This book has a complicated genesis. For many years, I have been interested in the problem of collective action. Discussions with Brian Barry and Russell Hardin helped me to see roughly where the main problems were located, but I never seemed to get them fully into focus. Concurrently with this preoccupation, and spurred largely by proddings from Amos Tversky and Fredrik Engelstad, I became increasingly puzzled by the relation between rational choice and social norms. I discussed this problem with Pierre Bourdieu, and together we organized a conference on the topic. Once again, I seemed to make progress up to a point, and then confusion descended on me. Clearly, I was going against the grain.

The catalyst for further progress came in 1985, when Nils Elvander of the Swedish Council for Management and Work Life Issues (FA-Rådet) asked me to write a report on bargaining and collective action in the context of their project on collective wage bargaining in Sweden. I accepted in the belief, mistaken as it turned out, that my earlier work on rationalchoice theory might help me explain the strategies, stratagems and outcomes of collective bargaining. It soon became clear that the complexity of these bargaining problems defies explicit modelling. My analytical skills, in any case, were not sufficient to reduce the moving, fluid process of collective bargaining to manageable proportions. In the Swedish system of collective bargaining, as I try to explain in Chapters 4 and 6, everything is up for grabs: the identity of the actors, the rules of the game, the set of payoffs, the range of acceptable arguments.