The architecture of belief

Abstract

This thesis examines the theoretical underpinnings of intentional realism-the view that cognitive science and psychology will quantify over states such as beliefs and desires. Intentional realism contains the following three elements. First, it attributes a certain type of state to a system whose behaviour it is attempting to explain. I introduce the distinction between Level One and Level Two analyses of systems. Intentional realism postulates Level Two states of the system. Second, the states intentional realism quantifies over are representational states of a certain sort. In chapter 5 I distinguish relatively coarse­ grained representational states, those that represent states of affairs, from fine­ grained ones, and it is the former which intentional realism quantifies over. Thirdly, in chapter 6, we see that these Level Two coarse-grained representational states are deemed to be the causes of, at least, non-verbal behaviour. I go on to argue that intentional realism is weak on two fronts. On the first front, I claim that it is under Level One analysis that one individuates the properties of a system that are criteria! of something's being a cognitive system, whereas The­ One-True-Cognitive-Psychology seems to postulate Level Two states of a system. I claim that the potential for the supporters of intentional realism making a Level mistake is very real. The second front has to do with whether or not there will be Level Two representational states of the sort that intentional realism requires. I argue in chapter 9 that there is a body of evidence which seems to count against intentional realism.