ABSTRACT
Dummett does not disagree with this conclusion. He says “I am wholly with [Brandom] in rejecting the idea of representation as the fundamental notion of a theory of meaning … the notion of representation is not available antecedently as that in terms of which we can build our theory of meaning.” And his reasons are essentially mine: what is expressed by terms such as ‘true’, ‘refers’, and ‘represents’ can in principle only be explained as part of a much more comprehensive story about the use of sentences to make assertions, and about what is involved in practically understanding such uses. What these classical semantic locutions express does not represent a sort of antecedent, unmovable conceptual Archimedean point that can be leveraged into an understanding of all the rest of discursive practice. That is why Dummett says such things as “I am at one with Brandom in rejecting a truthconditional theory of meaning … [T]o arrive at a satisfactory theory of meaning we have to ask what it is that someone needs to know in order to understand a given language and be able to use it.” And it is why he agrees:
Brandom rightly inveighs against the idea that there is an antecedently intelligible conception of reference or designation, for instance of the relation of a name to its bearer, that can be invoked in explaining meaning as representation. Antecedently, that is, to an explanation of the role that names, or other singular terms, play in sentences. He links this with the idea that whole sentences bear the same relation to states of affairs; but, even without this accompaniment, the conception of reference as a notion that we understand otherwise than as having a place within a whole semantic theory is as absurd as thinking that we can know what makes a card a trump in advance of learning the rules of some card game that has trumps in it.
