Chandrakīrti's Prasanga and the Constitutive Nature of Emptiness — Epoche C2
The reading of emptiness at issue Emptiness ( śūnyatā ) is standardly glossed as the absence of intrinsic nature ( svabhāva ) in things, and on the reading that dominates general accounts of Indian philosophy that is the whole of it: a corrosive anti-essentialism which dismantles substantialist descriptions and puts nothing in their place. This essay argues that the reading is wrong about Chandrakīrti — the seventh-century commentator whose Prasannapadā is the only commentary on Nāgārjuna's Mūlamadhyamakakārikā , the root text of the Madhyamaka or "middle way" school, to survive in Sanskrit — but wrong in a more particular way than the usual defences allow. Chandrakīrti's arguments are indeed purely destructive, and he insists on it more strictly than any other Mādhyamika, the term for an adherent of that school. The constitutive content comes from somewhere else: from an identification, textually explicit and argumentatively load-bearing, of emptiness with dependent origination. Locating that seam is the object of what follows, because the interest of the position lies exactly there, and both the negative reading and the more enthusiastic process-metaphysical readings miss it. Why Chandrakīrti will use only consequences, and never his own inference A prasaṅga is a consequence drawn from an opponent's own commitments and unacceptable to him: not "the vase does not exist" but "if the vase were as you say, then such-and-such would follow, which you deny". The restriction to this form is not stylistic. It is the outcome of a dispute internal to sixth- and seventh-century Madhyamaka, and Chandrakīrti's position in that dispute governs everything he can subsequently claim. The dispute begins with Bhāviveka (c. 500–570), who criticised the earlier commentator Buddhapālita (c. 470–550) for arguing by consequence alone. On Bhāviveka's view the Mādhyamika owes his opponent an independent ( svatantra ) inference, stated in the standard three-membered form — thesis, reason, supporting example — that the logicians of the period had made the currency of debate. Otherwise, he objected, a consequence can be turned round: refute "arising from self" by consequence and you appear to have asserted the contrapositive, "arising from other", which the Mādhyamika also denies. Chandrakīrti's reply in the first chapter of the Prasannapadā defends Buddhapālita and rests on two claims. The first is that a Mādhyamika has no thesis of his own to prove, for which he cites Nāgārjuna's Vigrahavyāvartanī : if I had any thesis, that fault would be mine; since I have none, I am faultless. The second is technical and decisive. An independent inference requires a subject term established in the same manner for both parties. But the subject of Bhāviveka's own specimen inference is the eye, and the realist's eye is an intrinsically existent eye, whereas the Mādhyamika grants no such thing. The two parties therefore have no common subject, the inference rests on an unestablished basis, and it is defective for one side or the other whichever way it is read. A Mādhyamika who states independent inferences is not being rigorous; he is smuggling in the very commitment he means to dislodge. Two clarifications belong here. The labels usually attached to the two sides — Prāsaṅgika for Chandrakīrti's party, Svātantrika for Bhāviveka's — are Tibetan doxographical coinages applied retrospectively, and Lopez (1987) shows how much of the later systematic contrast is a product of that classification rather than of the Indian texts. And the refusal of independent inference does not make Chandrakīrti an opponent of reasoning as such: he accepts, at the conventional level, the four means of knowledge the world acknowledges — perception, inference, testimony and comparison — declining only to give them the foundational role that Dignāga's epistemology had assigned to the first two. The paradigm argument: four cases, and why four The method's canonical application is the opening verse of the Mūlamadhyamakakārikā : nothing whatever arises from itself, from another, from both, or without cause. The enumeration is not a rhetorical flourish, and the reason there are four members is worth spelling out, since the argument's claim to completeness depends on it. Let $s$ stand for "arises from itself" and $o$ for "arises from another". The four cases are precisely the four cells of the table these two atoms generate: $s$ alone, $o$ alone, $s \wedge o$, and $\lnot s \wedge \lnot o$ — that is, $2^2 = 4$, exhaustively. Nothing is left over, provided that "self" and "other" between them cover every candidate producer, which the opponents grant because they are the ones who introduced the distinction. That proviso is the method working as intended: the exhaustiveness is borrowed from the opponent's framework, not asserted by the Mādhyamika. Each case is then refuted from the commitments of whoever holds it. From itself. The target is the Sāṃkhya doctrine that the effect pre-exists in its cause. If it does, its production is purposeless, since what already exists needs no producing; and if a thing can produce itself when it already exists, nothing stops it doing so again without end. From another. The target is the Buddhist realist who holds cause and effect to be distinct entities each with its own nature. If "otherness" is an intrinsic relation, it is unrestricted: a flame is other than a sprout exactly as a seed is, so anything could arise from anything. And since cause and effect are not simultaneous, at the moment of the cause the effect is not there to stand in any relation to it. From both. This inherits the faults of the first two rather than cancelling them. Without cause. This is refuted by what the opponent himself concedes: effects would then occur everywhere and always, and no undertaking would have any point. Notice what has and has not been shown. Every hypothesis about the arising of intrinsically existent things has been eliminated. No hypothe