Norenzayan and Gervais top off recent work studying religious and non-religious belief by trying to address distinctions between the cognitive mechanisms that promote these disparate forms of thinking. This topic is of immediate interest, I think, given the rapid growth in people who claim to disbelieve religious claims–albeit amid a world predominantly filled with believers of one stripe or another. Also, in the United States, the issue of religious belief and the rights of those who believe in a particular faith has become a major feature of the “culture wars”.
I had not been previously aware, but apparently there has been a budding discussion of the “evolutionary mechanisms of faith”, and a burgeoning growth of seemingly prominent hypotheses which, in Norenzayan and Gervais’ words, “see religious beliefs and behaviors as integral components of human nature” (e.g. see here and here).
In the present paper, however, the authors take what appears to be a much more reasoned approach, stating:
“We argue that disbelief arises from a combination of cognitive, motivational, and cultural learning processes traceable to both the genetic and cultural inheritance systems that are hallmarks of human evolution. As such, both religious belief and disbelief share the same underlying pathways.”
Along these lines, the authors first note a number of proposed cognitive mechanisms underlying religious belief. These include an intuitive (cognitive) sense that supernatural beings or gods might exist (e.g. Theory of Mind, an intuitive acceptance of mind-body dualism, a sense of purpose or teleology), motivational factors that might promote a belief in benevolent gods (e.g. awareness of mortality, randomness of life circumstances, social isolation), as well as cultural learning factors related to the actions of others, the prestige associated with religious avocation and the social consequences of disbelief. Thus, it has been concluded by others that these features of human mind, society and behavior inherently promote religious belief, making religious disbelief a difficult and labor-intensive task.
In contrast, Norenzayan and Gervais argue that religious disbelief is not simply an effortful process of over-coming this “natural” religious tendency, but occurs when a variety of predisposing religious factors are altered or disrupted. In particular:
“For a given person to believe in a given deity, he or she must (i) be able to form intuitive mental representations of supernatural agents; (ii) be motivated to commit to supernatural agents as realand relevant sources of meaning, comfort, and control; (iii) have received speciﬁc cultural inputs that… one or more speciﬁc deities should be believed in and committed to as real and important; and (iv) maintain this commitment without further analytic cognitive processing.”
For example, they discuss “mind-blind” atheism, which arises where individuals are incapable of or impaired at mentalizing or constructing a cognitive model of a “personalized” or anthropomorphic god. However, support for this instance of religious disbelief seems to be limited to cases of Autism, or small differences in mentalizing ability seen between men and women (men are less religious on average than women).
Other causes might be relatively good living circumstances (e.g. low levels of suffering, poverty and high levels of societal stability) which reduce motivational need for religious deities and a relative absence of socially-credible religious role-models, in many cases replaced by a safe secular society with non-religious role models (e.g. scientists and critical thinkers). A final determinant called “analytic atheism” was alluded to above. In this instance, the authors argue that disbelievers must utilize effort to overcome a more intuitive analysis of religious beliefs. On the plausibility of this final mechanism, I will note that while there may be group differences between believers and nonbelievers in how and when they use analytical skills, experimental studies investigating this relationship are not conclusive.
If there is an argument I take some issue with, it is the starting assumption that religious belief is akin to a “default value” of the human mind which can be “disrupted”. The authors don’t explicitly argue against this perspective, even while they note many instances where religious belief relies upon very malleable social circumstances as opposed to any indelible feature of (modern, western) human life.
This, of course, leaves us wondering exactly what is going on in the study of evolutionary models of religious belief. A point worth noting is that cultural transmission of religious belief need not be a direct driver of genetic selection promoting these beliefs, even if the timelines of cultural and genetic change overlap.
In response to this most recent paper, Michael Blume, writing on the Nature blogging site “Scilogs”, asks if “…human populations lacking any beliefs in superempirical agents [are] doomed to demographic extinction?”
He notes, for example, that individuals living in religious societies or who are members of religious groups tend to have higher birth rates, suggesting a potential form of religious selection. He further suggests that
“The more Atheism is flourishing numerically, the more Religion(s) are winning out evolutionarily.”
But I think the major monkey-in-the-room with strong evolutionary psychology theories of religious belief (and, correspondingly, religious disbelief) is that there is no evidence that the cognitive mechanisms underlying distinct systems of belief will segregate into separate populations. In other words, the offspring of both believers and non-believers are likely to have overlapping distributions of cognitive skills to promote either belief system.
Why should this be? Well, I taking as a starting point that the cognitive mechanisms which promote belief or disbelief are probably multi-purpose processes which support good performance on various tasks (e.g. theory-of-mind, social relationships, intuitive and analytical formats for learning and solving problems). In other words, they arise out of the functions of a diverse cognitive toolbox.
In order for these skills to segregate differently across populations of religious believers and disbelievers, we need to make a number of problematic assumptions. For one, given the likely utility of all of these cognitive abilities, we need to argue that religious societies actually select against transmitting skills that allow religious disbelief. Just because a person is a believer, doesn’t mean that lack the mechanisms to disbelieve.
We also need to assume that it is possible to select for genes that distinguish between these abilities, even though there is no evidence that such distinguishing genes exist or that they can be separated by selection (without detrimentally influencing other cognitive abilities). Most importantly, I highly doubt that there currently exists sufficient genetic differences between large groups of believers and disbelievers to allow any such segregation to occur based upon differences in the number of offspring produced by each group.
Rather than assume that there are heritable genetic differences across groups of believers and disbelievers, I am more in favor of the “belief/disbelief as an evolutionary by-product” model. A strong form of this model might propose that the cognitive mechanisms underlying religious belief are equally selected for across different cultures (or at least, those cultures which survive), which would allow cultural or social factors driving specific belief systems to influence what individuals believe right now, without impacting the ability of future generations to modify their beliefs.
In short, we must be very careful before proposing some powerful genetic selection factor with specific affiliation to religious or irreligious attitudes.