Rationality and Emotions in Ultimatum Bargaining: Comment (2)

Mr. Shmuel ZAMIR, who is renowned as both a game theorist, and an experimentalist, focused his lecture, given at the time of the First Annales Conference, on the correspondence existing between the proposals forwarded by non-cooperative game theory and the empirical results of work in experimental economics. More precisely, his objective has been to establish the conditions under which the use of experimental data can be made compatible, within the framework of non-cooperative game theory. From such a perspective, this article helps us nourish a debate in which SELTEN, one of the three 1995 Nobel Prize winners in Economics, sought to provide the conclusion by stating: Game theory is for proving theorems, not for playing gantes. SELTEN's position acknowledges the presence of a divide, which has been widely addressed in the recent literature and which can be illustrated in particular by the opposing points of view expressed by Guth [2000] and WEIBULL [2000]: Orthodox game theory relies on pelfect decision naionality, i.e. the unlimited cognitive and information processing capabilities of players. Even for finite games of perfect information like chess however, it is obvious that these requirements are far beyond what human decisionmakers can accomplish [...] then the game theoretical predictions are often not confirmed by experimental observations. Guto [2000, p. 2]