It has often been claimed that our believing some proposition is dependent upon our not being committed to a non-epistemic explanation of why we believe that proposition. Very roughly, I cannot believe that p and also accept a non-epistemic explanation of my believing that p. Those who have asserted such a claim have drawn from it a range of implications: doxastic involuntarism, the unacceptability of Humean naturalism, doxastic freedom, restrictions upon the effectiveness of practical (Pascalian) arguments, as well as others. (...) If any of these implications are right, then we would do well to have a precise statement of the nature of this phenomenon central to first-person doxastic explanations, as well as of our reasons for believing that it holds. Both of these are lacking in the literature. This paper is an attempt to elucidate and defend this claim. (shrink)