From Subjectivism to Projectivism

Abstract As an increasingly rigorous account of SCS has developed up to this point, the theory has moved further and further away from the crude sketch of it presented at the beginning—a crude sketch which is too frequently what opponents of subjectivism take it to be. Given the tendency among these critics to assume that any theory more sophisticated than that is ipso facto less subjectivist, it is time now to enquire more closely into the precise nature of the subjectivism of the developed theory. The Introduction argued, partly from the failure of the objectivist alternatives, that there is scope for a personalist interpretation of probability; and Chapters 1 and 2 presented the core of a theory which attempts to build such a personalist account on the single, minimal, constraint of coherence. But, apart from the objectivist competitors, there are of course within a personalist framework options other than the de Finettian approach I have been adopting—it is important to see how it relates to these options and to indicate, at least, why it is to be preferred to them. Several forms of pluralism in the interpretation of probability also need to be considered, before moving ahead with the positive project of locating our subjectivist commitments as commitments to a projective—better, a quasi-realist—approach to the problems of probability theory.