There are two important tenets of (Ernest Sosa's) virtue epistemology: one is the idea that knowledge is apt belief, the other is the claim that there is a distinction between animal knowledge and reflective knowledge. The approach of virtue epistemology helps us find a way out to both of the Gettier problem and the value problem (in its traditional version and its contemporary counterpart), both problems have been bothering epistemologists in the field for long time. Unfortunately, I will point out and argue in this paper, if we follow this approach, we will face a new nature problem as well as a new value problem of knowledge. Both of the problems come from some reflections on the alleged reflective knowledge deeply seated in Sosa's virtue epistemology.