How should we, as scientifically informed, twenty‐first‐century American philosophers, understand and come to terms with the Buddhist intellectual tradition? There are two basic ways in which we might approach this daunting project of cross‐cultural interpretation and dialogue. One way involves emphasizing the mythological and magical aspects of premodern Buddhism. By focusing on the extraordinary and miraculous aspects of Buddhist narratives, this approach brings out much of what was salient about the tradition to its ordinary lay followers in traditional Asian societies. We could call this approach the mythic reading of Buddhism. A clear example of the mythic reading would be Donald Lopez's () recent book The Scientific Buddha, which argues that the real, historical Buddha was anything but scientific.

The other way of engaging with the Buddhist tradition involves consciously looking for ideas which have some chance of making sense to us today. Setting aside the outdated cosmology and baroque mythology of the tradition, scholars following this approach examine Buddhist texts to see what aspects of their worldview and teachings are defensible in the context of a modern, scientific understanding of the universe. What emerges from this inquiry is then presented as what genuinely matters about the Buddhist tradition; the rest of what historical Buddhists believed is discarded as inessential. This approach could be called the naturalized reading of Buddhism. Owen Flanagan's The Bodhisattva's Brain is a leading recent example of the naturalized reading.

Advocates of each approach can criticize the other for not taking Buddhism seriously enough. Those favoring the mythic reading can point out the frequent occurrence and apparent centrality of what we would call supernatural beings and miraculous powers in the Buddhist texts that have survived and in Buddhist art, architecture and material culture. The view presented as Buddhism by the proponents of the naturalized reading, they could say, is so different from real Buddhism that if you explained it to seventeenth‐century Japanese peasants or Mongolian nomads, they would hardly recognize the naturalized view as being a version of their religion at all.

Of course, the Japanese peasants and the Mongolian nomads might, at least initially, have had some difficulty in recognizing each other's views as forms of the same religion. Buddhism takes many forms, syncretizing with the indigenous worldviews and cultural traditions of the societies that adopt it. Today, Buddhism is in the midst of a wrenching but immensely productive process of synthesis with the dominant worldview of Western culture: natural science.

More importantly, advocates of the naturalized approach can charge that the mythic approach turns Buddhism into a useless relic of the past, fit to be preserved only in museums. For when your presentation of Buddhism emphasizes tree‐spirits, hungry ghosts, and the Buddha's cosmically immense tongue, it becomes impossible for us today to take any of what is presented very seriously. By contrast, when you start with Buddhism and you remove everything supernatural, you end up with a substantive, interesting, and important set of claims about how things are, how we should aspire to be, and how, in practice, we can actually become what we should aspire to be. Drawing on his wide knowledge of scientific methodology, experimental psychology, philosophy of mind, and ethics, Flanagan gives us a very interesting version of this stripped‐down, naturalized Buddhism. In the process, he raises doubts about some incautious claims that have been made by naturalizers, while remaining optimistic about the relevance and significance of Buddhist ideas for today.

When it comes to the intellectual history of the Buddhist tradition, you can't always trust Flanagan to get the details right. But when it comes to the central truth‐claims and philosophical tenets of the tradition, Flanagan is mostly reliable, and better yet, effective at bringing out these views in forms with which we today can readily engage philosophically.

One scholarly area that Flanagan does get wrong, in my view, is the question of the logical relationship between the doctrine of no self and the traditional belief reincarnation. Flanagan writes that the details of how reincarnation could even begin to work for a being that has no‐self presents serious logical problems (Flanagan , 132). This passage may suggest to the reader that these two views contradict each other. But the conjunction of these two theses is not a contradiction, but merely empirically rather implausible—which is not the same at all.

Buddhists don't believe that what we call “personal identity” should be understood as the continued existence of an identical substance. Instead, what we conventionally call personal identity is based on the continuation of a complex, underlying process. What holds this process together, on the Buddhist view, is not memory, or psychological connectedness, or psychological continuity, but only the obtaining of a dense network of causal connections. Ultimately, I am not the same person I was yesterday, because ultimately, I don't exist. Conventionally, I am the same person I was yesterday only because so many of the characteristics I have now are causally explained by the characteristics I had yesterday.

What that means, in turn, is that reincarnation can exist in a selfless world if there is some kind of causal process that transfers large amounts of information from a dying sentient being into a sentient being that is coming into existence. It would then have to be the case that many aspects of that new sentient being's life are causally explained by the actions and characteristics of the sentient being who died. In saying “many aspects,” I mean, for example, that even though my actions and characteristics have many consequences for how my daughter turns out, nevertheless, the causal links between me and my postulated future self in another life are tighter than those between me now and how my daughter turns out. But these links don't have to consist in similarities. It would be enough, for example, if the fact that I am aggressive and brutal now causally explains the fact that my reborn future self is timid and fearful.

Several classic Buddhist texts express this point by distinguishing between two possible interpretations of reincarnation, affirming one and denying the other. They say things like, “One does not transmigrate, but one is reconstituted.” There is no thing, no substance that moves from one life to another; and yet there is a process that continues, mediated by causal connections. Flanagan reports that, in a private audience, His Holiness the Dalai Lama confessed to not believing in reincarnation in a literal sense (Flanagan , 132). I'm afraid it's possible that Flanagan might have misunderstood the Dalai Lama's meaning. He may have meant that, as an orthodox Buddhist, he denies that one is transmigrated, but believes that one is reconstituted. As Flanagan himself points out earlier in The Bodhisattva's Brain, the Dalai Lama, even while making numerous efforts to learn from science and to promote dialogue between Buddhist monks and scientists, has often tried to express reservations about the causal closure of the physical, precisely to make room for reincarnation. So I would suggest that he should be interpreted in the way I have sketched, as denying literal reincarnation only in the sense that there is no soul that transmigrates.

One of the ways in which classic Buddhist texts express their process understanding of the nature of sentient beings is by comparing a human life to a fire. Fires have obvious similarities to living things—they move, they breathe, they eat—but a fire is also an entity which we easily and intuitively understand to be a process as opposed to a substance. And we can use the metaphor of fire to explain rebirth, by comparing it to the transmission of fire from one candle to another.

Flanagan is aware of this metaphor, but protests that it does not help us to understand how my existence could continue, in any sense, when there is a gap “during the time between my (last) death and my (next) rebirth” (Flanagan , 133). A traditional Buddhist could respond by affirming that, given the structure of the account of personal identity on offer, the mere existence of enough causal connections is sufficient for my identity to continue, regardless of any temporal gaps. But there are two other approaches also available to the traditional Buddhist. One option, advocated by many Theravādins, is to deny that there is any temporal gap at all. Rebirth happens immediately after death. The other option, advocated by many Tibetans, affirms that there is a gap in my embodiment, but no gap in my conscious existence. The time between my death and my next birth constitutes the bar do, or intermediate state, during which I am fully conscious, though confused and terrified by hallucinations generated by my karma. The famous Tibetan Book of the Dead (Dorje ) is, among other things, an instruction manual for how to behave in the intermediate state. Vajrayāna texts even offer us a way to assert this without denying the supervenience of the mental on the physical: although beings in the intermediate state don't have physical bodies in the usual sense, their consciousness is supported by a “subtle wind” which is, they claim, made of matter of a special kind.

Obviously all of this looks like mythology, and the evidence that is sometimes offered for it falls well short of convincing. But there is nothing inconsistent about it. The doctrine of no self implies that one does not transmigrate, but in the sense explained, it doesn't contradict the claim that one is reconstituted. This is the precise meaning of the Buddhist claim to have avoided both eternalism—the view that you are an eternal soul—and annihilationism—the view that, because you are your body, death is the end.

Although we can't convict them of inconsistency, there are some embarrassing questions we could press against traditional Buddhists. Is the signal that carries the information from the dying person to the one who is born a physical signal, or a nonphysical signal? If it's a physical signal, why haven't scientists detected it yet? What enables the mechanism in the brain that sends the signal to know what direction to send it in, so that it will be picked up by an appropriate embryo recipient? If it's simply broadcast in all directions, what would guarantee that only one embryo picks it up? If, as many claim, the signal is received at conception, how can a fertilized egg store all that information? But if the signal is non‐physical, what evidence is available that is sufficient to overcome the very powerful scientific case for the causal closure of the physical? And won't the signal be subject to the very same problems that sink interactionist substance dualism?

Given that the traditional understanding of rebirth is, at least, implausible, we are in need of a new understanding of the meaning of karma. Flanagan proposes a “tame interpretation of karmic causation” (Flanagan , 77), understanding this form of causation as just consisting in all the actions of sentient beings and the results that we ordinarily understand to flow from them. This category is certainly tame enough for the naturalist to accept, but it is so far from the distinctive theses that Buddhist texts have in mind when they talk about karma‐vipāka‐phala, “the result of maturation of action,” that it is not an interpretation of that concept at all. So much has been stripped away that it would be less misleading to say that Flanagan has abandoned the concept of karma altogether.

But maybe we don't have to abandon this concept, even if we are strongly committed to naturalism. To be a form of karma in even a minimal sense, a causal process has to make the result of an action correspond, in some way, to the nature of the intention that led to it. So an intention to harm others needs to have at least some tendency to lead to harm to the agent, and an intention to benefit others needs to have at least some tendency to benefit the agent. For example, then, consider the fact that we humans are social animals, and that actions based on an intention to harm others in our social group will often lead the members of that group to treat us less well in future. This kind of process, which should be wholly acceptable to the naturalist, is at least a candidate to be an aspect of a tame account of karma.

We can say more, if we want to claim that human flourishing is constituted in part by virtues of character. Now if, as Aristotle claimed, virtues and vices are habits, and if we can get in the habit of doing wrong actions, that will constitute, in and of itself, a setback to our flourishing. Buddhists are likely to deny that virtues are habits, but they would agree that vices are—so the same argument would work for them. Virtues aren't habits because they aren't stereotyped and automatic, but fully conscious, spontaneous expressions of our connection to reality. But when we form intentions to act so as to harm others, those intentions leave psychological traces in our minds that predispose us to keep acting in similar ways. Every time you do something, it gets easier to keep doing that sort of thing: perhaps this is one of the basic lessons we can learn from research on neuroplasticity. But then the dispositions to act wrongly that we gain thereby will intrinsically set back our well‐being. Worse, if there are truths about our situation that only the virtuous person can perceive, then conversely, those who act wrongly may, by doing so, cause their own perspective on the world to become more and more distorted. Through claims like these, claims that flow naturally from eudaimonistic views, and without introducing anything supernatural or spooky, we can get much closer to something that has the right to be called a conception of karma.

In the context of his discussion of why he thinks it makes sense to describe Buddhist ethics as eudaimonistic, Flanagan offers a complex, and heavily hedged, discussion of how a theory that focuses on happiness, as classical utilitarianism does, can fail to be eudaimonistic (Flanagan , 142–43.) Now if we read these claims in the most straightforward way, we will make the following argument: Buddhist ethics is eudaimonistic; classical utilitarianism is not; therefore, Buddhist ethics is very different from classical utilitarianism. Now, as I think there are many important similarities between the normative perspective of certain Buddhist writers, such as Śāntideva, and the classical utilitarian view, I don't like this argument. Flanagan adds so many qualifications to his discussion that this simple argument is probably blocked. But as I think he could have done a better job of sorting through these considerations in any case, I'd like to look more closely at the relevant issues.

Flanagan suggests that there are three ways in which a theory whose central aim was to promote happiness could fail to be eudaimonistic. Actually, his presentation contains at least four distinguishable characteristics of such a theory that could stop it from counting as a form of eudaimonism. The first characteristic is that the theory so badly misunderstands “happiness” that its conception isn't worthy of being described by that name; his example is a crude hedonism. I would claim that Buddhist views cannot plausibly be accused of this mistake. The second characteristic is that a theory could fail to “provide an account of what makes for individual happiness, a good and contented individual life” (Flanagan , 142.) Perhaps some consequentialist ethicists, preoccupied with other issues, have given little attention to this question. As a result, their works, though not incompatible with eudaimonism, would not be examples of it. But Buddhists have obviously focused a great deal of their theoretical attention on these matters, and so such a description wouldn't apply to their texts.

Flanagan's third characteristic is that a theory is too focused on rule following at the expense of character. A theory like this, he says, “would not require moral personality to be structured by abiding personality features, other than whatever it takes to abide by the general‐purpose moral rule it advocates” (Flanagan , 143). This thought makes perfect sense in the abstract, but encounters difficulties on further reflection. What would it actually take to abide by the principle of utility in all our actions? What would it take to do, in each situation we encounter throughout our days, whichever of the available alternatives would be most effective at promoting the happiness and relieving the suffering of all sentient beings? After all, both the advocates and the critics of the classical utilitarian theory constantly reiterate how demanding it is: so demanding that almost no one has come even close to being able to follow it. Here's a description of just part of the emotional makeup of someone who would actually have sufficient motivation to behave in accordance with the principle of utility:

For example, suppose that a guild president or a leading merchant were to love his only child with love that goes down to the marrow. In just the same way, a bodhisattva who has attained great compassion loves all sentient beings with a love that goes down to the marrow. (Vaidya , 151, translation by the author)

In other words, we could consistently hold both that the definition of right action is fully captured by the principle of utility, and that a full description of the inner states necessary to be able to act rightly would consist of a specification and enumeration of an interdependent set of moral virtues. Suppose we added to these claims a Fortunate Coincidence Thesis, according to which someone with those virtues would be highly likely to exemplify human flourishing—or perhaps, according to which those virtues themselves actually help to constitute human flourishing. The result would be a view that we could plausibly call simultaneously eudaimonistic and utilitarian. (See Flanagan , 159.) And Buddhist ethics could be like that; indeed, I take it to be a live interpretive thesis that some forms of Buddhist ethics really are like that.

Although laying out his second characteristic, above, Flanagan mentions that a theory might not be eudaimonistic because “it aims at happiness in too impersonal a way” (Flanagan , 142). This, I think, is a fourth, independent criterion, and one that points in a different direction from the others. If a view involves the Fortunate Coincidence Thesis, that view invites us to ask a difficult question. Given that the actions and practices that lead to my own happiness will mostly, or possibly even entirely, coincide with the actions and practices that lead to the greatest overall happiness for everyone, what is the foundational justification for those actions and practices? Are they justified by the fact that they make me happy, and it's great that they also lead to happiness overall? Or are they justified by the fact that they lead to happiness overall, and it's great that they also make me happy?

I have argued before that, if we are thinking about the bodhisattva path, we should adopt the second interpretation. The best reason for this claim is one to which Flanagan repeatedly calls attention: that in Mahāyāna Buddhism, the realization that there is no separate, substantial self is supposed to make you morally better. But if we claim that what makes you morally better is the realization that the distinction between self and others has no intrinsic normative significance, how could we possibly also claim that the foundational justification for ethics is to benefit yourself and not others? This kind of reasoning could only lead to the conclusion that what ultimately matters is the welfare of all sentient beings. So, if this argument is sound, we should attribute to Mahāyāna Buddhists a view which, along the dimension marked out by our fourth criterion, is utilitarian as opposed to eudaimonistic. And that's a good thing, because being eudaimonistic in this fourth sense is the same as being egoist, and a defensible moral theory can't be a form of egoism.

Of course, this argument brings us back to a question about which Flanagan confesses he is unsure: What exactly is the relation, if any, between no‐self and compassion? At the end of chapter 5, Flanagan tentatively suggests we should adopt a kind of Aristotelian view that is between the Indian views of ātman and anātman, or as we might say, between soul and no‐self. In this context, he's interpreting no self to mean a kind of punctualism: I am just my present self, and the past and future segments of this life have nothing to do with me. And the Aristotelian view seems to be that, in virtue of having a character and a relatively stable physical embodiment, I am an enduring thing, even though I don't have an eternal, permanent, nonphysical soul. Flanagan imagines the Buddhist punctualist responding to him that the Aristotelian view can lead to egoism; and his response is that “there are many Buddhists who avow anatman but who are selfish creeps by any measure” (Flanagan , 162).

There are at least two serious problems with this whole discussion. One is that philosophically sophisticated Buddhists don't endorse punctualism. When it appears in their discussions, it is functioning only as a foil for the view of self, an intermediate perspective to be considered and discarded. Even now, there is no thing, no substance, that is me. Identifying with my present timeslice, regarding that as a self, is just as much a mistake as regarding anything else as a self. Conventionally, I exist as a process that unfolds over an entire lifetime, or on the traditional view, even beyond; ultimately, I don't exist at all. Because he doesn't focus on the distinction between these levels of truth, Flanagan can't fully capture the structure of actual Buddhist metaphysical views.

The second problem is that no Buddhist holds that simply avowing the view of no self is enough to stop someone from being a selfish creep. That's because, no matter what you say or believe at an intellectual level, you will still continue to relate to your life, moment to moment, from an instinctive, deep‐seated sense of self. With that sense of self comes self‐cherishing, the emotional disposition to favor yourself over others. But if, through a combination of resting meditation (Skt. śamatha) and insight meditation (Skt. vipaśyanā), you can start to see through the illusion of self, then you can begin to relate to the world in a selfless way. This would be something like going through much of your life in a flow state. All sense of a distinction between self and other drops away; agency in the usual sense stops functioning, and is replaced by spontaneous, skillful responsiveness.

Flanagan's reading of Buddhism leaves this kind of state out, and in doing so, it omits a concept that is absolutely central to the tradition: enlightenment, or Awakening (Skt. bodhi). And how can you really expect to understand the way of life of bodhisattvas if your account says nothing about the bodhi they seek? If you were to eliminate the subjective sense of self completely, so that all your time was spent in a state of spontaneous responsiveness, you would never act like a selfish creep. We may not have convincing evidence that it's possible to attain that level of perfection. If you were to spend some of your time in a state of spontaneous responsiveness, then we could say at least that you would act like a selfish creep only during some fraction of the rest of your time. And that might well be an improvement. Moreover, many experienced meditators report that such a state is possible, and when we watch how they behave, we may find we have reasons to agree.

Something else that experienced meditators regularly report relates directly to the major concerns of Flanagan's book: they say that meditation makes them happier. One of the significant contributions of The Bodhisattva's Brain is to cast doubt on the evidentiary basis of the common claim that practicing Buddhism will lead to happiness. Flanagan points out that the experimental evidence that has been offered for this claim seems to be quite thin and overhyped. But he goes beyond that to offer the important philosophical point that if the word “happiness” is being used in a sense that goes beyond a mere feeling—if the conception of happiness in play is eudaimonistic and involves virtues—then brain scans are unlikely to be able to detect it, and moreover, different ways of life that cultivate different virtues are likely to lead to different forms of “happiness.”

Moreover, once these points are made, we can distinguish a number of different possible claims that might be made about the relationship between virtue and happiness. Flanagan suggests that the most plausible claim for Buddhists to make would be

BL′′: Virtue (and wisdom and meditation/mindfulness) is the normal and reliable cause of happiness. (Flanagan , 186)

This thesis, BL′′, certainly captures some aspects of what Buddhists should want to say. But I think that we can get even more precise in understanding Buddhist views in this area, and by doing so, possibly make progress toward an interpretation of Buddhism that would make testable psychological predictions about happiness.

The key concept we'll need is a version of the idea of attitudinal pleasure: a pro‐attitude, in this case an attitude of acceptance, toward what is arising in my experience. There is some range of conditions I might be experiencing that I would be willing to accept. Call this range my acceptance window. If the ambient conditions are outside the window, I refuse to accept them, and as a result, I suffer. But within the window, I can take things as they come, and I feel no need to reject or struggle against my experience.

Now we could say that there are two kinds of improvements that might lead me to experience more attitudinal pleasure. First, my conditions might improve, so that I spend more time within my fixed acceptance window. Or, second, my window might expand to include some of the conditions I previously wasn't able to accept. Now I feel better about my life without changing my external circumstances. Moreover, my pro‐attitude toward experience is now more robust, as the range of situations in which I will be contented is broader.

This picture is probably much too simple, but it's worth exploring what the toy model implies. We could say that one of the major goals of Buddhist practice is to expand the acceptance window. Indeed, partially overcoming the poisons of attraction (rāga, lobha) and aversion (dveṣa) would precisely be expanding the window, and completely overcoming them by becoming Awake would mean expanding the window to include all possible experience. There's a sense of the word “equanimity” that refers to this kind of expansion.

Then we could say that Buddhism promises happiness in not one, but two senses. By expanding the acceptance window, you get attitudinal pleasure: something that just about everyone wants, and something that it wouldn't be too much of a stretch to see as a form of happiness. Then, by embodying the Buddhist virtues, you get what Flanagan calls eudaimonia^Buddha, a particular version of well‐being that is a straightforward alternative and competitor to eudaimonia^Aristotle.

It is only by embodying virtues that you are eligible for eudaimonia^Buddha; and what Flanagan calls the “normative exclusion thesis,” that good feelings produced in a morally unacceptable way don't count, will apply most straightforwardly to this meaning of happiness. Precisely for that reason, some people will find this kind of happiness unappealing. On the other hand, the normative exclusion thesis doesn't apply so easily to attitudinal pleasure. Nor is expanding the window the normal cause of attitudinal pleasure, because most people who have it, get it because, either through effort or through luck, their circumstances come to fall within their fixed window. But there's a clear sense in which expanding the window is a more reliable cause of attitudinal pleasure: it makes your acceptance of your situation more robust to perturbations in your circumstances.

It's hard to see how psychologists, at anything like their current state of knowledge about the brain, could even begin to try to measure eudaimonia^Buddha. But there's a real chance they might, sometime in the not too distant future, figure out how to measure attitudinal pleasure. And if it turned out, as Buddhists usually believe, that meditation practice really does widen the acceptance window, then that would go a long way to vindicating the kind of claim that modern Buddhists want to make: namely, that their practices and teachings have a genuine contribution to make to the peace and happiness of the modern world, and that this contribution does not in any way depend on blind faith in the supernatural, but fits comfortably in a naturalist worldview. If such a view ever is vindicated to the satisfaction of philosophers and scientists, then Flanagan's work will have represented a notable milestone on the path to a very important set of results.

Notes