Vagueness, supertranslatability, and conceptual schemes

A case is developed against Donald Davidson's thesis that any attempt to give meaning to the idea of conceptual relativism is doomed. Even under Davidson's assumption that untranslatability is a necessary condition of a difference between two schemes, it appears to be possible to conceive a situation in which one language is untranslatable for another, but still accessible for truth testing, which undermines Davidson's main argument. One further argument by Davidson is countered by the observation that even if we play the game of radical interpretation strictly on Davidson's terms, we may well encounter a situation in which the only rational choice is to allow for partial failures of translation, and thus for a difference between the two conceptual schemes.