Men keep their agreements when it is an advantage to both parties not to break them; and I shall so frame my laws that it will be evident to the Athenians that it will be for their interest to observe them.
-Solon, reported in George Shelley Hughs, Ancient Civilizations
The prisoner’s dilemma is the textbook example of a coordination failure in which rationality is an insufficient condition for bringing about the best state of affairs. Rather than ensuring the best state of the world, rational forethought actually brings about the worst. Rational forethought - prudence, supposedly a virtue - is the cause of malaise rather than prosperity. The prisoner’s dilemma has thus become a template for all coordination failures to be seen, whether that is the depletion of fishing stocks or the nuclear arms race between the United States and the Soviet Union.
One of the ignored features of the prisoner’s dilemma is that in the orthodox elucidation of the problem, the dilemma is not a dyadic relationship between the two prisoners, but a triadic relationship between the two prisoners as well as the police. Considering that the payoff matrix of a prisoner’s dilemma only provides payoffs for two players, one can easily forget that there are actually three parties at play: two who are interacting within the game and another who is setting up the parameters that the two prisoners are dealing with. In the prisoner’s dilemma, the police are offering the prisoners a deal no rational agent would refuse: “You can either confess your crimes and be a witness against the other guy in the room, get a plea deal, or go on trial for your crime. Oh, and your friend may tattle against you, so make your decision quick.” The two players aren’t prisoners of the dilemma; rather, they’re prisoners of a third party which has created the dilemma they are in. The prisoners have no choice but to play the game put before them.
Consequently, the prisoner’s dilemma is not a template for studying cooperation in general. We need to be mindful that game theory is a way of studying particular interactions, and that no game will ever provide a general theory of any phenomena. Whereas the prisoners are stuck in the position that they are within by virtue of being in custody, people across the world are working so as to ensure that cooperation does not take the form of a prisoner’s dilemma.
People are natural problem solvers. They are able to mold the context of social interaction in ways that violate the assumptions of the prisoner’s dilemma. Human interactions are more than the dyadic choice to cooperate or defect. People aren’t stuck with the choices that they are given by the police, but are instead free to bargain over the gains to cooperation, and to create deals which leave everyone at the table content. The entrepreneurial facet of human nature ensures that people are alert to opportunities for pure profit and that those opportunities open up the possibilities for better (we could even say Pareto-superior) outcomes.
Interactions can take forms like the invisible-hand game, in which the payoffs guide cooperation to better outcomes for all involved. For centuries, political economists have shown how, with the right rules in place, man’s selfish nature can even benefit social harmony. What enables the maintenance of such cooperation is mutualistic behavior: behavior in which it is in the best interest of both players to do and to see the other player doing.
The threat of defection does not threaten the integrity of the international division of labor nor would it threaten the integrity of a hunting-gathering band because culture structures the payoffs such that they don’t. Cooperation within such societies is mutualistic. Everyone’s best interest is served by cooperation within their respective societies. Cooperation therefore often doesn’t take the form of a prisoner’s dilemma, but rather some game in which cooperation is an evolutionarily stable strategy.
The prisoner’s dilemma does not describe cooperation in general. We would be fools for thinking that. Rather, the prisoner’s dilemma explains a very specific context of social cooperation, and a context which is actually artificially created by a third party, the police which are offering the deal to cooperate with the investigation.
The prisoner’s dilemma is certainly applicable to other circumstances, but when applying it to them, we need to be mindful about what are the circumstances that are preventing players from bargaining with each other to find better outcomes than the one they are initially faced with. Without that explicit constraint, the prisoner’s dilemma will largely fail to apply; after all, the history of human cooperation, from the Paleolithic on, has been a history of human beings adapting their social circumstances to the challenges they face, and that history of adaptation leading to the social conquest of the Earth.
Comments