supernatural punishment: what traits are being selected?

ReligionBrain&Behavior | In their critical analysis of supernatural punishment (SP) theories, Schloss and Murray (2011) tease apart two distinct but often conflated adaptationist approaches to religion: those that argue that religious belief enhances cooperation (cooperation enhancement, CE), and those that say that it helps people to withstand the temptation to cheat, helping them avoid the costs of being punished (punishment avoidance, PA). They also make a distinction between individual selection and group selection. However, they pay little attention to the traits that are the targets of selection: evolved psychological dispositions, flexible behavioral strategies, or culturally transmitted norms. These three types of traits roughly correspond to three styles of evolutionary approach to human behavior: evolutionary psychology, behavioral ecology, and dual inheritance theories (Smith, 2000). Each has a different expected temporal scale in which adaptive change takes place – this constrains the plausibility of particular hypotheses and their mutual compatibility.

Evolutionary psychologists explain our behavioral repertoire as a result of evolved psychological adaptations, which were shaped in ancestral environments. Let us examine PA accounts from this perspective. If belief in SP is an ancient adaptation, we would expect it to have emerged in small, egalitarian groups, where there is a tendency to punish all defection. The strong connection between detection and defection costs in such societies is said to favor the evolution of God-fearing psychological mechanisms. However, the purported egalitarianism of ancient human groups is to some extent an idealization, because evidence for social inequality, in the form of lavish beadwork in children's burials, dates back at least to the Upper Paleolithic (Vanhaeren & d'Errico, 2005). It is unlikely that in such societies everyone would have faced the same risk of being punished. Also, ethnographic parallels show that at least some small-scale egalitarian societies (e.g. Ju’/hoansi, Kalahari hunter-gatherers) practice mainly low-cost forms of punishment, including gossip and jokes at the expense of the offender (Wiessner, 2005). The low cost of punishment (for both offender and punisher) seems incompatible with PA. As Wiessner (2005, p. 135) says: “[this] informal means of punishment would be ineffective or insufficient in a larger-scale society with less mobility.” Taken together, it seems unlikely that PA emerged in ancestral environments. For the evolutionary psychologist, CE thus remains the only viable option, but it then still remains unclear, as Schloss and Murray aptly point out, why religion, rather than other forms of group-level activities, would enhance cooperation. Indeed, the plenitude of Upper Palaeolithic mobiliary art, such as Magdalenian Venus figurines, and their geographic clustering in distinct styles suggest that art may have been used to affirm group identity and group commitment (De Smedt & De Cruz, in press).

Behavioral ecology takes the current environment as the relevant context of adaptation, and builds models on the basis of expected utility of different behaviors. As organisms are expected to optimize their behavioral repertoire to their ecological and social contexts, this approach predicts little or no mismatch between expected and actualized fitness benefits. Forms of CE that see religion as costly signaling are compatible with this approach, because it expects individuals to make flexible, optimal choices, which could include a choice for a religious affiliation that allows for costly signaling. However, this places limitations on the explanatory scope of SP theories. They cannot explain, for example, why hunter-gatherers or medieval villeins would be religious, since there is little point in the costly signaling of one's membership in a religious community if there is no freedom of religious choice. After all, a “free market” of religious groups is a relatively recent and not globally widespread phenomenon, mainly restricted to northern America (Finke & Stark, 1989). Many European countries have state-funded churches with low levels of expected commitment, which makes competition by smaller high-commitment religious groups harder. Prior to the eighteenth century, religious choice was quasi nonexistent, as the treatment of religious minorities in Europe exemplifies – one need but think of the massacre of the Cathars in southern France, or the historical persecution of Protestants. There is still no freedom of religion for the present-day Iraqi housewife. So, although behavioral ecology allows for CE, it seems to be limited in explanatory scope.

Dual inheritance theories examine human behavior as a product of culturally transmitted norms that have effects on genetic fitness, and that can thus become part of a feedback loop. This approach is the most congenial to the possibility of group selection. In particular, groups must be distinct from each other and form cohesive wholes for group selection to occur. Group selection also requires that the fitness benefits of altruistic groups over selfish groups must outweigh the fitness benefits of selfish individuals over altruistic individuals within mixed groups (Sterelny, 1996). Human cultures, with their ethnic markers and distinct languages, do exhibit high between-group variation, and considerable within-culture homogeneity, allowing for group selection to occur. It is within this context that we can situate Norenzayan and Shariff's (2008) argument that belief in supernatural sanction is a group-level adaptive cultural response to life in large societies, where interactions between unrelated and unacquainted agents become increasingly important. However, as they themselves point out, the presence of large, cooperative and not very religious groups indicates that secular institutions like the police can be equally successful in instilling cooperation. From the perspective of dual inheritance theory, PA seems thus not very likely, since people in at least some societies (e.g. agnostic Scandinavian countries) can withstand the temptation to cheat when effective punitive mechanisms are present without belief in divine punishment.

While different SP theories are not all mutually incompatible, some of them may be so because of their divergent assumptions about the temporal scale on which selection acts and which traits are the targets of selection. Depending on the style of evolutionary approach one chooses, contrasting SP theories can be fleshed out. However, like most SP theorists, Schloss and Murray and remain inexplicit about whether psychological mechanisms, behavioral strategies or cultural traits play the most prominent role in their review of the evolution of religious behavior.