Printer Friendly

Buddhist reductionism and free will: paleo-compatibilism.

It is commonly accepted that different neural-state configurations might instantiate that same volition and that other instances of the same neural configuration might lack that volition if an alternate-worldinhabiting (atom-for-atom-identical) doppelganger could lack consciousness altogether. To the extent semantic dualism is akin to mind-body dualism, it shares such mapping problems. Anyone who believes in free will seems implicitly committed to the reality of conscious mental states, and neuroscience places a burden on the folk psychologist to come up with something like a supervenience relation. To Siderits's credit, he offers plausible arguments for reductive supervenience; but they are incredibly complex and subtle, and this issue is too problematic to resolve here.

Thus, just as it may beg the question (in the "looser" sense that the premise is as problematic as the conclusion, as opposed to question-begging in the tighter sense of circular reasoning), contextually, to appeal to the problematic two truths doctrine to support a position on free will, so too it may beg the question to appeal to supervenience to support semantic dualism. We now have stacked puzzles--free will, two truths, semantic dualism and supervenience. Such question-stacking appears as a weakness, but in fairness to Siderits it may be--and I think it actually is--indicative of the broadly coherent, multidimensional explanatory purchase of his theory. It ought not to be incumbent on readers of Siderits's individual articles on free will to digest his entire corpus on Reductionism in order to assess his amicus brief on behalf of paleo-compatibilism, and we can only devote so much space to that larger work here. Those who wish to more fully explore this larger project are encouraged to direct their attention to Siderits's brilliant treatise, Persons. Having noted the extent to which paleo-compatibilism is grounded in this larger dialectical and explanatory framework, then, let us focus the remainder of the assessment of paleo-compatibilism on its articulation in Siderits's articles on paleo-compatibilism.

Siderits says that at the ultimate level "there are events that correspond to what we ordinarily call deliberating and willing" ("Beyond" 155), and this is why he thinks the volitional analysis applies at the ultimate level. Does free-will-supporting conventionalese about volitional regulation map onto ultimatese about micro-level volitional phenomena, and does conventionalese about persons map onto ultimatese about person-series, such that conventionally responsible agents reduce to ultimate series that ground their conventional reality? This seems to be what Siderits intends, and that would be a welcome argument from the compatibilist perspective, but in light of the complexity of everything he says that bears on the subject, it is not entirely clear.

Siderits's suggestive approach has great merit, but some of that merit is obscured by the complexity of its many details:
   To see myself as capable of acting freely, I must view myself as
   the sort of entity that can endorse the actions it performs. This
   requires that I take deliberating and will ing as abilities or
   faculties that I possess. This possibility is denied us at the
   ultimate level of truth. (155)


Siderits thinks "this possibility is denied" because "persons" do not exist ultimately, though we exhibit these abilities. Siderits continues:
   At that level there are events that correspond to what we
   ordinarily call deliberating and willing; but there are no persons
   who are the authors of those events, for a person is a mere
   conceptual fiction that is constructed out of those and other
   events.... Thus, it is only at the conventional level that an
   action may be said to be performed
   freely. (155)


But if "a person is a mere conceptual fiction"--a mental entity--"that is constructed out of those and other events," where those events are mental, that is like saying shoes are not leather because there are no shoes, only leather-shoe parts. (41)

There seems to be an implicit asymmetry in the treatment of items here if, say, conventional volitions are said without qualification to correspond to ultimate events, but persons and free will are qualified as being "only" conventional. If there are ultimate events that correspond to conventional volitions that, arguendo, might satisfy some criteria for free will, then cannot ultimate events also correspond to conventional free will? If there are events at the ultimate level that correspond to deliberations, then there are arguably events at the ultimate level that correspond to the sorts of things that constitute responsible agency--that correspond to free-will-exhibiting agents (persons). Conversely, the same reason to think there is no ultimate free will or ultimate person may be used to conclude that there are no ultimate deliberations, as opposed to, say, deliberations-series. Persons are changing series of psychophysical events, just like deliberations are, so why is there reductive asymmetry between them? Asymmetry--per se--is fine, but not when things seem symmetrical.

The paleo-compatibilist might again appeal to the graded macro/micro, conventional/ultimate spectrum to explain any appearance of asymmetry: persons, person-series, deliberation, deliberation-series, volition, volition-series, and so on, down to dhammas and emptiness. Perhaps our shifting focus along this spectrum is what is generating the asymmetry, not the theory. Thus, let us examine this more carefully by focusing on just one centrally relevant item here, deliberation.

Deliberation is minimally dyadic: it minimally requires consideration of two alternatives. Even if superior beings apprehend many alternatives simultaneously, we do not. If each consideration corresponds to ultimate micro-events, the only way this micro-series can count as "deliberation" is if something links them together, even if that something is just a combinatorial or causal pattern/formula. Similarly, it seems there can be no "mother" if there is nothing that links together the bearer of a zygote and the woman who delivers a baby (Miln). But if this something does not exist at the micro-level, the claim that deliberation corresponds to ultimate events is problematic. If it does exist at the micro-level, then ultimate reals form micro-level deliberation-series that ground the conventional reality of deliberations; however, then ultimate reals could equally form micro-level responsible-agency-series that ground the conventional reality of responsible persons. At issue is this asymmetry in mereological reduction.

In his original article ("Beyond"), Siderits implicitly accepted mereological reductionism; now he explicitly asserts that "Buddhist Reductionists are thoroughgoing mereological reductionists" ("Reductionism"

34). This implies that "persons are reducible without remainder to completely impersonal entities because all partite entities are reducible without remainder to their parts" (34). But then the question is: Why are some reductions privileged, but others not? Buddhist mereological reductionism is articulated classically by Nagasena: chariot parts have no relationship with each other independent of those imposed by our needs, so chariots are not natural kinds (nothing inherent in them grounds the name "chariot"), but pragmatically justified linguistic convention names their configuration "chariot," as with the label for the psychophysical series "Nagasena," the configuration of elements of which lacks self-nature (Miln).

Mechanical artifacts intuitively lack any whole-level feature that is irreducible to their parts, but Nagasena's intentionality and consciousness seem irreducible to their parts. This apparently compositional difference would at least provide some ground for mereological asymmetry, but Buddhist Reductionism denies any such ultimate difference here, as Buddhism more generally decomposes even consciousness and the like into the interdependent interaction of sense-organs and objects of experience. If "all partite entities are reducible without remainder to their parts," however, then just as there is equally no chariot or person, so too there are no Buddhas, volitions, karmic merit, thoughts--eyes reading this--or anything apart from the scientific or Buddhist version of quanta, possibly also excluding quanta if they are divisible even into homogeneous parts. But as with any other reductio ad absurdum argument, the allegedly absurd conclusion may be embraced, and Buddhist Reductionists do embrace this conclusion. They dispel its apparent absurdity by saying that all these things that technically do not exist in ultimate reality nonetheless do exist conventionally. This suggests an ironic use of cliche: Buddhist Reductionists apparently can have and eat their (ultimately nonexistent) cake too.

Reductionists claim the person is a mere conceptual fiction, but maybe reductionism is in a similar category. Reduction involves identifying micro-constituents as ultimately real, but those micro-constituents must either be indivisible/impartite entities (atoms) (42) or divisible/partite entities (nonatoms). (43) Both options are deeply problematic. (44) If micro-reality is nonatomistic, then for mereological reductionists all macro- and micro-levels are partite--equally unreal--and this ontologically egalitarian option unveils no ontologically privileged ultimate reality. Thus, mereological reductionists--who affirm that all macro-level partite wholes are ultimately unreal and that only micro-level impartite entities are ultimately real--must reject either mereological reductionism or nonatomism. That leaves recalcitrant mereological reductionists one option, atomism.

Suppose there is an atomic level, where, say, all items are homogeneous quanta. However, as some Tibetan lamas say, (45) even the smallest particles can be distinguished into north, south, east, and west sides. This makes sense because if they have no sides, as Parmenides noted, there cannot be a plurality of them, but Abhidharma posits a plurality of dhammas. (Without anything to differentiate them, no less, it is logically impossible that they could aggregate to even appear as anything that may be differentiated.) Thus, they must have sides, so each such entity admits of regions (parts), each region itself admits of regions, and so on, ad infinitum. Thus, arguably, there is no true atom, thus no impartite, ontologically superior, ultimate reality. Where does that leave the recalcitrant mereological reductionist?

One way to try to evade this objection would be to say that truly impartite atoms are dimensionless points that have no sides, regions, or parts. That will not work because dimensionless entities are nonphysical, but nothing lacking physical magnitude can aggregate to form anything at the macro-level, and, conversely, nothing that has physical magnitude can be reduced to anything that lacks it. Thus, mereological reductionists affirming ultimate reality must reject atomism along with nonatomism. Thus, mereological reductionists seem required to admit that there is no mereologically ultimate reality.

Reasoning like this supports Mahayana antirealism, but perhaps paleo-compatibilism's atomistic "tropes" might circumvent the above logical dilemma, such as instances of whiteness (all of which are homogeneous and thus recognizable even if analyzed into identical segments); the Abhidharma suggests a large but finite number of them aggregate for a perceivable moment of (white-perceiving) experience. But if so, it is not that nothing partite/composite is real, but that nothing perceivably heterogeneous is real. A regular coffee, once blended, appears perceivably homogeneous, thus ultimate, though mereologically divisible into black coffee, milk, and sugar (suppose, arguendo, that these constituents are themselves homogeneous), just as green appears perceivably homogeneous, though mereologically divisible into yellow and blue. If Buddhist Reductionism is amenable to scientific discovery, it becomes problematic because photons, photo-receptive optic cells, and so on do not seem hospitable to the homogeneous-color-type model of trope-atoms.

Atomistic reductionism posits that reductive divisibility comes to an end with ultimate trope-like atoms of reality, but this posit--science aside--remains undermined by the logical dilemma that anything with magnitude is divisible and anything without magnitude cannot be aggregated. Because perceivable phenomena have magnitude, there are no atoms; thus, again, for the mereological reductionist there is no ultimate reality.

The Buddhist Reductionist can claim we hit rock bottom when we arrive at trope-quanta, the first micro-level populated by homogeneous entities, because further divisions only exist mathematically/conceptually, and Buddhist ultimate reality is what exists independent of our conceptualizations. However, it seems unprincipled--a distinction without an ontologically relevant difference--to ontologically privilege homogeneity (black coffee) over heterogeneity (regular coffee), particularly when we cannot conceive one without the other. Such a preference seems--contra the Buddhist Reductionist--to rest on some sort of pragmatic, psychological, hence conceptualized item that is intuitively nonprivileged in conceptualization-independent reality. Could this be the presupposition that grounds the intuition that ontologically differentiates persons asymmetrically from deliberations and volitions, as the paleo-compatibilist thinks the former "only" conventional and the latter ultimate? But even if this were not problematic, deliberation is dyadic, not homogeneous/monadic; so is volition, because it includes (a) conative/teleological impulse, toward (b) an intentional object (desired object, experience, or state of affairs). If these decompose into homogeneous/monadic parts, however, there can be no deliberation-trope or volition-trope among them, just as there are no mother-tropes, so it is a stretch to say "there are events that correspond to what we ordinarily call deliberating and willing" (or mothering) in ultimate reality.

Admittedly, similar difficulties challenge many competing Buddhist and non-Buddhist metaphysics, so some of these difficulties are not peculiar to Buddhist Reductionism. But to the extent paleocompatibilism rests on Buddhist Reductionism, it faces such difficulties, regardless of how widespread such difficulties may be. Siderits's more comprehensive four-staged Buddhist synthesis, however, likely resolves or circumvents most if not all of them (and actually rests its move from Buddhist Reductionism to Buddhist antirealism on similar considerations), but Siderits has yet to articulate the implications of that account for free will.

Turning to free-will-specific claims presented in Siderits's more refined iterations of paleo-compatibilism, Siderits distinguishes four paleo-compatibilist theses. The first two are libertarian; (46) the last two are hard determinist:

1. Persons are free in the responsibility-entailing sense.

2. Freedom requires that persons be originating causes.

3. Nothing could be an originating cause in the required sense.

4. All psychological states are the effects of prior causes. ("Buddhist")

Siderits argues that (1) and (2) are conventionally true and entail libertarianism, and (3) and (4) are ultimately true and entail hard determinism ("Buddhism"). The paleo-compatibilist considers ultimate hard determinism compatible with conventional libertarianism because both are true at different levels.

On my gloss, both free will and determinism are on the phenomenal/form level (conventional). A related Mahayana interpretation that may be used to articulate a Buddhist soft determinism is interdependent origination. (47) Mahayanists argue that there is no potter without a pot, no action without an agent (action-performer), and thus the agent/action pair is interdependently constituted by the action-performance. This conception provides a less inflated conception of the person, one that seems intuitively not in need of reducing. Let us thus define "[Agents.sup.M]" (where the superscripted "M" represents "in the Mahayana sense") as such interdependent action-performers. Thus, we can substitute "[Agents.sup.M]" for "persons" in Siderits's (l) and (2), yielding deflationary statements consistent with soft determinism:

1. [Agents.sup.M] are free in the responsibility-entailing sense.

2. Freedom requires that [Agents.sup.M] be originating causes.

As even Theravada scholars have argued, the Buddha thought volitional actions entail karma (merit) because they are voluntary, despite how strongly shaped by prior karma, and that they are initiating causes, contra the fatalists, because they create new karma and make a difference to the event series. (48) Thus, (3) and (4) are arguably false, and we may substitute the implied "persons" in (3) with "[Agents.sup.M]":

3. [Agents.sup.M] can be originating causes in the required sense.

Let us dub "paleo-soft-determinism" the conjunction of these modified theses (l)-(3). Ockham's razor seems to favor paleo-soft determinism over paleo-compatibilism, (49) insofar as libertarianism rejects scientific determinism and embraces mysterious causation. The paleo-compatibilist seeks to save the notion of free will implicit in the network of beliefs, judgments, and reactive attitudes that constitute our normative institutions and practices, (50) a fine intention consistent with the Mahayana notion (expressed in Siderits's final synthesis as semantic nondualism) that conventional reality is as it appears. I share this goal, but dispense with the unnecessarily heavy metaphysical "baggage," as Fischer puts it (Way), of libertarianism.

In his latest relevant publication ("Reduction"), Siderits condenses (l)-(4), and maintains free will is conventional and determinism ultimate:
   It can be true both that (l) persons are sometimes the originating
   causes of their actions, for which they are then responsible; and
   (2) each of the impermanent, impersonal elements in a causal series
   of psychophysical elements is causally determined by earlier
   elements. ("Reductionism" 36)


Let us call these "new-(l)" and "new-(2)," to set them apart from (l)-(4). Referring to new-(l)'s libertarianism and new-(2)'s determinism, Siderits says new-(l) "is conventionally true" and new-(2) "is ultimately true" (36). Because they are true in "fully semantically insulated" discourses, new-(l) and new-(2) "cannot be incompatible" (36)--inviting critics' suspicions of cage compatibilism.

As with (l)-(3), new-(l) and new-(2) admit of inflationary and deflationary readings. If read in inflated terms of libertarian agents and hard deterministic causation, as Siderits does, new-(l) and new-(2) seem contradictory, as new-(l) is indeterministic and new-(2) is deterministic. Someone, S, might believe some proposition, P, and also believe its negation, ~P, if they manage to partition (semantically insulate) these beliefs from each other enough to not notice the conflict, but whereas the belief statement "S believes P" is consistent with the belief statement "S believes ~P," (51) the simple statement "P and ~P" remains contradictory. But, if read in the deflated terms of interdependent [Agents.sup.M] and actions, new(l) and new-(2) are compatible in the semantically exposed (uninsulated) sense; then, paleo-compatibilism is otiose.

To support paleo-compatibilism, Siderits quotes the (ironically) Mahayanist Santideva, who presents a series of statements (22-31), only one of which I repeat here (22), instructing Buddhists to see others' behavior as--according to Siderits--determined by impersonal factors, the way bile is, for purposes of anger prevention. Santideva then entertains an inconsistency objection (32), to the effect that behavior's impersonal causation contradicts the personal agency (arguably, free will) that would be required for anger prevention. In reply (32), Santideva differentiates between anger's impersonality and the necessity of the assumption of personal agency for the Buddhist path.
   22. There is no anger in me toward bile and the like though they
   cause great pain.

   Why anger toward sentient beings? Their anger is also due to
   causes....

   32. [Objection:] Prevention [of anger] is thus not appropriate, for
   who prevents what?

   [Reply:] It [prevention] is taken to be appropriate with regard to
   dependent origination due to the cessation of suffering.
   ("Buddhist"; "Reductionism" 31) (52)


Siderits thinks verse 22 supports determinism: we do not get angry at impersonally caused things, and others' behavior is impersonally caused. This is consistent with hard determinism, but not an argument for it. (53) It is also consistent with soft determinism. Santideva might mean that most worldlings (54) are so afflicted by (the three poisons of) greed, hatred, and delusion that they are virtually hard determined, which warrants exculpatory reactive attitudes such as nonanger and compassion. Buddhists can change behavior upon Dharma reflection, an ability arguably better accounted for by soft determinism. (55) The worldling may alter behavior under similar circumstances, so virtually hard determined behavior can remain soft determined if it involves certain unexercised abilities. No statement consistent with two theories favors one over the other, ceteris paribus.

Siderits thinks verse 32 supports the paleo-compatibilist claim that ultimatese hard determinism does not conflict with the needs of unenlightened, conventionalese-speaking Buddhist practitioners, who must adopt the conventional notion of efficacious agency to follow the Buddhist path, thereby evidencing some authoritative support for paleo-compatibilism. However, if Santideva did mean that impersonally caused behavior was hard determined, then Santideva's response to the inconsistency charge arguably cannot be easily dismissed simply by reference to the Buddhist hope of reducing suffering. The idea that by doing something (contemplating bile-likeness) you can bring something about (anger reduction) implies anger is evitable; however, that implication arguably favors soft over hard determinism--if it actually does favor either hard or soft determinism, but it is enough here that it is, prima facie, logically consistent with either. The inconsistency objection Santideva entertains, therefore, does not obviously target hard determinism; that doctrine is not explicit in Santideva's writings. What it explicitly targets is impersonality (nonagential bile-likeness) in worldlings versus personality (agential bile-unlikeness) in Buddhists.

The Buddhist view of worldlings is that they are particularly overwhelmed by delusion, which implies their ignorance of dependent origination renders them on par with impersonal (nonagential) bile, which cannot control itself, whereas verse 32 implies those aware of dependent origination can control themselves to progress soteriologically. But the idea that knowledge of impersonal causal/volitional forces leads to agential self-regulation and freedom regarding those forces (however impersonal, determined) is arguably and intuitively more of a soft than a hard determinist idea, although some hard determinists claim that all such abilities are consistent with hard determinism.

It is one thing for data to be consistent with two theories; it is another for one theory to better accommodate that data. All determinists--soft and hard--agree that everything traces to impersonal causes predating agents' existence. But they dispute whether our satisfaction of agentproximal conditions--the data for the question of theory-superiority at issue--suffices for responsible agency. Consider this list of agentproximal conditions the satisfaction of which (by agents) counts here as data:

1. An agent can deliberate about competing choices that represent alternative futures and by selecting one of them she can be the central causal factor that brings about that future.

2. There is a certain harmony between her actions and considered judgments or values.

3. She can alter her values in light of relevant information.

4. She can form effective meta-volitions.

5. She can alter her volitions to have the sort of will she wants.

6. There is a certain agency-nonundermining history to her volitions and meta-volitions.

7. There is a certain mesh between her volitions and meta-volitions.

8. She exhibits self-regulating volitional control.

9. She exhibits counterfactual control over her volitional structure.

10. She would have done otherwise, had she wanted to do otherwise.

11. She would have done likewise, if she wanted to, even if she was able to do otherwise.

12. She exhibits reason-responsiveness.

13. Certain agency-undermining manipulative conditions are absent.

All determinists agree that many often possess such abilities, though some hard determinists might deny agents ever satisfy some such conditions (say, 9-11). (56) But all hard determinists likely accept a "generic hard determinist principle":

No matter what agent-proximal conditions are satisfied, agents never exhibit free will in the responsibility-entailing sense.

Soft determinists disagree over which conditions constitute responsible agency, but likely accept a "generic soft determinist principle":

If certain agent-proximal conditions are satisfied, agents exhibit free will in the responsibility-entailing sense.

Thus, Santideva's worldling may be overwhelmed by conditions that defeat her satisfaction of soft determinist responsible agency criteria, but this does not mean that no one ever satisfies them (as Santideva's objection implies, Buddhist aspirants do satisfy them), or that their satisfaction is always insufficient for responsible agency. In replying to the inconsistency objection, Santideva arguably employs implicitly soft determinist reasoning about how agents possessing knowledge of causal operations do not resemble helpless (bile-like) worldlings. This condition may be added to the list of conditions hard and soft determinists agree some of us satisfy but dispute whether such satisfaction constitutes responsible agency:
   14. The agent is aware of agent-proximal causal factors and general
   karmic and/or causal conditions/processes that shape her choice
   parameters, she can reflect on which choice is dharmic, (57) and she
   can make and effectively act on that choice, even in the face of
   phenomenologically powerful dispositional counter-tendencies.


As I have argued elsewhere ("Meditation"), to the extent Buddhist practitioners cultivate meditative discipline, condition 14 becomes increasingly true of them. The same holds for a variety of free-will-related agent-proximal conditions implicit in central tenets of pan-Buddhist doxography:

15. The agent has significantly dharmic views.

16. She has significantly dharmic volitions.

17. She has significant skill in selecting/performing dharmic actions.

18. She can exert appropriately calibrated effort.

19. She has significant one-pointedness skill.

20. She has significant mindfulness skill.

Item 14 is derivable from Santideva; items 15-20 are from the Eightfold Path. These items are not exhaustive, but are among those a Buddhist soft determinist would include in her list of conditions satisfaction of which arguably constitutes something like a Buddhist conception of free will. Thus, Santideva arguably favors soft over hard determinism.

Refining his analysis of the determinist thread of paleocompatibilism, Siderits unpacks the idea that tropes are the ultimatereality-level phenomena that determine our mental states: tropes may be described as abstract particulars, instantiations of--physical or mental--universals, "such as particular occurrences of white, sweet, cold ... desire, attention, etc." ("Reductionism" 36). Siderits illustrates these particulars with the case of a person eating vanilla ice cream who desires more. One might hope that whereas purely physical items subject the mereological reductionist to the dilemma of infinitely divisible magnitude, because tropes are mentality-involving they are not entirely physical, so they escape this dilemma. (58) But, setting aside the idea that mental states cannot exist apart from their physical instantiations, even putatively purely mental visual images of white ice cream--whether in the form of (at least once removed) mental representations or direct perceptions--have magnitude; so, they are infinitely divisible (they cannot be real atoms), despite being mental.

Siderits claims that because no statement seemingly about tropes can be conventionally true because trope-talk involves ultimate vocabulary, conventional trope-like-talk must be construed as "person-adjectival," whereas ultimate trope-talk is impersonal and "tropeatomistic" (36). But conventionalese whiteness is not a person-attribute just because it is discussed by speakers of conventionalese who accept personhood; (59) enlightened beings may use whiteness-talk. The whiteness beings perceive is, on my gloss, phenomenal, thus whiteness-talk is conventional. Paleo-compatibilists could say that conventionalese about volitions and perceptions is typically attributed to unreal persons, so its person-adjectival elements cannot be true ultimately, and ultimate events to which volitions and perceptions correspond are real but do not correspond to persons, but valid trope-talk can be both conventionally and ultimately true.

For Abhidharmikas, mental states are high-level aggregates of dhammas, indefinitely many (trillions) of which constitute a moment of consciousness (Dhamma and Bodhi "Introduction"). Thus, countless individually imperceptible white-trope momentary events ("trope-atoms") constitute visible white, and countless individually imperceptible desiretrope-atoms constitute felt desire. Any phenomenon that reduces to trope-atoms is conventionally real, but whatever does not reduce is not conventionally real. With "I want more vanilla ice cream," there are no Itrope-atoms or ice-cream-trope-atoms, but presumably desire-tropeatoms, sweetness-trope-atoms, and white-trope-atoms, and perhaps consciousness-trope-atoms. But saying "ice cream is an illusion but whiteness and sweetness are real" conjures the no-leather-shoes and the regular coffee objections.

In their "Introduction" to A Comprehensive Manual of Abhidhamma, (60) the most extensive canonical account of Buddhist atomism, Dhamma and Bodhi state:

Briefly, the dhamma theory maintains that ultimate reality consists of a multiplicity of elementary constituents called dhammas. The dhammas are not noumena hidden behind phenomena, not "things in themselves" as opposed to "mere appearances," but the fundamental components of actuality. The dhammas fall into two broad classes: the unconditioned dhamma, which is solely Nibbana, and the conditioned dhammas, which are the momentary mental and material phenomena that constitute the process of experience .... It is the dhammas alone that possess ultimate reality: determinate existence "from their own side" (sarupato) independent of the mind's conceptual processing of the data. (61)

The general justification for their view is the Buddha's claim to have directly perceived the dhammas with the penetrating insight of his enlightened mind, a claim confirmed by subsequent enlightened Buddhists and supported by meditative phenomenology, which reveals the ephemeral, pixel-like, micro-level nature of everything in the mind/body field that otherwise appears as a solid, permanent, macro-level object.

Buddhist tropes are experientially homogeneous aggregates, like whiteness, that decompose into parts, down to indefinitely many micro level dhammas/trope-atoms. This appears reasonable initially; recall, however, that whatever can aggregate (even micro-tropes) has magnitude and is divisible. Only what lacks magnitude is indivisible, but nothing lacking magnitude aggregates. Further, anything above the tropeatomic level, such as my wanting ice cream, say, "mental state x," already involves mereological fiction because there are no x-trope-atoms. For although some mental states (meditative trances) may be constituted by a homogeneous mass, say, of monadic/identical bliss-atoms, most mental states are constituted by a nonmonadic/heterogeneous variety of items only some of which seem composed of smaller quantities of identical atoms.

Mental state x is heterogeneous and minimally dyadic: it involves wanting and ice cream wanted (not to mention whiteness and sweetness). Siderits treats x as legitimate even though one of its parts, ice cream, does not exist (does not reduce to ice-cream-tropes) and though x is not made of x-tropes, perhaps because x's parts--wanting, sweetness, and whiteness, say, a, b, and c--reduce one-to-one to a-tropes, btropes, and c-tropes, respectively. But why is similar legitimacy denied to person "p" even though there are no p-tropes that compose p, and p arguably reduces by a process similar to the one in which x does not reduce to x-tropes but nonetheless reduces to a-tropes, b-tropes, and ctropes? That is, p arguably equally reduces to a series of mental states, x, y, z, each of which in turn (like x) reduce to a-tropes, b-tropes, and so on. If mental ("M") states are heterogeneous, they do not decompose into a mass of homogeneous M-trope atoms, so M-states are mereological fictions. There can be few ultimate M-states, if any, on the mereological reduction model (such as all-bliss trance states), and fewer ultimate psychological laws that govern them, because M-laws govern complexly nonmonadic causal relations between M-states.

Siderits states, "The illusion of incompatibilism may arise ... by smuggling the concept of a person into the ultimate level ... or by importing psychological determinism into our conventional talk of persons" ("Beyond" 155-156). But "psychological determinism" is the thesis that mental-state events ("that correspond to ... deliberating and willing") are determined. A generic principle of psychological determinism might be the practical syllogism rule:

If agent A desires z and believes doing y will bring about z, then, ceteris paribus, A will do y.

Laws of psychological determinism, if any, have this heterogeneous (triadic, quadratic, or increasingly complex) form (involving agents, beliefs, desires, and actions), or something like it, so they too are mereological fictions. Therefore, contra paleo-compatibilism, psychological determinism--not about micro-level trope-atomic homogeneous/monadic Mstates, but macro-level heterogeneous/nonmonadic M-states--cannot be ultimatese. Of course, the paleo-compatibilist can again appeal to the conventional/ultimate spectrum for wiggle room, but the more these appeals are made, the more support they provide for my Mahayanabased emptiness/form gloss, if not indirect support for Siderits's own (post-Reductionist) semantic nondualist synthesis (of Reductionism and antirealism).

Thus, for mereological reductionists, psychological determinism involves mereological fiction and is ultimately false, even if true in some nonultimate (conventional) domain. Garfield's omni-truth-levels model seems tailor-made to the conventional/ultimate (form/emptiness) spectrum, like Dennett's, for it suggests that there are many levelappropriate vocabulary/phenomena pairings and that all such pairinglevels are valid, but without reductive eliminativism. As he puts it, "let a thousand entities bloom, requiring of each that it genuinely toil and spin, accomplishing some real explanatory work" (Garfield "Nagarjuna" 512). If so, if they do explanatory work, responsible agents, M-states, their psychophysical constituents, and dhammas possess equal ontological value.

This line of reasoning is consistent with the sort of semantic nondualist view Siderits espouses as his own in Persons, which raises the question why Siderits has not simply offered the semantic nondualist view of free will. I cannot say, but I will speculate. In light of his explanation of why he does take the role of amicus theoria for the Buddhist Reductionist view (as an illustration of the philosophical potential of Buddhism), I suspect the answer has to do with the popularity of reductionism in the audience to which Siderits targets this illustration, namely, Western analytic philosophy. Another possibility is that as difficult as it is to grasp the meaning of Buddhist Reductionism, it is even more so with antirealism and semantic nondualism. Thus, perhaps it is wise to simply leave the matter so that any such interlocutors who are drawn in by the illustration may then go on to be drawn into the dialectical progression from Reductionism through antirealism to semantic nondualism. This would effectuate a philosophical bait and switch, but Siderits is explicit in his amicus theoria disclaimers. All of this is, admittedly, speculation. However, it bears repeating that it would be nice to hear Siderits's own articulation of the semantic nondualist view of free will.

Conclusion

Let us conclude our assessment of Siderits's amicus briefs on behalf of paleo-compatibilism. By applying Buddhist Reductionism to free will and determinism, paleo-compatibilism accommodates both libertarianism and determinism--two prima facie contradictory doctrines. Thus, paleo-compatibilism warrants the classification "semi-compatibilist," a term previously denoting only the view that determinism is "hard" because it is incompatible with metaphysical-alternatives-accessing autonomy but "soft" because it is compatible with moral responsibility. (62) But paleo-compatibilism is difficult to comprehensively assess because it stacks a number of puzzling doctrines atop each other, and readers must examine Siderits's other works to ascertain whether paleo-compatibilism is broadly coherent; arguably, it is not: to the extent his larger work in Persons dialectically critiques and absorbs (paleo-compatibilism's) Reductionism into its larger framework of semantic nondualism, that larger work suggests that paleo-compatibilism is not broadly coherent, though Siderits's larger framework is. Taken as a stage within that larger framework, it is coherent.

Although paleo-compatibilism raises questions, and some of its details seem at odds with each other, many such discrepancies are functions of external criticisms that shift focus between different nodes along the conventional/ultimate spectrum, rather than indications of internal inconsistencies in paleo-compatibilism. There is no doubt that elements of semantic dualism, mereological reductionism, and tropetheory do appear in various forms of Buddhism, especially pre-Mahayana Buddhism. Their seeming incompatibility, likewise, may owe more to the complexity of Buddhism than to paleo-compatibilism per se. To his credit Siderits seeks to integrate them all under one theoretical umbrella that ought to earn the respect of Western philosophers. Nonetheless, compatibilists and incompatibilists alike might consider paleo-compatibilism faux-compatibilism, for the generic reason they equally resist semi-compatibilism despite its offer of an olive branch to both sides, and perhaps also because the opacity of semantic insulation obscures the inconsistency between determinism and indeterminism--its cage compatible character.

Those Western philosophers who do look to Buddhism for support have questions for paleo-compatibilism. How can determinism be ultimatese if all nomological relations are minimally dyadic and moments-spanning, which contradicts the Buddhist idea that momentariness undermines the ultimate status of anything moments-spanning? How can there be ultimatese "moments" or "instants"--temporal atoms (indivisibles)--if anything with temporal magnitude is infinitely divisible? Abhidharma posits indefinitely many dhammas per blink of an eye, but even if these somehow manage to transcend logic and actually be magnitude-lacking indivisibles that can nevertheless aggregate to form magnitude-possessing perceptible elements of experience, anything involving more than one extremely small micro-moment--which is everything we experience and know about--does not exist. Any claim about moments-spanning relations or causal processes--sufficient for any perceivable iota--is ultimately unreal. The paleo-compatibilist reply is likely simple: this is correct, but do not forget the other part of the paleocompatibilist explanatory strategy, which is that all these things are conventionally real. And Siderits can always add that the full explanation may be found in his semantic nondualism, where a thousand flowers bloom.

My analysis suggests that determinism and free will are both conventional, so both are robustly compatible (even if located at different nodes along the conventional spectrum). But Sautrantrikas think whatever has causal powers is ultimately real. (63) Because natural kinds are defined by their causal powers, they must be ultimately real. Thus, laws about them, and generalizations about all laws (determinism), must be ultimatese as well. This "if causal, then ultimate" formulation supports the causal/counterfactual analysis of an agent-like process or person-series that exhibits the sort of self-regulatory (causal) control over itself (autonomy) depicted by satisfaction of some of our (arguably soft-determinist-favoring) items 1-20 (particularly 9-11 and 14). But then both free will and determinism would be ultimatese and robustly compatible (in the sense of being true on the same bivalent scale), or both conventionalese and equally unproblematic.

All such objections notwithstanding, it is to Siderits's credit that he sticks his neck out to plant fertile ground for humanists to try to salvage free will in the ever-encroaching face of increasingly threatening determinism, by reference to paleo-compatibilism. No complete account of the Buddhist understanding of free will can ignore Siderits's central contribution--the claim that paleo-compatibilism offers a philosophically rich way to understand the issue of free will and determinism--its unresolved implications notwithstanding. As Siderits himself insists, however, this is not necessarily "the" Buddhist view, "the" Abhidharmika view, or "his" view. Rather, it is just "a" possible view that a Buddhist Reductionist might adopt in response to the freedom-determinism problematic. But in spelling out the many ways even one kind of Abhidharmika model might be developed to this end, Siderits has, undoubtedly, significantly raised the level of debate, for any other kind of Abhidharmika, Buddhist, or non-Buddhist.

Although early-period scholars sought to identify a middle-path between "rigid" hard determinist and "chaotic" indeterminist libertarian extremes, but failed to clearly articulate their positions, Siderits's paleo-compatibilism seeks to salvage elements of both extremes by locating them on different levels of discourse one of which, the conventional (in which persons exist and have free will), reduces to the other, the ultimate (in which there are no persons but only person-series that are entirely determined by impersonal causes). His particular reductionism is mostly limited to pre-Mahayana Buddhism and thus it is perhaps unlikely to impress Mahayanists, compatibilists, or incompatibilists without further refinements. However, if those refinements develop (perhaps along traditional semi-compatibilist lines and those suggested by the dialectical progression in Persons), his strategy seems promising.

In the next article in this series, "Determinism", I discuss how Charles Goodman embraces hard determinism, arguing that Buddhism rejects autonomous agency because it rejects agency or selfhood and because it rejects moral responsibility, which latter presupposes an autonomous self. Siderits and Goodman embrace hard determinism, but in radically different ways, reflecting different reactions to and interpretations of the anatman doctrine. In recent-period scholarship (Repetti "Recent"), these divisions run more acutely along doctrinal lines, where scholars relying on Pali sources mostly accept determinism, but scholars relying on Mahayana sources seem to embrace indeterminism. Both groups agree, however, that Buddhism is compatible with free will even in the absence of a real self, a position I call "soft compatibilism" to contrast with "hard incompatibilism," the view that free will is incompatible with determinism and with indeterminism. In this sense, perhaps Siderits's semi-compatibilism, in trying to accommodate both sides, is not only consistent with early-period attempts at a middle-path toward free will, but prescient.

Bibliography

Baker, Alan. "Simplicity," The Stanford Encyclopedia of Philosophy. Online. Available: http://plato.stanford.edu/entries/simplicity/ (accessed February 21, 2012).

Bird, Alexander and Emma Tobin. "Natural Kinds," The Stanford Encyclopedia of Philosophy. Online. Available: http://plato.stanford.edu/entries/natural-kinds/ (accessed February 21, 2012).

Caruso, Gregg. Free Will and Consciousness: A Determinist Account of the Illusion of Free Will. New York: Rowman & Littlefield, 2012.

Cozort, Daniel and Craig Preston. Buddhist Philosophy: Losang Gonchok's Short Commentary to Jamyang Shayba's Root Text on Tenets. Ithaca, New York: Snow Lion Publications, 2003.

Dennett, Daniel. Elbow Room: The Varieties of Free Will Worth Wanting. Cambridge, MA: MIT Press, 1984.

Dennett, Daniel. Brainstorms. Cambridge, MA: MIT Press, 1981.

Dhamma, U Rewata and Bhikkhu Bodhi. "Introduction," A Comprehensive Manual of Abhidhamma: The Abhidhammattha Sangaha of Acariya Anuruddha, Bhikkhu Bodhi, editor. Online. Available

http://www.accesstoinsight.org/lib/authors/bodhi/abhiman.html (accessed February 21, 2012).

Federman, Asaf. "What Kind of Free Will Did the Buddha Teach?" Philosophy East and West 60, no. 1 (January 2010), 1-19.

Fischer, John Martin. My Way: Essays on Moral Responsibility. New York: Oxford University Press, 2006.

Frankfurt, Harry. "Freedom of the Will and the Concept of the Person," Journal of Philosophy 68 (1971), 5-20.

Frankfurt, Harry "Alternate Possibilities and Moral Responsibility", Journal of Philosophy 66, no. 23 (1969), 829-839.

Garfield, Jay L. "Nagarjuna's Theory of Causality: Implications Sacred and Profane," Philosophy East & West 51, no. 4 (October 2001), 507-524.

Gier, Nicholas F. and Paul Kjellberg. "Buddhism and the Freedom of the Will: Pali and Mahayanist Responses." In Freedom and Determinism, edited

by Joseph Keim Campbell, Michael O'Rourke, and David Shier, 277-304. Cambridge, MA: MIT Press, 2004.

Goodman, Charles. Consequences of Compassion: An Interpretation & Defense of Buddhist Ethics. New York: Oxford University Press, 2009.

Gyatso, Geshe Kelsang. Heart of Wisdom: An Explanation of the Heart Sutra. New York: Tharpa Publications, 2010.

Harvey, Peter. "'Freedom of the Will' in the Light of Theravada Buddhist Teachings," Journal of Buddhist Ethics 14 (2007), 35-98.

Harvey, Peter. An Introduction to Buddhist Ethics: Foundations, Values, and Issues. New York: Cambridge University Press, 2000.

Kasulis, T. P. Zen Action, Zen Person. Honolulu, Hawaii: University of Hawaii Press, 1985.

Nagarjuna. The Fundamental Wisdom of the Middle Way: Nagarjuna's Mulamadhyamakakarika, translated by Jay Garfield. New York: Oxford University Press, 1995.

Putnam, Hilary. "Meaning and Reference," Journal of Philosophy 70 (1973), 699-711.

Repetti, Riccardo. "Recent Buddhist Theories of Free Will: Compatibilism and Incompatibilism," Journal of Buddhist Ethics 18 (2012) (forthcoming).

Repetti, Riccardo. "Buddhist Hard Determinism: No Self, No Free Will, No Responsibility," Journal of Buddhist Ethics 18 (2012) (forthcoming).

Repetti, Riccardo. Response to James J. Stewart, "The Question of Vegetarianism and Diet in Pali Buddhism." Online. Available http://bl ogs.di ckinson .edu/buddh i steth ics/2010/05/15/vegetari an i sm and-diet-in-pali-buddhism/#comments (accessed February 24, 2012).

Repetti, Riccardo. "Earlier Buddhist Theories of Free Will: Compatibilism," Journal of Buddhist Ethics 17 (2010), 279-310.

Repetti, Riccardo. "Meditation and Mental Freedom: A Buddhist Theory of Free Will," Journal of Buddhist Ethics 17 (2010), 24-69.

Repetti, Rick. The Counterfactual Theory of Free Will: A Genuinely Deterministic Form of Soft Determinism. Saarbrucken, Germany: LAP Lambert Academic Publishing, 2010.

Schiffer, Steve. "Ceteris Paribus Laws," Mind 100, no. 1 (January 1991), 1-17.

Schiffer, Steve. "Physicalism," Philosophical Perspectives 4, Action Theory and Philosophy of Mind (1990), 153-185.

Searle, John. The Rediscovery of Mind. Cambridge, MA: MIT Press. 1992.

Sellars, Wilfred. "Philosophy and the Scientific Image of Man." In Frontiers of Science and Philosophy, edited by Robert Colodny, 35-75. Pittsburgh, PA: University of Pittsburgh Press, 1962.

Shantideva and Padmakara Translation Group. The Way of the Bodhisattva. Boston, MA: Shambhala, 2008.

Siderits, Mark. Panel. "Contemporary Perspectives on Buddhist Ethics" conference, Columbia University (October 7, 2011). Online Available http://www.cbs.columbia.edu/buddhist_ethics/panel-five.html (accessed February 21, 2012).

Siderits, Mark. "Is Reductionism Expressible?" In Pointing at the Moon: Buddhism, Logic, Analytic Philosophy, edited by Mario D'Amato, Jay L. Garfield, and Tom J.F. Tillemans, 57-69. NY: Oxford University Press, 2009.

Siderits, Mark. "Paleo-Compatibilism and Buddhist Reductionism," Sophia 47, no. 1 (2008), 29-42.

Siderits, Mark. "Buddhism & Free-Will: Buddhist Paleo-Compatibilism," Columbia Society for Comparative Philosophy (April 2, 2007). Online. Available http://www.cbs.columbia.edu/weblog/2007/04/buddhismfree-w.html (accessed February 21, 2012).

Siderits, Mark. "Buddhist Paleo-Compatibilism," handout at the Columbia Society for Comparative Philosophy (April 2, 2007). Online. Available http://www.cbs.columbia.edu/Paleo.pdf (accessed February 21, 2012).

Siderits, Mark. Buddhism as Philosophy: An Introduction. Burlington, VT: Ashgate Publishing Company, 2007.

Siderits, Mark. Personal Identity and Buddhist Philosophy: Empty Persons. Burlington, VT: Ashgate Publishing Company, 2003.

Siderits, Mark. "Beyond Compatibilism: A Buddhist Approach to Freedom and Determinism," American Philosophical Quarterly 24, no. 2 (April 1987), 149-159.

Thakchoe, Sonam. "The Theory of Two Truths in Tibet," The Stanford Encyclopedia of Philosophy. Online. Available: http://plato.stanford.edu/entries/twotruths-tibet/ (accessed February 21, 2012).

Wallace, B. Alan. "A Buddhist View of Free Will: Beyond Determinism and Indeterminism." In Neurosciences and Free Will, edited by Robert Pollack, 59-69. New York: The Earth Institute at Columbia University, 2008.

Riccardo Repetti

Kingsborough College, CUNY

(1) Department of History, Philosophy and Political Science, Kingsborough College, CUNY. Email: rrepetti@kingsborough.edu.

(2) I would like to thank Dan Cozort, Mark Siderits, Claire Gaynor, and an anonymous reviewer at the Journal of Buddhist Ethics for comments to the present article.

(3) Siderits also suggests this Humean model in Persons.

(4) Fatalists believe that if a certain event is fated, no antecedent interference can prevent it. Local fatalists think certain events are fated; global fatalists think all events are. Whereas fatalism is acausal, determinism is causal: every event is necessitated by causal laws. Global fatalists and determinists agree on a single, invariable, necessary series of events, but for acausal and causal reasons, respectively. Most hard and soft determinists agree that the invariable series runs through causally effective choices, but disagree about whether agent-proximal activities (such as belief, volition, choice, and action) are sufficiently "up to" the agent to count as responsible agency, although eliminativist hard determinists doubt mental states have any causal powers; see Caruso (Illusion). Some soft determinists assert that, had prior contingent conditions (volitions, say) been otherwise, the agent could have done otherwise; hard determinists insist that conditions are never otherwise, thus that the alleged ability is otiose.

(5) Siderits's work is classified as middle-period work because his first relevant article appeared in 1987 and, though his most recent paper presentation was at a 2011 conference, most of his relevant published work occurs between the early and recent periods.

(6) See Siderits ("Beyond"; "Buddhism"; "Buddhist"; "Reductionism"; "Expressible"; Panel).

(7) The Abhidharma ("higher Dharma" or teaching) is the third of three collections of early Buddhist texts (Tripitaka, "three baskets"), written in Pali, that constitute the early Buddhist Pali Canon; an Abhidharmika is a follower of the Abhidharma. The Abhidharma is a philosophical articulation of the doctrines implicit in the other two sets of texts, the Sutras (sayings of the Buddha) and the Vinaya (the monastic code). "Dharma" (Sanskrit; Pali: "Dhamma") is difficult to translate, and has many differences in meaning based on usage, but may mean any of the following, loosely: the way things are, the teachings of the Buddha(s), the universal pattern or way, the truth or the path to its realization, and so on.

(8) Siderits (Panel). Siderits takes this general posture--of offering Buddhist ideas as items of potential interest to Western philosophers, without asserting them himself--not only in his articles on free will, but in his major monographs, such as Persons and Philosophy. His writings on free will, therefore, must be understood within his larger curriculum vita, using that phrase in the broadest sense.

(9) Siderits ("Expressible"). In Sanskrit, "satya" (or "sat") seems sometimes to mean "reality" instead of "truth." Although reality and truth differ, truth-statements are convertible into reality-statements, and vice versa.

(10) Siderits (Panel).

(11) See Bird and Tobin ("Kinds").

(12) An alternative interpretation that avoids contradiction is to shift from ontology to epistemology, asserting two descriptions of one reality, but some interpretations of the two truths are pointedly ontological (Cozort and Preston "Buddhist" 54-55).

(13) Whereas many Western reductionists treat the reduced level as ontologically inferior, most Buddhist Reductionists treat both levels as almost equally real. Some Buddhists consider conventional reality delusional, however, and some consider both identical. See Thakchoe ("Theory").

(14) Contrary to most Buddhist thought, hedonic sense and sentience are noncoextensive, for certain primitive life forms with sensorimotor abilities experience no pleasure or pain; see Repetti (Response).

(15) Semi-compatibilists conceive the possibility of determined responsible agency as the determined ability of an agent to be reason-responsive, particularly with respect to moral reasons. I have described this reason-responsive ability more specifically along Buddhist ethical lines as "Dharma-responsiveness"; see Repetti ("Meditation").

(16) See Siderits (Persons) for the full account, but recall that he is not affirming this view, but rather using it as part of a larger dialectical progression.

(17) Cozort and Preston use the Mahayana ("Greater Vehicle") term "Hinayana" ("Lesser Vehicle") to refer to pre-Mahayana Buddhism (most early schools of Indian Buddhism, of which Theravada is the remaining living tradition), but many find this term loaded.

(18) The lower case term "dhamma" is Pali; it is "dharma" in Sanskrit. I use "dhamma" only to (acoustically) differentiate between it and the upper case term, "Dharma," only because "Dharma" is more widely used among Anglophone speakers.

(19) Thakchoe ("Theory").

(20) See Gyatso (Hea t) for an explanation of the Heart Sutra

(21) See Samyutta Nikaya (SN) 22.95 and SN 35.85, respectively. All references to SN and other elements of the Pali Canon are available online at the Access to Insight website, available free online at www.accesstoinsight.com (accessed February 21, 2012) and also the Pali Canon Online: The Original Words of the Buddha website, available free online at http://www.palicanon.org/ (accessed March 26, 2012).

(22) There is an exception for the unenlightened Buddhist who has conceptual but not experiential understanding (Thakchoe "Theory").

(23) The chariot and Nagasena examples are derived from the classic Buddhist statement of mereological reductionism, the Milindapauha ("Miln").

(24) Sellars ("Philosophy").

(25) Schiffer ("Laws"; "Physicalism").

(26) Even a single (divisible) hydrogen molecule lasts too long to be an Abhidharma dhamma, and quanta are too small. Thus, on my gloss, Buddhist Reductionists who accept physics must conclude that no conventional things (items of valid perceptual/phenomenal experience) have inherent natures that ground the use of names--there are no natural kinds--unless (like atomists who now say physics' "atoms" are not really atoms) they are willing to say Abhidharma "dhammas" are not really dhammas, but physics' quanta are.

(27) Siderits is espousing the pre-Mahayana view, but some Mahayanists, for example, Garfield, also consider dependent origination ultimate; see, for example, Nagarjuna (Mulamadhyamakakarika 24.18), although Garfield's interpretation of this passage has been seriously challenged, and the more standard Mahayanist view is that everything in form that is effable--including dependent origination--is conventional. (I owe this observation to Siderits.)

(28) Siderits (personal communication, February 2012). If free will were construed as whole-person-level action-origination and thus did not reduce to ultimatese, however, because, say, there are illusions about the whole-person-level composite-type entity (analogous to illusions about fire's nature as phlogiston), then free will would involve some conventional element incompatible with ultimate truth, reminiscent of [W.sub.x] and [W.sub.y.]

(29) Barnhart (conversation).

(30) Here is just one instance of my earlier claim to the effect that causation plays too heavy a role in Buddhism for too light a conception of causation.

(31) One may object that if there is no identity, nothing can prevent karma from person-series-Siderits from accruing to person-series-Repetti, and there could be no basis for claiming such transference was unfair. The reply would be that such transferences just do not naturally occur, just as pouring water on a basketball just does not ignite it. Arguably, however, only robust causation can prevent such deviant karmic or causal redistributions.

(32) Siderits (personal communication, February 2012).

(33) Two objections: First, there is a sense in which a causal stream may be said to be "free" if it is unobstructed, say, by a dam. In support of Siderits's claim, however, Frankfurt ("Freedom") argues that "free will" applies only to volitional beings that can suffer its privation. By analogy, rocks cannot be blind. Second, consider a flawed claim analogous to Siderits's claim that we neither have nor lack free will because only persons either have it or lack it: we neither experience pain nor do not experience pain because there are no selves (subjects of experience, experiencers) in ultimate reality (as if person-series that are not persons cannot experience pain). In Siderits's defense, it may be said that because there are no subjects of experience in ultimate reality, it is neither true nor false that such subjects experience pain. But there remains a sense that even if subjects are not selves in the sense denied by Buddhists, there may be pain experiences in ultimate reality.

(34) But on my gloss and perhaps also his, no questions arise at the ultimate level, so this prima facie plausible solution is somewhat misleading.

(35) Siderits (personal communication, February 2012).

(36) On Frankfurt's model, when an agent acts on a volition the agent approves of, this exhibits an effective meta-volition (a volition about a volition), and free will involves such effective (volition-controlling) volitional/meta-volitional harmony. For a Buddhist account along similar lines see Repetti ("Meditation" and Counterfactual).

(37) However, for some (Abhidharmikas), perceived redness is a dhamma, a basic pattern, and thus ultimate, though perceived apples are not dhammas, so they are conventional. (This observation illustrates the tedium that would otherwise ensue regarding spelling out all the doctrinal differences for each claim here, absent both my and Siderits's glosses.)

(38) Santideva (Bodhicaryavatara 6), referenced in the bibliography as Shantideva and Padmakara (Bodhisattva), discusses anger control.

(39) Frankfurt ("Freedom") defines a "person," roughly, as a being with meta-volitions and nonpersons as beings that lack meta-volitions, such as animals that simply act on whatever volition arises.

(40) Pudgalavadins (those affirming a "pudgala" or "person" doctrine) may have held this intermediary view (between nonreductive realism and reductionism); see Siderits (Persons 89-91).

(41) Arguably, this objection involves a straw man fallacy, where the intended claim, charitably interpreted, is that it is not ultimately true that shoes are made of leather because shoes are not ultimately real. But the objection may be restated more charitably: however empty shoes are ultimately, they are made of leather. The counterexample thus survives.

(42) Abhidharmikas assert that only dhammas are ultimately real, but this supports my gloss that everything else (all the phenomena of experience, except its perceivable dhamma fragments, if any) is, by default, conventional.

(43) Mahayanists deny that dhammas are ultimately real, insisting that the ultimate truth is that reality is empty, which also supports my gloss that everything else (apart from emptiness) is, by default, conventional. In Persons, Siderits seems to favor this conception insofar as it occupies a node along the progression of his four-staged argument closer to its final conclusion.

(44) This approach is intuitively physical, one might object, but Abhidharmikas are explicitly committed to the existence of nonphysical dhammas. Siderits's argument does not assert the existence of nonphysical dhammas, but neither does he deny their existence. Siderits makes clear, recall, that he is only extrapolating his model from Abhidharma, not that any Abhidharmika actually advocates paleo-compatibilism. Nothing in paleocompatibilism excludes the possible existence of purely mental (nonphysical) dhammas.

(45) I owe this reference to Tibetan lamas to Dan Cozort.

(46) Libertarians think free will and determinism are incompatible; agents are free because they are originating causes; and thus determinism is false.

(47) An authoritative source for the doctrine of interdependence is Nagarjuna (Mulamadhyamakakarika), an early-period source is Kasulis (Zen), and recent-period sources are Gier and Kjellberg ("Buddhism") and Wallace ("Buddhist"). Wallace discusses a Mahayana model of interdependence, consistent with physics, in which every quantum is holographically interconnected.

(48) See Harvey ("Freedom") and Federman ("Buddha").

(49) Ockham's razor, in Indian philosophy "the principle of lightness," is the principle of explanatory parsimony: always prefer the least risky hypothesis, or, in some interpretations, prefer the one with the fewest ontological commitments. See Baker ("Simplicity").

(50) Siderits (personal communication February 2012).

(51) Insofar as belief embeds propositions in semantically opaque contexts (propositional attitudes), it makes it possible for contradictory propositions to occur unnoticed; recall Jimmy's contradictory beliefs about Clark and Superman.

(52) Quoting Santideva (Bodhicaryavatara 6).

(53) Siderits says he does not think this is an argument for hard determinism, but that what Santideva develops in the subsequent verses is an argument for the conclusion that responsibility-entailing freedom cannot exist at the ultimate level; Siderits thinks that hard determinism would follow from this only if you were an eliminativist about persons, whereas the Reductionist is not an eliminativist. Siderits (personal communication, February 2012). But he does seem to take paleo-compatibilist theses (3) and (4) above to entail hard determinism, so either he does not draw the hard element from Santideva or I have misinterpreted him to take (3) and (4) to entail hard determinism. Another way to put this is: if one does not think the determinism in play is hard, then why would one think there's a need to show it is paleo-compatible?

(54) A "worldling" is a person who has not attained the first level of direct spiritual realization/ transformation, that of "stream entry," typically precipitated by Buddhist meditative discipline. Thus, most non-Buddhists and many (maybe most) Buddhists are worldlings, though non-Buddhists can be stream entrants. I stipulate that I use the term loosely to refer to most worldly folks and/or all who are not stream entrants.

(55) For a defense of this claim, see Repetti ("Meditation" and Counterfactual).

(56) As I have argued (Counterfactual), however, this denial presupposes what I termed "actualism," the view that only what is actual at some time is possible, and counterfactuals are never actual, so they are not possible. But determinism is an agglomeration of all deterministic laws, but deterministic laws are all counterfactual-supporting generalizations, in which case actualistic determinism is an oxymoron.

(57) The term "dharmic" means "Dharma-oriented" (skillful, relative to Buddhism's prime directive, liberation).

(58) This "hope" is not meant to represent the view Siderits describes, but just as a hypothetical reason to think there may be logically possible trope-types that escape the dilemmas facing physical tropes. Understanding Abhidharma's dhammas in the Sautrantika way as tropes, there will be mental and physical tropes. For them the latter would involve occurrences of yellow and sweet. For a modern, they might involve things like occurrences of a certain force at a certain point-instant. Siderits (personal communication, February 2012).

(59) Siderits thinks I've misunderstood his reasoning here because, on his reading, common sense denies that there can be the occurrence of whiteness without some thing--a substance--that is white. That, he adds, is also Candrakirti's view of conventional truth--and so what Madhyamaka implies. So even if I (using my gloss) take his words outside the Abhidharma context that he's formulated them within, I've missed the point. Siderits (personal communication, February 2012). I may be missing the point, but I doubt it is a given that common sense equates whiteness with substance, when students (who are not philosophy majors) discussing Descartes seem to know that red is a color and triangles are three-sided even if we are dreaming or in an illusory (insubstantial) world. Siderits also seems to evidence one of his glosses here, in implicitly equating common sense with conventionalese. Enlightened beings speak conventionalese, but insofar as common sense may be described as the sum of all mankind's prejudices, enlightened speakers of conventionalese do not possess common sense.

(60) In the title, "Abhidhamma" is Pali; "Abhidharma" is Sanskrit.

(61) "Nibbana" (as they spell it, but technically, "nibbana") is Pali for "nirvana" (Sanskrit). In this context, it refers to emptiness. (Likewise, what they spell as "sarupato" is, technically, "sarupato") Unconditioned dhamma, emptiness, is ultimately real for Abhidharmikas and on my gloss, but Abhidharmikas include conditioned dhammas, momentary phenomena, in ultimate reality; my gloss parses all conditioned phenomena, all form, as conventional, but also accepts that some conventional truths (their conditioned dhammas) are also ultimately true.

(62) See Fischer (Way).

(63) Cozort and Preston (Buddhist 55).
COPYRIGHT 2012 Journal of Buddhist Ethics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:p. 65-95
Author:Repetti, Riccardo
Publication:Journal of Buddhist Ethics
Article Type:Essay
Geographic Code:1USA
Date:Jan 1, 2012
Words:9753
Previous Article:Buddhist reductionism and free will: paleo-compatibilism.
Next Article:Buddhist hard determinism: no self, no free will, no responsibility.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |