Abstract

Belief is traditionally thought of as a two-place relation between a subject and a proposition. I want to argue that this is mistaken. Roughly put, belief should be indexed to relevant tasks the subject might engage in. This is required to give an adequate account of cases of conflicting behaviour – for example, implicit bias. No simple yes/no answer to the question ‘does S believe that p’ is acceptable in these cases – instead one needs to say ‘S believes that p relative to one index but doesn’t relative to another’. I’ll argue for this claim by looking at a number of examples that show that our functional structure is more complex that is often assumed. Usually, philosophers restrict their attention to cases in which verbal and non-verbal behaviour conflict; but there are many other ways it can come apart. For this reason, a correspondingly complex notion of belief is required.