Lawrence Solum's Legal Theory Blog has a fascinating piece on a recent review by Lawrence Blum of a book by John Doris: Lack of Character: Personality and Moral Behavior which draws on social psychology to argue that there is no empirical evidence that human beings actually have character traits, which pretty much puts paid to moral theories based on virtue and to ordinary language about character as well. I've followed most of the links, and I'm not entirely sure what to think about this. For one thing, I think that Solum's preliminary analysis is mostly on the money, both as to potential problems with the research itself and to many of the consequences for ethical and legal theory.
One thing that I don't think I've seen mentioned is that if one of the stronger forms of situationalism (that people's moral behavior is mostly determined by subtle specifics of their current situation and not by stable ongoing tendencies) is correct, it's not just bad for Virtue ethics, but pretty disastrous for Consequentialist ethics and probably Deontological ethics as well. For instance, if (as one experiment shows) whether someone stops to help a person apparently having a heart attack depends on nothing else except whether that person had previously been told that he or she had to hurry to perform some trivial task, then whether it is in some sense correct to assess the morality of an action based on its consequences or its conformity to certain moral rules is beside the point, since we can't actually expect anyone to perform the required assessment in practice. (This seems a bit worse for Consequentialism, since the Deontologist could still insist that practical or not the moral choice was correct, while Consequentialism about things that don't actually affect consequences seems a non-starter--but maybe that's just me.)