Saturday, July 25, 2020

Selection for slow life history?

One of several posters to promote family formation in the dwindling Parsi community (Jiyo Parsi)

In India, and overseas, the Parsis are renowned for their achievements, particularly in business but also in science, culture, and philanthropy. 

They are also known for something else: they’re dying out. From 114,000 in 1941, they were down to half that number by 2011. Today, more than 30% of Parsis don't marry, and an equal proportion are over 60 years old. Their fertility rate is 0.8—in other words, the average Parsi woman gives birth to less than one child during her lifetime. This existential crisis is worrying not only the Parsis but also the Indian government. In 2013, a program was set up to subsidize fertility treatments and promote family formation in the community (Dore 2017).

Extinction is irreversible. If the Parsis die out, the loss will be greatest in those things we don’t fully understand: the workings of the human mind. That is precisely where the Parsis have succeeded the most. How much of that success has been due to learning and how much to innate factors?

A few steps toward an answer were taken by Greg Cochran and Henry Harpending in their paper on Ashkenazi intelligence:

Since strong selection for IQ seems to be unusual in humans (few populations have had most members performing high-complexity jobs) and since near-total reproductive isolation is also unusual, the Ashkenazim may be the only extant human population with polymorphic frequencies of IQ-boosting disease mutations, although another place to look for a similar phenomenon is in India. In particular, the Parsi are an endogamous group with high levels of economic achievement, a history of long-distance trading, business and management, and who suffer high prevalences of Parkinson disease, breast cancer and tremor disorders, diseases not present in their neighbours. (Cochran et al. 2006)

Such disorders may be a side-effect of strong selection for intelligence over a short time in a small population. This was historically the case with Ashkenazi Jews. They are unusually prone to four neurological disorders: Tay-Sachs, Gaucher, Niemann-Pick, and mucolipidosis type IV. All four affect the brain by increasing the capacity of lysosomes to store sphingolipid compounds for axonal growth and branching. Furthermore, Tay-Sachs is caused in Ashkenazi Jews by two unrelated mutations and Gaucher disease by five. Random chance simply cannot explain why so many mutations exist in the same metabolic pathway and have reached such high frequencies.

Those mutations apparently spread through heterozygote advantage. Though harmful when two copies are inherited from both parents, they are beneficial when only one copy is inherited, as is more often the case. With a better supply of sphingolipids and no adverse effects, the brain can process information more efficiently.

Jared Diamond (1994) was the first to argue that chance cannot explain the high prevalence of so many lysosome storage disorders in a single population. He suggested the cause was selection for intelligence. His theory was then developed by Cochran et al. (2006). Other researchers have further confirmed Diamond’s theory by showing that Ashkenazim have high frequencies of alleles associated with educational attainment (Dunkel et al. 2019; Piffer 2019).

Frequent neurological/cerebral diseases among the Parsis

We see a similarly high prevalence of neurological or cerebral diseases among the Parsis. Parkinson’s disease is considerably more prevalent among them than among other Indians or even people of developed countries. Strokes are at least twice as common. Essential tremors are exceptionally frequent (Gourie-Devi 2014).

These diseases seem to have a genetic basis among the Parsis. A mitochondrial genome study found 420 unique genetic variants within that community, 178 of which are associated with Parkinson's disease. Others are linked to other neurodegenerative disorders, as well as colon, breast, and prostate cancer. A surprising number of these unique variants, 217, are linked to increased longevity. Finally, and perhaps curiously, some variants are linked to asthenozoospermia, i.e., reduced sperm motility (Patell 2020)

The above results are consistent with the findings of an earlier genetic study of the Parsis, specifically their autosomal, Y chromosome, and mitochondrial DNA. Signals of selection were strongest in SNPs associated with humoral immunity, cerebellar physiology, and neurological disorders like early epilepsy (Lopez et al. 2017).


The evidence is only suggestive, but it looks like the Parsis have undergone strong selection for intelligence over a relatively short time; hence, the high prevalence of neurological disorders.

In addition, this community seems to have adapted to its economic and social niche through a "slow life history" strategy. The Parsis are predisposed to live longer and thus learn more over a longer time. They may also be predisposed to longer birth intervals and higher parental investment in each child (K selection). Such a reproductive strategy is consistent with lower male fertility.

A slower life history, combined with higher intelligence, may have assisted the trading lifestyle of the Parsi community. Trade requires a high level of cognitive ability, particularly for literacy and numeracy, as well as lower time preference and a longer learning period. 

Ironically, low time preference may explain the demographic decline of the Parsis, and other people like them. If you’re future-oriented, you’re also keenly aware of future costs, particularly those of getting married and having a family. So you’ll postpone marriage and family formation until you’re financially ready. Unfortunately, that day may never come. Or it may come too late.

This problem was known to traditional societies, and there used to be social incentives to ensure that young people would marry before they got too old. Unfortunately, those incentives have disappeared in modern societies.

If you wait to check all the boxes, you may check into an old-age home … alone.


Cochran, G., J. Hardy, and H. Harpending. (2006). Natural history of Ashkenazi intelligence. Journal of Biosocial Science 38: 659-693

Diamond, J.M. (1994). Jewish Lysosomes. Nature 368: 291-292.

Dore, B. (2017). Glimmer of hope at last for India's vanishing Parsis. BBC News

Dunkel, C.S., Woodley of Menie, M.A., Pallesen, J., and Kirkegaard, E.O.W. (2019). Polygenic scores mediate the Jewish phenotypic advantage in educational attainment and cognitive ability compared with Catholics and Lutherans. Evolutionary Behavioral Sciences 13(4): 366-375.

Gourie-Devi M. (2014). Epidemiology of neurological disorders in India: review of background, prevalence and incidence of epilepsy, stroke, Parkinson's disease and tremors. Neurology India 62(6): 588-598.  

López, S., Thomas, M. G., van Dorp, L., Ansari-Pour, N., Stewart, S., Jones, A. L., Jelinek, E., Chikhi, L., Parfitt, T., Bradman, N., Weale, M. E., and Hellenthal, G. (2017). The genetic legacy of Zoroastrianism in Iran and India: insights into population structure, gene flow, and selection. American Journal of Human Genetics 101(3): 353-368.

Patell, V.M., N. Pasha, K. Krishnasamy, B. Mittal, C. Gopalakrishnan, R. Mugasimangalam, N. Sharma, A-K. Gupta, P. Bhote-Patell, S. Rao, R. Jain, and The Avestagenome Project. (2020). The First complete Zoroastrian-Parsi Mitochondria Reference Genome: Implications of 2 mitochondrial signatures in an endogamous, non-smoking population. bioRxiv preprint doi:

Piffer, D. (2019). Evidence for Recent Polygenic Selection on Educational Attainment and Intelligence Inferred from Gwas Hits: A Replication of Previous Findings Using Recent Data. Psych 1(1): 55-75.

Saturday, July 18, 2020

Religion as a force for natural selection

Buddhist temple, Singapore (Wikicommons – Cattan2011). In Buddhism, family is valued over self, but not strangers over family. Christianity goes farther: strangers are valued over self and family.

Originally, and even today in much of the world, social and economic activity is organized mostly by small groups of related individuals. As a result, a society cannot realize its full potential as it grows larger and encompasses people who are less related to each other. This is the “large society problem.” It has been alleviated by making kinship less important and, conversely, by encouraging forms of sociality that include everyone, and not just close kin. Northwest Europeans and East Asians have gone the farthest down that path.

In a large society, a social norm is less situational and more universal; it transcends the situation and its actors. The same applies to norm-breaking. To break a norm is to offend not only a particular person but also a general principle. In time, the principle may be sanctified, thus becoming even more inviolable. Offenders incur the wrath of the entire community, and not just the victim's family (Berman, 1974). Eventually, these reified norms coalesce into "religion."

Religion is key to the cultural environment of large societies and forces humans to adapt to it, just as they had to adapt to the natural environment of small societies. Unlike the natural environment, however, religion is a human creation. We have conceived an idealized vision of what humans should do, and this vision has rewarded those who do right and punished those who do wrong. In short, we have given religion the powers of natural selection. 

This point is made by the authors of a recent study:

Cultural evolution research on religion has highlighted the role religions play in enforcing large-scale cooperation [...]. Religious beliefs that expand what gods know and care about beyond local concerns and the local group, and increase gods' ability to punish rule breakers, may have contributed to sustaining cooperation at larger scales [...]. These beliefs create the perception that one's bad actions will be punished supernaturally, even if undetected by others, and can expand the circle of cooperation to anonymous strangers. Religions that lay out rules for cooperative behavior, and systems to enforce that cooperation, may create more stable and successful groups [...], perhaps increasing the ability of these groups and their religious beliefs to survive and spread. (Willard et al. 2020)

Thus, over time, there has thus been selection for individuals who respond to religion and willingly comply with its norms. Twin studies show that religiosity is 25 to 45% heritable. As is always the case, the non-heritable component includes everything else, like errors in understanding the question and collecting the data. We therefore have a substantial propensity to learn and obey social norms.

This propensity doesn't act alone. A growing child will develop it to a greater extent in a religious environment than in a non-religious one. The kind of religion also makes a difference.

Christianity encourages altruism

Gene-religion interaction has been shown at the gene DRD4. People are made more susceptible to social norms by the 2R allele or the 7R allele and less susceptible by the 4R allele (Sasaki et al., 2013). Furthermore, this susceptibility interacts with religion in the development of altruism. Using American and East Asian participants, Sasaki et al. (2013) found that 2R and 7R carriers were more altruistic than non-carriers if previously primed by the task of making a sentence from religious-sounding words. Priming had no effect on non-carriers.

These findings were partially replicated by Jiang et al. (2015). Among Singaporeans of Chinese descent, 2R carriers were more altruistic than non-carriers among male Christians, while being the same as non-carriers among women, Taoists, and Buddhists. The authors argue that men have more room for improvement because women start off caring more about others. The authors further suggest that Christianity better supports altruism by offering fellowship, comprehensible texts, and regular activities. Thus, if people are already more susceptible to social norms, they will become more altruistic if their environment is Christian. If, however, their environment is non-Christian, they will be no more altruistic than anyone else.

Because Christianity is better at developing this innate potential, and because this potential differs from one individual to another, altruism will vary much more among Christians than among non-Christians. A Christian society will have far more "super altruists" as a proportion of its population. This can be advantageous. Such people were once priests, pastors, nuns, philanthropists and the like. Before the rise of the welfare state, they provided valuable "collective goods," like education, moral guidance, and care for the sick and elderly. But what happens when a Christian society becomes post-Christian? For a while, there will still be lots of super altruists, but they will no longer be priests or pastors. They will become social justice warriors.

Taoism and Buddhism encourages veneration of ancestors and support for one’s in-group

Willard et al. (2020) studied how different religions affect prosocial behavior among Singaporeans of Chinese descent. Certain religious beliefs seemed to be key:

A moral afterlife

Buddhism, Taoism, and Christianity all share a belief that good deeds will be rewarded and evil deeds punished in an afterlife. When cued to think about the afterlife, participants from all three religions expressed a greater willingness to assist strangers.

Veneration of ancestors

Veneration of ancestors is an older stage of religious belief. It began in small societies but can be used to make large societies more workable, as long as everyone shares some common ancestry—or at least a belief in common ancestry.

Participants from all three religions believed that ancestors should be venerated, but this belief was supported much more by Buddhism and Taoism than by Christianity. When participants were cued to think about ancestor veneration, the Buddhists and Taoists expressed a greater willingness to assist their family and their in-group than did participants in the control condition. This effect was absent in the Christians: in fact, they became more willing to assist strangers:

When cued to think about moralized afterlife beliefs, Buddhists showed larger increases across almost all questions than Christians in what they believed was the normative amount to give. This effect is driven by Buddhists claiming weaker norms of giving than Christians in the neutral conditions on many of the questions. Though the moralized afterlife prime produced a greater change here, it brings both groups up to relatively similar normative amounts [...]. The ancestor condition generated weaker prosocial effects for Buddhists and Taoists on allocations to strangers than for Christians. In fact, the Christians showed a stronger effect here than anticipated, and the effects of the ancestor condition were stronger than those of the moralized afterlife condition on all questions. (Willard et al. 2020)


Religion encourages selflessness, but to varying degrees and to varying levels. In Buddhism and Taoism, family is valued over self, but not strangers over family. Christianity goes farther; strangers are valued over self and family. The Christian religion is thus more useful in large societies whose members don’t even pretend to share common ancestry.


Berman, H. J. (1974). The Interaction of Law and Religion. Nashville, Abingdon Press.

Bouchard, T.J. Jr., (2004). Genetic influence on human psychological traits: A survey. Current Directions in Psychological Science 13: 148-151.

Jiang, Y., R. Bachner-Melman, S.H. Chew, and R.P. Ebstein. (2015). Dopamine D4 receptor gene and religious affiliation correlate with dictator game altruism in males and not females: evidence for gender-sensitive gene × culture interaction. Frontiers in Neuroscience 24 September.

Lewis, G.J. and T.C. Bates. (2013). Common genetic influences underpin religiosity, community integration, and existential uncertainty. Journal of Research in Personality 47: 398-405.

Sasaki, J.Y., H.S. Kim, T. Mojaverian, L.D.S Kelley, I.Y. Park, and S. Janušonis. (2013). Religion priming differentially increases prosocial behavior among variants of the dopamine D4 receptor (DRD4) gene. Social Cognitive and Affective Neuroscience 8(2): 209-215.

Willard, A.K., A. Baimel, H. Turpin, J. Jong, and H. Whitehouse. (2020). Rewarding the good and punishing the bad: The role of karma and afterlife beliefs in shaping moral norms. Evolution and Human Behavior in press.

Saturday, July 11, 2020

Did women jumpstart recent cognitive evolution?

Scatter plots of frequencies of CASC5 variants by sex (Shi et al. 2017). During the last ice age, natural selection favored an increase in the gray matter of ancestral East Asians ... primarily in women.

Back in 2005 there was much interest in genes that regulate brain size, particularly in the ways they varied geographically within our species. It was found that two of these genes, Microcephalin and ASPM, continued to evolve as modern humans spread out of Africa. The latest variant of Microcephalin arose some 37,000 years ago in Eurasia and is still largely confined to the indigenous peoples of Eurasia and the Americas (Mekel-Bobrov et al. 2005). The latest variant of ASPM appeared even later, some 5,800 years in the Middle East (Evans et al. 2005).

Interest fell off when no association could be shown between the new variants and IQ or brain size (Mekel-Bobrov et al. 2007; Rushton et al. 2007; see also Frost 2020). Since then, we have learned that the new ASPM variant is associated with a larger cerebral cortex, and not a larger brain as a whole. Overall brain volume seems to be constrained in modern humans, perhaps by the breadth of a woman's pelvis during childbirth or simply by the high metabolic costs of brain tissue (Ali and Meier 2008; Frost 2020). As for the lack of an association with IQ, we now know that IQ correlates poorly or not at all with some cognitive abilities, like executive function and face recognition.

But what do the new variants actually do? Perhaps a specialized mental task.  It has been suggested that the new ASPM variant assists the brain in processing non-tonal language or alphabetical script (Dediu and Ladd 2007; Frost 2007).

CASC5, another gene for brain growth

Interest has since grown in another gene that regulates brain growth, CASC5. Like Microcephalin and ASPM, it has undergone recent evolution in the modern human lineage:

[...] the CASC5 gene contains mutations in modern humans, but not in Denisovans (Meyer et al. 2012) and this gene also shows distinct sequence divergence between modern humans and Neanderthals (Prufer et al. 2014). These data suggest that CASC5 is an important gene for human neurogenesis, and may harbor modern human specific mutations contributing to the recent evolutionary change of the human brain. (Shi et al. 2017)

Shi et al. (2017) found evidence of recent evolutionary change. Specifically, two nucleotides of CASC5 have been replaced with a new variant in all modern humans. Six other nucleotides have become polymorphic, with some people having the new variants and others not. These polymorphisms show regional differences:

- In four of the polymorphisms, the new variant has a much higher frequency in East Asians than in Europeans or Africans.
- In one polymorphism, it has a much higher frequency in Europeans than in the other two regional groups.
- The remaining polymorphism shows no differences in frequency between the three regional groups.

By and large, the new variants have been under strong positive selection, particularly among East Asians. When the authors examined the six polymorphisms, they found signals of selection for five of them in East Asians and for one in Europeans.

The new variants and brain characteristics

The authors then looked for correlations between the new variants and certain characteristics of the brain, specifically total brain volume, gray matter volume, and white matter volume. To this end, 267 healthy participants were recruited for brain imaging (Han Chinese, 178 females and 89 males, mean age 35.4 ± 12.5 years). All of them were free from mental disorders, drug abuse, alcohol dependence, and brain injury.

Gray matter was significantly larger in participants with the new variant than in those with the ancestral variant at five of the nucleotide sites, including the four polymorphic ones—the same ones that showed differences in variant frequency between East Asians and Europeans. When the authors examined the one polymorphism whose variants were equally common in East Asians, Europeans, and Africans, they found no brain differences between participants with the new variant and those with the ancestral one.

When the authors broke their data down by sex, they found that the new variants were significantly associated with a higher volume of gray matter only in women, not in men, although men seemed to trend in the same direction. The authors suggest that this effect would be significant in men if the number of male participants were larger. Probably. But it seems to me there would still be a sex difference, the number of participants being already large enough.

Ice age origin of the new variants

The authors say the new variants became prevalent "after modern humans migrated out of Africa less than 100,000 years ago." We can narrow down the time range further. The new variants are also present at high frequencies among the indigenous peoples of North and South America; therefore, they must have become prevalent before ancestral Amerindians crossed into North America some 12,000 years ago, apparently in a population that was ancestral both to Amerindians and to East Asians. That would be long before the time of recorded history and even before the Holocene, at a time when northern Eurasia was experiencing glacial conditions.

Did those conditions select for cognitive ability? Cold, seasonal environments did impose new cognitive demands on early modern humans, first by increasing their need to plan ahead over a yearly cycle and second by providing them with new tasks: garment making, needlework, weaving, leatherworking, and kiln operation. Women performed those tasks because the environment offered them few opportunities for food gathering—the usual female activity before the advent of farming. They thus moved into artisanal tasks that not only required greater cognitive ability but also offered much potential for further development. This was the "original industrial revolution" and it was led by women (Frost 2019a).

We can better understand this sexual division of labor by studying northern hunter-gatherers of recent times. According to a cross-cultural study, if women are less involved in food gathering, they specialize in activities unrelated to food procurement, i.e., house building, leatherworking, and burden carrying (Waguespack 2005). A study of two Inuit groups found the highest degree of technological complexity in garment making and shelter building, both of which are wholly or largely women's work (Oswalt 1976). Cold environments thus change the sexual division of labor among hunter-gatherers in a crucial way: while men continue to be food providers, women develop new technologies.

These findings may explain the recent evolution of CASC5: women were the focus of selection for cognitive ability during Ice Age times. But why was the selection stronger among ancestral East Asians than among ancestral Europeans? It looks like the climate at that time was more severe in northern Asia than in northern Europe. Europe benefited from the moderating influence of the Atlantic, which made for a milder and moister climate. Conditions were much colder and drier in northern Asia.

The evolution of human intelligence cannot be reduced to a single unified theory. Cold environments emancipated women from the mental straitjacket of food gathering, thus putting humans on the path to social complexity. That path, however, would take them to latitudes farther south in temperate and even tropical environments where they would be exposed to new cognitive demands. With the end of hunting, men moved not only into farming but also into the artisanal activities that women had developed. The same period saw a decline in brain volume that was greater in women than in men—an indication that cognitive demands were particularly high before the Holocene, and even more so for women (Frost 2019b).

The Holocene thus saw northern populations expand southward and eventually cover almost all of Eurasia, North Africa, Oceania, and the Americas. Those populations had a cognitive advantage that made them better able to exploit the social complexity emerging farther south. This point was made by Darwin's colleague Alfred Russel Wallace:

So when a glacial epoch comes on, some animals must acquire warmer fur, or a covering of fat, or else die of cold. Those best clothed by nature are, therefore, preserved by natural selection. Man, under the same circumstances, will make himself warmer clothing, and build better houses; and the necessity of doing this will react upon his mental organisation and social condition [...] a hardier, a more provident, and a more social race would be developed, than in those regions where the earth produces a perennial supply of vegetable food, and where neither foresight nor ingenuity are required to prepare for the rigours of winter. And is it not the fact that in all ages, and in every quarter of the globe, the inhabitants of temperate have been superior to those of tropical countries? All the great invasions and displacements of races have been from North to South, rather than the reverse.


Ali, F. and R. Meier. (2008). Positive selection in ASPM is correlated with cerebral cortex evolution across primates but not with whole brain size. Molecular Biology and Evolution 25(11): 2247-2250.

Dediu, D., and R. Ladd. (2007). Linguistic tone is related to the population frequency of the adaptive haplogroups of two brain size genes, ASPM and Microcephalin. Proceedings of the National Academy of Sciences 104(26): 10944-10949

Evans, P. D., Gilbert, S. L., Mekel-Bobrov, N., Vallender, E. J., Anderson, J. R., Vaez-Azizi, L. M., et al. (2005). Microcephalin, a gene regulating brain size, continues to evolve adaptively in humans. Science 309: 1717-1720.

Frost, P. (2007). The spread of alphabetical writing may have favored the latest variant of the ASPM gene. Medical Hypotheses 70: 17-20.

Frost, P. (2019a). The Original Industrial Revolution. Did Cold Winters Select for Cognitive Ability? Psych 1(1): 166-181

Frost, P. (2019b). Why did brain size decrease after the ice age? Evo and Proud, July 6

Frost, P. (2020). A second look at ASPM. Evo and Proud, April 14

Mekel-Bobrov, N., S.L. Gilbert, P.D. Evans, E.J. Vallender, J.R. Anderson, R.R. Hudson, S.A. Tishkoff and B.T. Lahn. (2005). Ongoing adaptive evolution of ASPM, a brain size determinant in Homo sapiens. Science 309: 1720-1722

Mekel-Bobrov, N., D. Posthuma, S.L. Gilbert, P. Lind, M.F. Gosso, et al. (2007). The ongoing adaptive evolution of ASPM and Microcephalin is not explained by increased intelligence. Human Molecular Genetics 16(6): 600-608.

Oswalt, W.H. (1976). An Anthropological Analysis of Food-Getting Technology, 1st ed.; John Wiley and Sons: New York, NY, USA.

Rushton, J.P., P.A. Vernon, and T.A. Bons. (2007). No evidence that polymorphisms of brain regulator genes Microcephalin and ASPM are associated with general mental ability, head circumference or altruism. Biology Letters-UK 3(2): 157-60.

Shi, L., Hu, E., Wang, Z. et al. (2017). Regional selection of the brain size regulating gene CASC5 provides new insight into human brain evolution. Human Genetics 136: 193-204.

Waguespack, N.M. (2005). The organization of male and female labor in foraging societies: Implications for early Paleoindian archaeology. American Anthropologist 107: 666-676.

Wallace, A.R. (1864). The Origin of Human Races and the Antiquity of Man deduced from the Theory of "Natural Selection." Journal of the Anthropological Society of London, 2, clviii-clxxxvii, Alfred Russel Wallace Classic Writings. Paper 6.

Saturday, July 4, 2020

Recent evolution of the British population

Frederick Morgan – Off for the Honeymoon (Wikicommons) Over the past 2,000 years, the British gene pool has shifted toward alleles that favor lighter hair, sunburn, and educational attainment. Was this because high-status men tended to mate with blonder, fairer women?

Have we evolved over the past two thousand years? Until recently, the answer was thought to be 'no.' Cultural evolution took over from genetic evolution around the time farming took over from hunting and gathering, some ten thousand years ago, thus putting our ancestors on a path to increasing social complexity: sedentary living, growth of towns and villages, formation of states, trade and specialization of labor, and so on. It was culture that changed during recorded history, not genes.

Well, things are not that simple. Genes and culture have coevolved with each other. Yes, culture has been changing rapidly over the past ten thousand years. But so have genes. During that time, our genetic evolution has been driven by adaptation not only to natural environments but also to cultural environments. Increasingly so. We live more and more in cultural environments of our making (Chen et al., 2016; Cochran and Harpending 2009; Hawks et al. 2007).

In what ways have we changed genetically during the past ten thousand years? In the ways we digest food. With the shift to dairy farming, and the resulting increase in milk consumption by adults, natural selection favored those who could digest milk sugar, an ability previously confined to infants.

We have also changed in the ways we think and behave. That kind of evolution is not difficult. A few point mutations may alter a behavior by changing its timing, its intensity, or its threshold of stimulation. Other alterations have been much more polygenic. Cognitive ability, for instance, seems to have increased through mutations at many genes, with each mutation causing only a tiny fraction of the increase.

Because recent evolutionary change has so often been polygenic, we need to examine it in relation to many genetic variants spread over the entire genome, i.e., by means of genome-wide association studies. Such studies can take many forms. A recent one, proposed by Stern et al. (2020), may be better than earlier versions, particularly in avoiding biases due to population structure and population stratification.

I nonetheless have a few reservation about this proposed method:

1. Population stratification can be a factor in evolutionary change. Let's take the work of Gregory Clark on the growth of the English middle class. He found it grew steadily from the twelfth century onward, its descendants not only growing in number but also replacing the lower classes through downward mobility. By the 1800s its lineages accounted for most of the English population. Parallel to that demographic growth, English society became more and more middle class in its values. "Thrift, prudence, negotiation, and hard work were becoming values for communities that previously had been spendthrift, impulsive, violent, and leisure loving" (Clark 2007, p. 166). Isn't that evolutionary change through population stratification? Or am I missing something?

2. The new method can reveal only evidence of directional selection. It thus fails to capture other interesting forms of selection, like diversifying selection.

How the British have evolved over the past 2,000 years

Stern et al. (2020) used their method to study how the British population has evolved over the past two thousand years. They found increases in the prevalence of lighter hair, in tanning and sunburn, in age at first birth, in bone mineral density, and in the risk of type 2 diabetes. They also found decreases in the risk of neuroticism and in the risk of high glycated hemoglobin levels.

Some of these changes correlate with each other. In such cases, we should step back and try to identify the common cause.

Lighter hair, more sunburn ... and higher educational attainment

Over the past 2,000 years, the British gene pool has shifted toward alleles that favor lighter hair, sunburn, and educational attainment. These changes in allele frequency correlate with each other, so what, exactly, was driving the overall change?

There is genetic linkage between light hair and pale skin, but it's weak. In fact, pale skin often coexists with dark hair. Moreover, we still have to explain the link to educational attainment. The common cause for all three changes may have been sexual selection mediated by social class. In other words, high-status men tended to mate with blonder, fairer women.

This form of sexual selection was observed in a Japanese study on social class and skin color. Upper-class men were shown to be fairer-skinned than lower-class men, even when the latter were factory workers and not farmers and even though the measurements were taken on unexposed skin. Wealthier men have a wider range of prospective brides and can thus choose the fairest women, for "skin color has long been regarded, by the Japanese, as one of the criteria for evaluating physical attractiveness, especially in young females" (Hulse 1967). Similarly, in India "[w]ealthy landowning families often have a tradition of seeking light-skinned brides among poorer members of their subcaste. It is very common to find a high concentration of lighter-skinned people among established land-owning families" (Béteille 1967).

Darwin discussed this sexual selection with reference to English social classes:

Many persons are convinced, as it appears to me with justice, that our aristocracy, including under this term all wealthy families in which primogeniture has long prevailed, from having chosen during many generations from all classes the more beautiful women as their wives, have become handsomer, according to the European standard, than the middle classes; yet the middle classes are placed under equally favorable conditions of life for the perfect development of the body. (Darwin 1936[1888], p. 892)

Until the 20th century, higher social status meant higher fertility (Clark 2007). Thus, the physical and mental characteristics of the upper and middle classes tended to displace those of the lower class.

Higher risk of Type 2 diabetes and glycated hemoglobin

Why would natural selection favor type 2 diabetes? Isn't diabetes harmful? It is, in a modern environment that lets you ingest calories almost without limit. That wasn't the case in Britain for most of the past two thousand years. During that time, food was scarce for most people, and natural selection favored the ability to get as many calories as possible out of our food.

Less neuroticism

This evolutionary change may be related to the demographic success of the middle class and associated mental and behavioral traits, particularly lower time preference and higher future orientation. The nascent English middle class valued being “calm, cool, and collected,” as opposed to reacting emotionally to negative outcomes.


Béteille, A. (1967). Race and descent as social categories in India. Daedalus 96(2): 444-463.

Chen, C., R.K. Moyzis, X. Lei, C. Chen, and Q. Dong. (2016). The encultured genome: Molecular evidence for recent divergent evolution in human neurotransmitter genes. In: J.Y. Chiao, S.-C. Li, R. Seligman, and R. Turner, Eds, The Oxford handbook of cultural neuroscience. New York, NY: Oxford University Press, 315-336.

Clark, G. (2007). A Farewell to Alms. A Brief Economic History of the World, 1st ed.; Princeton University Press: Princeton.

Cochran, G., and H. Harpending. (2009). The 10,000 Year Explosion: How Civilization Accelerated Human Evolution. Basic Books.

Darwin, C. (1936 [1888]). The Descent of Man and Selection in relation to Sex. reprint of 2nd edition, The Modern Library, New York: Random House.

Hawks, J., E.T. Wang, G.M. Cochran, H.C. Harpending, and R.K. Moyzis. (2007). Recent acceleration of human adaptive evolution. Proceedings of the National Academy of Sciences, 104(52), 20753-20758.

Hulse, F.S. (1967). Selection for skin color among the Japanese. American Journal of Physical Anthropology 27(2): 143-156.

Hysi, P.G., A.M. Valdes, F. Liu, N.A. Furlotte, D.M. Evans, V. Bataille, et al. (2018). Genome-wide association meta-analysis of individuals of European ancestry identifies new loci explaining a substantial fraction of hair color variation and heritability. Nature Genetics 50(5): 652-656.

Morgan, M.D., E. Pairo-Castineira, K. Rawlik, O. Canela-Xandri, J. Rees, D. Sims, A. Tenesa, and I.J. Jackson. (2018). Genome-wide study of hair colour in UK Biobank explains most of the SNP heritability. Nature Communications 9: 5271

Neel, J. V. (1962). Diabetes mellitus: a 'thrifty' genotype rendered detrimental by 'progress'? American Journal of Human Genetics 14: 353-362.

Stern, A.J., L. Speidel, N.A. Zaitlen, and R. Nielsen. (2020). Disentangling selection on genetically correlated polygenic traits using whole-genome genealogies
bioRxiv 2020.05.07.083402