It is well known that according to Davidson, it follows from the conditions of possibility of radical interpretation that the notion of belief is intrinsically truthful, in the sense that an agent can only have beliefs provided that most of them are true and coherent. In this paper, I offer some reasons to doubt that this thesis of the truthfulness of belief actually follows from the principles governing radical interpretation, unless the thesis is understood in a way that considerably trivializes it. If what I"m about to suggest is correct, I will have shown that one of the most popular objection to Davidson's program is in fact ungrounded; but I will also have provided some reasons to doubt the fruitfulness of such an enterprise.