Without offering you a lot of commentary at this stage, I present you Tim Dean's review (actually the first of two parts) of Philip Kitcher's The Ethical Project.
You can (once again) find my own review here if you want to compare notes.
I do agree that the pragmatic/functionalist approach to morality is the way to go, although one point that I'd want to make about this is that J.L. Mackie made the same points pretty strongly in Ethics: Inventing Right and Wrong (without, however, Kitcher's elaborate conjectures about evolutionary and cultural history). I suppose my other point is just that I wonder how this relates to moral semantics. I agree with Kitcher that morality can be seen as something we invented (with a bit of help from our prior evolution) to serve what can be loosely regarded as a function (there's a lot to say about the "loosely regarded" bit). This may be the most fruitful way to see morality, and it may be at least approximately right as an account of how morality actually came to be.
But the implication seems to be that moral norms are norms that they historically served the function and are now socially entrenched. Even if some moral statements are true in some sense (on a pragmatic conception of truth), they may not be true in the sense that the folk imagine they are, and if the folk are taken as asserting their truth in a sense in which they are not true ... well that's interesting, is it not? Error theory keeps lurking around, it seems to me, unless we are going to offer some very fancy moral semantics. I'd have liked to have seen Kitcher address this, and more generally to have addressed Mackie's views - and to have offered a bit more in the way of moral semantics. Given the considerable strengths of The Ethical Project this might have been a helpful and fruitful tack.