Davidson, Donald - Interpretation: Hard In Theory, Easy In Practice

12/12/2008

Paper given in 1998

The paper tries to establish the mutual dependence of language and thought, more aptly conceptualization. When comparing with a form of conceptual atomism (Fodor), author distinguishes between having a concept which is 'simply to discriminate objects or properties of one sort or another' and 'plac[ing] objects in[to] a category' (pg2), the latter being the approach he prefers. To place something into a category is to 'opine that it belongs there, and opnions are prone to error, they are true or false, and they are in part identified by their relations to other judgments' (pg2)

The paper wanders through the typical problems of interpreting another's propositional content from their speech.(pg3-4) With the example of a non-human animal that is able to discriminate between colors, author want to know how we can eliminate the possiblity that it 'isn't the activation of certain rods and cones in his eyes, or the firing of certain optic nerves...' instead of color. If we can't distinguish between the various possible causes for the parrot (Alex's) response, then we can't determine the content of the answer. But if we cannot attribute content to Alex, then we can't attribute thought, since there is no thought without content. (pg4-5)

Interestingly, in this paper Davidson retracts his Swampman argument, claming that 'science fiction stories that imagine thigs that never happen provide poor testing ground for our intuitions concerning concepts like the concept of a person' (pg6)

Author tours through a brief history of the various realist and nominalist positions from Plato to Quine, then concludes that there is a difference between 'showing, by one's behavior, that one merely responds to perceptual similarities, and that one has a criterion for grouping things found perceptually similar.' (pg10) To have a concept 'is to be able to judge, that is, believe, that something falls under the concept. Having a concept is like knowing what a predicate means... in other words, no concepts without propositional contents'. (pg10-11)

So how do we attribute these to people or non-human animals? We can know that propositional attitudes like belief are also related to intentions, perceptions, memories, desires, hopes, etc (pg11), so finding any will give us reason to assume the rest (pg11). Author briefly sketches his triangulation argument, that we're given reason to attribute conceptualization when we encounter not just animals that perceive (pg12) but also ones that recognize when they make mistakes, that is, can distinguish between objective conditions and their categorizations. Not only this, but 'it seems likely that a mind cannot accpt the idea of error without some notion of how error is to be explained. We know of no way all this can be shared by creatures except through the use of language' (pg13). Language and thought become interdependent, with true conceptualization arising 'only in a social setting, and in that setting conceptualization and thought emerge accompanied by the development of language' (pg14).