Mutual Aid

Abstract

This paper studies the concept of ‘mutual aid’ developed by Kropotkin, which implies cooperation as a strategic choice. We study this concept in a sequential prisoner’s dilemma in a non-cooperative framework and in an indirect evolution framework (with complete and incomplete information). We systematically compare this game with one that models Kant’s moral. In the non-cooperative framework both moral concepts yield different equilibria according to the parameters. In the indirect evolution framework with complete information Kropotkin’s moral concept leads to generalized cooperation, while Kant’s rules lead towards general defection. In the indirect evolution framework with incomplete information both moral approaches favor selfishness. However, if some agents have an imperfect detection technology cooperative behavior will not disappear in Kropotkin’s case, while it will vanish with Kant’s morality.