Since this thesi s i s concerned with issues in the phi losophy of possible worlds , and s ince there i s much debate about "what possible worlds are" , I feel I should say something about what the expression "possible worlds" means , at least in my mouth and hopefu l ly in the l iterature as wel l . One of the problems with "possible worlds" is that that expression is ambiguous, and this may lead to some confusion , although most peop le writing in this area are aware of the ambiguity, and so the negative effects of ambiguity are l imi ted. What fol lows then , is some conceptual analysis of the expression "possible world", where T di stinguish two rival senses, and make expl ic i t the most basic adequacy conditions on a theory which purports to explain the nature of possible worlds . Conceptual analysis i s a tricky matter: Wittgenstein's dictum "If one tried t o advance theses in phi loso phy, it would never be possible to debate them , because everyone would agree with them" (Wittgenstein 1958 s 1 28) i s certain ly incorrect as a c laim about ph i losophy, but it is more accurate when adapted as a c laim about conceptual analysis - for if a p iece of conceptual analysis is not agreed to by a l l of those who are competent in their deployment of the concepts , then this is ipso facto evidence that i t i s incorrect . So T should not hope my conceptual analysis w i l l be controversial . On the other hand, T hardly wish to bore the reader with banality either-so T must hope for an analysis which w i l l be i l l um inating through making the impl ic i t exp l icit rather than through dispel l ing error (except for error born of confusion) .