Direct Realism, Indirect Realism, and Epistemology

Abstract

The dispute between direct and indirect realism involves a combination of metaphysical and epistemological issues. According to indirect realism, perception is a triadic relation between a perceiver, a physical object that is perceived indirectly, and some private entity that is perceived directly. Standard objections to indirect realism derive immediately from the introduction of this third entity that supposedly stands between the perceiver and the external world. In particular, it is argued that if indirect realism were true it would be impossible for us to achieve knowledge of the physical world exactly because we are cut off from any direct perceptual access to that world. In other words, it is argued that an unacceptable epistemological consequence follows from the indirect realist's main metaphysical thesis. Direct realism eliminates this third entity and, it is widely assumed, thus leaves us in a better epistemic position than does indirect realism. In this paper I will challenge the claim that we are in a better position for learning about the physical world if direct realism is true than if indirect realism is true. It will be useful to begin by reviewing some of the central objections to indirect realism in a bit more detail. The most common forms of indirect realism hold that we directly perceive mental entities-variously referred to as 'ideas', 'sensa', or 'sense-data'-that are caused by a purely physical interaction between the perceiver and the physical world. But once such entities are introduced, the existence of illusions, hallucinations and dreams forces us to conclude that we cannot deduce the properties of physical objects-or even that physical objects exist-from the properties of the objects we perceive. At least since Locke, proponents of indirect realism have accepted this conclusion. In response, they have held that claims about the physical world must be justified by hypothetico-deductive argument and that perception