ABSTRACT
When we ascribe beliefs, by saying things like ‘she thinks that he is goodlooking’, there is a very delicate relation between the actual words we utter and the content of our ascriptions. What is the case with the people the belief is ascribed to, if the ascriptions say something true of them, depends in ways that are not easy to work out on a number of factors that we do not specify explicitly in our words. That much is familiar, though more detail is given below. In this chapter I show that among these factors are coordinations between belief-ascribers and belief-holders, determined by the needs of ascribers and believers to be intelligible to one another. In effect, we fine-tune what we mean by ‘believe’, from situation to situation in order that the ascriptions we make fit our shared projects. Usually this means that interacting
agents should mean the same by ‘believe’ (and ‘thinks’ ‘judges’ and so on). In fact, I shall describe two deep-rooted coordinations. There is a coordination of force, in which we agree to mean congruent things by ‘ believes’. And there is a coordination of content, in which we agree to take our beliefs as referring to things, real and notional, in ways that will facilitate our shared purposes.
