Is a name a mere meaningless tag, whose function is simply to denote its bearer, as Mill maintained? Or does it, as Frege thought, have, in addition to its reference, a sense, one its user can associate with it, understand, or "grasp"? An argument in favour of the second, Fregean, alternative many philosophers have found well-nigh irresistible is one that takes as its starting point the fact that co-denoting names are apparently not always substitutable salva veritate in sentences ascribing propositional attitudes. Only the postulation of senses, this argument -- which I shall call neo-Fregean -- goes on to claim, can explain this: only a difference in the senses of the names could account for (what is seen as) their different semantic contributions to the sentences in which they occur.¹