The metaphysics of virtue needs to be distinguished from the epistemology. The account of what a virtue is has been outlined. However, there is still the epistemological problem of how to determine what consequences of the trait count. Suppose that we were all grossly mistaken about the benefits produced by generosity. Suppose that generosity only produced good consequences in the short term, but the long-term consequences were devastating. If generosity toward the needy in the long run produced parasites, or persons whose characters had in some way been undermined, and if generosity did this systematically, then it would not be a moral virtue. Long-term consequences count. For this reason, I think it entirely likely that we are mistaken in calling some traits virtues precisely because we fail to see the harmful effects these traits produce. Those traits with good foreseeable consequences are the ones we regard as virtues – though the judgment could be mistaken. The more we know about the world, the fewer mistakes we will make. These observations provide a great deal of intuitive support for a consequentialist theory of virtue. The fact is that when we do see that we have misjudged the consequences of a trait, we change our judgment of the trait's status as a virtue.

Chastity may be an example of this. Chastity for women is not generally considered to be a moral virtue anymore, though it certainly used to be considered one.