How successful are recent evolutionary explanations of the persistence of religion?

A religion is a set of beliefs, held by many members of a religious group, concerning the cause, nature and purpose of the universe, and life within it. In Psychology, however, religion is "a human variable consisting of mental and physiological states and behaviour in relation to God or other religious or supernatural entities" (Petrovich, unpublished). Religion has persisted throughout human history in the same basic form and takes similar forms in all cultures. It has been shown to emerge intuitively in all modern societies (Rappaport, 1999), although in most cases it is very costly to the religiously practicing individual. Sacrifice, tithes, painful initiations and the effort involved in regular prayer, or travel all involve an element of cost (whether physical, mental or monetary) to the believer (Bulbulia, 2004), yet religion still persists. These costs suggest that religious cognition would have been deselected by the process of evolution, but it's persistence suggests either that there are other very beneficial results of religion belief, or that religion is not an adaption, but instead a by-product of ordinary cognition. These two views, the adaption and by-product views of the evolution of religion cognition are discussed below in an attempt to suggest why religion persists in modern society.

            Before considering the two opposing views of the cognitive science of religion, it is worth briefly touching upon two other, yet similar, views which attempt to explain the existence and persistence of religion. The first is the study of neurotheology, which suggests that the concept of God or religious beings and thinking in general is the result of misfiring of certain areas of the brain. Persinger (2003) famously induced a sense of a supernatural being in the room with the participant by stimulating these brain areas electromagnetically. Joseph (2001) also proposed the 'limbic marker hypothesis', which explains religious experience in terms of dysfunctional activity in the limbic system, which is responsible for emotional processing among other things. This view is perhaps quite closely related to the by-product view of the cognitive science of religion, however, and so this essay need not delve any further here.

The second is the idea of group selection (Wilson, 2002), where religion encourages prosocial behaviour within groups, causing these groups to out-survive other groups with worse social cohesion. Therefore, genetic information is passed on more readily and more widely by these prosocial, religious groups than by other groups, fuelling the persistence of religious thinking. This suggests that future generations may possess the biological predisposition to be religious, even if religion itself is not an innate concept or module. Atheism is a fairly new concept compared to the history of religion, and this perhaps is tied to the advances in medical science which allows a greater number of humans to survive when they would otherwise have not, allowing a greater number of people without the biological predisposition to religion to exist to challenge the status quo. Again, this group selection hypothesis, although not specifically suggesting that religion is an evolutionary advantageous adaption, is closely linked with this idea from the cognitive science of religion.

            The cognitive science viewpoint suggests the most evolution-focussed explanations, but is split in terms of the researchers in the field as to whether religion persists because it is an advantageous adaption in it's own right (and therefore perhaps a separate module in the mind) or simply a by-product of normal cognition, regardless of how useful or detrimental it may be. Any account of evolution must of course begin with Darwin, and in his 1897 book "The Descent of Man" he considered both views of religion, although post-Darwinian evolutionists only tend to take into account his by-product view. Darwin realised that when religion is defined as ritual, witch-craft and other practises, it makes sense to consider these as a by-product of evolution, as they serve no apparent purpose other than those suggest above of group selection, and the main criteria of an adaption, the idea that it is common to all members of the species, is lacking. However, when religion is considered as belief in unseen or spiritual agencies, this is seen in all cultures and societies and therefore could be considered an adaption. So how do we determine which view is correct? Many researchers have put forward suggestions in support of one or the other, and these are discussed and critiqued below, beginning with the adaptionist view.

            Research has suggested that religion may have an adaptive function in health benefits, ease of acceptance of death, and moral control and social cohesion (similar to Wilson's group selection theory). McCullough (2001) and Koenig (1997) both found that religious involvement is associated with lower death rates and better physical and mental health than seen in participants with less or no religious involvement. McCullough's study even took into account the respective benefits and costs of alcohol, smoking, social support and so on. The participants in these studies held mainly Judeo-Christian beliefs, however, so inferences about the effects of religion in general can't be made. Although a relationship has been found, health is not the main self-reported benefit of religion for the large majority of believers, and belief cannot prevent disease or prolong life more than is natural, so as supporting evidence for religion as an adaption, the health argument is weak.

            The benefit of religion in dealing with death is perhaps a more robust factor to consider. Humans are unable to represent nothingness or non-states, and therefore cognitive biases to believe in the immortality of the soul, or a different reality after death can be considered minimally counterintuitive (Bering, 2006). Bering (2005) believed that the human capacity to see meaning in natural and personal events must be adaptive, as it has persisted through the ancestral past and therefore must have enhanced genetic success. This inference does not necessarily follow, however, because persistence does not necessarily imply usefulness. Evolution and natural selection concerns whole organisms rather than specific traits, so plenty of non-beneficial trains may persist through evolution so long as the beneficial traits outweigh the non-beneficial in terms of survival (Fodor, 2005).

            Newberg et al. (2001) suggest that religion emerged to alleviate 'existential gloom' of our Palaeolithic ancestors, allowing them to struggle for survival while protecting them form self-defeating fatalism. The impact of religion on moral control is also a factor to consider. Bulbulia (2004) suggests that Gods provide an omniscient and omnipotent police force which ensures believers act beneficially towards other members of the species. This certainly seems like a beneficial adaption, but the suggestion does not imply it must be an adaption rather than a by-product, and there is not enough evidence to conclusively claim either way. Although these ideas can't be tested, and provide no explanation of the origins of religious belief, they do support the ideas of group selection discussed above. Bulbulia's (2004) suggestion that religion involves a high cost for believers may be explained by social psychological research into Cognitive Dissonance. For example, Aronson and Mills (1959) found that increasing the severity of an initiation procedure for a group increased the self-reported liking for that group once the participant became a member. Relating this to religion, the high cost of religious initiations or general practice and maintenance of belief may increase the sense of belonging to the religion, and therefore to the social group, thus in turn increasing social cohesion, with all the benefits that this produces.

            The by-product view suggests that religions is not a cognitive domain or module because there is no specific purpose for which it could have evolved that can nto be explained by a number of other modules or cognition. Therefore, religion originated and persists because of the beneficial adaption of other cognitive processes and the cultural transmission of attention grabbing concepts, particularly those that are 'minimally counterintuitive'. Boyer (2003) suggests that supernatural concepts, which are the bases of religious concepts, are 'parasitic' upon general assumptions from core domains such as the idea that a spirit is a special type of person. There are no modules which handle judgments about spirits, but many modules working in conjuction are able to make inferences based on an assumption that spirits exist.

            Guthrie (2002) notes that a common factor of religions is the attribution of life to inanimate objects or unobservable entities (anthropomorphism, or animism as Piaget put it). In 1980, Guthrie proposed the 'social brain hypothesis' which led Barrett (2002) to suggest that humans possess a 'hyperactive agency-detection device' (HADD) which detects supernatural agents when the cause of an event is unknown or unobservable. It is suggested that such a device may have evolved to detect predators and prey and for social development, but that over the course of human evolution, it became too good at its job. This explains why we see faces in clouds or hear voices in the wind. From an evolutionary standpoint, it is preferable for survival that we misinterpret false-positives, than we fail to interpret actual threats. Although there is no neural correlate suggested for this HADD, the suggestion of such a device is compelling support for the by-product view.

            Although it could be considered an innate pre-disposition, the cognitive predisposition of religion, as suggested by Pinker (2006) is theory of mind. Pinekr suggests that it is a short step form attributing a mind to biological entities (humans) to imagining a mind separate form a body (God). Adaptionists and by-product supporters alike would suggests that theory of mind is innate, or at least the ability to develop it is innate, but the difference is whether religion develops out of a combination of modules such as theory of mind and the HADD, or whether it is the result of an adaptive module in itself which interacts with these other cognitive modules to give rise to religious belief.

            The evidence presented above describes two conflicting views hoping to explain the persistence of religion, the adaption and by-product views. Both seem to agree that religious beliefs and practices stem from ordinary cognition, but they differ in whether religion is a distinct form of cognition, or a separate module as proposed by the adaption view, or whether religion arises out of the interconnected cognitive processing of multiple domains and modules which themselves have evolved as beneficial adaptations. Unfortunately, there is a scarcity of evidence in support of both views, but the evidence that is presented above suggests that a religion specific adaption would have been unnecessary, and therefore less likely to have persisted through evolution, while the cognitive modules which have more immediately apparent usefulness such as the HADD and theory of mind suggest that religion may have developed as a by-product of normal human cognition, and persists through social and cultural transmission, perhaps due to the beneficial effects on health, acceptance of death and moral and social cohesion, provided by this 'evolutionary accident'.