One of the goals of scientific theorising is to develop concepts which are adequate to the phenomena under study. In my view, things should work the same way in epistemology. We want to know what knowledge actually amounts to, not what our folk concept of knowledge is, since, just as with our pretheoretical concept of acidity, it might contain all sorts of misunderstandings and leave out all manner of important things.

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.

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.

The experimentalists think that we can only get at our concepts by way of empirical investigation, while the armchair philosophers think that we can skip the experiments and figure things out from our armchairs. What they have in common, however, is regarding our concepts as the targets of philosophical theorising, and I just don't think that, in the vast majority of cases, the subject matter of philosophy has our concepts as its target.

Chemists in earlier centuries were quite interested in the nature of acids. They had no interest in analysing their concept of acid. After all, they knew that their understanding of acids was at a fairly primitive level, and what they wanted to do was understand something about the world better - the nature of acidity - not something about their own concepts.

The role of empirical work in informing our philosophical theories, as I see it, is not that it gives us a better view of our folk concepts, but that it gives us a better view of knowledge, and the mind, and so on.

If we want to make sense of the possibility of successful inductive inference, and if we want to explain the possibility of laws of nature, we will need to appeal to something like natural kinds. This is, to be sure, a metaphysical commitment, but it is a metaphysical commitment that is implicit in science, as I see it.