It is widely supposed that evolutionary debunking arguments (EDAs) against morality constitute a type of epistemological objection to our moral beliefs. In particular, the debunking force of such arguments is not supposed to depend on the metaphysical claim that moral facts do not exist. In this paper I argue that this standard epistemological construal of EDAs is highly misleading, if not mistaken. Specifically, I argue that the most widely discussed EDAs (including those of Joyce, Kitcher, Ruse, and Street) all make key and controversial metaphysical claims about the nature of morality or the (im)possibility of moral truth that belie their apparently epistemological character. I show that the debunking force of these EDAs derives largely from metaphysical claims about morality and their (alleged) implications for the (im)possibility of moral reduction, rather than from epistemological worries associated with the existence of an (alleged) causal/non-moral explanation of our moral judgments. The paper briefly concludes with a dilemma that I believe confronts all EDAs such as those discussed in this paper: either such arguments are unsound, or else they prove too much, debunking our knowledge of science and the external world, as well as morality.