Published online by Cambridge University Press: 01 January 2020
In Making It Explicit (1994) Robert Brandom claims that we may distinguish those linguistic expressions with object-representational purport — the singular terms — from others merely by the structure of their inferential relations. A good part of his inferentialist program rests on this claim. At first blush it can seem implausible: linguistic expressions stand in inferential relations to each other, so how could we appeal to those relations to decide on the obtaining of what seems to be relation between linguistic expressions and objects in general (viz., x purports to represent y)? It is perhaps not surprising then that Brandom's proposal fails. But it definitely is surprising how it fails. The problem is that in order to specify the sort of generality there is to an expression's inferential role, one must appeal to some version of the traditional distinction between extensional and nonextensional occurrences of expressions, and there appears to be no way to draw anything like that distinction in inferentialist terms. For the inferential proprieties governing the different occurrences an expression can have are so varied that they do not determine a binary partition of those occurrences.