In this paper we give reasons to think that reflective epistemic subjects cannot possess mere animal knowledge. To do so we bring together literature on defeat and higher-order evidence with literature on the distinction between animal knowledge and reflective knowledge. We then defend our argument from a series of possible objections.

Internalists who argue against reliabilism usually construct thought experiments designed to show how reliability is not necessary or sufficient for justification. Defenders of reliabilism have responded with debunking explanations of the intuitions that people are expected to have when considering anti-reliabilist thought experiments. One defender is Jennifer Nagel, who argues that internalist counterexamples to reliabilism play off of a shift between belief-formation processes that are unconscious and those that involve self-reflection on the contents of one’s conscious states. Nagel aims to (...) show how thought experiments commonly appealed to by internalists rest on a mistake by carefully examining some of the most well-known examples in the literature. However, Nagel’s analysis does not apply on certain counterexamples to reliabilism, such as Cohen’s New Evil Demon. In this paper, I will show that when applied to NED-intuitions, Nagel’s analysis fails, so her argument does not extend to all anti-reliabilist counterexamples. (shrink)