Justin Dealy

On Giving Yourself a Sign, Journal of Ethics and Social Philosophy 28 (2):171-207 (2024)

I argue we can have subjective practical reasons to perform actions we believe are neither morally required nor a means to satisfy our intrinsic desires. These reasons are grounded in extrinsic desires. Specifically, my claim is that subjective practical reasons can be grounded in desires for signs (i.e., signatory desires), a species of extrinsic desire, together with means-end beliefs. These reasons act like any other subjective practical reason, except when they are trumped, which I argue can happen when they are competition with a subjective practical reasons grounded in intrinsic desire.

My dissertation explores the theme of realistic rational decision in three different ways. The first part is a study of a classic challenge to applying formal decision theory to real-world decision-making situations known as the Problem of Small Worlds. The second part (JESP 2024, see above) is about practical reason's relation to desire. The third part presents and defends a new Subjective Decision Theory based on a novel conception of conditional belief states.

On an "assimilationist" view, conditional beliefs and credences are ordinary propositional attitudes — what makes them "conditional" is just that their propositional contents are expressible by conditional sentences. On this sort of view, if we ask a question of the form 'Does S believe that C if A?', the answer depends on (1) what proposition 'C if A' expresses in our context, and (2) whether the subject S believes that proposition. I argue for an "anti-assimilationist" view that conditional beliefs and credences are sui generis. On the view I defend, when we ask a question of the form mentioned above, the answer depends on neither (1) nor (2), but rather on (3) what propositions 'A' and 'C' express in our context, and (4) whether S believes the C-proposition conditional on the A-proposition.

I defend a new decision theory that defines expected utility in terms of a conditional doxastic state that generalizes both standard conditional credence and imaging. I show that this theory can diffuse a range of well-known puzzle cases. A broader motivating contention is that the line between decision theory and epistemology needs to be redrawn in such a way that much of what has been thought to belong to the former discipline is in fact the purview of the latter.