Abstract

I wish to consider what can helpfully be meant by the phrase “rational to believe” as it might appear in the statement “It is rational for the person S in his circumstances at t to place more confidence in p than in q, provided his overriding interest at the time is to place confidence, among any propositions he is considering, in true propositions and not in false ones.” The reference here to the interest of the person is intended to avoid discussing persons for whom it is rational to believe something for extraneous reasons, e.g., for a man to believe his wife is loyal to him irrespective of the evidence. I shall take this term as paradigmatic of “epistemic terms”, examples of which philosophers take to be such terms as “is epistemically justified in believing”, “is warranted in believing”, “is credible”, and so on. I do not claim that all these terms have or should bear the same meaning, but I propose to ignore possible distinctions.