(In the last months, the IPhblog has hosted a long discussion on Evan Thompson’s last book, Why I am not a Buddhist. We are grateful to the participants, namely Amod Lele and Evan Thompson himself for their patience and accuracy in discussing so many details of the book and its written and unwritten assumptions and consequences. We aim at hosting further similar discussions in the future…although perhaps not that long! This also the reason for publishing this reply as a single post, although it in fact consists of an overall answer and of detailed points.–EF)
I had hoped to give Lele the last word in our exchange after my previous post, “Cherry Picking the Bodhi Tree.” This seemed only fitting, since he had begun the conversation by giving attention to my book at his Love of All Wisdom blog and had been willing to engage in a further exchange about it here at The Indian Philosophy Blog.
In his latest, seven-part follow-up reply, however, he declares that “[t]here is a great deal of conceptual confusion” in what I wrote and that “the goalposts may have been moved.” He also says that I’m “entirely wrong” about Śāntideva, and that I “disrespect” the text of the Bṛhadāraṇyaka Upaniṣad.
In this post, which will be my last one in our exchange, I will reply to these statements. But first I will explain why I think Lele has lost the thread of the argument and hasn’t squarely faced the fundamental issues that have arisen in the course of our exchange.
The fundamental issues are about karma, but they also encapsulate larger issues about Buddhist modernism. I say that the concept of karma is inherently eschatological-theodicean (I’ll say more about these Greek words below); Lele denies this, and says that the concept is inherently only eudaimonistic (another Greek word). Note that I don’t deny that the concept of karma has a eudaimonistic element. Rather, I say that this element by itself isn’t sufficient for what karma means, because the concept of karma necessarily includes eschatology and theodicy. So, the crucial issue is whether can you remove the eschatology-theodicy from the concept of karma and still legitimately claim to be talking about karma. Lele thinks you can, whereas I think you can’t.
What Lele does is to invent a new concept. He deletes the idea that karma operates across former and future lives. This immediately raises the problem of how karma can possibly work, given that many people act in good ways and experience bad results, and act in bad ways and experience good results. Lele’s answer is that karma works probabilistically: good actions are more likely to produce experienced good results in this life, and bad actions are more likely to produce bad experienced results in this life. Karma is “probabilistic this-world karma.”
I’d call this shwarma, not karma.
If you think I’m being tendentious, here’s an evolutionary analogy. Under certain selective pressures (modernity), adaptive genetic mutations occur (the deletion of the eschatology and soteriology gene-complexes and the addition of the naturalism gene-complex), and a new species is formed. “Probabilistic this-world karma” may descend historically and conceptually from karma, but it’s a new species of concept.
Let me make clear that I have no objection to conceptual evolution or invention. I’m also not arguing for ethicized rebirth—the existence of former and future lives determined by good and bad karma. (I’ll come back to this point later.) What I object to is saying that you’re preserving the core of the traditional concept when you’re actually engaging in radical, modernist, conceptual invention. This is what Buddhist modernists say and do. This is what Lele says and does. He invents a new concept while claiming that he’s also preserving the core of the traditional concept. It’s an example of what I mean when I say Buddhist modernists want to have their cake and eat it too.
Suppose Lele were to say the following (he does say many of these things but not all of them): “I consider myself to be a Buddhist, because the Buddha, his teaching , and his community are what I trust and rely on to guide me in life, especially in the face of impermanence and death. I’m a philosopher and want my thinking to be informed by the deep tradition of classical Indian Buddhist philosophy, especially Śāntideva. I’m inspired by his writing, and his understanding of good and bad mental states and their effects. But I find myself unable to accept rebirth, I don’t want to reject worldly life, I don’t think the removal of suffering should be our only ultimate goal, and I’m doubtful about nirvāṇa, at least if it’s understood in strictly otherworldly terms. This puts me in a difficult position with regard to traditional concepts, such as karma. Since my convictions rule out traditionalism, I’m going to have to pick and choose (and traditionalists pick and choose anyway). One way to do this, if you’re a philosopher, is conceptual engineering. This means you don’t just ask what concepts mean; you ask what they should mean. So, I’m going to ask, ‘Which concepts should I as a modern Buddhist embrace, and what do I think they should mean for us today’? In working out my answers, I’m going to radically alter some traditional concepts and thereby invent new ones. I’m going to cherry pick traditional sources. I’m going to combine what I cherry pick from Buddhist philosophy with what I cherry pick from other traditions and philosophers, such as Aristotle. This kind of cherry picking is unavoidable, given what I’m trying to do, which isn’t to give a faithful interpretation or reconstruction of classical Buddhist sources, but rather to use them as resources for my new, conceptually engineered inventions. Although these inventions aren’t going to be able to correspond to traditional Buddhist concepts, they’ll be descended from them and they’ll be part of an evolving Buddhist intellectual tradition with which I identify.”
If Lele were to say something like this, I’d have no objection (at least not yet). This is what philosophers do, in contrast to historians and philologists. He’d be owning up to cherry picking, and so wouldn’t be trying to have his cake and eat it too (to mix metaphors). He’d be thinking as a Buddhist and as a cosmopolitanist (in the philosophical sense of cosmopolitanism I advocate in Why I Am Not a Buddhist). I’d say, “I sympathize with your situation, more power to you, let’s see what you can do, and let’s see how others inside and outside the tradition respond.” For me, the proof would be in the pudding (to add another metaphor).
In the case at hand, the pudding is “probabilistic this-world karma.” This is the conceptual engineering we’re being offered. I don’t think it works.
“Probabilistic” properly speaking applies to a scientific causal model. When we have a causal model consisting of a set of variables, we can define a probability measure over propositions about the variables and their values. But what is the model in the case of eudaimonistic karma? What are the variables and parameters? Unless we can specify these, we don’t really know what we’re talking about. Are we supposed to use game theory and decision theory to answer these questions? But these theories typically abstract away from individual moral virtue, which is the key concept of eudaimonistic thought. These theories and the concept of karma don’t seem to fit together very well.
I worry that we’re dealing with the same kind of scientism that infects much of Buddhist modernism. By “scientism” I mean using scientific concepts where they’re not appropriate. “Eudaimonistic karma” is a normative concept. It refers to good and bad individual mental intentions and actions. Normative concepts operate in the logical space of reasons—the logical space of being able to justify what you say in relation to norms and values. The concept of probability, however, is a descriptive one that operates in the logical space of causes. When you say “probabilistic karma” you just code switch between the normative and descriptive languages without helping us to think about their relation.
There are other problems. As I wrote before, the proposition that good and bad actions are more likely to produce experienced good results, and bad actions are more likely to produce bad experienced results isn’t obviously true as a general empirical proposition about the world. Maybe it’s true, and maybe it’s false. Maybe it’s true for some people in some contexts, but not for others. So, appealing to it as the principal reason for being good doesn’t look very compelling (at least to me).
Suppose, however, the proposition is true. Why is it true? Is it true because of some inherent causal power on the part of good and bad intentions, as traditional Buddhists believe? (Mādhyamikas are an exception, because they deny that anything has inherent causal powers.) Or is it true because of social and political structures that regulate human behaviour? If the latter factors carry more of the causal weight, then eudaimonistic karma loses causal relevance in proportion.
Here’s another problem. Traditionally, karma is operative in this life from conception and continues until the moment of death (and then into the next life). But eudaimonistic karma, as Lele conceives of it, can operate only when we have virtuous agents. We don’t think of babies and very young children as virtuous agents. (Aristotle believed children couldn’t be virtuous agents.) So, exactly when does karma start? And does it end if we lose mental competence?
These questions about karma don’t make sense from a traditional Buddhist perspective. They do make sense, of course, if we substitute the Aristotelian idea of a virtuous agent or the Lockean idea of a person for “karma.” This is more evidence that we’re not really dealing with the concept of karma anymore but rather with an ersatz concept. We’re dealing with Greek and European ideas in Buddhist terminological wrapping. That’s a typical Buddhist modernist kind of package.
One more problem and then I’ll end this line of thought. If non-human animals aren’t capable of being virtuous agents (as is generally believed), then eudaimonistic karma can’t apply to them. This conflicts with the traditional Buddhist idea of the diversity of karma (karma in all the diverse realms of sentient beings, including animals). So, the concept Lele is offering is actually “probabilistic this-world human-only karma.” This is still more evidence that the meaning of karma has changed so fundamentally from what it traditionally means in Buddhism that we’re really talking about something else.
Given these problems, the idea of “probabilistic, this-world, human-only eudaimonistic karma” seems useless to me. It’s bizarre from the perspective of traditional Buddhism, and it raises more questions than it answers from the perspective of contemporary cross-cultural philosophy. In my estimation, it doesn’t work as conceptual engineering. It looks to me like another case of philosophically unsound Buddhist modernist thinking.
Let’s turn to Lele’s charges that I’m conceptually confused and I’ve been moving the goal posts.
In using the term “eschatology,” I was following Obeyesekere. Eschatos means “last” and eschatology is the part of theology concerned with “end things”—what happens to the soul at death, the afterlife, the final events of history, and so on. Lele says Obeyesekere is misusing the term, but this isn’t the case. Obeyesekere uses “eschatology” to refer to that part of a cosmology that describes what happens to the soul at death. This is a perfectly legitimate usage.
Here’s a quick summary of Obeyesekere’s theory. In some eschatologies, the soul goes to another world at death and stays in that world. In a “rebirth eschatology,” however, by definition the soul’s stay in the otherworld is temporary because the soul must eventually be reborn in the human world. The crucial factor that transforms a rebirth eschatology into a “karmic eschatology” is “ethicization.” The ethicization of religion involves the following interrelated things: understanding actions that are morally good or bad, according to societal norms, as also religiously good (merit) or bad (sin); the presence of priests (Vedic religion), prophets (the biblical tradition), or ascetics (Buddhism, Jainism) who enforce that understanding; the transformation of the otherworld into a world of reward and retribution; and, in the specific case of a rebirth eschatology, the idea of rebirth into the already ethicized human world according to “the principle of the contingency of reward,” whereby good and bad rebirths are based on the quality of the person’s actions during their previous lifetime, and where “good rebirth” means health, wealth, and high social status. In this way, given ethicization, a rebirth eschatology turns into a karmic eschatology.
By the way, the question “Why do bad things happen to good people?” is shorthand for the problem of why there is evil and an unequal distribution of happiness and suffering in the world. Why do bad things happen to good people, and good things happen to bad people, now, in the past, and in the future? Why are some people born into low and impoverished stations, and why are some born into high and privileged stations? Why are some people born sick and weak, and others are born healthy and robust? And so on. It seems to me that Lele is being literal-minded in how he understands the question.
Although Obeyesekere doesn’t state explicitly (to my knowledge) that a karmic eschatology explains why bad things happen to good people and vice-versa, it’s pretty obvious from everything he does say that this is one of the big explanatory upshots of a karmic eschatology. Indeed, the Buddha makes this kind of explanation explicit in the Cūḷa-kammavibhaṅga-sutta (“The Shorter Exposition of Karma,” Majjhima Nikāya 135).
Lele says Obeyesekere confuses eschatology with soteriology, but this charge is also incorrect. On the contrary, Obeyesekere carefully distinguishes between the two. He explicitly argues that the concept of salvation or nirvāṇa can’t be derived from just a karmic eschatology. Rather, salvation or liberation is an additional religious element, which, when joined to a karmic eschatology, must be conceived of as happening outside of the cycle of rebirths (saṃsāra) and as resulting in the cessation of rebirth and karma. Thus, nirvāṇa necessarily is otherworldly. (This is one reason why Nāgārjuna’s famous statement that there’s no difference between nirvāṇa and saṃsāra is radical and paradoxical.)
Lele says I confuse eschatology with theodicy. In fact, in my first post, I wrote: “The idea of karma functions analogously to a theodicy; it answers to question of the problem of evil.” Eschatology and theodicy are closely interrelated concepts: eschatology explains what happens at the end; theodicy deals with the problem of evil and suffering. Accounts of what happens at the end can help to give an answer to the problem of evil and suffering. This is precisely what a karmic eschatology does, among other things: it gives an answer to the problem of evil and suffering.
So, there’s no conceptual confusion on my part. I’m using “theodicy” and “eschatology” to talk about various aspects of one and the same package. For this reason, Lele’s remarks about eschatology being concerned with the future, theodicy with the present, and karmic theodicy with the present in light of the past are beside the point (namely, that karma is inherently eschatological-theodicean).
Lele says I “disrespect” the Bṛhadāraṇyaka Upaniṣad by treating it as concerned with eschatology and theodicy. He may disagree with my reading but the claim of disrespect is without merit. I refer Lele to Obeyesekere’s analysis of the emergence of karmic eschatology in the Bṛhadāraṇyaka and Chāndogya Upaniṣads. Again, although Obeyesekere doesn’t state explicitly that the Upanishadic karmic eschatology explains why bad things happen to good people and vice-versa, it’s pretty obvious from what he does say that this explanation is a significant upshot of the karmic-eschatological framework.
Some readers may be getting frustrated by all these Greek terms—eschatology, theodicy, soteriology, and eudaimonia. Why are we using Greek terms with their sedimented Aristotelian, Hellenistic, and Christian meanings to talk about Indian philosophy and religion? I agree this is unfortunate. If you want to propose better Sanskrit terms, I’m all ears.
One last point about karmic eschatology. As I said earlier, I’m not arguing for the existence of rebirth. Lele refers to psychiatrist Ian Stevenson’s research on purported memories of past lives, especially in young children. Stevenson claimed that these purported memories were strongly “suggestive” of reincarnation. Lele mentions them as a “potential anomaly” for the current scientific consensus that there’s no evidence to support the survival of consciousness after bodily death. Even if we took the Stevenson cases in the way Stevenson suggests, however, they wouldn’t necessarily be evidence that rebirth happens according to a moral principle. In other words, they wouldn’t necessarily be evidence for a karmic eschatology (ethicized rebirth).
Without getting into metaphysical issues about consciousness, I don’t think there is scientific evidence for the survival of personal consciousness after bodily death. In Waking, Dreaming, Being, I examined the Stevenson cases and near-death experiences as possible evidence for survival but concluded that they don’t warrant this conclusion. Here’s what I wrote about Stevenson:
Although Stevenson’s presentation of these cases often makes for compelling reading, all the evidence is anecdotal and derived from interviews where there is a large amount of room for false memory and after-the-fact reconstruction. The interviews weren’t conducted directly with the children when they first reported the memory, but only some time later, so there had been plenty of time for the child to assimilate information gotten from family members and to repeat it as if it were his or her experience. And sometimes the children weren’t interviewed at all; only family members were. Finally, it’s hard to know how to assess whether a memory report about a past life exceeds chance probability, and critics have pointed to a number of serious flaws in Stevenson’s statistical reasoning. For these reasons, I don’t find Stevenson’s evidence convincing, though it does seem possible in principle to investigate claims of past life memories using scientific methods (p. 290, with endnote references deleted).
On to goalpost moving.
Lele suggests I’m moving the goalposts when I say that exegesis of philosophical texts isn’t the right method for determining the core of the concept of karma. But my assumption was never that our discussion should be confined to Buddhist ideas as they’re presented in Buddhist philosophical texts. Our whole exchange started with Lele’s posts about my book, which is a philosophical discussion of Buddhist ideas as they’re embodied and expressed in many kinds of texts, communities, and meditation practices. A philosophical discussion need not be about just philosophical texts. In addition, a philosophical analysis of a philosophical text can make use of methods and data from other disciplines, such as anthropology (hence my use of Obeyesekere’s work), and history and cognitive science (which I use in Why I Am Not a Buddhist).
Of course, I’m not saying that philosophical analysis of philosophical texts isn’t important; I’m saying it’s not sufficient. (I think this is clear from the wider context of my whole post.) Karma isn’t like apoha (exclusion) or svasaṃvedana (reflexive awareness); it’s not a technical philosophical concept invented by philosophers. Philosophers can refine a concept like karma or God, and they can use it for their own ends. They can give us insight into what it means but they’re neither the creators nor final arbiters of its meaning. That meaning comes from the larger religious and cultural tradition. To get at that meaning we need other disciplines and methods besides philosophy.
This is the main reason why I linked to the social-psychology and cognitive-science-of-religion studies about how people think about karma. I think this kind of research provides important and philosophically relevant information about the cultural and religious significance of the concept of karma. I don’t have the space to talk about these studies here, so I will let readers come to their own judgements about their relevance. But I do want to take back one thing I said. I said that these studies “indicate” the concept of karma is fundamentally a way to handle the problem of evil. This is an overstatement. Rather, they support this claim, especially when taken in tandem with Obeyesekere’s analysis of the concept’s formation and function. They provide further evidence that karma is an eschatological-theodicean concept, so that if you remove eschatology and theodicy from karma, you’re no longer talking about the same thing.
The last three paragraphs have been about methodological issues, but there’s also a substantive issue in the vicinity. Lele thinks that Buddhist philosophers are to Buddhism, and Christian philosophers are to Christianity, as scientists are to science. I don’t share this viewpoint. Scientists are the ultimate arbiters of science, but Buddhist and Christian philosophers aren’t the ultimate arbiters of Buddhism and Christianity respectively (nor should they be simply by virtue of being philosophers). Of course, we read philosophers to learn about karma or God (nothing I wrote implies otherwise), but they’re just one voice in a much larger conversation that includes artists, counselors, social workers, meditation teachers, ritual masters, activists, politicians, and so on. Concepts such as “karma” or “God” span and interconnect these many areas of human life in ways that technical philosophical concepts typically don’t. I agree with Lele that we should learn from traditions “at their best,” but in the case of Buddhism and Christianity (and Hinduism, Islam, Judaism, etc.) there are more ways to be “best” than there are in science (even allowing for all the ways to be excellent in science). Philosophy is just one of them. Buddhism has Buddhaghosa and Basho; Christianity has Boethius and Bach.
Time now for Śāntideva. Lele says I’m “entirely wrong” that the eudaimonistic aspect of karma has no warrant and motivation for Śāntideva without the karmic eschatology. I discern two reasons in Lele’s counterargument, though he presents them as one.
First, Śāntideva states that good actions bring about good results, and bad actions bring about bad results, in this life. This is eudaimonistic karma in this life, and it doesn’t depend on rebirth.
We know, however, that sometimes good actions bring about bad experienced consequences instead of good ones, and bad actions bring about good experienced consequences instead of bad ones. So, we’re right back to the problem of evil and the unequal distribution of suffering and happiness in the world. As far as I can see, Śāntideva doesn’t have an answer to this problem without the karmic eschatology. Lele’s answer is probabilistic (this-world, human-only) karma, which I’ve already discussed.
Second, Śāntideva states that good actions generate pleasant mental states, and bad actions generate unpleasant mental states, in this life. This is eudaimonistic karma in this life, and it doesn’t depend on rebirth.
But can’t bad actions also generate pleasant mental states? Suppose someone were to say, “I like tormenting and killing small animals (or worse, young children) because of the thrillingly pleasing mental states it gives me.” What can Śāntideva say in response without relying on the karmic eschatology?
One response would be to say that the person is deeply deluded, because the mental states aren’t pleasant; they’re actually unpleasant. I’ve heard modern mindfulness teachers say this kind of thing: You only think anger is pleasant; you’d see it’s unpleasant if you were truly mindful. You only think smoking is pleasurable; you’d see the sensations are intrinsically unpleasant if you were truly attentive to them, and that would motivate you to quit smoking. It’s possible this idea is true, but I’m doubtful. To show it’s true you’d have to rule out the possibility that the sensations become unpleasant as a result of subjecting them to metacognitive attention and cognitive reappraisal according to the conceptual framework of mindfulness practice. I also find it hard to believe the Buddhist statement that all sentient beings fundamentally seek to be happy. I suggest Buddhists go read Freud on the death drive. In any case, what’s to stop our imaginary person from saying, “So what if I’m deluded by your lights? To deludedly take myself to be experiencing pleasure means that it seems to me that I’m experiencing pleasure, and that’s all I care about.”
Again, as far as I can see, Śāntideva doesn’t have a compelling response to this line of thought unless he invokes the karmic eschatology. It’s the karmic eschatology that enables him to say that the evil things you’re doing now will come back to you and make you absolutely miserable, far more miserable than the misery you’re inflicting on others, if not in this life, then in the long succession of lives to come.
Lele says that Bodhicaryāvatāra 6:42 on previous harm to sentient beings doesn’t depend on rebirth: “In the past, I too have inflicted such pain on sentient beings; therefore, I, who have caused harm to sentient beings, deserve that in return.” “In the past” includes this present life, so the statement doesn’t have to depend on rebirth.
But if we restrict the statement’s scope to just this present life, what grounds the “therefore”? In other words, how does it follow that I deserve to be harmed because I’ve harmed others? Of course, I know that, as a matter of empirical fact, if I harm others, it’s likely I’ll be harmed in return, but I can always take steps to try to prevent that. This is a prudential matter, not a moral one. The question is what grounds the ethical principle that I deserve to be harmed in return? Not much, for Śāntideva, it seems to me, without the karmic eschatology. It’s the karmic eschatology that enables me to see my actions—indeed, my very being—as inseparably interwoven with yours all the way back across the beginningless succession of lives and all the way forward until all sentient beings are liberated. From this perspective, not only must every harm I inflict come back to me, but also the very distinction between your suffering versus mine falls away, and there’s just suffering and the need to remove it.
Lele points out that what I call the mind-expanding Mahāyāna narrative in which every sentient being has been your mother doesn’t appear in Śāntideva but only in later Tibetan commentaries. He says that’s one way to interpret Śāntideva but it’s no less an innovative interpretation than naturalized Buddhism is.
I disagree, and think there’s a huge difference between the two interpretations. Śāntideva is a Mahāyānist. In the Mahāyāna universe, karma and former lives are beginningless, and we’ve all been related to each other in innumerable ways. If the Tibetans are innovating, they’re doing so by making explicit straightforward implications of what Śāntideva thinks. Naturalistic Buddhists, by contrast, are innovating by asserting things that are straightforwardly inconsistent with what Śāntideva thinks (no former and future lives, no consciousness that survives bodily death), and are adding things that don’t follow logically from what he says (probabilistic karma).
Lele asks us to imagine Śāntideva being presented with compelling evidence that there’s no rebirth, and hence no basis for a karmic eschatology. Do we really think he’d abandon Buddhism and either adopt another tradition or become a libertine? No, probably not. (Although a lot would depend on his personality, of which we know nothing.) But that question is easy and isn’t the relevant one. The relevant question is harder: What exactly would Śāntideva think? Would he embrace Lele’s version of apolitical eudaimonistic Buddhism? Or would he combine eudaimonistic Buddhism with socially and politically engaged Buddhism (as Seth Zuihō Segall does)? Would he reject eudaimonism and opt instead for consequentialism, as Charles Goodman and Mark Siderits do in their reconstructions of his thought? Or would he try to defuse the whole problem through Madhyamaka anti-realism or quietism? (Let’s not forget he’s a Mādhyamika.)
I have no idea what the answers are to these questions, and I dare say neither does Lele. I share Stephen Harris’ doubts about our being able to ascribe any ethical theory to Śāntideva using our familiar Western philosophical categories. As Harris points out, Bodhicaryāvatāra is best read as first and foremost a meditation manual, not as a constructive philosophical text like Aristotle’s Nicomachean Ethics.
Indeed, it seems to me that Lele misses the import of his own question. This is that there is such a profound incommensurability between the concept of karma and Śāntideva’s practice, on the one hand, and modern philosophical sensibilities and practice, on the other, that building philosophical bridges between the traditions is much harder than Buddhist modernists realize. Śāntideva would indeed be in a tough spot, and which way he’d choose to go forward is impossible to say.