In short

1. We act.
2. To act is to have reasons.
3. To have reasons is to be committed to concepts with moral content.

Adding: I really didn't mean this post as some kind of dig at Neal. On the contrary, the moral skepticism discussion in the Gaza thread below got me thinking about some of the meta-ethical problems that I work on, and those three statements occurred to me. Together, they seemed to me to be a succinct elaboration of a view that I've pretty consistently espoused for awhile now, and which I think is general enough to garner wide agreement. I thought that was neat, and worth writing down.

But just as Neal reacted more harshly that was warranted, so should I have toned down my response. Believe it or not, hours before Neal's broadside I abandoned a long comment about Mackie and error theory because diving into the literature seemed pedantic. Suffice it to say that I recognize that the three statements above don't constitute a proof for the existence of moral concepts. One could say, in something like the way Mackie does, that our commitments are in error. The weird thing about this line of thinking is that it assumes that the existence of moral concepts requires that they meet some standard beyond what is necessary for the truth of claims like 1, 2, and 3. Why do they need to meet such a standard? To me it seems more plausible to suppose that the standards for what a reason is, and what counts as a reason, are internal to the project of moral reasoning.

Which brings me to sociobiology. Again, "stupid" is too harsh. Let me first be clear what I'm not denying. Facts about human nature and biology certainly can inform our moral decisions. So, for example, we now know that lead poisoning is correlated with loss of impulse control and, subsequently, criminality. We also know that a leading cause of lead poisoning in the United States had historically been the ingestion of chips of lead-based paint by young children. These facts have some clear moral implications, and those implications have led to a variety of policies designed to reduce the incidence of lead poisoning.

Sociobiology is different. Supposing that a successful sociobiology were possible, what it would give us is an explanation of moral concepts which justifies them through a reduction to adaptive fitness. In reducing-complex-realities-to-syllogisms words, a successful sociobiology might give you statements like the following:

S1: People will act altruistically in situation S because doing so expresses a trait which has proven to be adaptive.

At most S1 will tell you whether people will act altruistically in S. But the moral question is, how ought people to act in S. The answer to the moral question, in turn, has to do with what reasons people in S have. But sociobiology doesn't address its explanations to those whose actions it describes -- the only reason it could give is, 'have reasons to act altruistically because having such reasons encourages the expression of genetic endowments quite similar to your own.' That wouldn't motivate most people. Which is to say that even if sociobiology is maximally successful in explaining altruism on its own terms, people in S would find themselves in search of reasons to act altruistically and sociobiology would offer no guidance. In other words, sociobiology neither answers moral questions nor removes the need to answer them.