SOME things stay the same, and some things change. We know more about some of the issues addressed in this book than we did when it was written. I have made changes, some small and some larger, to bring the discussion up to date. There are new footnotes and lots of new references, of course. Some larger modifications are made in the substantive discussions.

In Chapter 1 I say a little more to preview the importance of correlation that is to be the main theme in Chapter 3. I refer to correlation set up by local interaction in the form of bargaining with neighbors. This is work done with Jason McKenzie Alexander that was produced subsequent to the publication of the first edition. I also make a historical correction: Darwin himself really cracked the sex ratio puzzle.

I am more circumspect in Chapter 2 about the connection between rationality and backward induction, perhaps one might say evasive, since I do not want to plunge into the counterfactual reasoning involved. That is not what this book is about. I change terminology from “modular rationality” to “sequential rationality” to bring it into alignment with the standard terminology in the game theory literature. The empirical discussion of behavior in ultimatum bargaining now includes the work of anthropologists, which reveals great variability across small-scale societies. This fits well with a discussion of framing of social norms that was only hinted at in the first edition. There is a reference to new work with Kevin Zollman on a model of such framing.