The role of evolved predispositions in cultural evolution. Or, Human Sociobiology meets Pascal's Wager

Much human behavior affected by culturally transmitted ideas-e.g., religious sentiments- does not appear to make adaptive sense. Whether sound sociobiological explanations can or cannot be given to such behavior is at the heart of debates between human sociobiologists and their critics. We argue that the debate has been miscast as a clash between mutually exclusive hypotheses, when in theory and probably in fact there are many cases that mix elements of both. When culture, using Darwinian methods, is modeled as a system of inheritance that coevolves with genes, it exhibits both adaptive properties and a series processes that give rise to maladaptive variation. On the one hand, there is every reason to expect that cultural evolution is substantially shaped by evolved, genetically transmitted predispositions and that these predispositions often result in highly adaptive behavior. However, some kinds of beliefs are weakly affected by evolved predispositions and can have strong effects on behavior. For example, religious sentiments are carefully constructed to be difficult to challenge on the basis of empirical experience. Beliefs in rewards or punishments in an afterlife can substantially affect the behavior of people with quite utilitarian evolved goals. Once beliefs arise that are difficult to judge using evolved predispositions, processes such as group selection on cultural variation can be relatively strong. Patterns of behavior that could not be predicted without taking account of the evolutionary properties of culture can arise as a result of such "nonsociobiological" forces. Four coevolutionary scenarios capture much of the rich behavior that is possible when genetic and cultural evolution interact: 1) Culture may be kept on a "leash" by evolved predispositions, as classical sociobiological arguments would have it; 2) the tables may be turned, and genes may be leashed by the evolution of cultural norms that affect mate choice (we present a simple model of this process); 3) harmful cultural elements with an ability to attract imitators may arise, leading to a host-pathogen type of coevolution between genes and culture; and 4) genes and culture may evolve like a system of obligate mutualists. We speculate that the last scenario is the most generally important, but that the others are common. We view the application of population genetical methodology to cultural evolution as a friendly amendment to human sociobiology, but one that is essential to a complete Darwinian theory of human behavior.