Abstract

The guiding theme of the dissertation is that any theory of perception must also be a theory of intention. One cannot give a plausible theory of perception without an explanation of intention being front and center in the theory. The main argument presented is that current accounts of perception, both direct and indirect cannot succeed because of their inadequate accounts of perceptual intention. ;The first section of the dissertation looks at the idea that perception provides a foundation for knowledge about the world. The implications for perceptual theory of holding that knowledge must be certain is examined within in the context of two classic indirect theories of perception: C. D. Broad's theory of Sensa, and Bertrand Russell's account of sense-data. I argue that neither theory can account for the intentional content of perception, and the contrast of Broad's and Russell's account clearly brings out how their epistemological assumptions force the move to indirect theory. The structure of indirect theory, however, makes it impossible to give an account of intention that does not negate their claims to realism. Indirect realism, I argue, cannot support claims to knowledge of an external world. ;In the second section of the dissertation I examine Gibsonian perceptual theory, focusing on the work of Michael Turvey. I argue that while Gibsonian theory has made many inroads against indirect arguments, it ultimately fails to escape from the forces that shaped indirect theory because it shares the same epistemological assumptions. Their theory of intention faces an unmanageable choice between including incidents of 'perceptual error' to account for intentional inexistence, or leaving them out to preserve the nomic nature of the Gibsonian account of perceptual intentionality. ;In the third section it is argued that only by abandoning the epistemological assumptions shared by the examined perceptual theories that a realist theory of perception is possible. Doing so, however, prevents us from using any of the theories of intention advanced in the discussed positions. I argue that a robust direct realist theory of perception is possible by incorporating Ruth Millikan's theory of proper functions into the Gibsonian empirical framework.