The Bayesian Intentional Stance

Up: Book Index

Sections

• The Intentional Stance
• Problems with a Deductive System of Intentional Explanation
• Probability and Utility
• Objections to Real-valued Measures
• The Mortonian Agent
• Some Empirical Research
• The Dual Role of Consistency Requirements
• Beliefs and Opinions
• Interaction between Belief and Acceptance
• Varieties of Inconsistency
• The Normative and Descriptive Cox Proofs

Summary

In this part I have proposed and defended a theory of intentionality called Descriptive Bayesianism that was implicit in the writings of de Sousa and Dennett. The rest of this thesis is about Bayesian agents, and in the light of the arguments in this part we should regard any agent properly so-called as Bayesian. If you are not persuaded of the descriptive interpretation, then you can still read such results as being about a particular kind of (allegedly) scientifically ideal agent. Since we will see that such agents are capable of scientifically irrational behaviour, the subsequent argument can be read as a refutation of that notion of a scientifically ideal agent.

A philosophical project that says something about belief should at the outset have something to say about what belief is. In the course of defending Descriptive Bayesianism we have found that the everyday understanding of the term is ambiguous between at least two crucially different senses. Decision theory demands that we distinguish beliefs and opinions. As we have seen, philosophers have been converging on similar distinctions independently of any decision-theoretic considerations. In the rest of the thesis, I will be concerned not to defend Descriptive Bayesianism but to draw out some of its consequences. I will take the probability and utility functions as starting points, without going into how they are determined. The logical structure of this work is conditional, in that its conclusions are of the form, "If there are agents governed by this particular kind of BDT model, they will behave like this."