McDowell on Reasons, Externalism and Scepticism

This formulation of content externalism is very weak, in that most content externalists will want to be more robust—and, of course, more specific—in their construal of the role that environmental factors can play in determining the content of mental states, and will typically take environmental factors to be relevant to the determination of the content of just about any mental state. Nevertheless, the virtue of (CE) is that it highlights just what, at root, is being denied by content externalists, and this is the content internalist idea that the content of an agent’s mental state is purely determined by non-environmental facts—that is, by facts that are only concerned with what is beneath the skin of the agent. As Hilary Putnam might have expressed the point (but didn’t), content just ain’t in the head. Content externalism, if true, is clearly a thesis of vital philosophical importance. Moreover, one would expect content externalism to have some anti-sceptical ramifications. In particular, if content is at least sometimes determined, in part, by environmental factors, then it follows that insofar as we have mental states with ‘externally’ determined contents in this way then there must be an external world. Scepticism about whether there is such an external world is thus met. The more substantive issue, however, is whether content externalism is able to meet the sceptical challenge regarding whether it is possible for us to have widespread knowledge about the external world. Clearly it will have some antisceptical import in this regard. For example, if content is at least sometimes determined, in part, by environmental factors, then this will impose a constraint on acceptable forms of sceptical argument of this type by restricting the kinds of sceptical hypotheses that the sceptic can use in such an argument (we will consider two examples in a moment). Nevertheless, it is generally accepted that whatever construal one might endorse of content externalism, it will inevitably lack the resources needed to offer a full resolution of the radical sceptical problem. That is, that whilst content externalist theses might be able to rule-out a priori