An Alternative to the Proof Paradigm for Ground-Level Rationality

In Chapter 5, the author uses the clues from previous chapters to begin the construction of a theory of epistemic rationality in which mistake-correcting reasoning is the paradigm for reasoning. On his account, cognitive models are composed of scenarios, which are themselves sets of propositions that have expectations for experience. Models are tested not by propositions describing experience but by experiences themselves because scenarios holistically generate expectations for experience which the agent’s actual experience can either satisfy or fail to satisfy. On the author’s theory, only scenarios with full or partial necessitation hypotheses can earn rational confidence. He identifies two main conceptual frameworks, the ordinary object framework and the person framework, both of which are causal frameworks. The imagination plays a crucial role in rationality in the author’s theory. He ends with informal statements of two principles of epistemic rationality, of which the formal versions are stated in Appendix A.

Toward a Lockean Unification of Formal and Traditional Epistemology

A Lockean metaphysics of belief that understands outright belief as a determinable with degrees of confidence as determinates is supposed to effect a unification of traditional coarse-grained epistemology of belief with fine-grained epistemology of confidence. But determination of belief by confidence would not by itself yield the result that norms for confidence carry over to norms for outright belief unless belief and high confidence are token identical. We argue that this token-identity thesis is incompatible with the neglected phenomenon of “mistuned knowledge” – knowledge in the absence of rational confidence. We show how partial epistemological unification can be secured, even without token identity, given determination of outright belief by degrees of confidence. Finally, we suggest a direction for the pursuit of thoroughgoing epistemological unification.

Justification, Luminosity, and Credences

Chapter 8 discusses the repercussions of capacitism for the justification of beliefs, the credences we should assign to perceptual beliefs, and the luminosity of mental states. In light of this discussion, the chapter explores the consequences of capacitism for various familiar problem cases: speckled hens, identical twins, brains in vats, new evil demon scenarios, matrixes, and Swampman. I show why perceptual capacities are essential and cannot simply be replaced with representational content. I argue that the asymmetry between the employment of perceptual capacities in perception and their employment in relevant hallucinations and illusions is sufficient to account for the epistemic force of perceptual states yielded by employing such capacities. I show, moreover, why capacitism is compatible with standard Bayesian principles and how it accounts for degrees of justification. Finally, I discuss the relationship between evidence and rational confidence in light of an externalist view of perceptual content.