Posts by me will take on a more intermittent quality over the rest of the summer. I’ll be travelling some, but the main reasons are two. First, I’ve agreed to give some talks next year, and it’s always better to feel comfortable with your views in such situations (even if . . .), and that’s going to take some work. Second, I’ve seen my way through on a book project that ties together my interests in propositionalism (the anti-(reliabilism/properfunctionalism/virtueepistemology) view I’ve defended in several places), coherentist theories of justification, the nature of understanding, and value-driven approaches to epistemology. I think I now see how to defend a coherentist propositionalism about justification that is value-driven and forms a basis for an account of understanding. The key idea is to identify what I will call a “neutral” starting point: propositional contents that play the same role in the story no matter whether they are encoded in experiential states or in belief states. Combined with a suitable grasp of what makes justification matter to cognitive agents, such a neutral starting point provides a satisfying resolution to Sellars’ Problem and to Pollock’s Challenge. The former is well-known, and the latter is, to my mind, the strongest objection to propositionalism available. Pollock claims that justificatory relations are relations between mental states, not between their contents, since the same content can be the content of a variety of mental states. Whether that content functions so as to justify a given belief with a given content depends on what particular (kind of) mental state it is encoded in, for a belief with content p cannot justify itself, and yet an experience with content p can justify this belief.
Besides, Branden’s review of the Bovens-Hartmann book on bayesian epistemology has convinced me that I need to spend some time on it! (For the record, Branden, here’s my take: nobody should like L5; L4 looks right, and some form of L3 looks good as well; but L2 is not compelling, and L1 isn’t the right approach for a coherentist to take toward the issue of the truth connection, so they shouldn’t endorse it. There. I’ve said it. Now I’ll go read the book to see where I’m wrong!)