The problem I want to discuss has to do with believing as distinct from perceiving, imagining, positing, resolving, and hoping, as well as from knowing. Since these distinctions are not always observed, we must remind ourselves what ‘belief’ means when it is deliberately preferred to other intentional descriptions, and we ought to characterize it in such a way that we can see why it matters immediately, not just consequentially, whether one believes in something or not. I propose putting it this way: Believing in X (where X is what a belief-proposition is about) means taking X to be real, which in turn means accepting X as something to be dealt with. (A further implication of ‘real’, precluding private fantasy zones, is that any real X is dealt with not in isolation but in connection with everything else one has to deal with. ‘Real’ implies ‘the world’.) Unlike knowledge, which is originally conceived, so to speak, from the object to the subject – given that X is there, if I come along and see it there, then I am among those who know it is there – belief is conceived from the subject to the object: given that I am looking about me, if I see X there, then X is among the things I believe in, unless I mistrust it for some reason.