Skip to content
Coordinates of ThoughtHome of Philosophy
Join
Question · Explain

How is apoha supposed to work? ‘Cow’ means ‘not non-cow’ sounds circular to me

Vikram Sethi Fellow
· 5 replies · 194 views
Meaning & languageBuddhistIndianMimamsaNyayaDharmakīrtiDignāgaKumārila Bhaṭṭa#apoha#exclusion#nominalism#universals
Follow

I teach mathematics, and when I read that Dignāga's theory of meaning says a word like "cow" works by exclusion, meaning roughly "not non-cow", my first reaction was the one I would have to a student's proof: this is circular. To exclude the non-cows, don't I already need to know which things are cows?

I understand the motivation. Buddhist epistemologists did not want a real universal "cowness" sitting in every cow, which is what the Nyāya realists defended. And they did not want "cow" to be an arbitrary label either, because then why would we all agree on it? Apoha is meant to be a third way.

But I cannot see how it escapes the circle. In set terms, the complement of the set of non-cows is the set of cows, and to define one you need the other.

Could someone explain how apoha is supposed to work, and whether the circularity objection has an answer? I would especially like to know how Dharmakīrti developed it, since I gather he changed it.

LikeAnswer
Follow5 answers

Members reply here. Reading is always free.

Sign in to reply

5 answers

Accepted answer

  1. Meera Kulkarni

    ✓ Accepted answer Fellow

    They did, and the circularity charge rests on reading apoha as defining one class by its complement. That is not the idea.

    Start with Dharmakīrti's version. Only particulars are real, and each is utterly unique. But particulars have causal powers, and some particulars produce similar effects: these animals give milk, pull ploughs, and produce in us similar perceptual judgements. Others do not. His well-known illustration is medicinal: several different plants each reduce fever, without sharing any real "fever-reducing" universal. Their sameness is a sameness of effect.

    The concept "cow" is then a mental construction (vikalpa) that groups together things which are excluded from what does not produce those effects. Nothing positive is shared. What is shared is a difference, and that difference is grounded in real causal facts about each particular, not in a prior grasp of cowhood.

    So the order of explanation is: particulars with effects, then a construction that excludes what lacks them, then the word. The class of cows is an output of the process, not an input. Whether that fully disarms Kumārila is still argued over; later Buddhist logicians like Ratnakīrti refined the account precisely because the objection kept biting. But it is not the simple circle it first looks like.

    Helpful · 3

Other answers

  1. Govinda Prasad Mishra

    Fellow

    You are in excellent company. Kumārila Bhaṭṭa raised this objection, among several others, in the chapter of his Ślokavārttika devoted to apoha: you cannot grasp "non-cow" without first grasping "cow", so the exclusion presupposes the very positive content it was meant to replace. He also argued that a negation needs something positive to negate. As a Mīmāṃsaka I think the objection stands, but Meera will tell you the Buddhists had more to say.

    Helpful · 2
  2. Oskar Nyberg

    Fellow

    It may help to compare a European nominalism. Resemblance nominalists say things are grouped because they resemble each other, and Goodman in The Structure of Appearance pressed problems such as the "imperfect community": a set of things each of which resembles each other in some respect, though there is no single respect all share. Dharmakīrti's grounding in shared effects rather than shared appearance might avoid that problem, which I had not appreciated before Meera's answer.

  3. Vikram Sethi

    Fellow

    Thank you, Meera. So the mistake in my set picture is that I assumed the universe of animals is already sorted. On this view the sorting is what the concept does, using the causal facts. That is better than I expected. One worry left: "similar effects" sounds like a resemblance again, just moved from the cow to the milk.

    Helpful · 1
    1. Kenneth Leung Chi-ho

      Fellow

      That is the right worry, Vikram, and Buddhist logicians answer that the effects are also particulars, grouped by the same exclusion, so there is no point at which a real similarity is needed. Whether that is a regress or a virtue depends on your temperament. I would only caution against the set analogy: there is no given universe of discourse here. Apoha is closer to a theory of how a universe of discourse gets built than to a theory of complements inside one.

      Helpful · 2