No one would suggest that we can adequately investigate what makes something an acid, or what makes something aluminum, by bringing our pretheoretical intuitions about these things into reflective equilibrium by way of armchair theorising.

I am quite wedded to the view that epistemologists should concern themselves with knowledge rather than our concept of knowledge. The analogy I like to draw here is with our understanding of (other) natural kinds.

Bealer has a number of reasons for thinking that a naturalistic epistemology is self-undermining. Let me focus on one of these. (I've tried to take on all of them in the first chapter of Knowledge and Its Place in Nature.)

Here, there is simply no substitute for the kind of work that experimental psychologists do, work which shows some mechanisms to be quite reliable, and others to be quite unreliable.

My own reasons for favouring talk of natural kinds is just that I believe the best accounts of the success of scientific theories presupposes the existence of natural kinds.

When we recognise that reflective processes are no more outside the causal net than unreflective processes, and that they are bound by similar constraints, we may come to understand the nature of reflection for the first time.

If one's interest is not in some global question about the possibility of knowledge, but about some particular mechanism or inferential tendency, this fact about our evolutionary origin is of no use at all in addressing questions about reliability.