Unresolvable Plurality in Buddhist Metaphysics?
In discussion over my forthcoming article describing changes in the metaphysics of karma, I raised the problem of the moral force of karma in the absence of personal continuity. An interlocutor responded that the tradition had resolved this problem, but I'm convinced this is not the case. I think this problem is decidedly unresolved. Most writing on the subject of karma assumes that a unified view is intended and can be discovered in the early Buddhist texts, but I see plurality that seems unresolvable.
One specific problem is this. The moral force of karma derives from the notion that we must live with the consequences of our actions and that even death is no barrier to the consequences being visited on us (or someone linked to us in a way we care about). It is fundamentally fear of negative consequences, particularly a bad rebirth, that pushes the unawakened Buddhist to be ethical; and the prospect of liberation from repeated death that pulls them along. This implies that the person who lives out the consequences of my present actions must in some way still be me. I must feel a sense of ownership over my actions and their consequences. In other words karma implies some kind of personal continuity or it doesn't make sense in human terms.
I'm well aware that this is specifically denied in texts such as the Milindapañha. According to tradition the one who experiences the results is not the same as the one who acted, but not different either. That person is dependently arisen (I'll come back to this). However if this was intended to make people behave according to Buddhist norms I can't help thinking that it's a rather poor attempt at motivating people. Theoretically the problem is solved, but practically it still disconnects the actor from both their actions and the consequences and thus can hardly motivate anyone to do anything.
Teachings on karma emphasise this implication by telling stories which explicitly link past and present lives. Such stories as the many hundreds of Jātakas (both in the two books of the Pali Jātakas and spread throughout the Nikāyas, Vinaya and other collections such as the Avadāna). In Theravāda countries the Jātakas are the main vehicle for teaching morality precisely because they emphasise living with the consequences of actions performed in part lives. However this is not to say that karma is only presented in these terms in Buddhist texts. Compare also such texts as SN 15.1 which describes saṃsāra in terms of ancestors stretching back through beginningless time; and SN 15.10 which by contrast describes one person (ekapuggala) wandering through saṃsāra leaving a mountainous pile of bones behind them. Karma is also said to be quite specific. My actions determine my rebirth, they do not determine your rebirth and vice versa. Similarly your actions do not give rise to my suffering except where they directly impact on me. But direct impact is unnecessary for karma generally since it is intention that determines outcome for the actor.
Now contrast the metaphysics of paṭicca-samuppāda applied to the sense of selfhood. For the most part Buddhists seem to insist that, in reality, there is no self. There is a strong influence of the Two Truths teaching in such statements which use the language of existence or non-existence, i.e. the language of ontology. The Two Truths are a pervasive tool for dealing with the paradoxes and contradictions thrown up in Buddhist ontology. However if there is no self, then there is no continuity over time and Buddhist ethics simply does not work. The language of ontology is carefully avoided in many early Buddhist texts that emphasise the application of paṭicca-samuppāda to experience only (the locus classicus being the Kaccānagotta Sutta SN 12.15) which is why I do not find the Two Truths teaching, with a foot in the camp of existence, useful (See Not Two Truths).
But even if we reject the language of ontology as belonging to the wrong domain (avisaya; cf. the Sabba Sutta SN 35.23) we are still left with a denial of personal continuity. The one who is reborn is not the same as the one who died, though not different either – they arise in dependence on causes. As Nāgasena says to king Milinda when asked about this problem: “It is not he, nor is it another” (na ca so, na ca añño. MP 41; c.f. S ii.18ff). The idea that vedanā arises because of oneself or another simply misunderstands how experience arises. Indeed the very question "who suffers?" is deemed unsuitable (no kallo) (SN ii.13). This is clear enough. But it's not clear how morality would work on this basis. If it is not me that suffers (or enjoys) the consequences of my actions then what is my motivation for practising virtue and avoiding vice? I don't believe that a morality based on such an abstract notion of responsibility is viable. And I would argue that the Buddhist tradition, in a tacit acknowledgement of this problem, does not teach morality in this way. Most Buddhists or whatever time and place teach some variation on "your actions have consequences for you and the people around you." Cf Buddhanet, The Budddhist Centre (1st sentence in both cases), SEP (§1 sentence 2). The theme recurs in many introductions to Buddhist ethics.
There is a fundamental disconnect between the metaphysics of karma and the metaphysics of paṭicca-samuppāda. I cannot see how to resolve these two while preserving the essential features of both. On the face of it this problem ought to have produced a crisis in Buddhist philosophy, though to the best of my knowledge it never has.
Why Do We Seek a Singularity?
Is it possible to step back from the content of this question and ponder why it is important to frame problem the way it is framed? In other words I want to ask if it is essential to attempt to reconcile the irreconcilable. I said above that writing on the subject of karma assumes that a unified view is intended and can be discovered in the early Buddhist texts. The central historical narrative of Buddhism is that it all springs from a single individual, the Buddha. The Buddha is presented as having first cut his ties to society (i.e. to all the conditioning of his early life from family, clan, class). Then over an extended period he pursues practices designed to break his identification with his body (practices which were historically associated with Jain ascetics). Then, completely cut off from his antecedents, the Buddha produced a new insight (prajñā) and founded a new lineage of instruction (anuśāsana). All Buddhist traditions take their teachings to be the direct or indirect words of the Buddha, if not in his historical manifestation then of the Dharmakāya.
Conventionally we expect that all the lines of the development Buddhist thought ought to converge at some point in the past. We expect the teachings to be unified and systematic. Many scholars of Buddhism declare that, from their point of view, unity implying a founder figure is clear in the texts. However even a relatively obtuse reader becomes aware that all is not unified. There are apparent discontinuities. The twelve nidānas are sometimes ten and sometimes eleven. And sometimes other numbers. I don't know how the tradition dealt with this, but in modern scholarship we have the handy model of evolution. If a teaching exists in various different forms then we can line them up chronologically (if only relatively) and argue that as the Buddha lived 80 years he must have refined his teaching as he went on and what we have preserved are various versions of the teaching from different periods of His "career"; or that it was developed by later disciples. Thus anomalies that might make us question the story of "unity" are used to support it using evolutionary models.
Lately I've been questioning the applicability of the the linear models that result from applying simple Darwinian ideas to the development of Buddhism (see Evolution: Trees and Braids). The tree structure, with its linear, binary diverging leads back in time to a singularity. A braid allows for divergence and convergence that is not unidirectional. Might it be that because the tree metaphor dominates our view of evolution or development that even when we find inconsistencies we still perceive them as springing from a single source? Such a reaction would also be consistent with my theory of how people with strong beliefs handle counter-factual information.
The fact that we buy into the idea of an historical founder is also a cultural lens. It also predisposes us to see unity if we believe in a founder figure. But which comes first? Do we read widely with an open mind and discover a unity which demands that we accept the idea of a founder figure? Or does the idea of a founder figure cause us to read with confirmation bias and see only unity and ignore and explain away diversity? Is not Buddhism always presented as the teachings of the Buddha? And if our religious beliefs predispose us to believe very strongly in the historical reality? And if we identify him as someone who shares our religious, philosophical and cultural concerns as most modern Buddhists do?
If we extend the image of a braided river and look at the sources of the river we find that many tributaries contribute to the stream. It's not always clear which is the mainstream and which the tributary. If we follow one tributary to source it may resemble a singularity, but we must always keep in mind that there are many tributaries with equal claim. Indeed we can say that each source spring is reliant on a watershed, and that the hydrological cycle recycles water in complex ways. The metaphor is complex, dynamic, and undermined singularity thinking. And since the object is human culture, we require any metaphor to have these qualities.
In my writing about the Buddhist texts, over many years now, I have noted broadly Vedic influences, more specifically Brahmanical influences, Jain influences, animistic influences that I take to come from (one, many, or all of) the Austroasiatic, Dravidian and Tibeto-Burman speaking substrate populations inhabiting the foothills of the Himalayas (before Vedic speaking people even entered India). I've also noted some Iranian and/or Zoroastrian influences, some of which are certain and some speculative. Of course it is always possible that all these influences came together and were synthesised in the person of a founder. Possible, but not very likely. Cultures tend to be assimilated and synthesised by other cultures. And in our case this may well have extended both before and after the time of the putative founder.
My theory about early Buddhism has two aspects in relation to this problem. Firstly the morality we associate with Buddhism is probably (broadly speaking) the mores of the Śākya tribe. In my writing on the Śākyans (published and unpublished) I have argued that the Buddha and his contemporaries should be seen as representing the culmination of a process of synthesis that began with a group of Iranian tribes migrating into India. They becoming naturalised and then were forced by climate change to migrate again and so ended up on the margins of the kingdoms of Kosala-Videha and Magadha. And, contra Bronkhorst, I argue that Kosala was the cultural centre of gravity of this time and place (it is the setting for the debates with Yājñavalkya in the Bṛhadāraṇyaka Upaniṣad for example). Importantly the pan-Indian rebirth eschatology met and was synthesised with the Zoroastrian idea of single destination—Heaven vs Hell—eschatology based on morality. The result was most obviously the various Śrāmaṇa faiths, but there was also considerable influence on Brahmanism outside the Āryavarta as well. As yet no plausible explanation has been put forward for why the Brahmins suddenly became dissatisfied with being reborn amongst their ancestors. However, if eastern, Kosala based, Brahmins were interacting with Zoroastrian influenced tribes, they might have been attracted to an afterlife in an eternal heaven and adapted the idea to their own uses. The earlier appearance of karma in Brahmanical texts simply reflects a predisposition to encapsulating religious ideas in texts that early Śrāmaṇa groups did not share until centuries later.
If there was an influence from Zoroastrian eschatology then it would emphasise the impersonal inevitability of post-mortem judgement, and would suggest post-mortem personal continuity. Personal continuity was already a feature of pan-Indian rebirth eschatology. Thus personal continuity ought to surface as an element in early Buddhist morality since it is present in the substrate belief systems on which Buddhism is built. And this is what we find in the Jātakas. Many aspects of the Jātaka literature, in particular some characters and moral themes, seem to cross sectarian boundaries and reflect shared culture.
The other aspect of my account of early Buddhism, which is heavily reliant on Sue Hamilton's account of the khandhas, is that paṭicca-samuppāda was initially applied only to the process of having experiences. The basic description is vedanā (from √vid 'to know) arising from contact between sense object, sense organ and sense cognition, and the polarities involved in vedanā in turn giving rise to the mental processes by which we become infatuated and intoxicated with sense experience (papañca). My argument has long been that properly understood paṭicca-samuppāda only applies to this domain of experience.
However, it is only natural that having discovered a principle like paṭicca-samuppāda that Buddhists would want to see what light it could shed on all aspects of their lives. And after all, as I have pointed out, the impermanence of the world and human life is quite universally acknowledged (Everything Changes, But So What?) My account of early Buddhism predicts that paṭicca-samuppāda applied outside the domain of experience ought to produce metaphysical problems. Thus the principle applied to the sense of self is a powerful lever that can shift our perspective on experience, but when we start to ask whether our self is real or unreal and try to answer in the same terms, we end up with nonsense. Asked if your self exists, later Buddhists are forced to answer both yes and no. And the yes/no answer is still being debated almost 2000 years after it was proposed by Nāgarjuna and his contemporaries (Not Two Truths). The metaphysics of the ontology of the Buddhist self are enormously complex and confusing. The result is some of the most convoluted discourses that end up degrading into insoluble paradoxes and are presented as representing the ineffability of the truth. It's notable that early Buddhist texts which apply paṭicca-samuppāda in the correct domain (visaya) never seem to resort to paradox or convolution in this way. Experience has three salient characteristics: it is impermanent, unsatisfactory, and lacking substance. The experience of selfhood is just another experience, has the same three characteristics, and is subject to the same limitations. Whether or not the self is real is completely irrelevant to the discussion of experience and how to manage it for our well-being. All other ontological questions seem also to be set aside as unanswerable and therefore irrelevant.
In this view the application of paṭicca-samuppāda to an ontological question such as personal continuity, particularly post-mortem, will be unlikely to produce entirely consistent answers. Experience is inherently discontinuous whereas the afterlife requires some kind of continuity. And this leaves us with an unresolvable problem in the area of morality. Considerable ingenuity must be employed to join the Buddhist morality to the principle of paṭicca-samuppāda. It is no coincidence that the question is one that troubles King Milinda. That the first answer never satisfied Buddhists can be deduced from the fact that it has been constantly revisited and reshaped. And even so when teaching morality it was and is common for Buddhists to embrace the idea of personal continuity, as in the Jātakas, for pedagogical reasons.
The kind of dependent continuity proposed by Buddhists as underlying Buddhist morality can hardly motivate people to practice virtue or avoid vice. Without the experience of actions having consequences for oneself, particularly in personal relationships, that Buddhist morality hardly makes sense. Without continuity between actor and the experience of consequence we have no motivation to alter our behaviour. As social animals were keenly attuned to the impact of our behaviour on others and theirs on us. If there were no personal continuity, or even if it were experienced as the kind of nominal continuity that Buddhist theorists propose, then we could not correlate actions and consequences in this life, let alone across lifetimes. Thus, ironically, it is the sense of our self as an entity continuous through time that underlies morality, at least in the unawakened.
Or we could see this as part of a pragmatic program. Morality is best taught with a strong grounding in personal continuity. Morality is concerned with our personal relations and without continuity the word "relations" is meaningless. Approaching morality in this way helps to eliminate major conflicts and prepare the mind for meditation. When it comes to reflecting on the nature of experience in religious exercises designed to liberate us from suffering, its best to point to discontinuity and to emphasise that the sense of self is no different from other experiences. To me this suggests the marriage of two very different activities and modes rather than a unified teaching.
In fact what we see in the Pāli texts is a sea of partially integrated plurality, in crucial ways irreconcilable, and with a considerable amount of flotsam and jetsam from non-Buddhist systems of thought and practice. This isn't a problem if the Buddhist program is pragmatic rather than systematic. A lot of Buddhists are embracing pragmatism as an antidote to the idea of Buddhism as a systematic tradition. That said a surprising number of pragmatists are highly critical of, for example, teaching mindfulness to people who are anxious or in pain, precisely because it does not fulfil their criteria of Buddhism as a system. In the face of the plurality of doctrine, usually the best we can do is select a subset of the teachings that hang together and gloss over the discontinuities. A dense and complex jargon combined with an anti-intellectual discourse helps us to obfuscate such problems. Even those who study the texts more directly are doing so through cultural and historical lens that predispose them to see unity and continuity and to gloss over evidence of the opposite.
Del Toso, Krishna. (2008) 'The Role of Puñña and Kusala in the Dialectic of the Twofold Right Vision and the Temporary Integration of Eternalism in the Path Towards Spiritual Emancipation According to the Pali Nikayas .' Esercizi Filosofici 3, 2008, pp. 32-58. Online academia.edu.
I have some reservations about this article. It is presented in terms of what Gotama taught, e.g. "Gotama makes a dialectical use of Eternalism as means to eliminate Nihilism." which I think is indefensible unless by "Gotama" the author means "the Buddhist tradition" and he does not seem to make this distinction. However the observations about the rhetorical uses of eternalism are very interesting.
Del Toso also refers to:
And in this article Charles Hallisey highlights the Mahāmaṅgala Sutta, which gives a list of desirable practices. Hallisey points out that the list of duties in the sutta and Buddhaghosa's exegesis argue "against any attempt to find a single metaethical principle that would make sense of everything on the list with an account of the occasion on which the canonical text was first taught." Hallisey's point is that although the audience for the text could agree on the auspiciousness of particular acts, they could not identify universal criteria of auspiciousness and Theravāda Buddhist ethics is thus particularist rather than generalist.
"As a philosopher, my conceptual point is that only if an organism simulates itself as being one and the same across time will it be able to represent reward events or the achievement of goals as a fulfillment of its own goals, as happening to the same entity. I like to call this the "Principle of Virtual Identity Formation": Many higher forms of intelligence and adaptive behavior, including risk management, moral cognition and cooperative social behavior, functionally presuppose a self-model that portrays the organism as a single entity that endures over time."