||There are several things that we do not know about the world. Currently, for instance, I do not know whether it is raining in Canberra or not. So it is possible for me, epistemically speaking, that it rains in Canberra, and it is possible for me, epistemically speaking, that it does not. Most agree that epistemic possibilities are somehow related to knowledge: roughly, a possibility is an epistemic possibility just in case it is not known that it does not obtain. There are three central questions that philosophers ask about epistemic possibilities. First, what are epistemic possibilities? Some argue that they are akin to metaphysically possible worlds, others that they are akin to a priori consistent ways the world might be, and others again that at least some epistemic possibilities are akin to impossible worlds. Second, whose knowledge matters for determining what is epistemically possible and not? Some hold that it is the knowledge of a single subject that matters, while others hold that it is the knowledge of a group that matters. Third, what is the semantic analysis of epistemic modals? In natural language, epistemic possibilities are often expressed by sentences such as "It might be that P" and "For all I know, P", and people have proposed contextualist, relativist, contrastive, and non-truth-conditional analysis of such epistemic modals.