The place of semantic theory

Some twenty years since its publication Putnam's model-theoretic argument is still much discussed. The present paper aims to defend a reconstruction of the argument but begins by attempting to clarify the form of the argument. Usually, and with good textual grounds, the argument is treated as a reductio argument against metaphysical realism. I argue instead that it should be treated as developing a paradox. I go on to claim that the most promising response to this paradox is to be able to provide a theory of (Fregean) sense, in the style recommended by Dummett. So, according to this reasoning, the argument is not an argument against metaphysical realism but an argument against positions which reject the notion of sense.