What is apoha, and does defining by exclusion help with the frog-and-fly problem?

Question

I teach philosophy of language in Lund, and my hobby is reading Sanskrit philosophers in translation and noticing where they reached our problems first. I would like a clear explanation of one idea I only half understand.
Dignāga, the fifth- and sixth-century Buddhist logician, held that words and concepts do not latch onto real universals like "cowness". Instead a word works by apoha, exclusion: "cow" picks out its objects by excluding everything that is not a cow. The motivation, as I understand it, is nominalist: Buddhist epistemologists did not want to admit real universals, and exclusion seemed to let them have general terms without general entities.
Two things I would like explained. First, how is this supposed to work without circularity? To exclude non-cows, surely you already need to know what cows are, which is the objection I believe Kumārila made.
Second, and this is my own speculation: in recent Western philosophy of mind the disjunction problem asks how a mental state can be about flies and not about "flies or small dark moving dots". Does a theory that builds content out of exclusion have anything to offer there, or am I forcing a comparison?

Meera Kulkarni · Accepted Answer

I'll take both, Oskar, since this is my thesis area.
How apoha is meant to work. Start with perception, which for Dignāga grasps bare particulars, not kinds. Conceptual thought then constructs a class. The later and fuller version, Dharmakīrti's, gives this a causal basis: different particulars produce the same kind of effect in us (the same kind of perceptual judgement, for instance), and the concept "cow" is built by excluding everything that does not do that. So the exclusion is not of "non-cows" defined in advance, it is of whatever fails to produce that type of result.
The circularity charge. Kumārila's objection is exactly as you put it: you cannot exclude non-cows without already grasping cows. The apoha theorists replied in various ways over several centuries, and whether they succeeded is still debated. The 2011 volume edited by Siderits, Tillemans and Chakrabarti is the best place to see the modern discussion.
The comparison. Be careful. Apoha is a theory of how general terms and concepts classify, not a causal theory of how a mental state comes to be about one thing rather than another. But there is one real point of contact: both apoha and teleosemantics locate content partly in what the organism does with the input. That is a family resemblance worth noticing, not an identity.

Kenneth Leung Chi-ho · Answer

Meera's caution is right. I'd add that apoha has a negative structure that Western theories mostly lack. It defines by difference, which has more in common with Saussure's claim that in a language there are only differences than with causal semantics. If you want a Western partner for Dignāga, try structural linguistics before Fodor.

Oskar Nyberg · Answer

Thank you both. So the honest version of my comparison is: not an answer to the disjunction problem, but a different way of asking what a concept is, with an interesting overlap on the role of the user. I'll stop forcing it.

Govinda Prasad Mishra · Answer

For completeness, the Mīmāṃsā side: Kumārila thought apoha theorists were trying to avoid universals and had smuggled them back in. Whatever you think of his own realism about universals, the objection is a serious one, and you see it reappear in Western debates whenever someone tries to define a kind purely negatively.

What is apoha, and does defining by exclusion help with the frog-and-fly problem?

4 answers

Accepted answer

Other answers