Tuesday, May 27, 2008

Evans, Davidson and Quine

Both Evans, in his "Identity and Predication," and Davidson, in e.g. "Meaning, Truth, and Evidence," criticize Quine's views on reference while accepting what he says about the indeterminacy of translation. Neither thinks that Quine has established that we should understand others as possibly talking about rabbit stages as opposed to rabbits. They both want to emphasize the primacy of objects. The ways they go about this differ somewhat. Davidson argues on broadly epistemological grounds. To get a conception of evidence that will support our beliefs, we need a distal view of stimuli, which requires the distal end to be an object to act as the causal origin of a stimulus. Evans argues on semantic grounds. Given the speech behavior of a community with enough expressive resources to use a negation, we must understand them as talking about objects, material bodies, instead of something else.

Evans's argument doesn't seem to work since he draws conclusions from the descriptions of the speech behavior of the community that are not warranted by that description. In particular, his description doesn't support his ascription of negation and contradiction. Quine could respond to his argument in this way, denying that Evans has established that they are using the language in such a way that we must (can?) understand them as using logical language. As translators we haven't reached the stage of translation in which we can pick out the logical particles.

There is a certain affinity between Evans's argument for objects and Brandom's argument for the predicate/term structure of sentences. Both rely on the expressive power of a negation to argue that there must be a certain sort of thing on pain of contradiction. Brandom's argument is a bit more nuanced since he just needs a way of constructing a sentence with a reversed inferential "polarity". Using a conditional will do the job as well. Really, all Brandom needs is an operator with a tonicity (...,-,...), i.e. that is antitonic in some position, to get his result. Evans's argument needs the contradiction to result, so it seems like he needs the negation specifically. There are some further differences. Evans wants to establish an ontological conclusion, that there must be material bodies, while Brandom wants to establish a linguistic conclusion, that there must be a certain sort of linguistic structure. The particular linguistic structure, singular terms and predicates, naturally gives rise to thinking that there must be objects for the singular terms to refer to but this is an extra step; it is one that Brandom is, I think, not particularly disposed towards since he does not take reference as a primitive notion. Evans does not describe his argument as relying on the expressive capacity of negation, but it is an apt description. It is not until we attribute negation to the language users, translating something as a negation, that we are forced, according to Evans, to understand them as talking about material bodies. The stages of translation preceding that allow the possibility of understanding them as talking about, e.g., time slices or universals or some such.

It is another question whether there is some way in which Evans's and Davidson's arguments are related. Davidson understands his argument as being broadly semantic, even though I called it epistemological. I think he says that the picture he lays out, the distal theory of stimulus, is one way of doing semantics. This is because, I think, he sees investigating concepts and semantics as investigating the world in a way. In "A Nice Derangement of Epitaphs" he says something along the lines of knowing a language is no different than knowing one's way around the world generally. I don't know to what extent Evans would be on board with this. He does emphasize the importance of connecting conceptual capacities up with navigating the world, as in his different notions of space and their relation.

No comments: