Saturday, March 31, 2012

Dark coloration and male aggressiveness: Is there a link?

Red-winged blackbird (source). Is dark coloration directly linked to male aggressiveness?

For the past thirty years, psychologist Philippe Rushton has been using life history theory to explain human differences in many areas: IQ, sexual development, parental investment, mating system, time orientation, etc. Initially, he saw skin color as being incidental. In recent years, however, he has come around to the view that differences in skin color directly mediate these differences in life history:

The hypothesis that skin color is a genetic correlate of IQ was endorsed by Jensen (2006) who suggested that pleiotropy (genes having more than one effect) may underlie the relationship. Skin color became of greater theoretical interest after Ducrest, Keller, and Roulin (2008) reviewed the literature and reported that in 20 wild vertebrate species, darker individuals were more aggressive, sexually active, and resistant to stress than lighter individuals. Studied were three mammal species (African lion, soay sheep, and white-tailed deer), four fish species (mosquito fish, guppy, green swordtail, and Arctic char), four reptile species (asp viper, adder, fence lizard, and spiny lizard), one amphibian species (spadefoot toad) and 36 bird species. Darker individuals also tended to have a larger body mass and greater energy and physical activity such as grooming. Ducrest et al. (2008) confirmed the naturalistic observations using experimental studies such as the administration of melanocortins and the fostering of infants to non-biological parents. For example, the fostering studies found darker maned male lions are more aggressive and sexually more active, and darker barn owls mount stronger immune responses when their biological parents are darker, even though they had been raised by lighter foster parents. (Templer & Rushton, 2011)

This hypothesis is not a perfect fit for humans, as Rushton himself noted in an earlier study on human populations and behavioral differences: “We found the relationship between crime and IQ held (r=−0.35; Pb0.01), although the one between crime and skin color did not (ns). […] the East Asian countries had very low rates of crime but not the lightest skin color” (Rushton & Templer, 2009).

In humans, many things vary with latitude, and not just skin color. One of them is the means of sustenance, which in turn impacts parental investment, mating system, and male-male rivalry. In the tropics, for example, we see the following cascade of effects:

1. Year-round tropical agriculture enables women to provide for themselves and their children with little male assistance.

2. This greater female reproductive autonomy lowers the cost of polygyny for men. More men can afford to have second wives.

3. Because more men are competing for fewer available women, there is stronger selection for males with higher testosterone levels, more robust body build, and greater ability to fight off rivals.

It is hard to see how this cascade of effects could apply to nonhumans. Guppies, for instance, have not discovered farming. Nor is their coloration a climatic adaptation. It serves as a visual signal and not as a UV shield, as is the case with most of the other animals that Templer and Rushton cite.

Clearly, a cross-species correlation does exist between darker coloration and male aggressiveness. But there is an alternate explanation. Whatever the species, individuals are usually born with little or no pigment. Lighter coloration thus becomes associated with vulnerability and a need for parental care and protection. In contradistinction to this sign stimulus, adult males tend to evolve a darker coloration, especially in a context of intense male-male rivalry. This tendency was noted by Guthrie (1970):

Light skin seems to be more paedomorphic, since individuals of all races tend to darken with age. Even in the gorilla, the most heavily pigmented of the hominoids, the young are born with very little pigment. […] Thus, a lighter colored individual may present a less threatening, more juvenile image.

Our own species likewise shows analogous age and sex differences in skin pigmentation, as seen in a strong cross-cultural trend to associate darker skin with men and lighter skin with infants and women (Frost, 2011).

This mental association may influence human mate choice, especially under conditions of intense mate competition. Given that mate competition varies latitudinally among human populations, as described above, could it be that latitudinal differences in skin color result not only from selection for UV protection but also from sexual selection?

In humans, the polygyny rate does in fact significantly correlate with darkness of skin color (Manning et al., 2004). This correlation seems to hold up even if we control for latitude. In sub-Saharan Africa, high-polygyny agriculturalists are visibly darker than low-polygyny hunter-gatherers, i.e., Khoisans, pygmies, although both are equally indigenous (Bourguignon & Greenbaum, 1973, pp. 171-175; Cavalli-Sforza, 1986a; Cavalli-Sforza, 1986b; Weiner et al, 1964). Because year-round agriculture makes women more self-sufficient and polygyny less costly, fewer women remain unmated and men are less able to translate their mate-choice criteria into actual mate choice. Such criteria include a preference, widely attested in the African ethnographic literature, for so-called 'red' or 'yellow' women (van den Berghe & Frost, 1986). Less mate choice means weaker sexual selection for light skin in women and, hence, less counterbalancing of natural selection for dark skin in either sex to protect against sunburn and skin cancer (Aoki, 2002; Frost, 1994; Frost, 2007; Frost, 2008).

A higher polygyny rate might also, correspondingly, lead to stronger sexual selection for darker-skinned men, either because women tend to prefer them, or defer to them, or because such men can more easily intimidate rivals in a context of intense mate competition.


Aoki, K. (2002). Sexual selection as a cause of human skin colour variation: Darwin’s hypothesis revisited. Annals of Human Biology, 29, 589-608.

Bourguignon, E. & L.S. Greenbaum. (1973). Diversity and Homogeneity in World Societies, HRAF Press.

Cavalli-Sforza, L.L. (1986a). Demographic data. In Cavalli-Sforza, L.L. (ed.). African Pygmies, pp. 23-44. Academic Press.

Cavalli-Sforza, L.L. (1986b). Anthropometric data. In Cavalli-Sforza, L.L. (ed.). African Pygmies, pp. 81-93. Academic Press.

Ducrest, A., Keller, L., & Roulin, A. (2008). Pleiotropy in the melanocortin system, coloration and behavioural syndromes. Trends in Ecology and Evolution, 23, 502-510.

Frost, P. (2011). Hue and luminosity of human skin: a visual cue for gender recognition and other mental tasks, Human Ethology Bulletin, 26(2), 25-34.

Frost, P. (2008). Sexual selection and human geographic variation, Special Issue: Proceedings of the 2nd Annual Meeting of the NorthEastern Evolutionary Psychology Society. Journal of Social, Evolutionary, and Cultural Psychology, 2(4),169-191.

Frost, P. (2007). Comment on Human skin-color sexual dimorphism: A test of the sexual selection hypothesis, American Journal of Physical Anthropology, 133, 779-781.

Frost, P. (1994). Geographic distribution of human skin colour: a selective compromise between natural selection and sexual selection? Human Evolution, 9, 141-153.

Guthrie, R.D. (1970). Evolution of human threat display organs, Evolutionary Biology, 4, 257-302

Jensen, A. R. (2006). Comments on correlation of IQ with skin color and geographic–demographic variables. Intelligence, 34, 128−131.

Manning, J.T., P.E. Bundred, & F.M. Mather. (2004). Second to fourth digit ratio, sexual selection, and skin colour. Evolution and Human Behavior, 25, 38-50.

Rushton, J. P., & Templer, D. I. (2009). National differences in intelligence, crime, income, and skin color. Intelligence, 37, 341-346.

Templer, D. I., & Rushton, J. P. (2011). IQ, skin color, crime, HIV/AIDS, and income in 50 U.S. states. Intelligence, 39(20), 437-442.

van den Berghe, P.L., & P. Frost. (1986). Skin color preference, sexual dimorphism and sexual selection: A case of gene-culture co-evolution? Ethnic and Racial Studies, 9, 87-113.

Weiner, J.S., G.A. Harrison, R. Singer, R. Harris, & W. Jopp. (1964). Skin colour in southern Africa. Human Biology, 36, 294-307.

Saturday, March 24, 2012

The 'monkey people' we once knew

Remains of archaic hominins from southwest China (Curnoe et al, 2012). They were around when villages and towns were arising in the Middle East.

Recent findings have confirmed the ‘Out of Africa’ model of human origins, but only in part. The model diverges from actual prehistory on two main points. One is that modern humans picked up archaic admixture as they spread out of Africa and into Eurasia. Thus, modern Eurasians have 1-4% Neanderthal admixture, and Melanesians an additional 4-6% from the mysterious Denisovans (Reich et al, 2011). As for modern sub-Saharan Africans, they seem to be the most admixed of all. About 2% of their gene pool comes from a population close to Homo erectus and a further 13% from a population probably related to the Skhul-Qafzeh hominins (Hammer et al., 2011; Watson et al., 1997).

And the second point? It appears that modern humans didn’t immediately replace archaic hominins, at least not everywhere. Some of the latter held out in different places of refuge until the Holocene, and perhaps even later. At a time when villages and towns were arising in the Middle East, archaic hominins continued to hold out in western and southern Africa (Harvati et al., 2011; Stringer, 2011).

Now, we have evidence of another refuge area. Southwest China has yielded archaic cranial remains that date to ~14.3-11.5 thousand years ago. The remains actually show a mixture of archaic and modern traits, reminiscent of the Skhul-Qafzeh hominins of the Levant (120,000 – 80,000 BP) and other ‘almost moderns’ from North Africa.

Who were they? The authors offer two explanations:

Our analysis suggests two plausible explanations for the morphology sampled at Longlin Cave and Maludong. First, it may represent a late-surviving archaic population, perhaps paralleling the situation seen in North Africa as indicated by remains from Dar-es-Soltane and Temara, and maybe also in southern China at Zhirendong. Alternatively, East Asia may have been colonised during multiple waves during the Pleistocene, with the Longlin-Maludong morphology possibly reflecting deep population substructure in Africa prior to modern humans dispersing into Eurasia (Curnoe et al., 2012).

The two explanations aren’t that far apart. Different authors have alternately described the Skhul-Qafzeh remains as either late archaic or early modern. In the case of the Chinese remains, an obvious candidate would be the Denisovans, an archaic population that inhabited East Asia around the time that Neanderthals inhabited Europe and central Asia. But the authors evoke this possibility only in passing:

DNA extracted from a >50 ka hominin fossil from Denisova Cave in Central Asia belonging within the Neandertal lineage shares features exclusively with Aboriginal Southeast Asians and Australasians. This has been interpreted as: 1) evidence for interbreeding between the ‘Denisovans’ and the earliest modern humans to colonise the region; and 2) implying occupation of Southeast Asia by this archaic population during the Upper Pleistocene.

In fact, we have good evidence that Denisovans were present in Southeast Asia. Reich et al. (2011) found Denisovan admixture in some but not all of the oldest indigenous peoples of Southeast Asia. Such admixture was present in Aboriginal Australians, New Guineans, and a Negrito people from the Philippines. It was absent, however, in Negrito groups farther west. The authors thus concluded:

Our finding that descendants of the earliest inhabitants of Southeast Asia do not all harbor Denisova admixture is inconsistent with a history in which the Denisova interbreeding occurred in mainland Asia and then spread over Southeast Asia, leading to all its earliest modern human inhabitants. Instead, the data can be most parsimoniously explained if the Denisova gene flow occurred in Southeast Asia itself. Thus, archaic Denisovans must have lived over an extraordinarily broad geographic and ecological range, from Siberia to tropical Asia.

If the Denisovans lingered on into historic times, the same might be true for other archaic groups, like the Neanderthals in Europe. Perhaps those stories about hairy wild men were not pure imagination.

Deusen (2001) mentions that the Tungus peoples of far eastern Siberia remember the existence of ‘monkey people’ in their region. One folk-tale describes how these monkey people abducted a man:

So the older sister took the shaman's drum. She started to sing and then said, "Brother, when you go hunting in the taiga tomorrow, you're going to meet two people. Check out their breasts, and then marry them."

The next day, he woke up and set out to go hunting. He walked and walked and came to a hill, a mountain. There were big rocks. He looked up, and then went on. Suddenly he saw two people sitting there. He approached and at that time the ties on his skis broke.

He came up to those people and felt their breasts and they were women. And they took him along with them.

At home time went by. A day passed and another, and still he was gone. Many days went by. And then the younger sister said, "Sister, you made this happen. Now you bring him back. Those two monkeys in the mountain came and took him away and now they are keeping him in the mountains, sucking his blood. He's become just skin and bones."

… So the younger sister sang and drummed, flying to her spirits, but she couldn't get there. She tried a second time and still didn't have the strength. The third time she gathered all her strength and flew to those rocks. She took her brother and dragged him out of there. He flew, looking thin as a shirt. They got him back and healed him. And that's how the younger sister brought her brother back from those monkeys.

… So that's it about the monkeys. They lived in the rocks and when they rolled back and forth, they called, "Tsyoo, tsyoo, papandasyoo!!"
(Deusen 2001:126-128)


Abi-Rached L, Jobin MJ, Kulkarni S, McWhinnie A, Dalva K, et al. (2011). The shaping of modern human immune systems by multiregional admixture with archaic humans. Science, 334, 89–94.

Curnoe D, Xueping J, Herries AIR, Kanning B, Taçon PSC, et al. (2012). Human Remains from the Pleistocene-Holocene Transition of Southwest China Suggest a Complex Evolutionary History for East Asians. PLoS ONE 7(3): e31918. doi:10.1371/journal.pone.0031918

Deusen, K.V. (2001). The Flying Tiger. Women Shamans and Storytellers of the Amur. Montreal: McGill-Queen's University Press.

Hammer, M.F., A.E. Woerner, F.L. Mendez, J.C. Watkins, and J.D. Wall. (2011). Genetic evidence for archaic admixture in Africa, Proceedings of the National Academy of Science (USA), 108, 15123-15128,

Harvati, K., C. Stringer, R. Grün, M. Aubert, P. Allsworth-Jones, C.A. Folorunso. (2011). The Later Stone Age Calvaria from Iwo Eleru, Nigeria: Morphology and Chronology. PLoS ONE 6(9): e24024. doi:10.1371/journal.pone.0024024

Reich D, Green RE, Kircher M, Krause J, Patterson N, et al. (2010). Genetic history of an archaic hominin group from Denisova Cave in Siberia. Nature, 468, 1053–1060.

Reich D, Patterson N, Kircher M, Delfin F, Nandineni MR, et al. (2011). Denisova Admixture and the First Modern Human Dispersals into Southeast Asia and Oceania. Am J Hum Genet, 89, 516–528.

Stringer, C. (2011). The chronological and evolutionary position of the Broken Hill cranium. American Journal of Physical Anthropology, 144(supp. 52), 287

Watson, E., P. Forster, M. Richards, and H-J. Bandelt. (1997). Mitochondrial footprints of human expansions in Africa, American Journal of Human Genetics, 61, 691-704. 0024024

Saturday, March 17, 2012

What makes hair color "hot"?

The ‘hot’ hair color this year (source). While there seems to be a general trend to prefer average physical characteristics, this doesn’t seem to apply to hair color. People seek colors that are uncommon or even unnatural.

Europeans have departed from the species norm of black hair and brown eyes by evolving a wide range of bright hair and eye colors. What is the selective advantage of these new hues? Or are they merely a side effect of something else?

I’ve argued that these new colors were selected for … their newness and colorfulness. To be precise, their selective advantage lay in their novelty and brightness. These eye-catching qualities enabled women to improve their mating prospects at a time when the operational sex ratio was skewed toward a female surplus and a male shortage.

This is the logic of advertising. Visual merchandising matters most in saturated, highly competitive markets that offer too many interesting choices (Lea-Greenwood, 1998; Oakley, 1990). Such a context rewards products that stand out because of their bright or novel look, as seen in colors for home interiors. This market has grown more competitive over the past half-century, and the novelty factor has correspondingly grown more important: preference for one paint color rises until satiated, then falls and yields to preference for another (Stansfield & Whitfield, 2005).

In the natural world, and under conditions of intense sexual selection, this same logic leads to a color polymorphism. A new color appears through mutation and spreads through the population until it is as common as the established color. This equilibrium will then last until another color variant appears. The total number of colors thus grows over time.

This aspect of sexual selection can be demonstrated under controlled conditions. In an American study, male participants were shown pictures of attractive brunettes and blondes and asked to choose, for each series, the woman they would most like to marry. One series had equal numbers of brunettes and blondes, a second series 1 brunette for every 5 blondes, and a third series 1 brunette for every 11 blondes. Result: the scarcer the brunettes were in a series, the likelier any one of them would be chosen (Thelen, 1983).

The same trend appears in popular culture. On American TV programs, women are four and a half times more likely than men to have red or auburn hair and five times more likely than men to have blonde hair. Conversely, men are four times more likely than women to have gray hair and 40% more likely than women to have black hair (Davis, 1990). A similar trend has been observed on Turkish TV programs:

Women were more likely than men to have red (5.3%) or blonde (15.6%) hair. In fact, no primary male characters in this sample had red or blonde hair at all, but female characters did. (Ikizler, 2007, p. 39)

This sex difference undoubtedly reflects the use of artificial hair coloring, although female hair color is naturally more diverse than male hair color (a legacy of the female-directed nature of sexual selection in Europe). Interestingly, women are using hair dyes to give themselves less typical hues, rather than more typical ones. Such colors may be uncommon but naturally occurring, such as platinum blonde and red. Or they may not exist at all in nature, such as green, purple, and magenta.

This year, the leading hair colors are forecasted to be “red, burgundy, strawberry blonde, copper brown and auburn shades.” Among celebrities, the hottest colors will include “bright reds, vibrant blues and pastel pinks” (Fall Hair Color Trends 2012).


Davis, M. D. (1990). Portrayals of women in prime-time network television: Some demographic characteristics. Sex Roles, 23(5/6), 325-332.

Fall Hair Color Trends 2012

Ikizler, A.S. (2007). Gender role representations in Turkish television programs, Submitted as a St. Mary's Project in Partial Fulfillment of the Graduation Requirements, St. Mary's College of Maryland for the Degree of Bachelor of Arts in Psychology

Lea-Greenwood G. (1998). Visual merchandising: a neglected area in UK fashion marketing? International Journal of Retail & Distribution Management, 26, 324-329.

Oakley M. (ed.) (1990). Design management. A handbook of issues and methods. Oxford: Basil Blackwell.

Stansfield J., & T.W.A. Whitfield. (2005). Can future colour trends be predicted on the basis of past colour trends? An empirical investigation. Color Research and Application, 30(3), 235-242.

Thelen, T.H. (1983). Minority type human mate preference. Social Biology, 30, 162-180.

Saturday, March 10, 2012

Why are redheads less common than blondes?

Gilbert Blythe tormenting Anne Shirley (Anne of Green Gables)

Male preference for female hair color seems to be frequency-dependent. The less common a hair color becomes, the more it is preferred. When male subjects are presented with a series of photos showing blondes and brunettes, preference for any one brunette is inversely proportional to the number of brunettes in the series (Thelen, 1983). This frequency dependence may explain why blonde hair is less strongly preferred in England than in France.

England is, however, somewhat fairer generally than most parts of Europe; so that, while it may be said that a very beautiful woman in France or Spain may belong to the blondest section of the community, a very beautiful woman in England, even though of the same degree of blondness as her Continental sister, will not belong to the extremely blonde section of the English community. (Ellis, 2007[1905], p. 160)

Hair color is normally a minor factor when men select mates. It becomes a major factor only when the level of sexual selection is intense, like the situation of a movie producer who has to choose one actress from a number of excellent candidates. Under such conditions, relatively unimportant factors can make a big difference, especially those that can attract and retain attention.

I’ve argued that such conditions once prevailed among ancestral Europeans, specifically European women (Frost, 1994, 2006, 2008). One result was an ever broader range of eye and hair colors. Whenever a new color appeared through mutation, it would be favored by sexual selection until it had become as frequent as the other colors. Although the pressure of sexual selection was on women, these changes in physical appearance spilled over on to men, thereby creating a new phenotype in both sexes.

But we still see some evidence that this selection pressure had targeted women. A twin study has shown that hair is, on average, lighter-colored in women than in men, with red hair being especially more frequent in females. Women also show greater variation in hair color (Shekar et al., 2008).

Do redheads have less fun?

Red is the least common of these new hair colors. It also seems to have a lower tipping point where preference gives way to non-preference—if not dislike. Indeed, in comparison to blondes, redheads become more easily the butt of ridicule and prejudice.

How come? Several reasons appear in the literature. One is that Judas, the betrayer of Christ, was red-haired. Yet this association of ideas may have originated in a much older one that the Church later recycled for its own purposes:

There can be little doubt that this tradition is simply the application of the old belief—much older than Judas Iscariot—that red-haired men are treacherous and dangerous, to the Arch-traitor, sometime during the early Middle Ages, when the popular imagination was busy making up biographies and biographical details for the saints and martyrs of the Church. (Baum, 1922)

This older folk belief had its basis in the idea that redheads are hotheaded and not to be trusted:

We must look further for an explanation of Judas color, and specially in the ill-omen of red hair. This itself took its beginning no doubt, like so much else of popular tradition, in the shrewd observation of natural phenomena. The common German proverb, "Roter Bart, untreue Art," represents a condensed popular judgment. Even to-day a red-haired man is assumed to be hot-headed and quick-tempered, and so not quite to be counted on. (Baum, 1922)

Is this true? Perhaps. Of all the different hair colors, red is the one that seems to have the most pleiotropic effects, e.g., skin type, freckling, beard color, etc. (Flanagan et al., 2000). There are also physiological consequences. Redheads seem to be more sensitive to painkillers, the effect being stronger in women than in men (Mogil et al., 2003). This increased sensitivity was noted over sixty years ago:

Dr. Paul M. Wood, Secretary-Treasurer of the American Board of Anaesthesiology, writes, " ... in twenty-two years of personal experience I have discovered that many difficulties have occurred with persons who have red hair. They do not seem to take anaesthetics in the same way that others do. Most of them are much more sensitive to the anaesthetic."

Dr. David M. Levy states that red headed persons in general go under anaesthetic more readily than do persons of other pigmentations.

Dr. Marco Nunez confirms these observations that red heads are affected more readily than are other persons.
(Keeler, 1947)

We may have here one reason why red hair never became as common as other hair colors. There were too many side effects, and natural selection has probably not had enough time to iron them all out.

But culture has probably been just as important in determining the frequency of red hair. Whereas attitudes to red hair tend to be negative in Latin cultures, they are generally positive in Celtic cultures, as Curry (1916, pp. 18-19) notes in his study of medieval literature: “It is worthy of note, however, that in the Old Irish and Welsh red hair seems to be appreciated, tho to a less extent than the blonde.”

The last point is interesting because the frequency of red hair is much higher in the Celtic regions of Europe (Wikipedia). Has this prevalence been maintained at a higher level by a more “redhead-friendly” environment?


Baum, P.F. (1922). Judas's red hair, The Journal of English and Germanic Philology, 21, 520-529.

Curry, W.C. (1916). The Middle English Ideal of Personal Beauty, as found in the Metrical Romances, Chronicles, and Legends of the XIII, XIV, and XV Centuries. Baltimore: J.H. Furst Co.

Ellis, H. (2007 [1905]). Studies in the Psychology of Sex, Volume 4 (of 6), Teddington, Echo Library.

Flanagan, N., E. Healey, A. Ray, S. Philips, C. Todd, I.J. Jackson, M.A. Birch-Machin, & J.L. Rees. (2000). Pleiotropic effects of the melanocortin 1 receptor (MC1R) gene on human pigmentation, Human Molecular Genetics, 9, 2531-2537.

Frost, P. (2008). Sexual selection and human geographic variation, Special Issue: Proceedings of the 2nd Annual Meeting of the NorthEastern Evolutionary Psychology Society. Journal of Social, Evolutionary, and Cultural Psychology, 2(4),169-191.

Frost, P. (2006). European hair and eye color - A case of frequency-dependent sexual selection? Evolution and Human Behavior, 27, 85-103.

Frost, P. (1994).
Geographic distribution of human skin colour: A selective compromise between natural selection and sexual selection? Human Evolution, 9, 141-153.

Keeler, C.E. (1947). Coat color, physique, and temperament. Materials for the Synthesis of Hereditary Behavior Trends in the Lower Mammals and Man, The Journal of Heredity, 38, 271-277.

Mogil, J.S., S.G. Wilson, E.J. Chesler, A.L. Rankin, K.V. S. Nemmani, W.R. Lariviere, M.K. Groce, M.R. Wallace, L. Kaplan, R. Staud, T.J. Ness, T.L. Glover, M. Stankova, A. Mayorov, V.J. Hruby, J.E. Grisel, & R.B. Fillingim. (2003). The melanocortin-1 receptor gene mediates female-specific mechanisms of analgesia in mice and humans, PNAS, 100, 4867–4872.

Shekar, S.N., D.L. Duffy, T. Frudakis, G.W. Montgomery, M.R. James, R.A. Sturm, & N.G. Martin. (2008). Spectrophotometric methods for quantifying pigmentation in human hair—Influence of MC1R genotype and environment. Photochemistry and Photobiology, 84, 719–726.

Thelen, T.H. (1983). Minority type human mate preference. Social Biology, 30, 162-180.

Wikipedia. Red hair

Saturday, March 3, 2012

Puzzle of European hair and eye color

Hair and eye color diversity is unusual in two ways. It’s confined to Europeans. And it seems to be linked to prenatal feminization.
Europeans are distinguished from other humans by a diverse palette of eye and hair colors. I’ve argued that these color traits arose from intense sexual selection of women in ancestral European environments (Frost, 1994, 2006, 2008). Until 10,000 years ago, Europe had vast expanses of continental tundra—an environment where male hunters provided almost all the food and where long-distance hunting caused more deaths among young men than among young women. With fewer men overall, and fewer who could shoulder the costs of polygyny, the mate market had a surplus of unmated women. The pressure of sexual selection was therefore on the female sex. They were the ones who had to compete for mates.

East of Europe, the tundra zone swung north into colder and more arid territory. Humans were present there too, but in fewer numbers and less continuously. Only in Europe did this zone have a continuous human presence throughout the last ice age, and only in Europe could humans pass on and steadily accumulate the genetic legacy of intense female-directed sexual selection.

But if sexual selection had created European hair and eye colors, wouldn’t we see some sex linkage? Wouldn’t these color traits be expressed more in women than in men?

When I initially wrote on this subject, I had some evidence of sex linkage. Blond hair darkens with age more slowly in women than in men (Olivier, 1960, p. 74). I had also read an unpublished study that found higher digit ratios, and thus higher prenatal exposure to estrogen, in individuals with blond hair or non-brown eyes (Mather et al., unpublished).

But that was it. So I fell back on the explanation that sex linkage would have taken too long to evolve, especially because men incurred little cost in having unnecessarily showy eye and hair colors.

I’ve since come across more evidence of sex linkage. First, a twin study has shown that hair is, on average, lighter-colored in women than in men, with red hair being especially more frequent in females. Women also show greater variation in hair color (Shekar et al., 2008).

Second, blue-eyed men seem to have a more feminine face shape. This was the unintended finding of a rating study of male facial photos, which initially found that brown-eyed men were perceived to be more dominant than blue-eyed men. As a control, the authors repeated the experiment after altering the photos to give the brown-eyed men blue eyes. The altered photos were still rated as more dominant. Careful study revealed that the brown-eyed men had more masculine faces with broader and more massive chins, broader mouths, larger noses, larger eyebrows, and closer-set eyes. The blue-eyed men had smaller and sharper chins, narrower mouths, smaller noses, and a greater span between the eyes. It was thus face shape and not eye color that made the brown-eyed men seem more dominant. The authors denied the possibility of ethnic differences between the two groups, stating that the photos depicted only university students of Czech origin (Kleisner et al., 2010).

Finally, a study of preschool children suggests that blue eyes are sex-linked to shyness:

In the present study, 152 Caucasian preschool-aged (Mage=54.09 months, SD=5.84) children (77 males) with either blue (n=84) or brown (n=68) eyes, were compared in terms of parental and teacher ratings of social wariness, social play, and aggression. A significant Eye Color×Gender Interaction was found in terms of indices of social wariness; blue-eyed males were rated as more socially wary than brown-eyed males, while blue- and brown-eyed females did not differ in this regard. (Coplan et al., 1997)

This finding seems robust in the sense that other authors (cited by Coplan et al.) have found similar results. Although shyness in itself is not more common in girls than in boys, girls show a stronger interaction between loneliness and aggression (Coplan et al., 2007). In a male brain, this interaction might lead to a higher likelihood of social withdrawal, given the generally higher level of aggression in boys than in girls.

But how does this sex linkage operate? It may be mediated by prenatal exposure to estrogen, as Mather et al found in their unpublished study. This pre-natal estrogenization would be over-determined in women, i.e., almost all women are fully exposed to estrogen before birth regardless of their eye color. In men, however, it would be limited to blue-eyed individuals. Thus, blue-eyed men may be partially feminized not only in their face shape but also in their behavior.

Could other behaviors be partially feminized in blue-eyed or fair-haired men? Sexual orientation comes to mind, but that’s one thing that natural selection should have quickly rectified (as Greg Cochran has often pointed out). Indeed, Ellis et al. (2008) failed to find any significant relationship between sexual orientation and eye color or hair color.

Alternate explanations

- Neanderthal admixture?

Not many alternate explanations are still in the running. For a while, a plausible one was Neanderthal admixture. We’ve since managed to retrieve the Neanderthal MC1R allele and it doesn’t match any existing allele in modern humans. In any case, the estimated Neanderthal admixture of 1 to 4% in modern Europeans is well below the frequency of non-black hair and non-brown eyes.

- Genetic linkage to lighter skin?

Perhaps eye and hair color diversity is genetically linked to lighter skin. Eyes and hair have diversified in color in the same part of the world where skin color has lightened the most. Perhaps this lightening involved genes that also affect hair and eye color.

Yet the genes are different in each case. European skin has lightened mainly through replacement of alleles at two genes: SLC45A2 (AIM1) and SLC24A5. European hair color has diversified through a proliferation of new alleles at the MC1R gene. European eye color has diversified through a proliferation of new alleles in the HERC2-OCA2 region and elsewhere.

Lighter skin is associated with a few of the new alleles, namely the ones for red hair or blue eyes. Conceivably, these two color variants are a side effect of selection for lighter skin. But why would such selection increase the total number of alleles for hair and eye color? This is especially strange because many of the alleles have little or no effect on skin pigmentation. And why have neither red hair nor blue eyes reached fixation in any population, even those with milk-white complexions?

- Cochran’s theory

Greg Cochran has argued that European hair and eye color diversity reflects an underlying behavioral polymorphism, which in turn is due to a process of self-domestication that resulted from the advent of agriculture. As Europeans formed larger and more sedentary communities, they had to become more obedient to authority. An extra dose of prenatal estrogen might have been the necessary quick fix to tame European males.

Greg’s theory is a mirror image of my own. I argue that sexual selection of women was driving the evolutionary change. The effects on male behavior are thus merely collateral damage. Greg argues that the evolutionary change was driven by natural selection for male submissiveness. The new hair and eye colors are thus merely a cute side effect.

To my knowledge, Greg has presented this theory at length only in the book he published with Henry Harpending The 10,000 Year Explosion. Even there, the presentation is a bit hazy. The two authors start off well enough:

[…] selection on genes affecting skin color, eye color, and hair color somehow created lots of variety in Europeans: redheads and blondes, blue eyes and green eyes. Nowhere else in the world is that sort of variety common. In most parts of the world, even in temperate regions, everyone has dark eyes and dark hair. (Cochran & Harpending, 2009, p. 94)

This pattern points to “something fundamentally different in the selective forces.” The next few pages suggest that this “something” is selection for submission to authority, which in turn is incidentally linked to eye and hair color. But Cochran and Harpending never spell out the link. The closest they come is a paragraph eighteen pages onward:

Selection for submission to authority sounds unnervingly like domestication. In fact, there are parallels between the domestication in animals and the changes that have occurred in humans during the Holocene period. In both humans and domesticated animals, we see a reduction in brain size, broader skulls, changes in hair color or coat color, and smaller teeth. (Cochran & Harpending, 2009, p. 112)

That’s it. There are no references to the literature linking eye color to behavior. Why? Maybe it’s bad karma to imply that blue-eyed men are more civilized. That sounds a bit …. well, you know.

Personally, I don’t care whether or not a theory is socially acceptable. Only one question concerns me. Does it make sense?

Let’s walk through it step by step. Agriculture created a selection pressure for more obedient men. This selection in turn led to blue eyes and fair hair as a side effect. We know that agriculture spread through Europe in stages. For a long time, the wave of advance stalled along a line stretching from the Low Countries to the Black Sea. Then, around 7,500 years ago, agriculture resumed its northward spread. The last refuge of European hunter/fisher/gatherers seems to have been in the East Baltic, where farming became dominant only 3,000 years ago.

These last holdouts were presumably less affected by the process of self-domestication via agriculture. They should therefore have lower incidences of blue eyes and fair hair. Is that what we see?


Greg’s theory has other weaknesses. Are farmers really more submissive than hunter-gatherers? Not necessarily. It isn’t so much farming that makes men less unruly as State formation, specifically the State’s monopoly on the use of violence. That’s a later historical development (Frost, 2010).

And blue eyes are linked not so much to male submissiveness as to male shyness. The two aren’t really the same thing. Can one confidently argue that male shyness is more adaptive among farmers? Aren’t social relations more extensive and intensive in farming communities than in small bands of hunter-gatherers?


Cochran, G.M. & H. Harpending. (2009). The 10,000 Year Explosion, Basic Books.

Coplan, R.J., L.M. Closson, & K.A. Arbeau. (2007). Gender differences in the behavioral association of loneliness and social dissatisfaction in kindergarten, Journal of Child Psychology and Psychiatry, 48, 988-995.

Coplan, R., B. Coleman, & K. Rubin. (1998). Shyness and little boy blue: Iris pigmentation, gender, and social wariness in preschoolers. Developmental Psychobiology, 32, 37–44.

Ellis, L., C. Ficek, D. Burke, & S. Das. (2008). Eye color, hair color, blood type, and the Rhesus factor: Exploring possible genetic links to sexual orientation, Archives of Sexual Behavior, 37, 145-149.

Frost, P. (2010). The Roman State and genetic pacification, Evolutionary Psychology, 8(3), 376-389.

Frost, P. (2008). Sexual selection and human geographic variation, Special Issue: Proceedings of the 2nd Annual Meeting of the NorthEastern Evolutionary Psychology Society. Journal of Social, Evolutionary, and Cultural Psychology, 2(4),169-191.

Frost, P. (2006). European hair and eye color - A case of frequency-dependent sexual selection? Evolution and Human Behavior, 27, 85-103.

Frost, P. (1994).
Geographic distribution of human skin colour: A selective compromise between natural selection and sexual selection? Human Evolution, 9, 141-153.

Kleisner, K., T. Kočnar, A. Rubešova, & J. Flegr. (2010). Eye color predicts but does not directly influence perceived dominance in men. Personality and Individual Differences, 49, 59–64.

Mather, F., Manning, J.T., & Bundred, P.E. (unpublished). 2nd to 4th digit ratio, hair and eye colour in Caucasians: Evidence for blond hair as a correlate of high prenatal oestrogen.

Olivier, G. (1960). Pratique anthropologique. Paris: Vigot Frères.

Shekar, S.N., D.L. Duffy, T. Frudakis, G.W. Montgomery, M.R. James, R.A. Sturm, & N.G. Martin. (2008). Spectrophotometric methods for quantifying pigmentation in human hair—Influence of MC1R genotype and environment. Photochemistry and Photobiology, 84, 719–726.