Debajyoti Gangopadhyay: Let us now turn towards the concept of śūnyatā in Buddhist Philosophy. Śūnya had been famously advocated by Nāgārjuna as going even beyond the catuṣkoṭi —the four possible state of affairs advocated by the Buddha, so that given any question, the Buddha used to advocate four possible answers, namely Yes, No, Both, and Neither, as in the Greek Tetralemma. The Buddha doesn’t seem to have endorsed explicitly any possibility, but trace of his departure from common sense ontology can often be traced in his advocacy of the ‘Both’ option. Yet, Nāgārjuna advocated to go even beyond the framework of four possibilities!
Accepting Nāgārjuna’s standpoint, what about the possibility to understand śunya in terms of any so-called logical characterization? Or is śūnyatā forever beyond any kind of epistemic qualification, is it a Reality beyond any possible categorization?
Mark Siderits: My own reading of the uses of the catuṣkoṭi in the Indian Buddhist tradition is rather different.
The Buddha himself seems to have used the device in order to show that with respect to some question all four possibilities involve a false presupposition, thus rendering any of the four possible answers meaningless and so neither true nor false. The rejection of each involves use of a commitmentless form of negation that does not entail affirmation of any alternative. In effect it functions like ‘Do not say …’ (Note that when p has a false presupposition, no contradiction results from asserting both ‘Do not say that p’ and ‘Do not say that not-p’.) As for the apparent violation of non-contradiction and excluded middle in the third and fourth koṭis, these are avoided by parametrization. As for instance the third claim the Buddha rejects, ‘The enlightened person both exists after death and does not exist after death’, is to be understood as ‘exists in one sense’ and ‘does not exist in another sense’.
Nāgārjuna’s use of the catuṣkoṭi follows the same pattern. The one key difference is that whereas in early Buddhism and in the Abhidharma schools it is applied only to things like persons and chariots that are deemed to not be ultimately real, in Madhyamaka the same logic is applied to everything that is thought to be ultimately real. So if the arguments in support of this use are sound, the upshot is that there is nothing that is ultimately real. The only things that can be said to be real are things that other Buddhists consider merely conventionally real. Nāgārjuna is not affirming an inexpressible ultimate reality. He is instead trying to get us to stop supposing there could be any such thing as how the world is anyway. Nāgārjuna is not a Hegelian-style absolute idealist. He is a global anti-realist.
If this interpretation is correct, then any attempt to turn physicalism into a form of ‘serious ontology’ will be incompatible with Madhyamaka, and Madhyamaka arguments can be deployed against any table-pounding variety of physicalism. In philosophy of science this would amount to something like a global operationalism. What it would most definitely not do is align Madhyamaka with anything like Ladyman’s structural realism, let alone with anything calling for wholesale adoption of a deviant logic. Of course a Mādhyamika might concede that our best theories about such things as quantum mechanics might turn out to require using something other than classical logic. But they would add that this is not to be taken as showing that ‘reality is ultimately non-classical’.
DG: The crux of the present day debate (championed notably by Graham Priest and a few others) about whether Buddhism implied contradiction (paraconsistency in modern term) as an essential part of understanding Reality or not evolves around this point – his implicit advocacy for the ‘Both’ option. Do you feel sympathetic to Priest and the others who are trying to figure out paraconsistency in Buddhism as well as in Jain Logic in this way?
MS: I’m not competent to judge the case of the Jainas (though scholars I trust assure me that they reject paraconsistency). But as will be clear from what I’ve already said, I reject Priest’s interpretation of Buddhist uses of the catuṣkoṭi. Indeed I’ve often pointed out to Graham passages where Buddhist philosophers explicitly affirm non-contradiction, excluded middle and the like; even a passage where the Mādhyamika Candrakīrti calls anyone who utters a contradiction ‘crazy’ (unmatta).
DG: Above considerations lead us once again to one of the most important aspects of metalogical issues, namely the way Indian mathematics and astronomy were affected by the dependence of logic on metaphysics. Of course this issue is important in context of the Growth of Knowledge in the Indian Philosophical context as a whole. So far Buddhist teachings are concerned, certainly the Buddha had an implicit agenda against Metaphysics. We need not get here into the social perspective of this agenda in Mithila where Buddhism and Jainism developed, or other contemporary doctrinal traces of this agenda in Yājñavalkya’s teachings or in Sāṅkhya, but this agenda (against metaphysics) was surely an assignment to defend for Nāgārjuna as well as for the later Buddhist Logicians. How would you like to relate today in modern logical terms the Buddhist defenses of their kṣaṇavāda metaphysics or no-metaphysics/ontology) with the catuṣkoṭi?
MS: If by ‘metaphysics’ you simply mean using philosophical methods to try to determine facts about the ultimate nature of reality, then I am not sure the Buddha is against all metaphysics. He certainly opposes all ‘speculative’ metaphysics: all attempts to use reason alone to prove the existence of things that are by nature imperceptible. But his anātmavāda as well as his rejection of the existence of an Īśvara and his apparent acceptance of mereological nihilism look to me like metaphysics. Nāgārjuna could be called anti-metaphysical insofar as his global anti-realism might best be understood as leading to a kind of quietism.
Kṣaṇavāda was not taught by the Buddha, who merely claimed that everything relevant to the continued existence of persons is impermanent. Kṣaṇavāda was, however, accepted by all the philosophical schools that later developed from his teachings, with the probable exception of Madhyamaka. And this doctrine certainly looks like a case of a metaphysical theory that is in principle not empirically verifiable. Now the standard argument for kṣaṇavāda does not use the catuṣkoṭi, it only examines two options: the possibility that anything subject to destruction does so due to extrinsic causes, and the possibility that destruction is intrinsic to anything subject to destruction. And the argument seeks to prove the latter is true by showing the former is false. So even if the catuṣkoṭi did involve the use of deviant logic, their defense of kṣaṇavāda does not. So this would not count as a case of adopting a logic based on its utility for establishing a certain metaphysical theory that supports one’s doctrinal commitments. In fact I’m not sure any Buddhist ever ‘adopts’ a logic; since they all seem committed to classical logic, they would be better described as simply using the logic they take to be commonly accepted.
The one possible exception here might be Madhyamaka. Although they never say anything like this, I suspect that if they were presented with a convincing case for adopting an alternative logic in order to account for the phenomena we observe at the quantum level, they might agree that logical deviancy is acceptable in this specific area; that is, they might accept the utility of ‘quantum logic’ or some such. But they would hasten to add that this can tell us nothing about ‘the logical structure of ultimate reality’. There being no such thing as the ultimate nature of reality, the question of its logical structure simply does not arise.
DG: In fact, metaphysics had been tried to be exorcised, from time to time, from Logic or even from Science as a whole. We are aware of the famous Viennese attempt against Metaphysics led by Carnap and his colleagues; but none of these attempts has been successful ultimately. Compared to the Buddhist agenda against Metaphysics, what message do you suggest to read out of all these failures as well as the recalcitrant Metaphysics at the same time?
MS: If metaphysics is just bookkeeping for physics, then there should not be any incompatibility between naturalism and the practice of metaphysics. Doing metaphysics typically requires the assumption that there is such a thing as how the world is anyway, but then since the proponents of naturalizing usually share this assumption, there is no source of conflict there. What the Vienna Circle seems to have been objecting to is Hegelian-style speculative metaphysics. The sort of metaphysics one finds in contemporary analytic metaphysics as well as in much of the Buddhist and Nyāya material is more like ‘bookkeeping for physics’: using philosophical tools to lay out the ontological commitments of our best theories.
Metaphysicians run into opposition when their practices lead to results that outrun empirical verifiability. The trouble is, it is not at all easy to say exactly why doing such a thing should be a problem. Vienna Circle anti-metaphysicians tried to do this using assumptions derived from the doctrines of epistemological internalism, semantic internalism and internalism about mental content. But externalists question all those assumptions so those arguments are not decisive. This difficulty of saying why verification-transcendence is a problem may be why periods of anti-metaphysics are followed by the resurgence of metaphysics: Kant’s strictures are followed by Hegel, the logical empiricist orthodoxy is followed by the rise of analytic metaphysics, Madhyamaka quietism is followed by the return of Abhidharma-style metaphysics with Diṅnāga and Dharmakīrti, etc. There is evidence both that bookkeepers can have important things to say about how the firm should be run, and that when a bookkeeper is put in charge the firm suffers.
(the first part of this interview can be read here).
*Edited for the IPhB by Matthew Dasti and Elisa Freschi