Abstract

Many debates in philosophy of language are driven by examples in which two expressions have the same meaning, in some sense, yet fail of intersubstitutability in some of their occurrences. The usual move in response is to postulate a kind of meaning different from that which is shared by those two expressions. I argue that that the resulting semantic theories nevertheless typically cannot explain such failures: the explaining is not done entirely by the postulation and individuation of the new meanings. It is done partly by accompanying metaphysical and epistemological claims about them. Making this clear requires distinguishing between intersubstitutability salva veritate and intersubstitutability on grounds of logical form, and getting straight on what kinds of facts secure the obtaining of each.