UC Berkeley

I describe a debate among metaethicists about how to explain our reliability about value, showing that it is no mere coincidence. The “epistemic challenge” is to give such an explanation.It can appear that some metaethical views—that is, views about what values and norms are, most basically, like—have an easier time than others in answering the epistemic challenge. This makes the challenge relevant to evaluating such views, since metaethicists agree that successfully giving this explanation is one demand on a satisfying metaethical theory. This consensus is premature, however. Metaethicists have not adequately explained why we should want an explanation of our evaluative reliability at all. Until we have clarified whether and why an explanation of our evaluative reliability is worth having, we should not make such an explanation a goal of a metaethical theory. Metaethicists sometimes suggest that, unless there is such an explanation, knowledge and epistemically justified beliefs about values and norms would be impossible. However, they are not explicit about why achieving knowledge and epistemic justification in evaluative thought matters. To make this explicit, we should focus on valuable features associated with knowledge and epistemic justification. Our question is whether our evaluative beliefs would lose these features if we lacked an explanation of our evaluative reliability. I consider three such features: the admirability of our evaluative beliefs, their stability, and their reasonableness. In each case, I argue that the relevant feature is not threatened by a failure to answer the epistemic challenge. Even if a failure to answer the challenge does not threaten the valuable features of our evaluative beliefs, we might still have a reason to give the sort of explanation desired by the challenger—an explanation that makes this reliability no coincidence. I consider two arguments on behalf of the epistemic challenger for the claim that it is good to have such an explanation. On the one hand, commitment to falsehoods is a bad thing, and we consider some things highly likely to have an explanation. If our evaluative reliability is such a thing, then metaethicists have a reason to answer the epistemic challenge, because otherwise they would be committed to a likely falsehood. I argue, however, that no one is yet in a position to say that our evaluative reliability is highly likely to have an explanation. So we cannot yet claim that this is a reason to explain our evaluative reliability. On the other hand, we often take explanations to have great intellectual value. It seems straightforward to conclude that if we cannot answer the epistemic challenge, we miss out on an explanation, and so on something of intellectual value. I argue that this is not as obvious as it seems, by showing that on one prominent view of explanatory value, we cannot draw this conclusion. If the value of explanation lies in unification, as this view suggests, it is not obvious that explaining our evaluative reliability in a way that makes this “no coincidence” would yield greater value than an alternative explanation. To appeal to explanatory value to motivate answering the challenge, we cannot simply invoke that value, but must show how the relevant explanation better promotes that value than its competitors. One result of this project is that, for all we have seen, a key question that has interested metaethicists may have no importance. We may have no reason to engage with the epistemic challenge. This should inspire further thought about the proper tasks of metaethics, and about what we should want in a metaethical theory.