What I argue is that talk of knowledge plays an important role in theories within cognitive ethology. The idea is this. First, one sees cognitive ethologists arguing that we need to attribute propositional attitudes to some animals in order to explain the sophistication of their cognitive achievements.

By putting the first-person point of view in a naturalistic perspective, I believe that we may genuinely come to understand it for the first time.

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.

Ethologists thus have an interest in looking at these capacities for the reliable acquisition of belief, and it is not surprising that they have a name for the true beliefs which are the typical product of these reliable capacities. They call them items of knowledge. So I argue that talk of knowledge may thereby be seen to be embedded within a successful empirical theory.

Externalists reject any such view. I think that the idea that we can tell, simply by way of reflection, whether our beliefs are justified, is deeply commonsensical. More than that, the idea that responsible epistemic agents ought to reflect on their beliefs, and hold them only if they somehow pass muster, is utterly natural.

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.

The worry that unreflective belief acquisition may be unreliable, after all, applies equally to reflective belief acquisition: it too may be unreliable. To my mind, the plausibility of internalist views about justification is dramatically decreased when one becomes vividly aware of what introspection and reflection actually achieve.