I'm reading Philip Kitcher's book, The Ethical Project. Tim Dean has an extensive review of it at Ockham's Beard (starts here and goes on for several posts), so I'm not going to try to "blog the book." But I do want to make a few remarks.
Kitcher is quite the polymath, having written on everything from Finnegan's Wake to the philosophy of mathematics. Unfortunately, he isn't a very good writer - the book isn't one of those that grab you and pull you in. (And that's not just my opinion, several commenters at Tim's blog also found it to be a slog.) It is, however, worth the effort. Kitcher's approach is grounded in evolutionary theory, game theory (Kitcher has himself published papers on game theoretical simulations), anthropological and primate studies, and archaeological evidence.
The first surprise for me was that Kitcher sees ethics as a cultural phenomenon - not an instinct. The reason for its existence is to remedy "failures of altruism." For humans to live in large groups - and gain the benefits of division of labor, etc, that come with such groups - they needed to invent mechanisms that led them to be more altruistic than (for example) chimpanzees are. The goal of the first part of the book is to show how this could possibly have happened. He emphasizes that we will likely never know how it actually did happen; he only intends to refute the claim that "you can't get there from here", i.e., there's no way to derive ethics from evolution.
The second surprise was that Kitcher sees the invention of religion as a crucial step in this process. As rules for behavior developed it became necessary to invent some sort of supernatural enforcement for those rules. Most societies believe in a supernatural enforcer: not necessarily a god, though (he mentions one group that believes breaking the rules brings bad luck).
If he's right, then those who say "no morality without God" may, in fact, be correct - just not in the way they think!