This paper takes a look at a variety of attempts to account for the apparent existence of a priori knowledge: knowledge that is independent of evidence from sense experience. The classic views can be helpfully classified in terms of three different responses they provide to a single key question. Recently Alan Gibbard and Hartry Field have made space for a fourth, ‘evaluativist’, response. I highlight a few difficulties faced by this new camp and indicate where I think a fifth camp can be set up. I conclude by discussing recent arguments due to John Hawthorne and Timothy Williamson to the effect that the a priori/a posteriori distinction is not (always) a helpful one to draw at all.