead either to a justificatory infinite regress, or to the incoherent notion of an inherently existent ground for our epistemic practices.30 In the Ratnavali Nāgārjuna warns: "Beyond good and evil, profound and liberating, this [doctrine of emptiness] has not been tasted by those who fear what is entirely groundless."31

The great Prāsa ṅgika Candrakīrti takes up some of these themes in chapter 6 of his Madhyamakāvatāra .32 His analysis of persons there utilizes an elaborate sevenfold negation of the self. According to this analysis, the self (1) is not essentially other than the aggregates, (2) is not identical with them, (3) does not inherently possess them, (4) is not inherently dependent on them, (5) is not the basis upon which they inherently depend, (6) is not the mere collection of them, and (7) is not their shape.33 Candrakīrti says this is analogous to the case of a cart, in that a similar sevenfold analysis of the relation of a cart to its parts is possible. The implication in both cases is the same: the identities of the self and the cart are mere verbal conventions, "dependent designations" (praj ñaptir-upādāya); both they and their parts are conventionally existent, but empty of inherent existence. Moreover, in order to preempt further reification, Candrakīrti states explicitly that emptiness is itself empty.34 In other words, emptiness is not different from conventional reality, but just the fact that conventional reality is conventional.

--------------------------------------------------------------------------------

p379

According to Candrakīrti, then, we cannot ground our ascriptions of identity in inherently existent facts about personal identity or in those inherently existent facts to which personal identity can supposedly be reduced. However, he also makes it clear that we can still preserve our ordinary conventional beliefs about what we are, provided that they are understood as merely conventions:

Things such as jugs, cloth, tents, armies, forests, rosaries, trees, houses, trolleys and guesthouses should be understood to exist in the way they are commonly spoken of by people because the Buddha did not argue with the world over these matters. Furthermore, by applying the analysis of the cart to part-possessors and their parts, quality-possessors and their qualities, people with attachment and their desires, bases of characteristics and their characteristics and fuel and the fire it burns, one finds that they do not exist in any of the seven ways. But as long as they are not subjected to such analysis, they do exist in another way: namely, in terms of their being well known to the world.35

Living in terms of this insight requires wisdom (praj ñā). According to the Buddhist tradition, such wisdom is a matter of both intellectual understanding and action. The enlightened bodhisattva, then, is characterized by both intellectual discernment (knowledge-that) and a non-inferential actualization of what has been discerned (knowledge-how).36

These Prāsa ṅgika themes are obviously much closer to Minimalism than to Reductionism. Indeed, I want to claim the Prāsa ṅgika Madhyamaka position on personal identity as a Buddhist analogue of the Western Minimalist position on personal identity.37 Both believe that any metaphysical view that we may have of persons is not indispensable to the practice of making judgments about personal identity and organizing our practical concerns around these judgments. Both believe that the presence or absence of "deep facts" about personal identity is largely irrelevant to justifying our ordinary normative practices because these are founded not on a metaphysics of persons but on our circumstances and needs. Both are, in this sense, deflationary about the normative pretensions of metaphysical Reductionism and Non-Reductionism. Both acknowledge, however, that this does not mean that the philosophy of personal identity must leave everything as it is. Our everyday practices are open to cr- iticism and revision, even if neither they nor their alternatives are groundable in the inherent existence of things in themselves.38

I suggest, then, that Parfit was correct to have thought originally that his Reductionism was analogous to certain Buddhist views about "no self": there were, indeed, many, many Indian Buddhist Reductionists. But this was not the only influential Buddhist view about personal identity present in India: the Prāsa ṅgika Madhyamikas held a view more closely comparable to Minimalism. Thus, they rejected both Buddhist Reductionism and Hindu Non-Reductionism in order to tread a middle way between these opposed extremes.

III

Why did only the Prāsa ṅgika Madhyamikas advocate Minimalism in India and reject the Indian Grounding Assumption about personal identity common to both Buddhist

--------------------------------------------------------------------------------

p380

Reductionists and Hindu Non-Reductionists? By way of reply I offer the following conjecture.39

It is obvious enough that we can think about ourselves in two rather different ways: as subjects or as agents. According to the former conception, we are the detached subjects of our experiences, transcending these experiences and their contents. According to the latter conception, we are doers, psychophysical beings both in the world and of it. (The Kantian argument about the significance of the differing perspectives on the self taken by theoretical and practical reason seeks to build an elaborate-and contentious-philosophical structure upon this intuitively obvious point about our two rather different self-conceptions.)

The intuitive distinction between these two points of view was not unknown in ancient India. Thus, the Hindu philosophers recognize a tension between what they call the pravṛtti and nivṛtti traditions. The pravṛtti tradition is the activist strand of Hindu thought, exemplified in the Vedic ritualistic tradition; the nivṛtti tradition is the quietism exemplified in the later Upanisadic renunciant tradition. It is also true that the Hindu philosophers acknowledge that the human person is characterized both by agency (kartṛtva) and by enjoyment (bhoktṛtva). However, although both the activist and quietist strands are present in Hindu thought, in general it is the quietist tradition that is valorized by the philosophical tradition.40 The self (ātman) is most typically conceived of by the Hindu philosophers as a pure subject, detached from the objects of its consciousness, enduring and changeless amidst the flux of our mental states. (Curiously enough, this is broadly true even of Mimāṃsā, the philosophical school most concerned with ritual action. To be sure, Mimāṃsā vehemently insists that the eternal self is an agent as well as an enjoyer, and even that knowing is an action. However, Mimāṃsā also claims that the self is in no way essentially related to the world, a fact about it that is supposed to be realized in liberation.)

The picture of the self as witness subject, rather than doer, thus dominates Hindu philosophy, and liberation is typically identified with the realization of the self's true nature as pure subject. Correspondingly, it is reasonable to infer that the Hindu Non-- Reductionists would not be sympathetic to any suggestion that the demands of practical reason should place constraints on the metaphysics of persons, nor would they be willing to align themselves with the Kantian-style critique of Reductionism.

At first sight, the case of the Buddhist Reductionists looks rather different from that of the Hindu Non-Reductionists. After all, the Buddhists deny that there is an enduring atman, reducing the self instead to the psychophysical states and the causal relations between them. But, in an important sense, the same valorization of the person as subject, rather than agent, persists in early Buddhism. This is particularly evident in the emphasis on the detachment of Buddhist meditators from their own mental and physical states. Of course, there is a well-known tension within the early Buddhist tradition between two types of meditation: one involving the cultivation of tranquillity (samatha-bhāvanā), the other involving the cultivation of insight (vipas-sanā-bhāvanā).41 The former emphasizes the pursuit of liberation through enstatic techniques designed to destroy the passions by withdrawal from all contact with the

--------------------------------------------------------------------------------

p381

external world; the latter emphasizes the pursuit of liberation through analytic techniques designed to remove ignorance by the cognition of the way things really are. The enstatic techniques clearly reinforce the view of the person as subject, rather than agent. But so, too, does the analytic stress on cognition, with its attendant picture of the meditator as a knower detached from the objects of knowledge.

I conjecture that this convergence between the Hindu Non-Reductionists and the Buddhist Reductionists on the matter of the person as primarily subject is to be explained by the common historical origins of both traditions. Following Dumont, it is sometimes suggested that the dominance within philosophical Hinduism of the quietist strand over the earlier activist strand is to be explained as a response to the challenge to Brahmanical authority represented by the rival renunciant tradition, especially Buddhism.42 But if that is right, then it should not be surprising that we find within early Buddhism what is, in one sense, a not too dissimilar picture of the person from that of the Hindu quietists.

Of course, there is indeed a crucial metaphysical difference between Hindu Non-Reductionism and Buddhist Reductionism about the nature of personal identity: this is the subject of the fiercely contested debate about ātmavāda and anātmavāda that the texts of both parties make so much of. However, what this debate obscures is a shared quietistic assumption that the person is properly to be seen as detached from his or her experiences, a witness rather than a doer. This assumption is arguably, in its own way, as embedded in early Buddhist meditation theory as in Hindu quietism. It is because of this deeper, shared assumption about the nature of persons that the Indian Grounding Assumption is unquestioned by both Hindu Non-- Reductionists and Buddhist Reductionists: both parties to the Indian Reductionism/ Non-Reductionism debate assume that a person is most properly conceived of as a subject rather than an agent. Hence, they assume, too, that "the four features," which are a! ssociated with personal agency and which need to be linked to personal identity, are in reality derivative features. Although the Hindu Non-Reductionists and the Buddhist Reductionists disagree strongly about precisely which facts about persons ground the four features, they nevertheless agree that the agent-centered features of persons must be grounded in the more fundamental subject-centered features of persons. Moreover, whatever facts about personal identity serve to ground the four features must, on pain of a regress, be self-grounding (i.e., inherently existent).

Why do the Prāsa ṅgika Mādhyamikas dissent from this common assumption? Basically, I suggest, because Prāsa ṅgika Mādhyamaka is so closely associated with the systematization of Mahāyāna Buddhism and its attendant religious ideal of the bodhisattva, a being tirelessly active in the compassionate service of suffering sentient beings. Candrakīrti's Madhyamakāvatāra, for instance, explicitly integrates Madhyamaka philosophy into the Mahayana spiritual path. Similarly, the major work of Śāntideva-that other great Prāsa ṅgika-is the Bodhicaryāvatāra, a statement of the bodhisattva's path to enlightenment, which includes a chapter expounding a Prāsa ṅgika Mādhyamika understanding of the emptiness of all phenomena, including the self. In other words, Prāsa ṅgika Madhyamaka combines a Mahāyānist commitment to the kind of activism associated with the bodhisattva with

--------------------------------------------------------------------------------

p382

a philosophical commitment to the absence of inherent existence. The two commitments fuse to create a system that acknow! ledges the primacy of practice (even prajna involves knowledge-how) without feeling any need to seek to ground our practices in anything metaphysically "deeper" than these practices themselves.

Prāsa ṅgika Madhyamaka, then, is much more a Buddhist Minimalism than a Buddhist Reductionism (or Non-Reduction ism). But it also resonates with the Kantian argument that Reductionism is insufficiently attentive to the constraints that the agent-centered concerns of practical reason place on an adequate theory of personal identity. In this way Prāsa ṅgika invites us to tread a middle path with respect to the issue of personal identity: a path between the extremes of both Reductionism and Non-Reductionism, dwelling in emptiness and fearless of groundlessness.

Notes

In the writing of this essay I have benefited from the assistance of Jay Garfield, Chakravarthi Ram-Prasad, and Mark Siderits. I thank them warmly and hereby exonerate them from responsibility for any shortcomings that may have persisted in the face of their good advice. An ancestral version of this piece was presented as part of the symposium "A Dialogue on Reductionism: Buddhism and the Philosophy of Derek Parfit," held at the International Association of Buddhist Studies Conference, University of Lausanne, August 1999.

1 - See John Perry, ed., Personal Identity (Berkeley: University of California Press, 1975), pt 1.

2 - Cf. Derek Parfit, Reasons and Persons (Oxford: Oxford University Press, 1984), p. 210.

3 - Vinit Haksar's Indivisible Selves and Moral Practice (Edinburgh: Edinburgh University Press, 1991) is particularly apposite for my present purposes since Haksar explicitly defends Non-Reductionism against both Buddhist and Parfitian Reductionism. Other contemporary defenders of Non-Reductionism include Roderick M. Chisholm, Person and Object (London: George Allen and Unwin, 1976); Richard Swinburne, "Personal Identity," Proceedings of the Aristotelian Society 74 (1973-1974): 231-248; Geoffrey Madell, The Identity of the Self (Edinburgh: Edinburgh University Press, 1981).

4 - Marya Schechtman, The Constitution of Selves (Ithaca: Cornell University Press, 1996), p. 2.

5 - Parfit, Reasons and Persons, pp. 307-311. 6 - Ibid., p. 312.

7 - Derek Parfit, "Comments," Ethics 96 (1986): 838-843.

8 - See, for instance, Haksar, Indivisible Selves and Moral Practice.

--------------------------------------------------------------------------------

p383

9 - Peter Unger, Identity, Consciousness and Value (New York: Oxford University Press, 1990).

10 - Susan Wolf, "Self-Interest and Interest in Selves," Ethics 96 (1986): 704-720. 11 - Christine M. Korsgaard, "Personal Identity and the Unity of Agency: A Kantian Response to Parfit," Philosophy and Public Affairs 18 (1989): 101-132.

12 - Simon Blackburn, "Has Kant Refuted Parfit?" in Jonathan Dancy, ed., Reading Parfit (Oxford: Blackwell, 1997).

13 - The papers by Mark Johnston include "Is There a Problem about Persistence?" Proceedings of the Aristotelian Society Supplementary Series 61 (1987): 107135; "Reasons and Reductionism," Philosophical Review 101 (1992): 589618; "Human Concerns without Superlative Selves," in Dancy, ed., Reading Parfit.

14 - Indeed, Minimalism about the metaphysical problem of persons might be taken as confirmatory of Arthur Danto's general thesis that "philosophical problems arise in connection with indiscriminable pairs, the difference between which is not a scientific one ... [or a] difference that need ever be imagined as revealing itself in experience at all.... Life could go forward perfectly well without these distinctions ever needing to be thought about" (Arthur Danto, Connections to the World [New York: Harper and Row, 1989], p. 11).

15 - For independent criticism of four such Parfitian arguments see Brian Garrett, Personal Identity and Self-Consciousness (London: Routledge, 1998), pp. 88-94. 16 - Johnston, "Human Concerns without Superlative Selves," p. 167.

17 - Ibid., pp. 167-168.

18- Derek Parfit, "The Unimportance of Identity," in Henry Harris, ed., Identity (Oxford: Clarendon Press, 1995), p. 33.

19 - Frank Jackson, "Epiphenomenal Qualia," Philosophical Quarterly 32 (1982): 127-136; "What Mary Didn't Know," Journal of Philosophy 83 (1986): 291 295.

20 - See Lawrence Nemerow, "Physicalism and the Cognitive Role of Acquaintance," and David Lewis, "What Experience Teaches," both in William G. Lycan, ed., Mind and Cognition (Oxford: Basil Blackwell, 1990). For other responses to the Knowledge Argument see Ned Block, Owen Flanagan, and Guven Guzeldere, eds., The Nature of Consciousness: Philosophical Debates (Cambridge: MIT Press, 1997), pp. 557-617.

21 - T.R.V. Murti, The Central Philosophy of Buddhism (London: George Allen and Unwin, 1955), chap. 1.

22 - Advaita Vedanta is a slightly tricky case to classify here since while they are Non-Reductionists about the absolute Self (ātman), the precise ontological status within their system of the empirical self (jīva) is, notoriously, a matter of intramural controversy.

--------------------------------------------------------------------------------

p384

23 - Cf. James Duerlinger, "Reductionist and Nonreductionist Theories of Persons in Indian Buddhist Philosophy," Journal of Indian Philosophy 21 (1993): 79-101; Mark Siderits, "Buddhist Reductionism," Philosophy East and West 47 (1997): 455-478.

24 - Parfit, Reasons and Persons, pp. 273, 502-503.

25 - Derek Parfit, "An Interview with Derek Parfit," Cogito 9 (1995): 123.

26 - Pace Jim Stone, "Parfit and the Buddha: Why There Are No People," Philosophy and Phenomenological Research 48 (1988): 519-532. For more on this point see Siderits, "Buddhist Reductionism."

27 - Duerlinger, "Reductionist and Nonreductionist Theories," pp. 83-84.

28 - This Tibetan way of distinguishing between Svātantrika- and Prāsa ṅgika Madhyamaka is not to be found in any extant Indian text. However, when the Tibetan Mādhyamika Tsong kha pa utilizes it, it is clear that he takes himself to be expounding the position of the Indian Prāsa ṅgika Candrakirti.

29 - See Kenneth Inada, Nāgārjuna: A Translation of His M ūlamadhyamakak ārik ā with an Introductory Essay (Tokyo: Hokuseido Press, 1970); Jay L. Garfield, The Fundamental Wisdom of the Middle Way: Nāgārjuna's M ūlamadhyamakak ārik ā (New York: Oxford University Press, 1995).

30 - Kamaleswar Bhattacharya, The Dialectical Method of Nāgārjuna (Vigrahavy āvartan ī), 2d ed. (Delhi: Motilal Banarsidass, 1986). See also Mark Siderits, "The Madhyamaka Critique of Epistemology I," Journal of Indian Philosophy 8 (1980): 307-335.

31 - Ratnav āli 1.79, as quoted in C. W. Huntington, Jr., and Geshe Namgyal Wangchen, The Emptiness of Emptiness (Honolulu: University of Hawaii Press, 1989), p. 26. Cf. Jeffrey Hopkins, The Precious Garland and the Song of the Four Mindfulnesses (London: George Allen and Unwin, 1975), p. 28.

32 - See Geshe Rabten, Echoes of Voidness, trans. and ed. by Stephen Batchelor (London: Wisdom Publications, 1983); Huntington and Wangchen, The Emptiness of Emptiness; Peter Fenner, The Ontology of the Middle Way (Dordrecht: Kluwer, 1990).

33 - Madhyamakāvatāra 6.120-164. On the development of Candrakīrti's sevenfold analysis in Tibetan Madhyamaka see Joe Wilson, Chandrakirti's Sevenfold Reasoning: Meditation on the Selflessness of Persons (Dharamsala: Library of Tibetan Works and Archives, 1980).

34 - Madhyamakāvatāra 6.185-186.

35 - Madhyamakāvatāra 6.166-167; Rabten, Echoes of Voidness, p. 82.

36 - Cf. Praj ñakaramati's distinction between two types of praj ña, discussed in Huntington and Wangchen, The Emptiness of Emptiness, pp. 90-92.

--------------------------------------------------------------------------------

p385

37 - Duerlinger, in "Reductionist and Nonreductionist Theories," claims Chandrakirti is instead a Buddhist Non-Reductionist, but at least part of our difference on this is traceable to our somewhat different accounts of what the Reductionism/ Non-Reductionism distinction consists in.

38 - Johnston makes this point about his own Minimalism ("Human Concerns without Superlative Selves," pp. 175-176); Madhyamaka's role in the reshaping of local cultural practices in its spread through Asia (and now the West) suggests something similar about Buddhist Minimalism.

39 - What follows here owes a debt to the perceptive remarks in Jitendra Nath Mohanty, Reason and Tradition in Indian Thought (Oxford: Clarendon Press, 1992), pp. 192-205, and the use Mohanty makes there of his distinction between "subject" and "person."

40 - This generalization assumes an austere characterization of "the Hindu philosophical tradition" as the dar śana tradition of speculative philosophy. The epic and theological traditions are often much less quietist.

41 - Cf. Paul J. Griffiths, "Concentration or Insight: The Problematic of Therav āda Buddhist Meditation-Theory," Journal of the American Academy of Religion 44 (1981): 605-624; On Being Mindless: Buddhist Meditation and the Mind-Body Problem (La Salle: Open Court, 1986), pp. 13-16.

42 - See Louis Dumont, Homo Hierarchicus, rev. ed. (Chicago: University of Chicago Press, 1980).