22 May 2015

Critiquing Buddhist Karma

In researching karma and rebirth I came across an interesting article by well known scholar of religion, Paul J. Griffiths. It dates from 1982 and while there was some immediate response from one scholar and the article has been cited a number of times, the ideas in the article seem to end up going nowhere. I read recently that 90% of humanities articles are never cited!

One of the main points made in the Griffiths article is that there is very little critical evaluation of Buddhist philosophy or Buddhist ideas more generally. My own reading on the subject of karma has turned up few critical articles (see Bibliography). What's missing, according to Griffiths, is a critical engagement with Buddhist ideas. By this he specifically means a genuine attempt to assess the validity of truth claims made by Buddhists (1982: 277).  

Scholarly publications on Buddhism seem to fall into a number of categories:
  1. Descriptive works which are concerned with continuing to flesh out the history of Buddhism and to describe the intricacies of this complex subject. The opening up of Chinese Buddhist studies and comparison of Chinese, Gāndhārī and Pali versions of the early Buddhist texts mean that this descriptive phase of Buddhist studies will continue for the foreseeable future. 
  2. Apologetics, that is works whose aim is to defend a Buddhist worldview in some form. We have both religious apologists who seek to retain the traditional elements of Buddhism, and secular apologists whose views are broadly aligned with a Buddhist tradition and who write in such a way as to bolster traditional readings, particularly of history. 
  3. Polemics of the field of Buddhist Studies itself, which are aimed not at Buddhism per se, but at the hubris of scholars making claims based on texts which contain far more uncertainly, ambiguity or down right incoherence than Buddhist Studies scholars like to admit. At best these result in more sophisticated articles of type 1.
  4. Ideological polemics aimed Buddhism, which essentially criticise Buddhism for not being, for example, Christian, Vedic, or Marxist enough. 
There's quite a bit of work which is comparative, especially recently comparative ethics, which seeks to find points of cross-over between Buddhist thought and the Western Intellectual tradition. These seem to combine descriptive and apologetic modes of writing. It's a mode of positively engaging with Buddhism, but it never really gets down to assessing the truth claims made by Buddhists.

There is a large gap in the market when it comes to the kind of critical attention that Griffiths has in mind. And in an age where the claims of religion are increasingly challenged head on, it is surprising that Buddhism appears to be escaping the kind of scrutiny that atheists are directing to Christianity (which may simply be a measure of how unimportant Buddhism is in the West). Although Griffiths was writing more than 30 years ago, very little appears to have changed. There is almost no critical engagement with Buddhism in the Academy and yet at the same time attention is lavished on Buddhism. This might be because disproving the truth claims of Buddhists would be killing the goose that laid the golden egg for scholars of Buddhism (though the survival of the field since Griffiths is a counterargument to this). Many people see the criticisms that I propose, for example, as "destroying" Buddhism (at the very least I regularly have people tell me that I am not really a Buddhist). Another factor is that many scholars of Buddhism are clearly in love with their subject and have all of the objectivity of the love struck. They can hardly be expected to criticise their love interest.

Another explanation is that we have adopted the anthropologist's disinterested, objective stance. I recall discussing this with an anthropologist friend who did her post-graduate work on stupa building projects in New Zealand and Australia. She showed me pictures from her Australian group which showed physical relics of the Buddha and the previous Buddha. I scoffed, since at best the previous Buddha is a myth, but she told me that her stance was to accept that this is what they believed and to focus on how this belief affected their lives and actions. The fact that they venerated such "relics" was more interesting than the truth of their claims. Assessing the belief played no part in her approach to research. Her work is thus mainly descriptive. Such is also true for those whose focus is historical or linguistic.

My impression is that it's not just Buddhism. The whole field of Indology skirts around the problem. Of course criticising Indian thought is a political minefield. Another problem for Western scholars who positively engage with Indian thought is that they risk being labelled as racists or cultural imperialists. Professors Michael Witzel and Wendy Doniger for example regularly suffer these canards, largely because their work does not satisfy the criteria of nationalists.

Doing Philosophy with Buddhism

Problems emerge when we engage with Buddhist ideas as philosophers. There is a great deal published now on the subject of Buddhist (or Indian) philosophy and the history of Buddhist ideas. As philosophers we cannot be content with a descriptive approach. We have an imperative to weigh the claims of Buddhists to see if they are true, or to what extent they are true. And if they are not true, then we have an obligation to say so, and to make a case for abandoning the claim. It's very difficult for Buddhists to do honest philosophy when we are in love with Buddhism. We are too strongly subject to cognitive bias. The same is true for many scholars of Buddhism. Personally, I found the intellectual weakness of Buddhist teaching, and in particular the teaching I had received directly, quite shocking once I began to study the history of Buddhist ideas in earnest (I think of Professor Gombrich's 2006 Numata lectures as a watershed in this sense).

Some Buddhist bloggers have taken up the challenge, to some extent: e.g. myself, David Chapman,  and Glenn Wallis (and others who used his blog as a vehicle; with whom my relationship is complicated). But we are not always consistent, or always coherent, and we all have different approaches and agendas. One thing we all have in common is limited success in engaging mainstream Buddhists. There's very little interest from the wider Buddhist community, who almost universally prefer to read confirmation of their beliefs rather than challenges. The most popular Buddhist blogs simply reflect Buddhists beliefs back to them. Thus those of us who write critically about Buddhism, are either preaching to the converted or to the birds.

What makes Griffiths interesting is that he is trying to do philosophy with Buddhist ideas, rather than trying to justify a religious view. And this means that his paper is one of the most interesting articles ever published on Buddhist philosophy. It is certainly a relief from the steady stream of (re)interpretations of Nāgārjuna's impenetrable jargon-filled jumble. Not only do the least interesting philosophers of Buddhism seem to get all the attention, but we seldom seem to get to the nub of the issues they were grappling with.

There was a response to Griffiths (1982) from White (1983) and then a rebuttal from Griffiths (1984), but little beyond that. Bronkhorst has tackled the problem of teleology in Indian conceptions of karma across the board (2000). More recently Cho (2014) has joined the discussion from an interesting angle. In response to attempts to jettison karma by secularists he argues that Westerners have failed to understand how traditional cultures make use of karma. Meanwhile apologetics continue to multiply: e.g. ThanissaroBodhiSangharakshita, and Vessantara's resent essays Some Problems with Not Believing in Rebirth & More on Rebirth.

Griffiths acknowledges that there is a difficult apprenticeship to grapple with the subject. It requires at least some familiarity with a number of scriptural languages. Griffiths's own article suffers from a common complaint, which is over-reliance on Vasubandhu's Abhidharmakośabhāṣya as a source book on sectarian Buddhist views. This general over-reliance on the Bhāṣya is because most of the surviving source material is only available in Chinese translation. Sanskritists and Palists have historically not had the Chinese language skills to match. It's not surprising as mastering these languages takes years of effort (it takes about 5 years to get really confident in Sanskrit), though this is changing.

Griffiths Thesis

Griffiths distinguishes three major functions of karma in Buddhism (280):
  1. as an "explanatory cosmogonic hypothesis"; the universe as created by volitional acts.
  2. as an "explanatory hypothesis for the varied states and conditions of sentient beings"; why we are human, why some have good/bad fortunes.
  3. as a means of social control in Buddhist societies; "acting as a powerful mechanism for regulating and enforcing the essentially hierarchical structure of Buddhist societies and of providing a rationale for Buddhist soteriological practice..." 
From these functions Griffiths derives seven truth claims, seven propositions that have to be true if the Buddhist account of karma is to be judged true as a whole (in the article they are labelled P 1-7). Griffiths is at pains to say that he has no space for a full treatment of these propositions. However, for our purposes even a superficial analysis is useful. In this essay I will focus on the propositions that Griffith deduces from the second major function. 

In my previous essays I have already showed that Buddhist cosmogony cannot be taken literally, thus the propositions (P1 & P2 n Griffiths notation) that derive from this function are not very interesting since they are patently false. If it once provided Buddhists with a satisfactory account of the origin of the universe then it does not now do so. Griffiths emphasises this when he shows that for the karma theory of the origin of the universe to be true, it means sentient beings must precede the origin of the universe, which is nonsensical.

The propositions (P6 & P7) associated with the third, regulatory, function probably do help to regulate Buddhists societies as long as they are treated as being true. Traditionally Buddhists have believed in a supernatural function of the universe, which correlates past actions to present vedanā and future punarbhava or rebirth. In fact my own work has already shown that Buddhists did not come up with a completely coherent account of this function. But a coherent narrative is perhaps less important than a compelling narrative when it comes to motivating people to ethical behaviour. 

This brings us to the truth claims for the second function. These seem to me to be the most important claims made by Buddhists. As an explanation for our present state, Griffiths suggests that the Buddhist theory of karma makes the following truth claims (282, paraphrased):
P3 Each individual undergoes more than one life.
P4 The parameters for any individual are set at conception and result from actions in previous lives.
P5 There is no undeserved suffering.
As critical readers we must not only assess the validity of the author's conclusions, but also the strength of the reasoning involved and the validity of this starting propositions. We need to be clear that the basis of this account of karma is Vasubandhu's Abhidharmakośabhāṣya. I have a serious objection to P5, but let us work through Griffith's argument. 

Multiple lives
P3 Each individual undergoes more than one life.
Griffiths argues that if define an individual as having physical continuity in time, then P3 is incoherent since death clearly disrupts physical continuity. A soul inhabiting many bodies is unacceptable to Buddhists. Buddhists conceive of an individual as a series of psycho-physical events (khandhas). In this view death is not an interruption to the series, even though the individual body ceases. I've already discussed Vasubandhu's approach to rebirth to some extent. He believed that rebirth occurred an appreciable time after death if only to account for the time taken for vijñāṇa to travel from one physical location to another. Vasubandhu was against instantaneous arising of vijñāṇa elsewhere. 

As such Griffiths reformulates P3 as
P3' Any given caused continuum of momentary states exhibiting sentience (i.e. an individual) does not cease with death (283).
This allows Buddhists to eliminate any tendency to postulate a more substantial something which dies and is reborn. Griffiths notes, as I have done, that Buddhists are likely to assert P3' when they are concerned with social control and as the basis of morality. If actions have consequences, but the consequences are lived by someone else, then that is no motivation for morality. Thus Buddhists discussing morality emphasise personal continuity. Whereas Buddhists discussing metaphysics and identity stress mere processes. 

A corollary of P3' is that all moments of being are simply moments in a series. In the strong form of this idea, identity is always merely contingent. In fact there are no individuals. It's a moot point to claim that at any two moments any two "individuals" are linked at all. Without the sense of continuity, the concept of an individual, the concept of individual responsibility for actions, breaks down. 

I would add that this was Nāgārjuna's ultimate argument against mainstream Buddhist karma theories:
karma cen nāsti kartā ca kutaḥ syāt karmajaṃ phalam |
asaty atha phale bhoktā kuta eva bhaviṣyati || MMK_17.30 ||
If there is no agent and no action, could their be result born of action?
In the absence of a fruit, how can there be one who suffers the result? 
kleśāḥ karmāṇi dehāś ca kartāraś ca phalāni ca |
gandharvanagarākārā marīcisvapnasaṃnibhāḥ || MMK_17.33 ||
Defilements, actions, forms, agents and fruits;
Are like a Gandharva city, like a mirage or a dream. 
All the talk of morals is just an upāya, a lie that is justified by compassion. However Griffiths argues that neither P3 nor P3' stand up to criticism and are in fact both false, because the whole idea of an individual is false in the strict reading of Buddhist metaphysics. Taken to it's logical extreme we get to Nāgārjuna's conclusion, that ultimately none of these ideas constitutes a substantial reality. Nothing is what it seems and the idea that anything could persist long enough to earn the title "individual" is simply wrong. Ergo, there is no one to reap the fruit of actions, no one to be reborn. Here we see precisely why other Buddhists considered Nāgārjuna to be a nihilist. However, barred from any form of Realism (Sarvāstivāda and Pudgalavāda notwithstanding) Buddhists seeking an alternative to Nāgārjuna's perceived nihilism, fell into forms of idealism, specifically citta-matra, the idea that there is only mind.

Time and again when this idea comes up, Buddhists cite researchers into paranormal phenomenon such as Ian Stevenson or Jim Tucker. I've dealt with some of the problems of this kind of research previously: Rebirth and the Scientific Method (1 Oct 2010) and Rebirth is Neither Plausible nor Salient (27 Jan 2012). What Stevenson in particular thought he had evidence for was reincarnation. The same soul returning to be incarnated in a new body. If this were true then the whole edifice of Buddhist metaphysics would be broken, since we vehemently deny that existence of a soul and the very possibility of reincarnation (Tibetan tulku's notwithstanding).

Influence of Previous Lives

P4 The parameters for any individual are set at conception 
and result from actions in previous lives.

By parameters Griffiths means state of congenital health, intelligence, physical appearance, i.e. all the factors that we would now ascribe to genetic inheritance from our parents. Not included, is what we will make of our lives within these givens. Clearly if the propositions P3 and P3' are false, then P4 must also be false, since it is predicated on P3 (previous lives). However, as Griffiths notes, Buddhist karma theory is not strictly deterministic. On this basis he is willing to grant that actions in a previous life might have influenced this life, with the caveat that the mechanisms suggested by Buddhists are hardly credible. 

It may be that Griffiths was not writing from a very informed position on genetics, or that in 1982 genetics seemed a less promising science. This was before the sequencing of the human genome for example. But from my point of view, Griffiths is far too generous on this point. Even if we take in the newish field of epigenetics, which (re)opens the door to Lamarckian passing on of passing on learned characteristics or at least parental responses to environmental conditions, the idea of influence from one life to another is no longer credible.

I've outlined Sean Carroll's argument against this and have yet to see any Buddhist attempt a refutation. Basically, if the body is made of atoms that hold the information which makes up our personality and memories, the second law of thermodynamics means that the information cannot survive our death. Additionally if there were other forces that could affect matter in any perceptible way we would have found them by now. The argument against this is typically dualistic. Mind stuff is different from matter stuff. However for mind stuff to be interesting it would have to interact with matter stuff, since mental events always have a material correlates. Such interactions would be detectable and thus since we do not detect them we have a strong argument against a separate mind stuff. There is only one kind of stuff. 

Thus Griffiths was too soft on Buddhists in this part of his argument. Hereditary does indeed set the parameters for our present lives, but there is no evidence that this has anything to do with having lived before. Our parameters are set by our particular recombination of our parents DNA, a few epigenetic factors and the environment we develop in. The combination of hereditary and environment explain everything about us without the need to invoke previous lives.

When one looks at afterlife beliefs they are almost always tied to the idea that the universe is moral or ethical. An afterlife is required to deal with the patent unfairness of life. And this brings us to proposition five. 


P5 There is no undeserved suffering.

This proposition is not logically contradictory or incoherent in itself. However, in Buddhism it is always tied to the previous propositions. One could hold this view independently of Buddhist truth claims (many conservatives appear to hold this view for example), but according to Griffiths, traditionally Buddhists always combined them. Indeed the attractiveness of Buddhist karma as an explanatory power is precisely for the kinds of extreme circumstances cited by Griffiths: e.g. the suffering of infants who have had no chance to commit the kinds of deeds that might warrant suffering a punishment (there's an assumption here that suffering can be deserved or undeserved, but we'll have to take it on face value for now). Explaining undeserved suffering has been a major feature of all religious moral narratives, which are necessarily predicated on the idea of deserved suffering. However, I think Griffiths has missed some subtly here. Buddhists have a variety of responses. While Tibetan Buddhists appear to believe that everything that happens is a result of karma, and this accept that the suffering infant must have been evil in a past life; the Pāḷi texts make it clear that karma does not account for illness, only for birth in the human realm where one is subject to illness and suffering.

I'm not aware of any traditional narratives from the Pāḷi to explain infant suffering, but there are several stories which purport to show how one might deal with the death of a child.  I'm thinking particularly of the Piyajātikā Sutta in which a man has lost his child and the Buddha simply tells him: that's just how it is (evametam evametam). The other stand out example is the story of Kisā Gotamī who loses her child, but is brought to a gentle understanding that death, even of an infant, is simply part of life one just has to learn to accept. None of the texts I am aware of attribute apparently undeserved suffering, such as the suffering of infants, to actions in a previous life. The one example of an infant suffering that comes to mind, is in the story of Aṅgulimala in which the eponymous character relieves the suffering of a mother having a difficult birth by using the magic of truth (I discuss this in Attwood 2014). Nothing in the story, in either it's Pāḷi or Chinese versions, suggests blame was apportioned to the infant. Aṅgulimala's suffering is a result of evil deeds in this life!

In my research on the inevitability of karma (Attwood 2014) I drew attention to a major change in how karma worked from the early Buddhist texts to the later texts. At first karma is absolutely inescapable. One must always live with the consequences of one's actions. This is very strongly emphasised. But gradually this criteria of karma is abrogated and ways to mitigate the effects of karma and to avoid them all together become mainstream Buddhism. The acme of this idea is the Tantrika chanting the Vajrasattva Mantra to eliminate any and all bad karma.

The fact is that even were the Buddhist theory of karma correct, there would be no way to link present suffering to past actions, because we do not have knowledge of those past actions. However, this limitation has not stopped Buddhists from constructing narratives which attribute present suffering to past actions in the form of Jātaka stories. These are mainly pious homilies which draw on the wider Indian culture (some of the stories also occur in Jaina and Brahmanical texts). Precisely this ability to see how past actions contribute to present suffering is one of the supernatural abilities which are attributed to the Buddha. This god-like ability is necessary for the Buddha to function as Buddhist saviour. In one view of this, the Buddha cannot be in the dark about this as we are, else we could not break free of suffering. He has to know how to act in order to not cause suffering. Since natural sources of knowledge cannot reveal this, the Buddha has to have supernatural knowledge. This is an example of a teleological argument of the type critiqued by Bronkhorst (2000).

Ultimately Griffiths rejects P5 on the basis that it is connected with P3. He also suggests that it is repugnant to most Western eyes and "certainly to Christians". However I find the latter an extremely weak argument. The karmic explanation is no less repugnant than the idea that an omnipotent God allows an infant to suffer, or causes that suffering, as part of some cosmic plan. Centuries of Christian arguments over Theodicy show that many Christians found there own narratives of apparently undeserved suffering equally repugnant. Griffith's objectivity has slipped here. 


Griffiths' own conclusion is that the philosophy of karma, as represented in the Bhāṣya does not stand up. He concludes:
"The empirical falsification of P1 and P2, the partial incoherence of P3 and its variant P3', the falsity of P4 in so far as it depends on P3/P3', the empirical falsification and moral repugnance of P5, and the vacuousness of P6 and P7 -- all these mean that Buddhist karmic theory as expounded in the major theoretical works devoted to it must be false." (291. Emphasis added)
I am in broad agreement with Griffiths, with some caveats as stated above. In some cases I find the case has strengthened over time. The empirical evidence against any kind of afterlife is much stronger in 2015 than it was in 1982. This is not to say that it is not possible to formulate a Buddhist theory of karma that is true, but that the traditional accounts are not true. Also any new formulation of karma must deal with the objections raised in the article.

As Griffiths points out, this may "pose many urgent questions for Buddhists". I suggested at the outset, such criticisms have largely been ignored. Those who write, for example, about Buddhist ethics do not seem to take a critical stance on traditional Buddhist moral philosophy. A great deal is written for example on whether Buddhist ethics is a virtue ethic or a consequentialist ethic, but very little about the fundamental validity of the worldview. And because we never really come to grips with the flaws in Buddhist thinking, we can never move on. Those who do write about it are marginal, if not marginalised. Dayāmati has asked why it even matters what kind of ethics Buddhism has in relation to the Western intellectual tradition.

One response is to quietly drop the subject of karma because of the supernatural aspect of it, which some people reject out of hand. Having rejected karma, one can then describe a secular humanist ethic with a Buddhist flavour: retaining the five or ten precepts, but explaining them in secular humanist terms. This suffers from the problem that many people identify in my own writing. A secular humanist account of ethics, albeit with a Buddhist flavour, is secular humanism rather than Buddhism. Isn't it?  If the underlying account of ethics is humanists, then the Buddhism is just window dressing or marketing. Certainly actions having consequences is no more a revelation in the Western intellectual tradition than is "everything changes". It's another case of "So what?" Defining what is Buddhist about Buddhist ethics in the absence of the supernatural elements is difficult. Buddhist ethics is predicated, as Griffiths suggests, on certain truth claims. Truth claims that turn out not to be true.

Coming back to the broader point about assessing truth claims, we can of course point to the efforts to research the effects of the practices known collectively as "mindfulness". On his blog, Justin Whitaker has written a useful summary of the latest round of recriminations against mindfulness, Buddhist mindfulness, morality, and Protestant presumptions. There is also criticism of the research into the effectiveness of mindfulness along the lines that it suffers from confirmation bias (unlike the rest of Buddhism?). There is a growing body of research and as long as it does not seal itself off from the outside world, like say paranormal research, then the scientific method will eventually sort out any kinks. These things take time. It is mildly interesting to see Buddhists attacking innovation and genuine attempts at scientific assessment, since this is exactly what we expect from a religious community. It confirms the problems that religious style thinking produces and highlights the clash with Enlightenment thinking.

If Buddhism is to have a future then we need to create an intellectual culture of open minded questioning and testing. At present, Buddhists are nice enough, but on the whole they don't ask the hard questions and they appear to dislike being asked hard questions by outsiders. And those who do ask hard questions are treated as apostates and outsiders. The problem is that a lot of what we take for granted as timeless truth is at best "a skilful means" and at worst simply false. But rather than face up to this and think about how to respond, most Buddhists are hiding their heads in the sand and pretending that such discussions are not happening. Of course the level of investment many Buddhists have in old ideas is enormous. For some, their whole identity is built on propounding "ancient truths" and all that. Scholars of Buddhism appear to be colluding with Buddhists in this. But it means that there is a huge amount of inertia. This calls for patience and compassion on the part of critical philosophers as we proceed to have this discussion that so many would rather not have.


Attwood, Jayarava (2014). Escaping the Inescapable: Changes in Buddhist Karma. Journal of Buddhist Ethics, 21. http://blogs.dickinson.edu/buddhistethics/2014/06/04/changes-in-buddhist-karma/ 
Bronkhorst, Johannes (2000). Karma and Teleology: A Problem and its Solutions in Indian philosophy. Tokyo: International Institute for Buddhist Studies. 2000. (Studia Philologica, Monograph Series, XV.) http://is.gd/Z85330
Cho, Francisca. (2014) Buddhism, Science, and the Truth About Karma. Religion Compass. 8(4): 117–127. http://dx.doi.org/10.1111/rec3.12103
Griffiths, Paul J. (1982) Notes Towards a Critique of Buddhist Karma Theory. Religious Studies 18: 277-291. http://dx.doi.org/10.1017/S0034412500014128
Griffiths, Paul J. (1984) Karma and personal identity: a response to Professor White. Religious Studies 20(3): 481-485. http://dx.doi.org/10.1017/S0034412500016334
Hayes, Richard P. (1989) Can Sense be Made of the Buddhist Theory of Karma? [Paper read at the Dept of Philosophy, Brock University]. http://www.unm.edu/~rhayes/karma_brock.pdf
White, J. E. (1983) Is Buddhist Karmic Theory False? Religious Studies. 19(2):223-228. http://dx.doi.org/10.1017/S0034412500015055

Related Posts with Thumbnails