If someone knows something then there is something in virtue of which she knows it; and if someone justifiably believes something then there is something in virtue of which she is justified in believing it. That much is relatively uncontroversial. Only slightly more controversial is the claim that our having an epistemic achievement, such as knowing something or being justified in believing something, depends on how we are in non-epistemic respects. That is, our instantiating of epistemic properties depends on our instantiating non-epistemic properties. In this paper, I argue that epistemic/non-epistemic dependence should be given a central place in epistemology, and that doing so has significant consequences.