Empirical work on and common observation of the emotions tells us that our emotions sometimes key us to the presence of real and important reason-giving considerations without necessarily presenting that information to us in a way susceptible of conscious articulation and, sometimes, even despite our consciously held and internally justified judgment that the situation contains no such reasons. In this paper, I want to explore the implications of the fact that emotions show varying degrees of integration with our conscious agency—from none at all to quite substantial—for our understanding of our rationality, and in particular for the traditional assumption that weakness of the will is necessarily irrational.

The paper has two targets: the proximal target is the claim that in choosing the incontinent action rather than the continent one, the agent necessarily does something irrational; the distal target is the dominant naturalistic conception of how to justify norms of rationality. I'm taking aim at the latter through the former. The naturalist project aims to articulate and defend norms of rationality that are norms for the kind creatures we are—that is, for finite, embodied, social beings, with a specific cognitive architecture, functioning in particular environments. On the standard way of developing that project, norms of rationality are not a priori knowable and none can be viewed as privileged. Even norms as wellentrenched as norms prohibiting incontinence must await empirical support and recent work on the emotions raises the possibility that they might fail to get it.