Abstract

I begin with one way of making a distinction between something that can be called Realism, and its alternatives. It is a semantic distinction--about whether sentences of the relevant part of language have truth conditions, where truth is correctness of representation. This kind of truth is not warranted assertibility, and is not guaranteed by acceptance or agreement. The issue to be addressed comes from the combination of two things. The first is the naturalness of the view that sentences have truth conditions, the naturalness of the corresponding interpretation of science, and the apparent inadequacies of the alternatives. The second is that there are problems for that view, as shown by arguments that correctness for use of language can't go beyond correctness according to the norms of a linguistic community--and this kind of correctness is warranted assertibility, not truth. ;I address one attempt to show that sentences can have truth conditions. That attempt invokes an aspect of the "Causal Theory of Reference". I argue that the part of the Causal Theory of Reference that might have been a solution to the problems I've raised is wrong. ;In the final part of the dissertation, I give a solution to the problem raised. I contrast two standpoints from which the situation of language use can be described. One is the standpoint of someone external to a linguistic practice, who is engaged in the task of giving an account of meaning for that language. The other standpoint is that of someone in a position to use the language. ;I argue that in the position of someone who is able to use the language, there is a kind of correctness for language use that applies, which is not guaranteed by acceptance, agreement, nor by satisfaction of operational standards. Because this kind of correctness possesses these properties, I suggest that this kind of correctness is truth as correctness of representation