Thinking Outside the Altruistic Box

Abstract

Two theories currently share prominence as explanations for the near universality of organized religion. Theory 1, the costly signalling hypothesis and its extensions have not to date generated predictions about the central question of why religion is religious; that is, why does religion invoke the gods? Theory 2, supernatural punishment, predicts that religion would be religious, but it requires group selection to stabilize its proposed evolutionary dynamics. We should not immediately dismiss group selection hypotheses, but given its rarity in the rest of nature, asserting group selection in humans requires extraordinary evidentiary support that at present is not enjoyed by the supernatural punishment hypothesis. Researchers studying the evolution of religion should consider more fully alternatives to these two currently popular hypotheses. Alternatives include the hypothesis that standardization of religious rituals and beliefs for signalling social group membership but potentially without group selection, that religion might function primarily for emergence of mutualism rather than prosocial altruism, and that group selection might apply to religious systems only during punctuated bursts of denominational diversification and death.