Religion as a Mediator of Social Exchange: An Evolutionary Psychology Perspective

            When I hear the words “evolution” and “religion” in the same sentence, my immediate instinct is to cringe at the classic conflict between creationist and evolutionary perspectives that has pervaded Western society and American courtrooms, starting with the historic Scopes Monkey Trial. But there is an even more far-reaching and bracing relationship between evolution and religion, with the field of evolutionary psychology bridging the gap between these two worlds. Evolutionary psychology seeks to identify causal mechanisms for psychological phenomena, using evolutionary processes as a foundation for explaining human behavior. Religious belief is one of the most common elements of human culture, with nearly every modern civilization having some sort of supernatural belief system in place.

Evolutionary psychologists agree that there must be an evolutionary basis for religion, an almost universal element of human society, apart from the actual existence of supernatural beings. However, they have not been able to reach a consensus about the specific evolutionary significance of religion. Considering the various forms of interpersonal cooperation, sharing of resources, and mutual conflict that often result from religious belief in society, is religious belief best explained as a direct adaptation for social exchange or an indirect exaptation, or byproduct? Evolutionary psychologists have proposed several theories to explain why religion developed and gained prevalence as a part of human societies, including its adaptive significance for individuals and groups, as well as a byproduct of other social mechanisms. Of these, the group selection model seems the most plausible, as it clearly explains how religion could improve the fitness of a group or society and why this trait would have evolved by natural selection.

            Religion is a nuanced, complex concept that requires a clear definition before one can hope to explain its evolutionary significance. In general, religion can be defined as a system of beliefs in supernatural agents with some degree of omniscience and/or omnipotence. Religion must be shared between individuals, resulting in communal beliefs and practices. In social contexts, it acts as an organizing force for believers, and involves some sort of social stratification to give elevated status to religious leaders, like shamans or priests. As a part of its social nature, it also requires that individuals participate in rituals that involve altruism, such as sacrificing one’s livestock for the gods (Bulbulia 2004). By this definition, belief in a supernatural power is not sufficient to qualify as religion. The system of beliefs must be shared and must induce some form of ritual commitment that lowers an individual’s own fitness for the sake of the group (religious altruism).

In order to view religion from an evolutionary perspective, one must define an ‘ancestral’ and a ‘derived’ state that can draw a line between the historical lack and presence of religion in society. It is hard to find a historical example of the ancestral state, as religious belief is essentially universal among hunter-gatherer societies and modern civilizations alike (Bulbulia 2004). If the derived state is defined as an organized, shared system of supernatural beliefs that fulfills all of the conditions listed above, then the ancestral state can refer to the lack of an organized set of beliefs that is shared within a community. For instance, a prehistoric population of Neanderthals who, despite having some predisposition to believe in the supernatural, had not yet organized or codified a set of beliefs common to their community, would still be considered to be in the ancestral state. Then the derived state, religion, cannot exist without the social aspects of rituals and stratification.

            Specifically, religion can be described as mediator of social exchange, in that it entails moral principles and ritual practices that govern interpersonal relations in religious communities. Supernatural agents may directly govern some aspects of social exchange, as many religions invoke gods that “police social contracts” (Bulbulia 2004). In short, religion promotes the morality of followers through the intrinsic reward and punishment mechanisms of the gods. Bulbulia additionally argues that “costly religious displays convey information that solves prisoner’s dilemmas,” and that this can account for the omniscience and omnipotence of gods (Bulbulia 2004). Basically, rituals like celibacy and self-sacrifice harm an individual’s personal fitness (giving them a ‘handicap’) in order to demonstrate commitment to a religion. Participation in these rituals involves altruism, and is thus subject to the ‘prisoner’s dilemma,’ in which an individual must essentially decide between altruistic or cooperative actions and selfishness, based on the possible effects on his/her fitness. Religion’s many complex effects on social exchange have prompted three major explanatory theories of its evolutionary significance.

            Some adaptationists have argued that religious commitment can improve an individual’s fitness when he/she is part of a community of other believers, and is thus subject to natural selection at an individual level. Belief in supernatural agents can “alter the relevant payouts for exchange” in social contexts (Bulbulia 2004); omniscient gods cause individuals to be morally accountable. If a believer defects from his commitment, he may risk punishments sometimes as severe as eternal damnation for acts of selfishness. If religion is an adaptation, it must have evolved to solve a particular ancestral ‘problem,’ and it must satisfy the four postulates of natural selection. Religious belief and commitment must vary between individuals, this variation must be heritable, the fitness (survival and reproductive success) of individuals must vary, and the heritable variations in religious belief must affect an individual’s fitness. The individual adaptation model proposes that religion fulfills these requirements by fostering cooperation and reciprocal altruism, as well as individual benefits for religious believers. Clearly, individuals show variation in their levels of religious belief, degrees of commitment to social exchange, and in the types of reward and punishment systems in which they believe. Heritability, in this model, is slightly more complicated. Individuals inherit the innate psychological mechanism of cost-benefit analysis of actions, through basic socialization processes. They also acquire knowledge of the costs and benefits of certain actions, which can be passed on through cultural inheritance. Fitness varies in the sense that those who are more strongly committed to social exchange forfeit more material resources, but in their minds, see themselves as achieving the immaterial success of salvation (reward), as opposed to damnation (punishment). Therefore, an individual who is more committed to social exchange will be more positively rewarded within a religious framework, and will not incur a significant loss of material resources if surrounded by other religious people.

This framework, though it accounts for why individuals may be inclined to believe, does not provide a clear reason why this would increase their fitness relative to nonbelievers. Natural selection can only act upon phenotypic differences in success while living, not on mental pictures of eternity. Even if an individual believes that he/she will be granted eternal salvation, this will not positively affect physical survival or reproductive success. In fact, the individual will probably incur a loss from his altruistic actions, which would decrease fitness and cause selective pressures to act against religion. There is no clear reason that natural selection would favor an individual who is less selfish, even if the individual himself believes that he is increasing his fitness. The model does explain why being surrounded by other religious and altruistic individuals would increase success, but does not explain why religion would persist among individuals rather than be eliminated completely. Theoretically, a non-religious person would have the highest fitness in a religious community, because he/she could partake of others’ resources without giving up any of his/her own resources in return (Bulbulia 2004).

            Many, like me, are dissatisfied with this account of natural selection for religious belief, and instead look to its adaptive benefits at the group level. David Sloan Wilson is a proponent for group selection: the view that natural selection can act on groups, just as it acts on individuals, through differences in group fitness and inter-group competition (Wilson 2002). He argues that “the best explanation for religious thought and behavior is that it facilitates strongly integrated and functionally adaptive groups” (Bulbulia 2004). The group adaptation model seeks to explain non-reciprocal altruism, such as non-violence, vegetarianism, charity, and other forms of sacrifice, as beneficial adaptations for the survival of groups. Group selection models argue that the increased fitness of religious groups as units can overrule the decreased fitness of individuals within a group. In other words, strong positive selection for religious altruism at the group level can preserve religion even if individual selection favors selfishness. The existence of religious ritual could improve the mutual coordination of groups, thus rendering them more successful than non-religious groups.

Again, this model must be examined in relation to the four postulates of natural selection, this time viewing religious groups as the unit of selection rather than individuals. Groups show variation in their religiosity (ranging from devout to secular), as well as the relative number of altruistic rituals that members of the group are expected to perform. Heritability is still social and cultural in nature: religious heritage is passed down vertically between generations and horizontally to other groups of individuals (via proselytism). Certainly, some groups are more successful than others, in terms of economic success, survival, et cetera. Richard Sosis conducted an analysis that showed that “religious communes are more likely than secular communes to survive at every stage of their life course” (Sosis 2000). Religious communities are more likely to foster intragroup cooperation and loyalty than their secular counterparts, which gives them higher fitness. In addition, for religious communities, having more ritual requirements is correlated to having a longer group lifespan (Sosis & Bressler 2003). This implies that the effects of religious ritual on social exchange are highly beneficial to a group’s fitness. The same trend, however, does not apply for non-religious rituals (for instance, hazing in modern fraternities) in secular communities (Sosis & Bressler 2003). This is very interesting, because it indicates that altruistic behavior on its own is not sufficient to make a group more successful. In addition, religious communities that do not invoke religious altruism are not very successful either. The communities that participate in religious altruism are far more successful; religious altruism is the group trait that would theoretically be selected for.

            The evidence for the adaptive benefits of religion and ritual on a group level seems very convincing. However, there are some concerns with the group selection model of religion that deserve some attention. Religion on a group level can only be explained by group selection, and comes into conflict with the concept of inclusive fitness, by which individuals seek to increase their fitness by helping the survival of their relatives. Religious families as a whole will often risk their well-being for religious commitment (Bulbulia 2004). However, the whole basis of the group selection argument is that selection at the group level can act against selective pressures at the individual level. Also, it is not certain how a sociological adaptation would appear in individual psychology (Bulbulia 2004). Shared religious belief would somehow have to appear within several individuals of a group in order for selection to act upon the group. Perhaps groups of early humans who happened to begin sharing their supernatural beliefs with each other better withstood the tests of time, eventually causing organized religion to form. Another slight complication arises from the levels of stratification within religion: “religious groups” are hard to define (Bulbulia 2004). In a modern sense, this could refer to individual churches, sects, denominations, whole religions, or even the entire community of ‘believers’ in the world. Though it does make it harder to characterize the effects of group selection, it’s plausible that selection is occurring at several or all of these levels, according to the different selective pressures acting at each level.

            However, there are others who remain unconvinced that all of the effects of religion on social exchange can be explained by a single adaptive mechanism, and argue instead for religion as a byproduct of other psychological phenomena. Gould & Vrba term these byproducts “exaptations,” in which an original adaptation is co-opted to a second purpose after it evolves, conferring increased fitness (Gould & Vrba 1982). Some evolutionary psychologists observe that religion is not a necessary component of group sociability, and that groups divided along secular lines, such as ethnicity, sex, or political affiliation, produce similar results (Bulbulia 2004). Therefore, group-oriented sociability could have evolved independently, and then become co-opted as a mechanism for supporting religion in society. However, this seems less likely in light of Sosis and Bressler’s analysis. Secular groups that imposed altruistic requirements on their members were not significantly more successful than other secular groups, and religious groups that did not invoke religious altruism were not much more successful than secular groups. From an evolutionary perspective, religion seems to coincide directly with group sociability, rather than one preceding the other.

Another concern is that religion does not always control morality: “large societies are more often characterized by a belief in moralizing gods…[and] religious beliefs are correlated with existing social relations in a given society” (Roes & Raymond 2002). In other words, moral pressures coincide with the existing social structure of society; hunter-gatherer religions did not involve religious moralism or economic stratification between shamans and laymen. Therefore, religion may not have initially evolved with a moral agenda, or to enforce social contracts. However, even if religions are shaped to the societies from which they arise, and not vice versa, religious altruism would still play a major role in the success of the societies’ social structures over time. Ancient religious groups still involve religious altruism in the form of rituals, and thus could still differ in group solidarity and functionality.

            Clearly, none of the theories presented thus far are foolproof, and all require further investigation. However, the group selection model provides a clear and logical explanation for why scattered beliefs in the supernatural would evolve into organized religion, in the context of altruistic rituals. The individual selection model does not really show how individual religious believers would have increased fitness over non-believers in a group. The group selection model, by contrast, develops religious ritual as an adaptation for group success, and implies that religion evolved for the purpose of facilitating religious rituals, which in turn act to facilitate successful social exchange and increased group fitness. It is now for you, the reader, to consider that ritual altruism could be the real evolutionary purpose of a system so complex and so important to nearly every human civilization: religion.