Accurate Updating

Abstract A number of authors have observed that epistemic externalists seem to face a dilemma: Either deny that Conditionalization is the rational update rule, thereby rejecting traditional Bayesian epistemology, or deny that the rational update rule maximizes expected accuracy, thereby rejecting accuracy-first epistemology. Call this the Bayesian Dilemma. I’m not convinced by this argument. Once we make the premises explicit, we see that it relies on assumptions the externalist rejects. In this paper, I argue that the Bayesian Dilemma is nevertheless a genuine dilemma. My argument does not make any assumptions that the externalist rejects.