Fragmented Models of Belief

This chapter motivates a fragmentationist research program by identifying a cluster of problems that such a research program is better positioned to address or resolve than a unified model—all instances of the phenomenon of subjects having information that’s available to them for some behavior-guiding purposes but that isn’t available for every purpose. It also identifies some of the challenges and research questions that the fragmentationist program will need to address and where the space of possible answers is not yet well charted. One cluster of such problems is about how to construct fragmented models of belief, and another cluster is about normative questions that arise concerning the rational evaluation of fragmented beliefs and believers.