Religion, Context, and Constraint toward Strangers

Although it is widely believed that religion can constrain egoistic behavior, this has not been tested with behavioral data. This article provides such a test, using prisoner's dilemma data collected in Logan, Utah, and in Eugene-Springfield, Oregon—contexts that differ sharply in both the incidence of religious affiliation and the extent to which one religious group dominates that context. There were three major findings: a widespread belief, shared equally by religious and nonreligious people, that religious people will cooperate more than nonreligious people; no relationship, in fact, between religious affiliation and cooperation; and an increase of cooperation with church attendance but only among Mormons in Logan. Consistent with standard experimental method, subjects in these experiments were assigned to experimental treatments randomly, meaning that the people they confronted were randomly met strangers. It is proposed that involvement with a religious institution will constrain behavior toward strangers only when the religious group dominates the ecology—and when there is, therefore, a high probability that such a randomly met stranger shares one's own religious affiliation.