CHAPTER 2: SELFLESSNESS AND NORMATIVITY ŚĀNTIDEVA AND EMPTINESS
ETHICS
2.1 Introduction
My central thesis here is that the No-Self view fails to provide us with a clear account of agency, particularly, ethical agency. In chapter 1, I argued in a very general way that our intuitions regarding what it means to be an ethical agent are better accommodated by a conception of the self as the owner of its mental and physical states. Furthermore, I pointed out that influential early and late 20th Century thinkers like Henry Sidgwick and Derek Parfit argued—albeit from two different standpoints—that normative, rational, and ethical considerations are importantly shaped by our conception of selfhood, particularly with respect to how we conceive of the distinction between
self and other. I also, in a very broad way, reconstructed a version of the Buddhist No-Self view (anatman or nairatmyavada), which likewise places considerable ethical weight on how we conceive of the self. I then argued that entertaining such a view in the practical domain might not only prove to be unnecessary for making sense of other-regarding considerations and altruistic motivations, but may even collapse or greatly destabilize a vital distinction that underlies such considerations: the distinction between self-regarding and other-regarding motivations. The metaphysical self-effacer may end up failing to distinguish between being selfish and being altruistic, and thus prove incapable of being altruistic. Practical unselfishness may turn out to be incompatible with No-Self metaphysics. While this, on its own, might not be a highly original claim, I wanted
to better clarify what morally relevant “selflessness” might mean. In doing so, I showed how egocentric considerations might in fact be consistent with unselfishness (just as abolition of the ego can be practically consistent with selfishness of the sort that, in popular imagination, Yashodhara must have charged Prince Siddhartha, who with the aim of achieving enlightenment, abandoned her and their baby).
The crucial issue to examine here is the relationship between the Ownership view of self, which compellingly accommodates many of our firm intuitions with respect to agency and ethics, and the various theoretical programs that revise our belief in an owner of mental occurrences. In section 2.2, I address some of the glaring challenges No-Self views face in terms of providing both a compelling, general account of agency, and a specifically convincing account of ethical agency. I argue that No-Self views require re-description of some fundamental features of forward-looking agency and language-use, and I show how such re-descriptions face formidable problems that are better met by an Ownership view.
I only broadly covered this problem in chapter 1. The problem is that agency and ethics crucially appeal to notions like volition, voluntary action, freedom, intention, and persistence and duration. The view that we are owners of our mental lives nicely accommodates the deeply held intuition that we can be rational or irrational agents who are responsible for our lives, and that some sense of endurance justifies prudence and concern for the welfare of our nearest and dearest. These are basic aspects of the lived, human experience, and these notions are ineluctably stubborn organizing principles that have been emphasized to greater or lesser degrees throughout the history of civilized life. Moreover, these principles make social living conceptually intelligible. Theoretical advancements, ontologically revisionary insights, and compelling reductionist
programs either force us to abandon some of these basic notions, or proceed along an insulated domain of discourse that simply remains neutral to the reality of agency and ethical phenomena. The radical implications of some these revisionary models for our ethical intuitions and for models of agency and practical rationality can be crudely compared to the difficulties quantum mechanics presents for Newtonian physics and Relativity Theory. These physical models, certain aspects of which are mutually inconsistent, are all indispensable to the advancement of physics—none have so far been abandoned for the other. Similarly, the social aspect of the human condition that produces ethics cannot be “transcended” or dispensed with merely in order to accommodate ontological models that dispense with person-involving and ethical terms. And this is precisely why Buddhist views, and the work of Parfit, are so attractive: they provide possibilities of a sort of “grand” or “unifying theory” in which the theoretically compelling project of reductionism may be able to provide us with normative-ethical insights. Thus, it is crucial for ethicists to take these projects seriously.
While various No-Ownership and/or No-Self views might distance themselves from normative discourse, those views that do attempt to develop an ethical orientation—or to derive normative principles from their ontological commitments—would be attractive to those who want to retain the normative gravity of “I should do the right thing” while also adhering to eliminativism or reductionism that construes “I” as an empty term. The ethical orientation of the Buddhist and Parfitian projects are just the sorts of views that want to derive moral conclusions from ontological commitments (or, ontological revisions in the form of metaphysical non-commitments). I sketched one way in which metaphysical views about the self—especially revisionary metaphysical views—can importantly impact what it makes sense to do in the practical domain. This was in response to the claim that
metaphysically deep categories do not matter so much in ethical questions as do considerations of practical identity. The counter-argument I offered in chapter 1 claimed that if we can achieve a certain impersonal view about selfhood, then such descriptive views might in fact have normative implications. Moreover, we might achieve an impersonal view through theoretical dialogue with metaphysics. For example, one version of the No-Self view in Buddhism, the “reductionist view” that we find in Vasubandhu (c. 4th-5th C. CE), which is arguably rehearsed in Santideva’s argument for rational altruism in chapter 8 of the Bodhicaryavatara (passages 101-103), claims that under analysis we can view the psychophysical stream in totally impersonal terms. We can do this without appealing to any sort of further underlying and unifying thread. The idea here is that psychological
continuity, and various forms of causal relationships between psychophysical events, underlies the erroneous view that there is a persisting self that owns its mental and physical states. When we unburden ourselves of the belief that we are persisting owners of our experiences, we not only free ourselves up for other-regarding considerations in the conventional, practical domain, but also recognize that altruistic motivations are rationally justified.
I also briefly discussed a variant of this view. Acknowledging that we need a more robust view of how we develop a conception of ownership and “mineness,” Buddhists like Candrakirti (c. 6th C. CE)—on Jonardon Ganeri's reading —argued that the sense of self is forged through an appropriative activity in the practical domain. I gloss “appropriative activity,” as involving the capacity for the stream to “grasp” or “take ownership” of (or fuel and feed itself) various sensations, volitions, and physical
objects. Appropriation is a gloss on the term, “upadana,” which is a word for fuel. The psychophysical stream is “fueled” by what it wants and does not want, which is reflected in its positive pursuits and its various aversions. The stream operates relationally and fluidly like a fuel appropriating oxygen. Without hypostasizing the stream, we can say that psychophysical elements show up as intimately related by the sorts of intentional objects at which such states aim, and by which the stream feeds itself. The fuel of desire determines the directionality of those aims. Desire produces a kind of closure wherein states of consciousness knit a seemingly unified tapestry out of the
sentiment, “I want, therefore I am.” The “I” appears to be singular and self-same as it is a term that arises out of desire for existence and for sensual objects. To want is to make the object of want one's own, and this process holistically produces the sense that “this is mine,” and “I am.” There is lust for the lover's body, or desire to smell and taste something sweet. There is pride in the supple movements of the body, and a certain experience of power in exerting physical force and maintaining social equilibrium.
Contiguity and continuity between desires and volitions produce a first-person sense of unity, which is neither a mere sum of parts nor a real whole. We might think of this in terms of a sort of “vortex” of desires and perceptions that forge a sense of unity amid the “grasping” action that Buddhism believes fuels sentient existence. While this produces an egocentric closure, I independently suggested that a more other-centric manifestation of the appropriative activity could be one way of creatively glossing “selflessness.” When we learn to more consciously view the self as a dynamic activity with porous borders (not a substantive thing), and dispense with the belief that the sense of self is the result of a real, persisting entity, we may better learn to identify with the needs and sufferings of other people. We understand that no real
metaphysical limits sustain a hard and fast individuation between the more intimate needs we experience, and the needs that others attest to having. The very boundary between the intimate and the objectified “other” grows blurry, and the result is a sense of “exchange” and dynamic unity between self and other. On the practical level, the so-called appropriative self is one built out of the various mental and physical occurrences that are either appropriated and identified with, or disavowed or transcended. This sort of “self-activity” or “self-ing” is more or less coherent, but always porous and dynamically in flux, and it is fueled by the mental occurrences and values that it endorses. The
idea here is that with the right practice we can expand the view of self to radically incorporate others (and by “others,” I mean other constellations of mental and physical occurrences available for appropriation). When this is supported by the Buddhist soteriological goal of liberation from cycles of suffering—which the Buddhist argues are chiefly produced by attachment, egoism, and narrow obsession with one's own desires and projects— we can cultivate a deeply other-centric orientation. This is presumed to be a liberating and salvific orientation—an “enactment,” if you will, of non-egocentric being. I will develop a slightly different, but I think, more attractive version of this argument in chapter 3. However, in chapter 4, I will point out some of its shortcomings, and argue that a more normative gloss on the “appropriative self” offers a contemporary and respectably embodied, naturalistic view. The important difference between the two versions is that the latter supports the belief that authorship of one's life through practical identity and embodied avowals and disavowals is not an illusion to be seen through.
In section 2.3 of the present chapter, I will lay out the conceptual territory of contemporary views on selfhood. My goal is to better situate No-Self views. I specifically set up the conversation in light of the Madhyamaka-Buddhist “emptiness” thesis—the view that nothing in reality has intrinsic existence or inherent essence (svabhava). I focus on Madhyamaka, because its ethical champion, Santideva provides one of the more systematic ethical treatises in the Mahayana corpus. Moreover, I need to situate Madhyamaka and Santideva’s altruistic project in terms of competing Buddhist No-Self views.
In 2.4, I cover with greater detail the utilitarian-leaning ethics we may extract from Santideva's Bodhicaryavatara and his Siksasamuccaya. Santideva provides us with a version of the No-Self view that presumably aims to derive moral conclusions from ontological insights. Furthermore, as theorists like Charles Goodman have contended, Santideva's work aims to justify a version of act-consequentialism, which promotes altruism and agent-neutral ethics. This should be taken seriously, because there are independently compelling reasons to disavow the belief in a real self or a real owner of experiences. So if we can develop a workable ethics out of such insight, then we may have an ethically compelling reason to dispense with our belief in real selves.
Goodman provides a nice overview of some of the independent reasons Buddhists have for rejecting belief in the self. For example, Buddhists are fond of appealing to vagueness arguments. Mereological vagueness, which drives various Sorites problems, destabilizes a realist account of composite objects, thereby problematizing our capacity to individuate objects and identify parts that belong to a whole. Moreover, it implies that identifying P as a part of W is a function of conceptual construction, and not an isomorphic depiction of fixed structures in nature. Thus, all composite items—like the complex stream that Buddhists believe account for the self—are conceptual imputations that disappear under analysis. As Goodman writes, “Plants and animals, artifacts and natural objects, all have vague spatial boundaries; once we look at them on the atomic level, they are clouds of tiny particles, which constantly exchange matter with the environment in such a way that there is often no fact of the matter about whether a
particular particle is part of them or not.” And of course, at the normative level, vagueness drives many of the controversies involved in the abortion and assisted-suicide and euthanasia questions. Parfit famously capitalizes on vagueness to argue that there is not always a fact about the matter as to whether or not someone has survived a process of change. This, of course, has important implications for theories about personal identity, and for the ethics of punishment and responsibility. So, if Santideva's revisionary work can provide us with normative conclusions, then it's a theory worth taking very seriously. However, to better situate Santideva's work—and show how it may differ in important ways from Parfit's project—we need a more thorough taxonomy of views about the self, particularly No-Self views, since not all Buddhists mean the same thing by denying the existence of a self.
I will tackle this, as I said, in 2.3. In doing so, I will utilize Ganeri's helpful taxonomy. Accordingly, I will assess Santideva's work in light of this taxonomy. Finally, in 2.5, I will also assess some of the leading reconstructions of the crucial passages, 101-103, in chapter 8 of Santideva's Bodhicaryavatara. In these passages, Santideva argues that we must endorse robust altruism. I will ultimately conclude, like Harris, that denying the existence of the self fails in any straightforward way to justify an altruistic moral imperative. There may be better ways to reconstruct Santideva's work. However, I cannot currently devise a strategy for rescuing Santideva's argument (although
I provide some original takes on the leading critiques of his argument, while providing a metaphysical option that might bolster Santideva's thesis). It seems to me that the most compelling reason I might have for helping someone—or for being committed to altruism—is through some form of identification with that person. At some general level of empathic identity, your pain might as well be my pain: this is something we can share. At some level of metaphysical description “we are all one,” and so any suffering is suffering for all. At some level of social generality, I would want to be free of pain, and I believe that this is a right that anyone who is sufficiently unbiased would have to extend to everyone in a Rawlsian-contractualist sort of way. However, in all these scenarios—pace the Buddhist emphasis on “emptiness”—the important distinction between self and other is always maintained. The key to altruism, I believe, rests somewhere in the paradoxical notion of forging real identity amidst accepted difference.
2.2 Ownership and No-Self Views
South Asian thought, from its rich philosophical inquiry into the nature of the self in the Upanisads, to early Buddhist practical philosophy, and medieval Buddhist epistemological-metaphysical speculation, has viewed the question of the self as one of the most central philosophical and spiritual-religious questions to pursue. A general belief pervading the constellation of diverse texts we monolithically call, “Indian philosophy” (darsana) is the belief that understanding the nature of the self is necessary for commanding a clearer sense of what one should do, in both the social-normative sphere, and in one’s private life. Moreover, and particularly in the case of Buddhism,
understanding the nature of the self is thought to be necessary for achieving maximum well-being, for understanding one’s duties and social roles (dharma), and for freeing oneself from the existential suffering that presumably arises from holding mistaken views about the self. In fact, one way of viewing the Buddha’s mission, and the philosophical thought that has grown out of that mission, is to see it as a deeply pragmatic labor aimed at healing various “psychological sicknesses” that arise—not out of aberrant mental disorders and neuroses, but rather, out of the ordinary experience of living under erroneous, but dominant practical and philosophical paradigms. So the Buddha may be viewed as a kind of “psychological doctor,” or “philosophical healer.” Especially with such foundational Madhyamaka thinkers as Nagarjuna and Candrakirti, Buddhism becomes a philosophical therapeutic and corrective in its critique of the reifying habits and practices of language-users, a move we’ve taken very seriously in the Euro¬American tradition following Wittgenstein’s lead. Also, in a move paralleled by Hume in British philosophy, the Buddha’s project might be viewed as a kind of empirical
philosophical-psychology, or, proto-psychology. This is especially visible in the Abhidharma tradition, which aimed to provide metaphysical support for the leading conceptual tenets of Buddhism: the teachings of nonself (anatman), impermanence (anitya), and suffering and its healing (duhkha and nirodha). Abhidharma articulated what it believed were the fundamental building blocks (dharmas), or, matrices of basic experiential data that underlie our mental lives and express the structures of our consciousness. Thus, Buddhist philosophy, with its pragmatic-empirical and therapeutic orientation, and the equally therapeutic but deeply metaphysical Yoga tradition, emerging out of Patanjali’s c. 4th century CE, Yoga-Sutras, along with the analytic and logic-driven discourse of the Nyaya- Vaisesika realists all set the stage for what Ganeri calls, “substantive truth-directed debate about the self” in the Indian tradition. The Buddhists were particularly concerned with locating the source of error that perpetuates erroneous thinking about the self. One does not exaggerate in saying that all the major schools of Indian philosophy believed that getting it right about the self is essential for achieving a fully realized life (construed in both transcendent terms, and with respect to social harmony and the proper view of one's social roles and duties).
Mahayana-Madhyamaka Buddhism, in the likes of Nagarjuna, Candrakirti, and Santideva’s 9th chapter of the Bodhicaryavatara, “Perfection of Wisdom”
(prajna-paramita), adopts a radical global-irrealism by attacking epistemological foundationalism, and deconstructing substance ontology altogether. If the Sautrantika-Abhidharma philosophy of Vasubandhu (c. 4th-5th C. CE) sought to locate and analyze the basic experiential stuff out of which our mental lives are constructed, the Madhyamaka School, growing out of Nagarjuna's Mulamadhyamakakarika (“Root Verses of the Middling Path”), sought to problematize the epistemological assumptions that grounded such a pursuit. This inspired the Madhyamika philosopher to develop a radically relational philosophy, eschewing any references to either real natural kinds, or foundational knowledge obtained through allegedly certified and truth-producing instruments (pramana). Thus, it is necessary to clarify the nuanced iterations of the No-Self view that arise out of the trove of diverging texts called, “Buddhism.” It is also necessary to provide a broader scope of the different concerns that drove the evolving critique of belief in a persisting, substantial self. Helpfully, Ganeri has distilled two basic research targets running throughout the variegated strands of the Buddhist project:
1. Identify false constitutive accounts of the concept of self, and provide instead a true account of what our concept of self consists in.
2. Consider whether the concept of self, correctly described, has any place in a properly constituted mental life.
No-Self views may be oriented against the background of these two trajectories. First, one must provide a convincing account of what is conceptually constitutive of selfhood. Second, one must determine the practical and ethical upshot of employing this correct, or, corrected account of self. With that in mind, I'd like to articulate several philosophically relevant iterations of the No-Self view in order to more closely assess, in light of chapter 1, whether a more compelling account of ethical agency and justified altruism can be developed by departing from the view of the self as an embodied owner of its mental and physical states. I will begin by adopting a useful taxonomy from Ganeri's work.
But before turning to that taxonomy, I'd like to emphasize that in order for a No-Self view to be convincing, especially those views claiming that even a correctly described concept of self is a moral ill that somehow impedes one from achieving optimal ethical clarity and maturity, it has to overcome some formidable ontological and ethical obstacles surrounding the concept of agency. As I've claimed above, chapter 1 was a broad and very generic rehearsal of what is ethically at stake when we consider the metaphysics of selves. But the epistemology of selves is equally important when considering agency. How I come to know that certain plans and preferences are in your interest, and how I come to relate to the various beliefs, desires, and intentions that fashion my plans and interests, as being mine, require that I employ some conception of selfhood. Ownership and an attendant conception of self are key components that make agency and self-regarding and other-regarding considerations intelligible. As Ganeri argues, the concept of self “is important because it enables me to think of my ideas and emotions, my plans and aspirations, my hopes and fears, as mine—as belonging to me, and the proper focus of my self- interest.” The discussion of conflicts between self-interest and considerations of the greater good does not get off the ground with out some working conception of selfhood. In chapter 1, I showed how the strategy of collapsing the distinction between self and other does not necessarily defuse the conflict between self and other. It may either: (1) leave us with no motivations, or, (2) disguise a deeper selfishness that comes with apathetic acts of omission. Moreover, lacking a concept of a self is tantamount to lacking a robust sense of agency: one would lack the capacity to consider how intentions and interests coincide or conflict. Altruism would not make sense unless one had a clear sense of whose interests are at stake, and whose interests are either advanced or impeded.
Also, as Marya Schechtman has pointed out, some sense of duration and persistence is conceptually essential in a convincing notion of selfhood, and this is straightforwardly supported by an Ownership view of the self. Duration and connectedness is required for making sense of talking, listening, working, acting, having beliefs, desires, goals, intentions, and for thinking, and vacillating, and being inconsistent. Indeed, as Arindam Chakrabarti has pointed out, the self has been used to conceptually make sense of the forward-looking nature of desire, and to make sense of our capacity to recognize persons and things, and to communicate and use language. So, in summary, a No-Self and/or No-Ownership view must address three general problems.
1. In terms of ethical agency, the No-Self view, at least on the surface, threatens to destabilize the distinction between self-regarding and other-regarding considerations, and this is a basic distinction that makes altruism intelligible.
2. Again, in terms of ethical agency, the equanimity that the Buddhist believes is achieved in fully realizing the No-Self view, seems, at least on one interpretation, to produce a dilemma. On the one hand, when the importance of conventional selfhood is stressed, the close and personal conventional relationship to one's own future-self seems to justify at least moderate egoism (or a privileging of one's own future welfare). On the other hand, by de-emphasizing conventional selfhood, equanimity may be achieved, but equanimity is not logically inconsistent with apathy and neglect; so de-emphasizing conventional selfhood is consistent with moral nihilism. Clearly, moderate egoism and nihilism undermine the demanding commitment to altruism that texts like Bodhicaryavatara endorse.
3. There are all sorts of problems for the No-Self view that concern broader notions of agency. For example, everything from developing a full-fledged intention, to entertaining self-doubt, anticipation, repentance, or engaging in inconsistent or illogical thinking requires some sense of persistence and duration. We require some sense of persistence and duration to even connect the phonemes that make up words, which then produce intelligible sentences. In short, the future-directed, or forward-looking nature of agency requires a sense of duration and persistence that the Ownership view more easily accommodates.
So I've basically boiled down the problem with No-Self and/or No-Ownership views to two general problems concerning ethical agency, and one general problem concerning the forward-looking nature of agency and linguistic communication. However, in fairness to the No-Self view, which has several nuanced versions, we cannot simply assume that No-Selfers deny the importance of conventional selfhood and ownership, especially when doing so produces a glaringly incoherent account of everyday, transactional reality. And yet, as stated in (2) above, it remains to be shown how a No-Self view that emphasizes an impersonal perspective escapes the egoism/nihilism dilemma. Moreover, it remains to be shown just how normativity can be derived from the revisionary metaphysical assertions various Buddhist No- Selfers endorse. As Harris has shown, this is a formidable problem, and we've yet to produce a satisfactory account of how the ontological claim of No-Self and No-Ownership entails the moral imperative that we should be committed to altruism. Accordingly, I will develop a broad sketch of Ownership and No-Self views, and examine where we might place the arguments for altruism that Buddhist philosophers like Santideva employed. I want to specifically examine how
Santideva’s argument holds up against some of the conceptual requirements that an Ownership view of self more easily accommodates.
2.3 Ownership and Individuation: Varieties of Selflessness
Ganeri has provided one of the most succinct contemporary taxonomies that organize the intersection and conversation between contemporary views of self and classical Indian No-Self views. He employs a crucial organizing principle extracted from the Indian tradition: the distinction between “place” (adhara) and “base,” or, “resting place” (asraya). These are terms of art that address both the issue of ownership, that is, the question of who owns a set of mental occurrences, and addresses the issue of metaphysical dependence, that is, the issue of accounting for that upon which a mental life is metaphysically dependent. “Place,” answers the ownership question, and “base” answers the question of what individuating ground supports an individual’s experiences. Put still another way, we can ask where we should place mental occurrences, like glee or anxiety, and how we should establish the basis of individuation. Ganeri organizes views about self into four types, and eleven basic views.
Figure 1.
Type-I: One-dimensional Views
1. Cartesian View
§ The self is both the place and the base.
§ The self is both the owner of the experience, and that upon which it adjectivally depends; just like a knot depends adjectivally upon the existence of the rope out of which it is a knot, so mental experiences are not possible without a basic substratum, or bare particular, which is the self.
2. Materialist View
§ The body is both the place and the base.
§ This ontology is monistic with respect to the underlying entities that account for mental and physical properties. However, materialism also accommodates type-identity theory and property dualism (for example, Strawson’s M and P predicates, and other forms of non-reductive materialism).
3. Reductionist View
§ The stream of psycho-physical occurrences is both the place and the base.
§ Hume and Parfit provide examples of this view: a stream of experience collectively provides mental states with both an individuating identity and an owner.
Type-II: Real Self Views
4. Ownership View
§ The self is place, and the body is base.
§ Constitution View: the relation between body and self is a constitution relation, like that between gold and a ring; that is, the body constitutes the self. However, the self is not identical with or reducible to the body.
§ Natural Self View: the self is the basis upon which there can be a mental life. It is not a Cartesian res cogitans, that is, it is not something whose essence is thinking and consciousness. Our mental lives, however, are metaphysically dependent upon a body.
§ Minimal Ownership View: the self is more aptly characterized as an embodied and invariant sense of reflexively attuned presence. o What unites all these threads is that the self is metaphysically dependent upon the body, but the self is not reduced to the body.
5. Phenomenal View
§ The self is place, and the stream is base.
§ The self may be emergent from, or constituted by, a stream of mental occurrences, but it is neither identical with nor reducible to the stream, which is its individuating base.
6. Pure Consciousness View
§ The self is place, but there is no base.
§ Self is pure consciousness, self-owning, and metaphysically dependent on nothing! For example, the “witness consciousness” (saksin) of Advaita Vedanta.
Type III: No Place (i.e., No Ownership) Views: Buddhist Schools
7. NP-1: The body is base, but there is no place (Abhidharma).
8. NP-2: The stream is base, but there is no place (Yogacara and Sautrantika).
9. NP-3: There is no base, and there is no place (Madhyamaka).
Type IV: Stream is Place
10. Tornado View
§ The body is base, and the stream is place.
§ The self is an emergent phenomenon that arises from dynamic and sufficiently complex systems. Supervening over the body, it displays autonomy. In the Tornado view, “We would then think of the mind as a tornado-like occurrence within a dynamic flow of experience. What characterizes the Tornado View, above all, is the thought that mind is an emergent macrostate of the non-linear dynamical behavior of aggregated particles, here mental events.”
11. Flame View
§ The stream is place, and there is no base.
§ The Flame View dispenses with the idea that the self is emergent from an underlying microstate of shifting patterns of aggregates. Instead, on the model of a flame, the self emerges out of a constant fusion and mutation of non-persisting particulars.
Figure 1.
The advantage of using “place” and “base” terminology is that it nuances and complicates the conversation about selves. The notion of an enduring self disappears in the Materialist and Reductionist views. However, these views embrace the idea that some sort of ownership, or “place,” is operative in the processes that produce the sense of self. In other words, “it is open to stipulate that what the term ‘self' really means, in any of these views, is whatever it is that the view takes to be the place: the body or the stream.” So Ganeri's point clarifies the radical nature of the Buddhist No-Self views. It is not just a reductionist project that revises and stipulates a distinct place of
ownership. These views, articulated as they are here in terms of “place,” deny that it ultimately makes sense to speak of any ownership at all. However, not all Buddhists maintained such a view. The Vatsiputriya Buddhists argued that some sort of “person” (pudgala) supervenes over micro¬elements, and is emergent from, but non-reducible to such micro-states. They would not embrace NP1-NP3. When Vasubandhu summarily discounts this view, of which we have a paucity of texts to draw from, he does not envision the more complex emergentist, systems-dynamical views that would gain such popularity today. Like the reductionists, the Vatsiputriya Buddhists wanted to minimize ontology by eschewing the view that an immaterial soul exists distinct from the body. However, they were not willing to reduce mental life to just a functionalist-causal model of impersonal
psychophysical occurrences. For the Abhidharma Buddhists, particularly the Vasubandhu of Abhidharmakosabhasya, sequences of tropes supervene on the physical body, and this accounts for the erroneous belief in an enduring, Cartesian self. The “mind-only” (Citamatra/Vijnanavada) schools, also known as “Yogacara,” are interpreted to be idealists of one sort or another. They are represented by NP2, in that they believe a sense of self arises out of relations internal to a mental stream, and does not supervene on a physical body. Finally, Madhyamaka Buddhism, which denies that anything has independent and intrinsic existence (svabhava), rejects any sort of real supervenience or real determining relations.
This latter view may be read in an eliminativist sort of way. For the Madhyamika, in order for X to be metaphysically independent, it needs to be free of any sort of causal or ontological dependence on anything else. Showing that nothing conceptually fits the bill of having an essence, or, having metaphysical independence reveals that all things are essentially “coreless” (asara), and appear as co-relational terms in an interdependent field of “emptiness” (sunyata). However, in a contemporary context, we can say that the Churchland eliminativist believes that we can abandon talk of the theoretical entity “eliminated” by a compelling reduction discourse. Once we find an impersonal and purely physical-functional way of describing the mind, for example, we need not speak of minds at all. That is, we can replace talk of minds with talk of neurons and/or functional
systems. I doubt that we can read the Madhyamaka project in this way. In fact, some theorists have suggested that we can read Madhyamaka philosophy as a form of deflationary- minimalism. What’s nice about this reading is that the radical, and often inscrutable, nature of Madhyamaka is reconstructed as a philosophy that prizes epistemic humility, while eschewing the need for deep metaphysics to provide justification for our practices. For the Madhyamika, all metaphysical views, particularly the belief in an enduring self, ensnare us in dichotomous and angst-inducing thought; thus, divesting ourselves of such metaphysical commitments is the key to liberation. Indeed, even the claim that all is “empty,” or, interdependent and without essence, should not be entertained as a positive metaphysical claim. This clearly supports the deflationary reading. We can embrace the conventional understanding of phenomena without believing that the entities we have commerce with arise unmediated by conceptual imputation. Of course, revealing that we impute concepts on our experiences does not mean that we have to embrace radical skepticism. It's one thing to say that something is mediated by concepts, and quite another
thing to say that, as a result, we've falsified our experience. The deflationary and minimalist reading tempers Madhyamaka by viewing it as a rejection of the traditional realist project of describing things from a purportedly God's-eye view, or, a “view from nowhere.” So it views justification as solely a product of convention and conceptual construction. All entities arise out of a web of interdependent conceptual schemes, and seeing the scheme qua scheme is presumed to be a liberating achievement that frees us from the pangs of “views” (drsti). Thus, we can read Madhyamaka as a therapeutic and self-corrective philosophy along the lines of Wittgenstein. However, all this notwithstanding, it seems uncontroversial—or problematically ambiguous—that Madhyamaka philosophy eschews the belief in any sort of metaphysical ownership or firmly-fixed individuating base, and this becomes a problem when we're trying to make sense of agency and the altruistic program endorsed by Mahayana Buddhism. Accordingly, I will now turn to Bodhicaryavatara—a text glossed as an example of Madhyamaka philosophy—and assess with more detail whether or not Santideva’s project can withstand the conceptual obstacles I addressed above.
2.4 Bodhicaryavatara: Altruism, Egoism, and Utility
The 8th century CE Buddhist philosopher, Santideva composed one of the most lyrical and sincere expressions of the Mahayana-Buddhist ethos in the Sanskrit language. In doing so, he also provided a more systematic and ethics-driven treatise that more closely parallels the generality, strategies, and aims of contemporary ethics than do other texts in classical Buddhism and Indian philosophy. In fact, Charles Goodman argues that the shift in the ethical orientation between early Theravada Buddhism and later Mahayana Buddhism lies in the latter’s more systematic articulation of values that parallel universalist-consequentialism and classical act-utilitarianism. In support of this claim,
it’s worth citing in whole the passage from Santideva's Siksasamuccaya that Goodman believes epitomizes the Mahayana project. Through actions of body, speech, and mind, the Bodhisattva sincerely makes a continuous effort to stop all present and future suffering and depression, and to produce present and future happiness and gladness, for all beings. But if he does not seek the collection of the conditions for this, and does not strive for what will prevent the obstacles to this, or he does not cause small suffering and depression to arise as a way of preventing great suffering and depression, or does not abandon a small benefit in order to achieve a greater benefit, if he neglects to do these things even for a moment, he is at fault (italics mine; translation, Goodman). Bracketing the intricate distinctions and philosophical differences that characterize the various Mahayana schools since c. 100 BCE to the present day, we can be sure that all Mahayana Buddhism converges on the general belief that cultivating “great compassion” (mahakaruna) is partly constitutive of the goal of achieving Buddhist enlightenment and Buddha status. For the Mahayana Buddhist, achieving Buddhahood is the supreme goal. While the Buddha could have remained a lonely forest-dweller, saved by his own transcendent accomplishments, he instead chose to teach out of compassion for the existential suffering of other seekers. Not only motivated to seek personal liberation, the Mahayana Buddhist-heroes, or, Bodhisattvas delay their own release in order to liberate all sentient beings from the throes of samsara (cyclic existence), and the various “hell realms” believed to exist in classical Buddhist cosmology. Bodhisattvas strive to cultivate the “awakened mind” (bodhicitta), which amounts to the cultivation of noble attainments, or,
“perfections” (paramita). By attaining unshakeable dhyana (meditative one- pointedness of mind and yogic detachment), which is the key paramita developed in chapter 8 of Bodhicaryavatara, Bodhisattvas recognize the interdependence of self and other, and they practice an exalted “exchange” or identification with all suffering beings. This is in fact so exalted a state of achievement that it is regarded as a sort of “secret” (guhya) drawn out of a cavernously inward, obscured, and not fully articulated sense of fellow-feeling. Santideva notably captures this in the BCA at 8:120: Whoever longs to rescue quickly both himself and others should practice the supreme mystery (paramam-guyam): exchange of self and other (paratma- parivartanam).
The merits that Bodhisattvas accrue through these practices are devoted to the salvific release of those who are still corrupted by the pangs of mortal existence. The revisionary/reformative Mahayana schools viewed the so-called lesser vehicle Buddhists (hmayana) of the early Theravada tradition (a designation fashioned by Mahayana) as insufficiently concerned with the altruistic spirit of the Buddha. While Theravada presumably emphasizes individual liberation in the image of the Arhat, Mahayana emphasizes the altruistic compassion of the self-sacrificing Bodhisattva. Mahayana schools further distinguish themselves by viewing achievement of Buddhahood as the greatest goal. Part of what it means to be a Buddha, they believe, is to devote oneself to universal salvation. This dedication is expressed through altruistic and consequentialist-oriented acts of self-sacrifice.
For one who fails to exchange his own happiness for the suffering of others, Buddhahood is certainly impossible—how could there even be happiness in cyclic existence? (BCA, 8:131)
Hey Mind, make the resolve, ‘I am bound to others’! From now on you must have no other concern than the welfare of all beings (BCA, 8:137). If the suffering of one ends the suffering of many, then one who has compassion for others and himself must cause that suffering to arise (BCA, 8:105). Through my merit may all those in any of the directions suffering distress in body and mind find oceans of happiness and delight. As long as the round of rebirth remains, may their happiness never fade. Let the world receive uninterrupted happiness from the Bodhisattvas (BCA, 10:2-3).
This exalted sense of concern for all beings resonates with the agent-neutral ethics and impartialism found in utilitarian and consequentialist discourses. Rather than placing emphasis on the owner of suffering, the utilitarian-minded philosopher places ultimate import on the total quantity of suffering. In chapter 1, I briefly discussed Santideva’s argument for viewing suffering in an egalitarian, impersonal, and therefore impartial light. The rational and motivational economy required to practice such an ethic is what early 20th century utilitarians like Sidgwick called, “rational benevolence.” In achieving an impersonal view, one recognizes that “from the point of view of the Universe,” her own suffering is not privileged over anyone else’s. Given that more good is better than less, one recognizes a reason to take particular interest in the total sum of good and bad in the world. And in order to actually act upon that insight, one must cultivate the capacity to sacrifice one’s own egocentric considerations for considerations of the greater good. Likewise, Bodhisattvas are called on to delay their own release until all sentient beings achieve enlightenment. They selflessly devote themselves to the good of all.
The philosophical question that arises for me is this: How, and under what conditions, would it make sense to set aside my optimal advantage for another’s advantage? In an applied-ethical context, this would have to be assessed on a case by case basis. But I’m asking how we can make sense of this structurally, that is, with respect to a general framework of reasons for doing things. As far as cliches go, we can always say, “To help others, you must first help yourself.” But we can always ask why, as a general rule, or in a normative sense, we must be motivated to help others. Being motivated to help my nearest and dearest is not such a mystery to me, but how far-out does rationality demand I extend this concern? Now, if we reverse the cliche, we might say, “To help yourself, you must help others.” We may even interpret the logic of Santideva along this line.
Those who become oceans of sympathetic joy when living beings are
released, surely it is they who achieve fulfillment. What would be the point in a liberation without sweetness? (BCA, 8: 108).
Some sort of carrot must motivate the reasonable person to delay her valid needs and aspirations for the sake of someone else’s well-being. When I act prudently, I delay gratification for myself right now for the sake of a greater balance of happiness on the whole. I do this because, ultimately, I will benefit from this in the long run. Would it make sense to delay my valid needs now for the sake of someone else, without pursuing a state of affairs in which I am also well off in the long run? What am I supposed to ultimately aim for when I engage in acts of charity and altruism? One consequentialist-minded response is that I should aim for a larger total quantity of happiness or good. But when I consider the manifest fact that I am one individual among many individuals striving for a more pleasing and fulfilling life, I cannot help but wonder why that quantity of overall good should not necessarily include my own welfare. Let me begin to develop an answer to these questions by examining some of the leading reconstructions of passages 101-103 in Bodhicaryavatara.
2.5 Bodhicaryavatara: Altruism and the Anatman Dilemma
Chapter 8 of Bodhicaryavatara may be reconstructed to endorse the following views: (1) impartial concern for the welfare of all is rationally justified; (2) all suffering, regardless of whose suffering, ought to be eradicated. Taken together, these provide a working definition of “altruism,” glossed in normative-ethical terms, rather than purely socio-biological or game-theoretical terms. The more recent cottage industry emerging out of chapter 8 of Bodhicaryavatara centers on passages 101-103. This is largely due to Paul Williams's seminal study of the text, and his controversial claim that Santideva's position not only fails, but also unwittingly undermines the Bodhisattva path it aims to endorse. A key controversy that philosophers like Williams, Siderits, Pettit, Clayton, and Harris have engaged is whether or not the No-Self view provides rational justification for altruism. While this has spawned an extensive conversation in the literature, I’d like to provide only a brief account of some key philosophical reconstructions of the said passages, which have been succinctly tackled in Harris’s essay. I’ll examine whether or not the text can stand up to some of the conceptual problems I’ve addressed above. So let me begin with the passages themselves:
The continuum of consciousness, like a queue, and the combination of constituents, like an army, are not real. The person who experiences suffering does not exist. To whom will that suffering belong? (BCA, 8:101)
Without exception, no sufferings belong to anyone. They must be warded off simply because are suffering. Why is any limitation put on this? (BCA, 8:102) If one asks why suffering should be prevented, no one disputes that! If it must be prevented, then all of it must be. If not, then this goes for oneself as for everyone (BCA, 8:103).
Verse 101 addresses the orthodox Buddhist view that person-involving concepts ultimately refer in one way or another to an impersonal continuum or “stream” (santana) of psychophysical aggregates (skandhas). But the verse also warns us not to hypostasize this dynamic complex or view the stream as an enduring and real composite entity (samudaya). Santideva is enumerating a premise here accepted by nearly all of his Buddhist interlocutors, that is, that a person is nothing over and above the individuating, stream-like psychophysical constituents, and the “stream” is not a real composite entity. Just like a string of beads or an army, a whole is merely a conceptual imputation (prajnapti), derived from ultimately independent elements conceived to appear as parts of a whole. In fact, the Abhidharmika Buddhist who adopts the "momentariness" thesis
(ksanabhangavada) believes that every dharma, or, metaphysically atomic entity arises as an independently ephemeral point-instance. The arising of a new dharma, which in some way or another appropriates certain aspects of its predecessor, follows from the perishing of a preceding dharma. These dharmas, which are species of the five basic psychophysical aggregates, serve as the reduction-base for all entities. Given their radically ephemeral nature, we can attribute the appearance of persistence of these seemingly real composite entities to conceptual construction. These dharmas are not parts of a real whole, nor are they the properties of an underlying subject; thus, mental occurrences are not ultimately properties, because no enduring subject exists to bear them. What we have here are ephemeral property-tropes. So, if the self is a composite entity constituted by, or emergent from the five skandhas, and composite entities are only conceptually constructed fictions (mrsa), then the self must in turn prove to be a sort of fiction or error.
So, as Santideva claims in 102, without a real self, there is no owner (svamika) or ultimately enduring locus of suffering, and if suffering ought to be removed, then it ought to be removed simply because it is suffering. Santideva is addressing the egoist who believes that a quantity of suffering being "mine"
relevantly restricts practical considerations, and justifies privileging one’s own future welfare. The fact that my burnt hand provides me with compelling practical reasons to tend to myself in a way that hearing about a stranger burning her hand in New Mexico does not might compel me to believe that, in general, my own interests are especially privileged. Santideva counters that without the restriction of real ownership that comes with the existence of enduring subjects, there are no ultimate limitations (niyama)
that justify partial concern for one’s own or another’s suffering. If suffering ought to be prevented, then it must be prevented because it is inherently bad. If one were to object to that point, as Santideva points out in verse 103, then he or she would be committed to the view that no suffering ought to be prevented. However, this latter option is a reductio that allows Santideva to derive his conclusion that one must remain committed to preventing all suffering. As in all sutra literature, these pithy passages require detailed unpacking. I will therefore provide a brief sketch of leading reconstructions of the argument in order to better situate my own gloss on Santideva’s project. For Paul Williams, Santideva commits a basic fallacy. He makes the ontological, descriptive claim that real selves are nonexistent, and then tries to derive the normative conclusion that we must be committed to altruism. On the surface, then, Santideva wants to appeal to our rationality, but cannot successfully bridge the is/ought gap. As Harris has pointed out, “Drawing upon the Tibetan commentarial tradition, Williams interprets Santideva’s argument as an appeal to our rationality. Once we understand reality correctly, and accept the self is nonexistent, it will no longer be rational to remove one’s pain before removing the pain of others.” A general problem I have with Williams point is that I think he exaggerates and oversimplifies the ontological weight of the argument. The idea that selves do not exist at the “ultimate” level of analysis is not necessarily a purely descriptive claim that carries no normative force. The Buddhist-hermeneutical device of distinguishing between
conventional truth (samvrti-sat) and ultimate truth (paramartha-sat) should not be so straightforwardly glossed as the Platonic distinction between appearance and reality. It’s not so clear that in these passages Santideva simply wants to make the truth-functional claim that persons in fact do not exist, and that a presumably more accurate description of reality will not include person-involving concepts. It’s not just that when we talk in person-involving ways we’re not “getting it right” in an absolute sense. A more literal interpretation of the concept, “samvrti,” would be that of a “concealing” or a “cloaking” truth, thus admitting some degree of concept-mediated universality. Ganeri has provided an excellent alternative to the typical subjective/objective cum appearance/reality dichotomy often foisted on to these Buddhist concepts. The cloaking truth, and its cousin, the conceptually fashioned truth (prajnapti-sat) are not necessarily “subjective” or purely falsifying truths, so much as they are indicative of observer-dependent, conceptually-mediated positionality. Viewing an impressionist painting from a distance, I see a grassy field fashioned out of vivid colors and delicate forms. I see couples strolling along under parasols. However, as I close the distance, what seemed like solid forms become tiny points of color. I lose the image of couples enjoying the bright and cloudless day. The images from the painting can be said to lose “stability under analysis.” But that does not mean that the images emerging at a distance are purely personal and subjective images that falsify reality. If any fiction (mrsa) is involved, it’s the belief that the person-involving level of description is a solid, unanalyzable fact that
stands on its own. Once we achieve a non-person-involving level of analysis, we come to appreciate an impersonal perspective that levels off viewing things in terms of owners and bearers of properties. We come to recognize an impersonal field of relations. The self, which is an experienced aspect of reality when one “zooms out” at a certain cognitive distance, is not as much a perniciously falsifying concept as the rigid view (drsti) that the self is an enduring substance that exists with an essence (svabhava), independently of its causes and conditions.
Some of the leading interpreters of the text, while diverging on important details of the argument, agree that Santideva, the Madhyamika, is provisionally adopting the Abhidharmika position here. Remember, the reductionist-minded Abhidharmika, like Vasubandhu, believes that there is no self over and above the stream of psycho-physical constituents. However, for these philosophers, the constituents like form (rupa) and affective-coloring (vedana), for example, remain stable under analysis. When we shift our cognitive frame, these constituents remain unanalyzable and basic. Santideva, the Madhyamika, adopts the Abhidharmika’s position as a dialectical tactic. Showing that from both the Madhyamaka and Abhidharma perspective one must be committed to altruism allows Santideva's interlocutor to embrace altruism without necessarily rejecting his prior doctrinal commitments. If we accept this reading of Santideva, then things like suffering—at least in the context of the argument in chapter 8—are fixed aspects of unanalyzable dharmas such as affective-coloring (vedana). Thus, from the cloaking level of reality, selves exist, while from the stable or unanalyzable level of description they disappear. Suffering, however, remains a stable aspect of the unanalyzable dharmas.
It’s important not to confuse Ganeri’s concept of stability and non-stability under analysis with Madhyamaka anti-foundationalism and irrealism. If, through a purely deconstructive analysis, Madhyamaka claims that the ultimate truth is that there is no ultimate truth, Vasubandhu’s distinction between the two truths is not irrealist and anti-foundationalist, nor is it eliminativist in nature. Rather, his view is reductionist, with the caveat that person-involving concepts, and therefore, concepts concerning bearers of suffering, provide a useful enough glimpse of reality that universally shows up at a particular level of cognitive-framing. However, from the analytically stable perspective, owners of suffering completely disappear. This means that instead of solely examining whether or not one view is “more objective” in nature, which is of course
appropriate in certain contexts, we should also be asking ourselves what the pragmatic upshot of viewing the world in one way rather than the other amounts to, and we can do this without succumbing to full-blown relativism or Madhyamaka anti-foundationalism. In other words, one perspective is not simply "better" than another just because it captures more basic and unanalyzable aspects of reality—it's not as though we should claim that the physicist is "more important" than the biologist, because she deals with more basic items. The biologist deals with complex, organic systems that are not amenable to talk of quarks, neutrinos, and black holes. Pragmatically, we can ask ourselves what we get when we operate from the unanalyzable conceptual perspective as opposed to the conceptually unstable perspective. Now, it's part of the basic message of Buddhism that the person-involving level generates attachment and grasping that produces suffering. Buddhist interlocutors already accept the normative claim that suffering is something bad and ought to be overcome. So any Buddhist will agree that we should strive to avoid seeing the world from a perspective that contributes to suffering—that's just part of
the commitment that comes with asserting that all suffering is bad. To be sure, under the Abhidharma banner, there are things that really exist, and things that really do not exist. However, the point of emphasis is not solely on the ontology. Buddhism always remains pragmatic to one degree or another. So where does this leave Williams's claim that Santideva fallaciously tries to derive an ought from an is? If we can achieve an impersonal view, then the egoist's claim that a distinction between self and other justifies egoism does not apply at this level of analysis. When Santideva ontologizes here, he does so to undercut the egoist's sticking point that the distinction between self and other—particularly the first-person experience of pain versus a third-person description of another's pain—justifies privileging one's own future welfare. From the metaphysical level, no such distinction applies, and given the commitment to end suffering, we ought to maintain this perspective
to the best of our abilities. The normative force of the Buddhist view on suffering allows Santideva to derive his desired conclusion. The speech-act of sincerely asserting X comes with a commitment, namely, a commitment to act in light of that truth, and accept any truths derivable from it. So, if we find that the most stable analysis does not include person-involving concepts, then we are committed to accepting any truths derivable from that assertion. Likewise, if we accept that suffering is intrinsically bad, and ultimately real (that is, stable under analysis), then we're committed to accepting any derivable claims from that assertion. Part of what it means for something to be bad is that it should be eradicated if possible, and it's part of the Buddhist worldview that we should avoid perspectives that produce suffering. Thus, we can adopt an impersonal view, and given the Buddhist thesis on suffering, we ought to maintain that perspective since it arguably defuses suffering. The egoist's primary justification for egoism disappears from the analytically stable perspective. Santideva is not deriving a normative from a descriptive claim. Rather, he's addressing an interlocutor who argues that the only sensible limiting condition that justifies partialism with respect to suffering is ownership. But from the conceptually stable perspective, ownership disappears. However,
from this same, stable perspective, suffering presumably remains intact, and the concept of suffering already has normative force, and carries with it a host of commitments. Nevertheless, Williams has pointed out a much deeper problem with Santideva’s argument, and Siderits has in turn come to Santideva's defense in a strategy similar to what I’ve sketched above. I will briefly develop these points, and then assess Harris’s analysis of the debate, because I think ultimately Harris cum Williams have outlined a dilemma in Santideva's argument that is not so easily defused. What I would add is that, specifically with respect to other-centric and altruistic considerations, bridging the gap between self and other in a normative context is best served by concepts of identity and real relationality, and not by concepts like “emptiness” and non-ownership. In other words, the integrity between self and other must be maintained—and not just as a matter of “conventional reality”—in order to impel one to find a motivating and ethically-relevant bridge between self and other.
Now, Williams and Siderits argue that Santideva is employing the Abhidharmika perspective as an example of skillful, or, pragmatic means (upaya). Abhidharmikas are mereological irrealists, that is, they do not believe that a whole exists over and above the parts that are thought to make it up. However, they do believe that there are ultimate physical and mental events, or, dharmas that are the really existing building blocks over which conceptual imputation operates. As Ganeri has more recently shown, Abhidharmika philosophers like Vasubandhu may very well believe in the conventional validity of medium-sized objects along with the validity of the conventional self. Vasubandhu would instead argue that our belief in the self as an enduring entity, which exists independently of its causes and conditions, is a pernicious fabrication that produces suffering.
With this in mind, Williams argues that Santideva must, nevertheless, deny the validity of the conventional self. He argues this on the basis that all we need to potentially prioritize the interests of person A over person B is the ability to distinguish between A and B. In the conventional context, we can make this distinction. Thus, accepting the
validity of the conventional self is enough to justify at least moderate egoism, where one believes that it is sometimes perfectly justified to prioritize one’s own welfare. Put still another way, moderate egoism claims that it is not immoral to prioritize one’s own future welfare. This is enough to undercut a demanding ethics of universal altruism. If Williams is right, then Santideva must reject the validity of the conventional self in order to maintain the view that we ought to be unwaveringly committed to altruism. This means that Santideva cannot just attack the belief in an enduring metaphysical self, but must also deny the validity of the conventional and transactional self. But this is a glaring problem for Santideva's argument. The conventional, transactional world is where things like altruism and egoism make sense. Agency in general—and ethical agency in particular—draws on the sorts of conventions that only make sense at the transactional level of reality. This is the level of reality in which I do not just see suffering, but I see that you are suffering, and that I am in a position to help you, or sacrifice my interests, or simply ignore your pain. Moreover, I am in a position to identify the fact that not all pain is my pain, and that you have a stake in your life that is worthy of my attention and respect. Thus, by denying the validity of such a world, Santideva renders altruism unintelligible.
Harris tentatively objects that we need not foist this view onto Santideva. He offers a quick gloss on Barbra Clayton's interpretation of passages 101-103 to illustrate the point. Clayton sees a parallel between Santideva's argument and anti-discrimination arguments. If all beings equally desire happiness and freedom from suffering, and it is ethical for me to ultimately prioritize my own welfare, then I need to provide an ethically relevant distinction that justifies my bias. If my future self that I am concerned with turns out to be identical to my current self, then that would provide the distinction that justifies my temporal neutrality with respect to my own happiness, and my bias with respect to the happiness of others. But in the light of Santideva's argument, my current and future selves are not really identical, but only conventionally so. Thus, I cannot provide the ethically relevant distinction that would justify my partiality with respect to the common good of happiness. If I regard suffering as an ill to be removed
in my own case, then I ought to apply that view to all cases of suffering. As Harris writes, “Santideva can be seen as claiming that although conventional selves exist, when we realize they are only convenient fictions we will accept that we should not prioritize our own conventional self's welfare above that of other persons.” But, as Harris points out in the context of Williams's argument, when we emphasize the conventional nature of the self, we might not only learn to become less fixated on our own suffering, but also learn to become less fixated on anyone’s suffering. In other words, altruism is not the logical outcome of such a belief. One may lose all concern for suffering in general. Verse 103 may be interpreted as a response to this sort of problem. Suffering is not something that can be viewed as morally neutral: it is not something we can ever really feel justified in ignoring. As Bodhicaryavatara puts it: “No one disputes that!” So Santideva appeals to commonsense. But, as Harris points out, the opponent “can respond that most people agree to this because they believe their self, and the selves of the ones they care about, endure.” While it might be the case that anatman provides altruistic motivations for some, it may very well provide no such motivations, or even apathy, for others. If Santideva were to respond that psychological connectedness and physical
continuity motivate us to care about our conventionally established future suffering, then he would play into the hand of the egoist, who claims that close ties to our future-self warrant special concern for our future welfare. Thus, Santideva’s argument faces a dilemma.
However, the dilemma developed in Williams’s argument relies on a crucial assumption: our obligation to remove pain requires that there be selves to experience this pain. In particular, the apathy/moral nihilism horn of the dilemma assumes that without the belief in an enduring self, and without emphasizing the importance of psychological connectedness and physical continuity, we lose motivation to care about any suffering. But why do we need these views to motivate the cultivation of compassion and other-centric action? According to Siderits,
Santideva rejects this key assumption. Here's a reconstructed version of Siderits's interpretation:
1. The person, being an aggregate, is ultimately unreal.
2. Hence if pain is ultimately real, it must be ownerless or impersonal.
3. We are in agreement across the board that pain is bad (even though concern over pain is often restricted to one's own case).
4. Either pain is ultimately and impersonally bad, or no pain is ultimately bad.
5. But it's not the case that no pain is ultimately bad; therefore, it is impersonally and ultimately bad.
So given (5), we can say that pain is impersonally bad, which just means we have no reason, other than pragmatic considerations, to restrict our concern to ourselves or our nearest and dearest. As Harris glosses Siderits, "It is our commitment to removing this pain, which ultimately exists impersonally, that leads to us adopting personhood conventions in which we identify our current and future collection of causal constituents as being the same persons." Siderits makes the original and controversial point that personhood conventions are useful for removing suffering, since, often times, we are in a unique position to successfully alleviate the suffering that appears closest to us. Thus, personhood conventions become a stratagem for removing pain, which is ultimately impersonal. We can gauge the extent to which we must pay attention to ourselves and our nearest and dearest by employing a consequentialist calculation of one sort or another.
Harris points out two problems with this interpretation. First, as Williams has claimed, it is questionable that most Buddhist philosophers would buy that personhood conventions are useful devices employed to remove suffering. The orthodox view is that belief in an innate conception of self, that is, identification with the ego and its ambitions, produces self-attachment and grasping, and this contributes to existential suffering. This sort of self-attachment motivates, and/or, works in tandem with the belief in a “further fact” or metaphysical presence underlying the flow of mental occurrences. Personhood conventions do more to entrench self-attachment and ego-identification than they do to facilitate the removal of pain. However, a Buddhist following Siderits’s strategy might object that the misidentification of the stream as an enduring person, which
contributes to fixation on the ego and its projects, causes suffering. But the practice of Buddhist detachment and “right views” prevents one from clinging to selfhood. Without such clinging, suffering does not arise. This is perfectly consistent with Siderits’s claim that personhood conventions are stratagems for removing pain and suffering. The second and more controversial problem that Harris points out is that it is questionable whether or not pain is inherently bad. One might distinguish pain from suffering, and claim that impersonal pain is not really bad. Only the grasping that comes with believing in a really enduring self—rather than recognizing that the self is a conceptual imputation—transforms pain into suffering. This line of thought relies on accepting the distinction between pain and suffering. On the one hand, pain is not ultimately negative
—only the interpretation of pain as belonging to a really existing and enduring person is negative, and this sort of belief transforms pain into suffering. On the other hand, suffering is illusory, since it disappears upon realizing the right view. So there is no real need to eliminate pain, and if we can achieve this remarkable realization, then suffering, as it were, simply falls away. This line of thought would be damning to Santideva's argument, since it relies on the normative claim that all pain/suffering is inherently bad.
However, even if this quasi-Stoic line of thought is somewhat convincing, it's hard to argue that no pain is intrinsically negative. Mental pain like depression, for example, is not so clearly analyzable as a relationship between some initial quantity of physical pain, and a personal interpretation that then generates angst and full-blown suffering. Depression is just depression, even if it is the result of cognitive framing of one sort or another. In other words, the pain and suffering of depression are co-temporal. Unlike the case of physical pain, it is conceptually difficult to distinguish between the suffering that comes with seeing the world through melancholy eyes, and the raw data of a more cognitively neutral sensation. Until that more neutral data is drawn into a melancholy portrait, it is inappropriate to call it “pain” at all. So we can at least assume with Santideva and Siderits that some mental states are intrinsically negative, regardless of whether or not they belong to persons.
But nevertheless, as Harris points out, bridging the is/ought gap still remains a problem for Santideva's argument. Certain mental states being inherently negative do not necessarily entail the moral conclusion that we ought to remove them. Harris believes that Santideva anticipates this point, and thus appeals in 103 to ethical intuition. “Everyone agrees” that suffering needs to be removed. In other words, this basic belief is not up for debate. In response, Harris levels a cherry-picking argument. While endorsing anatman purportedly undermines the ethical intuition that supports egoism, it also threatens to undermine the ethical intuition that all suffering is inherently bad, and we cannot just cherry-pick which ethical intuition to take seriously.
I believe this point needs to be assessed a bit further. First, when appealing to commonsense, we can only dispense with ineluctable intuitions on a case by case basis. Harris claims that ethical intuition supports two principles: the Principle of Moderate Benevolence (PMB), and the Principle of Moderate Egoism (PME). PMB agrees that we ought to remove anyone's suffering provided that it is not overly demanding, while PME claims that it is not immoral to prioritize one's own future welfare. One can consistently maintain both principles. PME just says that it is not necessarily immoral to prioritize our own welfare, not that we must always prioritize our own welfare. Harris's argument is that if the truth of anatman is meant to undercut the intuitions supporting PME, then it can likewise do the same to PMB. At the analytically stable level that frames the anatman view, PME may fall away, but the fact that suffering is bad no longer seems intuitive at this level of analysis.
It’s not that Santideva is just cherry-picking between ethical intuitions. Rather, Santideva is cherry-picking between the levels of analysis he draws from. PMB is most intuitive from the conventional or “unstable” level of analysis. When Santideva undercuts PME by moving to the more analytically basic level of analysis, there’s nothing that seems to prevent both PME and PMB from falling away as intuitive claims.
However, Harris does not seem to have adequately entertained the possibility that the intuition that impersonal suffering ought to be removed is more resilient than either PMB or PME at the analytically basic level of analysis. Remember, PMB and PME are “moderate” and mutually consistent principles. However, the impersonal suffering thesis is a demanding principle that claims that all suffering, no matter whose, ought to be removed. Could this make any sense at the impersonal or analytically basic level of analysis? Harris claims that he is not clear “that we have any intuitions about impersonal suffering.” But if we accept that Santideva employs “skillful means” through embracing Abhidharma metaphysics, there is nothing to stop us from interpreting him to embrace other Buddhist metaphysical principles that would strengthen his argument and bolster its rhetorical force. The metaphysical view that comes to mind for me is the view that all mental dharmas are inherently self-reflexive (sva-prakasa), and in this sense, self-
aware. What this means is that any mental occurrence is both aware of an object, and aware of its own awareness of the object. So, if there really were quantities of suffering “just hanging there,” without a real subject who bears them as properties, then we can see how each suffering moment would to some extent be aware of its own suffering. The self-reflexivity view is, among other things, a metaphysical device that helps make sense of the phenomenal feel of “mineness,” which then, along with other theoretical principles not pertinent to the discussion here, better explains the psychological connectedness of various mental occurrences. In short, there is a sense or trace of ownership, but that does not mean that we have to believe in a really existing and enduring subject who owns that suffering. More importantly, we can imagine point-instances of suffering that take on a first-person feel. In other words, there is some form of self-awareness that accompanies instances of suffering. Perhaps, contrary to how Williams and Harris have stated things, endurance and psychological connectedness are not the most important factors that motivate compassion and concern for instances of suffering. Self-awareness seems to be the most basic fact about suffering that makes it a moral ill. In fact, suffering (as opposed to damage or corruption) seems unintelligible when we do not
include self-awareness as a basic component of the phenomenon. The self-reflexivity principle makes sense of self-awareness without appealing to any real or conventional ownership. While PMB and PME may fall away as intuitive concepts at the impersonal level of analysis, the nastiness of instances of self-aware suffering seems resilient to whatever level of analysis we operate from. Harris is correct that there's a sort of cherry-picking going on here, but I do not think that we lose all intuition about our view of suffering when we operate from the impersonal level, so long as we are committed to other metaphysical principles that would allow us to believe that self-aware suffering can arise as a point-instance in a flow of ephemeral tropes. Now, this may seem to play into the hands of the dilemma, because Harris or Williams might argue that by stressing ownership, continuity, psychological connectedness, and other such conventional notions, we may embrace, at very least, PME if not full-fledged egoism. But the only “ownership” and connectedness employed here is one where we can make sense of the reflexive awareness of each momentary flash of suffering, and that means that as long as there is the
experience/awareness of suffering, we have room to compassionately care about any point-instance of suffering. While this might not be intuitive, it provides a theoretical and sophisticated bit of machinery that can support our already entrenched intuition that suffering is a nasty thing no matter who endures it. However, the same cannot be said for the egoistic intuition. If there is no enduring entity that ultimately reaps the benefits and endures the pain of a set of current actions, then it seems less important to be fixated on who exactly endures the suffering. That does not mean that we do not care about suffering whatsoever, we simply care more about the overall quantity of suffering and good in the world, which is a view that aligns nicely with consequentialism and utilitarianism. Bear in mind that this assumes, as Harris allows, that it makes sense to say that suffering is inherently bad. To say that it is impersonally bad is just to say that we do not need person-involving concepts to justify our concern for suffering. We can simply think in terms of quantities of a type of mental occurrence that, being inherently self- aware, perpetuates a larger quantity of inherently negative experiences. Harris’s problem is that “Santideva has shifted the burden of proof to his opponent by pointing out that there is a common acceptance that suffering ought to be removed. The opponent, however, may respond that once anatman has been accepted, it is irrational to accept any intuitions that arise in dependence on the belief in enduring persons" (110). Presumably, then, the belief that suffering is inherently negative is one such intuition that depends on some investment, however attenuated, in the endurance of persons. I'm taking exception with that claim. Instead, I'm focusing on the awareness and self-awareness of suffering, no matter how attenuated, as a motivating reason to prevent suffering.
While I think that Santideva's argument may be able to withstand the cherry-picking critique, I think that read as a Madhyamaka text, it does not supply the sort of systematic ethical clarity that Goodman imagines it does. By chapter 9 of Bodhicaryavatara, Santideva endorses the radical teaching of Madhyamaka irrealism. Remember, this view eschews any notion of "place" or "base" (see figure 1). And yet, he claims that we need to embrace a conventional view of self to make sense of altruism and Mahayana compassion (see chapter 1). This seems to indicate that when we graduate past the conventional view of things into the impersonal view, things like altruism and ethics disappear. So, in some sense, Williams and Harris are correct that the ultimate level of ontological analysis, at least from the Madhyamaka perspective, does not come embedded with any normative
implications. Santideva, himself, seems to agree by emphasizing that the altruistic motivations of the Bodhisattva require a conventional view of things to make sense of agent-driven concepts like karma, retribution, and kindness and compassion. If my response to Harris above is at all satisfying, then it relies on a great deal of metaphysical machinery, including a "base" of individuation (see figure 1) that the Madhyamika would not ultimately endorse. Without ownership (“place”) and without any metaphysical individuation (“base”), it's not clear what guiding principles we should appeal to when considering the treatment of others, and the value of their suffering. This may not be as much a problem for figures like Hume and Parfit, since some notion of “base” is still in play. However, the same cannot be said for Santideva. Santideva needs a conventional set of transactional notions to make sense of ethical requirements, and that is precisely why he recommends that we embrace the “delusion” of conventional reality for the sake of being guided by altruistic principles (see chapter 1). But this does not clarify what sort of principle would motivate him to take the fabrications of conventional reality
seriously to promote universal salvation. Any principle that would do that sort of work would only make sense in conventional reality. As a purely transcendent and amoral philosophy, this would not be a problem; but the same cannot be said for an ethically-oriented philosophy that has any normative teeth. Finally, although I appealed to self-awareness rather than endurance to make sense of the normative force of suffering, and to make sense of why we might be motivated to care that anyone is suffering, endurance and persistence are still key concepts that make sense of agency construed more broadly. From reasoning, to language acquisition and linguistic communication, to anticipation and responsibility, we need some account of ownership and persistence. The Madhyamika might be able to appeal to convention to make sense of how we come to operate in a domain of transactional reality, but the state of “perfect wisdom” he ultimately recommends does little to make sense of why we ought to care one way or the other about the suffering that we find in that domain.
2.6 Toward Ownership and Other-centricity
I have devoted considerable attention in this chapter to the Mahayana philosopher, Santideva, because his Madhyamaka text provides a novel approach to an ethical theory of altruism. We can call this, “emptiness ethics.” While we can certainly find consequentialist-utilitarian streaks in Santideva’s text, aspects of his approach are unique and not so readily accommodated by modern and contemporary Euro-American ethics. The analytic and decision-procedural schools require a rather robust notion of agency and practical identity to justify their emphasis on universal principles, practical rationality, freedom, responsibility, and punishment. While Santideva’s concerns may be read in a way that
is consonant with some of those aims, he does not provide enough resources (or the usual sorts of resources we find in Euro-American ethics) that would support such an endeavor. His uniqueness lies in trying to make “emptiness” (of ownership and individuation) do the work that contemporary principles-based ethics does through theories of real agency and practical identity. On the other hand, while the non-principles based schools—like the situational and care-ethical schools—emphasize relationality and interdependence, the self is not so effaced in these schools, because that would undercut the very partial concerns that prioritize intimacy, family, and friendship. Nor can it be so effaced that it threatens to undercut one’s capacity to respect and take the other on his or her terms. And yet, Santideva tries to build other-regarding concern and
ethical responsibility out of a radical effacement of the self and its practical identity, effacement so radical that partialism toward lovers, family, friends or neighbors is rejected —thus, preventing Santideva's ethics from conversing productively with care-ethics. Santideva explicitly aims for an impersonal, impartial, and agent-neutral ethics. So, on the principles-based side, he neglects a robust enough account of agency to credibly make sense of responsibility and ethical action, and on the non-principles side, he effaces the self so radically that it's not clear what (or who) can be self-possessed enough to invest in its immediate relationships of care and its situatedness in a nexus of caring relationships.
Santideva, and more recent No-Selfers like Parfit, deny the belief in a Cartesian soul or substance, in order to better accommodate and respond to the needs of the other. Cartesian dualism threatens to produce an egoistic self- enclosure—both a metaphysical and an ethical solipsism—that is only bridged by reducing other-regarding considerations to an extension of egocentric, self-regarding considerations. This sort of metaphysics generates the problem of making sense of altruism and other-centric considerations from an inescapably egocentric vantage-point. The selflessness thesis in Santideva aims to undercut the sort of metaphysics that generate inescapable egoism. Madhyamika Buddhism
does this by denying the reality of either ownership of cognition and volition, or a base of individuation. While Derek Parfit orients his ethics in a similarly non-Cartesian direction, his reductionist project still remains grounded in an individuating base— namely, the body—which helps orient the ethical act by providing us with enough of a distinction between self and other to make sense of the terms involved in an ethical relationship. In the next chapter, I will further explore the sort of “emptiness ethics” we find in Madhyamaka philosophy, and examine whether or not this sort of research project has some more promising insights and moves available to it for addressing the critique I’ve provided so far.
I have argued that ethically altruistic action requires the concept of a self, which owns its cognitive, affective, and volitional states. I make this claim on two accounts. First, I have argued that we need a clear self/other distinction in order to both generate other-orientedness, and intelligibly place our acts of compassion and generosity in a space not fully describable in egocentric terms. Without the Ownership view of the self, the self/other distinction is compromised, and the other for whom we become responsible
loses his or her independence. Instead, the other becomes a complicated extension of the ego. Second, I’ve tried to show that the No-Self strategy destabilizes our sense of genuine agency, including the agentive will to act against the basic egoistic inclination of the natural, self-possessed person. We need a self that owns its mental states to make sense of both agency, and the capacity to recognize the validity of another’s needs independently of immediate self-interest and desire. In short, the No-Self view, while sounding friendly to unselfish action, fails to provide us with genuine other-orientation and genuine agency. Therefore, other-orientation, which is the ground of altruistic action, requires an Ownership view of the self.
Are there other Non-Ownership views that not only make sense of altruistic action, but also show how the Ownership view comes up short? I will tackle this question in the next chapter, by providing a hybrid No-Self view. This will be my attempt at preserving ethical obligation, and the lived experience of caring for another, without appealing to a real self or an Ownership view.