"The Fool Says in His Heart That Atheists Are Mutants"

Poor science underlies claims about atheism resulting from adverse mutations

Posted Mar 17, 2018

There has been debate about the role of religious beliefs in human evolution that has produced some very interesting theorising. Unfortunately, there has also been a fair share of very poor theorising. As an example of the latter, a recent paper with the provocative title 'The Mutant Says in His Heart, “There Is No God”: the Rejection of Collective Religiosity Centred Around the Worship of Moral Gods Is Associated with High Mutational Load' (Dutton, Madison, & Dunkel, 2017) set out to show that religious views outside the mainstream – disbelief in a god as well as belief in paranormal phenomena – result from genetic mutations that have allegedly occurred due to relaxation of natural selection for belief in a moral god that has occurred in these degenerate times we live in. The authors claim that atheism and paranormal belief are “deviations” associated with indicators of mutation load, including poor health, autism, fluctuating asymmetry, and left-handedness. However, this theorising is poorly thought out and largely unsupported by evidence. As their article covers a lot of ground, my response will be in several parts. This post will discuss whether religiosity is selected for by evolution and whether atheism might be adaptive. Subsequent posts will discuss claims about the relationships of religiosity with health, supposed markers of mutation load such as left-handedness, and with paranormal belief. (For shorter more sarcastic takes on this paper, see this post by Neurocritic and this one by Skeptophilia.)

To provide some context, advocates of the "social epistasis" model propose that since the Industrial Revolution in modern Western countries there has been a relaxation of natural selection for many human health and fitness-related characteristics that has resulted in an increase in deleterious genetic mutations. The argument is that throughout history, infant mortality was strongly related to genetic mutation load, and with the decline in infant mortality in modern times there has been a corresponding accumulation of genetic mutations, with such consequences as increased rates of medical and neurodevelopmental disorders such as autism and left-handedness (Dutton et al., 2017). For example,  a study by Michael Woodley and colleagues, who are proponents of this theory, claimed that general intelligence has widely decreased since Victorian times, even though IQ scores appear to have been rising since that time (Woodley, te Nijenhuis, & Murphy, 2013). (For an amusing and insightful critique of this study, see this blog post.) Personally, I am rather sceptical of this theory as it rests on many questionable assumptions and some rather shaky evidence. In this article I don’t intend to go into too much detail about this theory, but wish to focus on its application to religious (non)belief in a recent paper (Dutton et al., 2017). Specifically, the authors argue that a very specific form of religiosity was selected for in complex pre-industrial societies, that is, belief in a moral god apparently helped early complex societies to function by encouraging people to cooperate with each other and obey social rules. This is in contrast to the beliefs of earlier hunter-gatherer societies that did not usually feature a moral god, but tended to emphasise belief in a wide range of spirits and ghosts. The authors argue that paranormal beliefs, such as belief in ghosts would have been selected against in complex societies “because it is, in fact, comparable to the kinds of beliefs held by hunter-gatherer societies, and does not involve a moral god.” However, relaxation of selection in modern times would have led to “deviation from this very carefully selected religious norm” in the form of either paranormal belief on the one hand, or rejection of a moral god (i.e. atheism) on the other. The authors argue that these two types of belief are only superficially different as they are both underpinned by “increasing genetic mutations affecting the mind.” Furthermore, the authors hypothesise that these two types of belief “would be associated with the same genetic correlates.” They claim that their study tests this hypothesis by examining the genetic association of these “deviations.” As I will show later, they do a very poor job of testing their hypothesis.

Wikimedia Commons
The gods of the Greeks and Romans were not particularly moral, yet they somehow managed to create highly successful civilizations.
Source: Wikimedia Commons

This paper starts off in a very odd way by citing non-academic sources such as social media posts and Bible verses that supposedly suggest that there is a stereotype that religious people are more attractive and healthier than others, whereas those inspired by other gods or even Satan are more likely to be autistic or even left-handed. The authors then suggest that these claims may have a degree of truth that they propose to uncover. One reason I found this very odd ─ apart from a scientific paper making claims about alleged stereotypes based on such questionable sources ─ is that the Bible verses cited do not even make these claims and mostly were not even relevant. For example, I Samuel 16:18 is cited as saying that “those who intensely fear Yahweh… are more physically attractive.” However, the verse in question simply mentions that a servant stated that one of his sons was “a fine-looking man. And the Lord is with him.” One example of a person who happens to be good-looking is hardly equivalent to a claim that believers in general “are more physically attractive.”[1] I mention this because it foreshadows a pattern with these authors in which they will cite an article as evidence for a particular claim, but closer examination will show that the article says something quite different.[2] That is, their level of scholarship is often quite sloppy and their arguments crumble under close scrutiny.

Dutton and colleagues’ argument assumes that religious belief, especially belief in a moral god, has been favoured by both individual and sexual selection.[3]  This also assumes that religiosity is associated with better physical and mental health and that the relationship between religiosity and health reflects common genetic factors. Additionally, they claim that religious people are preferred as mates because religiosity “functions as a marker of socially desirable traits” such as in-group commitment and trustworthiness. Unconventional beliefs including both paranormal beliefs and atheism are “deviations” that reflect deleterious mutations resulting in developmental instability. Therefore, in preindustrial times, “atheists and believers in the paranormal would, disproportionately, never have reached adulthood or never have been born, because these beliefs, though very different, are partly an expression of the breakdown of selection and thus of rising mutational load.” All these assumptions are flawed as I will show.

Firstly, the claim that atheists “would, disproportionately, never have reached adulthood or never have been born” in premodern times is rather ahistorical because disbelief in gods has been attested since ancient times in a variety of civilizations (e.g. Greece, Rome, India, and China). Another paper co-authored by Dutton (Dutton & Van der Linden, 2017) even points out that in Classical Greece and Rome it was widely remarked that “fools” tended to be religious while the “wise” were skeptics, noting that, “Euripides (440–406 BC) has his eponymous hero, Bellerophon, ask: ‘Doth someone say that there be gods above? There are not; no, there are not. Let no fool, led by the old false fable, thus deceive you.’”

Considering that throughout history there has actually been considerable variation in how religious people are, the evidence that religiosity is selected for is rather weak. For example, geneticist Razib Khan points out that if a continuous trait is under directional selection it will quickly become fixed. However, religiosity generally, and god belief specifically, are highly variable. Additionally, a study on twins and their relatives suggested that religious attitudes are influenced by a combination of genetics and shared environment (Eaves et al., 2012). On the other hand, this study showed that shared environment had almost no influence on personality traits, which were much more strongly related to genetic factors. This suggests that religiosity is not as strongly under genetic control as other individual characteristics such as personality traits.

Sexual selection occurs when an individual has characteristics that makes them more attractive to the opposite sex, which increase their reproductive fitness and consequently the likelihood of these characteristics being passed on to their offspring. Dutton et al. state that, “Religiosity is sexually selected for, because it functions as a marker of socially desirable traits (Dunkel et al., 2015a), for example that a person follows moral rules and has access to a useful network of co-religionists (Blume, 2009; Figueredo et al., 2006).” Note that of the three references cited, only one of them (Blume, 2009) actually makes an argument for sexual selection, based on findings that religious people tend to have more children, and that women tend to be more religious than men. The study by Dunkel et al. (2015) found that the general factor of personality, a hypothetical construct consisting of socially desirable traits, was higher in religious people compared to atheists and agnostics. However, this was based on a sample from one country, and did not test whether having these traits made their owners more attractive. International research has found that the relationship between personality traits and religiosity varies between countries, depending on how important religion is in a given place (Gebauer et al., 2014). Specifically, personal religiosity was associated with higher levels of agreeableness and conscientiousness in countries where religion was generally considered important, but in more secular countries this relationship tended to disappear. Hence, religiosity is not necessarily “a marker of socially desirable traits.” Furthermore, Figueredo et al. (2006) merely stated that religiosity is associated with sexually restricted behaviour and made no claims about sexual selection or even about religion providing “access to a useful network of co-religionists.” In any case, it is unlikely that sexual selection particularly favours religiosity. Sexually selected traits are those that are universally considered attractive. For example, high intelligence is considered desirable in a mate by both sexes around the world (Penke, Denissen, & Miller, 2007). On the other hand, most people prefer a mate who has about the same level of religiosity as themselves (Eaves et al., 2012). This is an example of assortative mating, that is, highly religious people prefer highly religious partners, whereas people who are less religious or unreligious tend to prefer their partners to be the same. If sexual selection applied, people would generally prefer highly religious mates even if they were not very religious themselves, in the same way that most people would prefer mates who are highly physically attractive even if they not that attractive themselves. Furthermore, contrary to Blume’s argument, having more children is not necessarily an indicator of sexual selection. Having more children may instead be an indicative of a reproductive strategy that occurs under harsh conditions involving high infant mortality. Under more benign conditions, people tend to have fewer children and invest in them more intensively. Considering that religion is more prevalent in countries with harsher living conditions (Diener, Tay, & Myers, 2011) this is a more likely explanation. Furthermore, if religiosity was subject to sexual selection it would be hard to explain why women tend to be more religious than men, as the latter would have more incentive to become religious to attract mates. On the other hand, it may be that men and women (on average) favour different reproductive strategies, where women tend to be more sexually restricted in their behaviour than men, and find religion more appealing because it usually encourages sexual restraint.

Although Dutton et al. argue that atheism is an “aberration,” others have argued that atheism might be adaptive in certain environments. For example, atheism tends to be associated with certain individual characteristics such as preference for logical reasoning and scepticism over intuition, along with less sociality and higher individualism (Caldwell-Harris, 2012). These might be adaptive in environments where there are advantages to independent learning and judgment rather than conformity and following tradition. According to Caldwell-Harris, atheism and non-belief may well be responses to evolutionary mechanisms. This is in line, with a view of evolution that considers that some traits are subject to “balancing selection,” that is, particular traits may be adaptive in some environments and not others. This is in contrast with a view that Dutton et al. seem to espouse in which evolution is considered unidirectional so that natural selection promotes only a narrow range of traits and anything else is a deviation from the norm. Furthermore, Caldwell-Harris states that, “Viewing atheism/non-belief as an individual-differences variable promotes tolerance for diverse views.” This is in marked contrast to Dutton et al. who regard non-normative and atypical attributes as markers of deleterious mutations, and therefore aberrations. Furthermore, in terms of individual adaptiveness, atheism is associated with more personally beneficial outcomes than uncommitted religious belief, indicating that religious belief in itself is not necessarily an unalloyed good (Galen, 2015).

In my next post, I will discuss the religiosity-health relationship, and show that it is more complicated than it first appears, and that, contrary to what Dutton et al. claim, there is no evidence that atheism is associated with poor health.

© Scott McGreal. Please do not reproduce without permission. Brief excerpts may be quoted provided a link to the original article is provided. 


[1] Astute readers might be surprised to hear that the Bible mentions autism. It doesn't. The passage in question, Mark 9:25, refers to a boy possessed by an unclean spirit since childhood, that has rendered him mute and prone to fits of falling on the ground while foaming at the mouth, gnashing his teeth, and becoming rigid. I think it is a huge interpretive stretch to refer to this as autism. Why not epilepsy? Additionally, the passage in question does not refer at all to "those inspired by other gods or by Satan," and does not even say anything about the boy's religious ideas. Dutton et al. seem to be clutching at straws to suggest that there is a religious precedent for linking atheism with autism.

[2] In a previous post, I examined an article co-authored by Edward Dutton (Dutton, van der Linden, & Lynn, 2016) in which the authors cite an article as evidence that Europeans have more body hair than Asians because the former have more Neanderthal DNA than the latter group. However, the article cited (Sankararaman et al., 2014) actually stated that Asians have a higher proportion of Neanderthal DNA and did not say anything about this being related to body hair at all. This suggests remarkable carelessness about checking the accuracy of their research.

[3] Additionally, the authors invoke the scientifically concept of group selection, i.e. that selection acts not just on individuals but on societies. The existence of group selection is highly controversial at best. 

Image Credit

Two Satyrs by Peter Paul Rubens


Blume, M. (2009). The Reproductive Benefits of Religious Affiliation. In E. Voland & W. Schiefenhövel (Eds.), The Biological Evolution of Religious Mind and Behavior (pp. 117-126). Berlin, Heidelberg: Springer Berlin Heidelberg.

Caldwell-Harris, C. L. (2012). Understanding atheism/non-belief as an expected individual-differences variable. Religion, Brain & Behavior, 2(1), 4-23. doi:10.1080/2153599x.2012.668395

Diener, E., Tay, L., & Myers, D. G. (2011). The religion paradox: If religion makes people happy, why are so many dropping out? Journal of Personality and Social Psychology, 101(6), 1278-1290. doi:10.1037/a0024402

Dunkel, C. S., Reeve, C. L., Woodley of Menie, M. A., & van der Linden, D. (2015). A comparative study of the general factor of personality in Jewish and non-Jewish populations. Personality and Individual Differences, 78(0), 63-67. doi:http://dx.doi.org/10.1016/j.paid.2015.01.014

Dutton, E., Madison, G., & Dunkel, C. (2017). The Mutant Says in His Heart, “There Is No God”: the Rejection of Collective Religiosity Centred Around the Worship of Moral Gods Is Associated with High Mutational Load. Evolutionary Psychological Science. doi:10.1007/s40806-017-0133-5

Dutton, E., & Van der Linden, D. (2017). Why is Intelligence Negatively Associated with Religiousness? Evolutionary Psychological Science, 3(4), 392-403. doi:10.1007/s40806-017-0101-0

Dutton, E., van der Linden, D., & Lynn, R. (2016). Population differences in androgen levels: A test of the Differential K theory. Personality and Individual Differences, 90, 289-295. doi:http://dx.doi.org/10.1016/j.paid.2015.11.030

Eaves, L., Heath, A., Martin, N., Maes, H., Neale, M., Kendler, K., . . . Corey, L. (2012). Comparing the biological and cultural inheritance of personality and social attitudes in the Virginia 30 000 study of twins and their relatives. Twin Research, 2(2), 62-80. doi:10.1375/twin.2.2.62

Figueredo, A. J., Vásquez, G., Brumbach, B. H., Schneider, S. M. R., Sefcek, J. A., Tal, I. R., . . . Jacobs, W. J. (2006). Consilience and Life History Theory: From genes to brain to reproductive strategy. Developmental Review, 26(2), 243-275. doi:http://dx.doi.org/10.1016/j.dr.2006.02.002

Galen, L. (2015). Atheism, wellbeing, and the wager: Why not believing in God (with others) is good for you. Science, Religion and Culture, 2(3), 54-69.

Gebauer, J. E., Bleidorn, W., Gosling, S. D., Rentfrow, P. J., Lamb, M. E., & Potter, J. (2014). Cross-Cultural Variations in Big Five Relationships With Religiosity: A Sociocultural Motives Perspective. Journal of Personality & Social Psychology.

Penke, L., Denissen, J. J. A., & Miller, G. F. (2007). The evolutionary genetics of personality. European Journal of Personality, 21(5), 549-587. doi:10.1002/per.629

Sankararaman, S., Mallick, S., Dannemann, M., Prufer, K., Kelso, J., Paabo, S., . . . Reich, D. (2014). The genomic landscape of Neanderthal ancestry in present-day humans. Nature, 507(7492), 354-357. doi:10.1038/nature12961 

Woodley, M. A., te Nijenhuis, J., & Murphy, R. (2013). Were the Victorians cleverer than us? The decline in general intelligence estimated from a meta-analysis of the slowing of simple reaction time. Intelligence, 41(6), 843-850. doi:http://dx.doi.org/10.1016/j.intell.2013.04.006

Be sure to read the following responses to this post by our bloggers:

More Posts