Causation in Commonsense Realism

We think of perceptual experience as a source of propositional knowledge of the world around us. So we think of perceptual experience as a reliable source of true beliefs. Therefore, we must be thinking of perceptual experience as causally dependent on perceived objects. This is a rough sketch of a line of reasoning that plays a significant role in P. F. Strawson’s defence of the causal theory of perception. In Strawson’s words, it provides the ‘rationale’ for the causal theory of perception. The ‘rationale’ has been criticized by Paul Snowdon in a series of papers pioneering a ‘disjunctivist’ account of perceptual experience. Snowdon’s central complaint is that the causal requirement has not been shown to be necessary, given that the causal theory is not without alternatives. For example, one might account for the reliability of perception by invoking a relational view of perceptual experience, on which mind-independent objects are constituents of the experience we enjoy in perceiving them. Some philosophers have argued that the causal theory of perception and the relational view of experience are not in fact in conflict. I think they may well be right, but the point does not resolve the dispute between Strawson and Snowdon. Even if we can consistently hold both views, we still face the question whether Strawson is right that a causal requirement on perception is somehow implicit in everyday explanations of what someone knows in terms of what they perceive. If he is, this would help to motivate, or even lend support to, the idea that such conditions are in some sense part and parcel of our very concepts of perception, or, as Strawson sometimes puts it, that they are a ‘pre-theoretical commitment of commonsense realism’. It would present a challenge to sceptics, such as Snowdon, who think that an interest in the causal requirements on perception is, as it were, an acquired taste, not