The prisoner's dilemma assumes that the players are selfish; that is, they do not care what happens to the other player. This assumption is not always true. Sometimes people do worry about others. Suppose, for example, that the two prisoners are fellow revolutionaries dedicated to the ideals of their cause, or they are husband and wife. In these cases, even with the payoffs given in the first table, it is unlikely that either prisoner will confess. Each prisoner in these cases considers not just what will happen to him as a result of his decision, but also what will happen to the other prisoner.
However, one should not argue that "good" motives necessarily lead to good results. Suppose that there is hard evidence that links two criminals to a crime and that each cares a great deal about the other. If neither confesses, both may serve ten years. Under these circumstances, each may confess to the crime, claiming that he alone was responsible in an effort to help his fellow prisoner. But if both try to help the other by admitting guilt, they both may end up serving more time than if each had maintained silence. Of course, this would not happen if they could communicate and plan their strategies, but often in life both information and communication are lacking. "Good" motives with poor knowledge can result in harm to the group welfare.
If one ignores problems of information and knowledge, a conclusion one can draw from the stories about prisoners is that there are two ways for groups (including that largest group, the nation) to get good results. Groups can take motivations or goals as given and try to find incentives (payoffs) that yield desired results, or groups can take incentives as given and try to find motivations or goals that yield desired results. However, a closer look at the assumption of self-interest suggests that the problems of information and knowledge cannot be ignored, and that they place important limitations on the ways groups can organize.1
In discussing self-interest, one must distinguish between people's actions and motives in small-group situations and in large-group situations. In small-group situations, as in a family, tribe, or a pre-industrial village, self-interest seems to be only one of many motives determining actions. At least as important are motives such as loyalty, duty, affection, and compassion. In such situations narrow self-interestignoring the interests of othersis a vice. Small groups often depend on selfless motives to get good results. The most basic of groups, the family, could not exist without selfless motives. In a family, good intentions usually lead to good results because family members have extensive knowledge about each other. In small groups, people often consider how their actions affect others in the group.
It is dangerous to assume that large groups can be organized in the same way as small groups. As groups get larger, more and more knowledge is needed if good motives are to lead to good results, and less and less is generally available. In addition, as groups grow larger, those affected by a person's actions change from identifiable people to anonymous strangers and this can affect behavior.
When one makes a distinction between small groups and large groups, one finds that the assumptions needed to get "self-interested" behavior are extremely weak. In fact, it is possible to assume that people are perfectly altruisticthey care only about the well-being of othersand end up with behavior that appears self-interested. One need only assume that the altruism is very strong toward the small groups with which people are associated (such as families or friends) and weak or nonexistent toward strangers (the large-group world). For example, a man who is interested in the well-being of his family acts in a way that, as far as the large group is concerned, may be indistinguishable from actions motivated by narrow self-interest. He will demand as much pay as he can get for his work and will try to pay the lowest price he can for the goods he purchases. An appeal made to him to work at a lower pay for the common good is likely to be resisted because it will harm the group he is most interested in helping, his family.
1Adam Smith thought deeply about this problem over two centuries ago. Smith recognized that many things, such as love, friendship, envy, and sympathy, motivate people to act, but he concluded that only self interest would generate the cooperation with strangers need for large-scale specialization and division of labor. For Smith, greater division of labor meant a higher standard of living. For more on Smith and self interest, see Ronald Coase, Essays on Economics and Economists
, especially Chapter Seven.
Copyright Robert Schenk