Abstract

The first part of the paper argues that there is no analysis of (or statement of sufficient conditions for) "justification." The problem is that as an analysis it would have to be lawlike. But if it were lawlike it could be used to support a counterfactual conditional to the effect that if I were justified in believing it I would satisfy its analysans. I argue that since no analysis can be so used there is no such analysis. In part II I show how the conclusion of part I supports naturalism and clarify one version of naturalism