Abstract

In veridical perception things are revealed as they are, but looks can be deceiving. This is characteristic of intentional phenomena generally: the object of the intentional act need not exist, or exist as intended. In the first half, I argue that the problem such phenomena present is that of accounting for their normativity. I take Descartes to show that a conception of belief on the model of informing cannot clarify the distinction between what we believe and what we ought to believe and so fails to provide an adequate account of intentionality. Kant, I argue, makes the analogous point for a conception of action as motivated by desires. More positively, Descartes argues that perception must be understood to be a form of mental scrutiny; and action, according to Kant, has its intentional character by virtue of the fact that action is motivated by respect for the moral law. For both, normativity is inherent to the mental realm . As I argue following Sellars, however, nothing can be given to reason in the way required by the Cartesian conception. Following Wittgenstein, I argue that the Kantian conception generates a vicious regress of rules for interpreting rules. I also suggest that various contemporary approaches make either the "Aristotelean" error of failing to account for the fact that action and perception are normative phenomena, or the Cartesian-Kantian error of taking an entity to have normative significance by virtue of its objectively characterizable properties. ;In the second half, I argue that an individual's obligation to act in accordance with a practice arises out of a particular form of interaction between functional systems. My problem then is to account for properly intentional action and perception which can fail even where what one does or believes is what anyone would do or believe. In the last chapter, I indicate how such an account might go. The result is less a theory than a program but it is one that promises a new, and intelligible, model of the mind