The Virtues of Inconsistency

Abstract

I will argue for the thesis that by knowingly adopting a set of proposi tions which is logically inconsistent, an epistemic agent violates no valid epistemic rule.1 Beyond that, I will suggest that every knowledgeable per son's complete set of fully justified beliefs is logically inconsistent and that this fact is epistemically praiseworthy.2 No doubt, the claim that logical consistency is an absolutely necessary condition of epistemic acceptability has seemed obviously true; and that probably accounts for the paucity of arguments for it.3 I propose to ex amine the arguments which could be given against the acceptability of in consistent sets of propositions in order to show that they do not succeed in ruling out such sets in general. Thus, I hope, at the very least, to provoke more discussion about an issue which has received very little. The epistemic acceptability of inconsistent sets of propositions has rather disastrous consequences for most standard coherence theories of justification. For such theories take consistency to be the sine qua non of coherent sets of beliefs. Hence, one way of viewing the thesis here is to see it as a challenge to any theory of justification which adopts this mild coheren tist reqirement.