Abstract

There are two main approaches used to evaluate someone’s epistemic accuracy. On one hand, we have the approach that evaluates the truthlike-ness of some propositions and orders them by their closeness to truth. On the other hand, we can look at someone’s probability assignments and then calculate the accuracy using the Brier score. Recently, Oddie tried to show that these two methods are in tension, because a measure that obeys probabilism will fail to hold Proximity. I argue that the Unified Approach can be saved and, moreover, has some important advantages over the isolated Truthlikeness.