Saturday, June 28, 2014

How universal is empathy?

Bronislaw Malinowski with natives on the Trobriand Islands (1918 - source). Pro-social behavior seems to be a human universal, but is the same true for full empathy?

What is empathy? It has at least three components:

- pro-social behavior, i.e., actions of compassion to help others

- cognitive empathy, i.e., capacity to understand another person's mental state

- affective or emotional empathy, i.e., capacity to respond with the appropriate emotion to another person's mental state (Chakrabarti and Baron-Cohen, 2013)

In their review of the literature, Chakrabarti and Baron-Cohen (2013) conclude that all three components are moderately to highly heritable, although the affective component seems to show the highest heritability (68%). This is in line with Davis et al. (1994), who found significant heritability for the affective facets of empathy (empathic concern and personal distress) but not for non-affective perspective taking.

All three components can vary from one individual to another, although studies to date have focused on pathological variation:

For example, it is suggested that people with psychopathic personality disorder may have intact cognitive empathy (hence being able to deceive others), but impaired affective empathy (hence being able to hurt others), whilst people with autism may show the opposite profile (hence finding the social world confusing because of their deficit in cognitive empathy, but not being over-represented among criminal offenders, having no wish to hurt others, suggesting their affective empathy may be intact) (Chakrabarti and Baron-Cohen, 2013)

People with depression may suffer from too much empathy, i.e., being too sensitive to the needs or distress of others (O'Connor et al., 2007). In short, these disorders seem to be the tail ends of a normal distribution. By focusing on these extremes, we forget that most of the genetic variability in empathy occurs among healthy individuals (Gillberg, 2007). 

Using research findings on autism and Asperger syndrome, Chakrabarti and Baron-Cohen (2013) have identified nine candidate genes that seem to promote empathy. They fall into three functional categories: sex-steroid synthesis and metabolism; neural development and connectivity; and social-emotional responsivity. The first category includes the degree to which a fetus is androgenized or estrogenized before birth, as shown by digit ratio (Frost, 2014).

Variation among human populations

If the genes associated with empathy vary among healthy individuals, do they also vary among human populations? This would be expected because populations have differed in their needs for different components of empathy, particularly since hunting and gathering gave way to farming some 10,000 years ago—when genetic change speeded up over a hundred-fold. At that time, humans were no longer adapting to new physical environments. They were adapting to new cultural environments that differed in social structure, in division of labor, in means of subsistence, in norms of conduct, in future time orientation, in degree of sedentary living, and so on. Our ancestors were now reshaping their environments, and these human-made environments were now reshaping them—in other words, gene-culture co-evolution (Hawks et al., 2007).

Humans have been transformed especially by the shift from small bands of hunter-gatherers to larger and more complex groups of farmers and townsfolk. With social relations expanding beyond the circle of close kin, kinship obligations were no longer enough to ensure mutual assistance and stop free riding. There was thus selection for pro-social behavior, i.e., a spontaneous willingness to help not only kin but also non-kin.

Pro-social behavior is attested across a wide range of cultures. It is the subject of a recent book about the nature and limits of empathy in Oceanic cultures. The Banabans of Fiji for instance express the idea of pro-sociality through the term nanoanga, which they normally translate into English by "compassion" or "pity."

[...] compassion is the basis for their capacity to bond socially with others, even compassion to the point of readiness to take strangers into their community. Their empathy therefore relates causally to how they act socially toward others. Here compassion or pity embraces both understanding and fellow feeling: the islanders understand that the stranded mariner is at the end of his strength, which is why they succor him and treat him as one of their own. They understand him because he, like them, is a human being, a person. [...] Thus, for example, when someone passing by a house does not belong to the immediate family of those inside, it is customary to welcome the passer-by by calling out the words mai rin! (Come in!), which carry the implication that food and drink will not be found wanting inside. (Hermann, 2011, p. 31)

This desire to help non-kin is not unconditional. The author notes that prior experiences with an individual in distress can determine whether compassion will be given or withheld. Moreover, Barnabans can "proceed strategically when deciding whether to extend trust to others or to keep thoughts and feelings to themselves" (Hermann, 2011, p. 31). This is not the affective empathy of entering another person's mind to feel his or her pain.
When the Barnabans compare themselves with others, and when by their behavior toward the stranger they show that they understand him and feel with him, they do not, however, equate themselves fully and entirely with him. (Hermann, 2011, p. 32)

Another contributor to the same book writes similarly about the inhabitants of Vanatinai, in the Trobriand Islands.
On the island of Vanatinai, when someone, including an ethnographer, privately asks a trusted confidant, "Why did she/he act like that?" "What was she/he thinking?" the common answer, often uttered in tones of puzzlement and despair, or anxiety and fear, expresses one of the islanders' core epistemological principles: "We cannot know their renuanga." Renuanga is a word that refers to a person's inner experiences, both and inseparably thought and emotion.

[...] And their psychic states, their inner thought and feelings, are inherently unknowable. It may never be clear why they were angry or sympathetic, and what caused them to act and influence an event in someone's life [...] (Lepowsky, 2011, p. 44)

In short, Oceanic cultures display hospitality but not full empathy, which would be considered undesirable anyway:

The philosophical principle of personal opacity, the interiority of others' thoughts/feelings (renuanga), is closely bound to the islanders' fierce insistence on personal autonomy, both as cultural ideology and as daily social practice (Lepowsky, 2011, p. 47)


From pro-sociality to full empathy

Whereas pro-sociality is attested across a wide range of cultures, full cognitive/affective empathy is more localized. The difference is like the one we see between shame and guilt. Most cultures primarily use shame to enforce correct behavior, i.e., if other people see you breaking a rule, you feel ashamed and this feeling is reinforced by social disapproval. In contrast, only a minority of cultures—largely those of Northwest Europe—rely primarily on guilt, which operates even when only you see yourself breaking a rule or merely think about breaking a rule (Benedict, 1946; Creighton, 1990).

Northwest Europeans have thus undergone two parallel changes in behavioral control: 1) a shift from pro-sociality to full cognitive/affective empathy; and 2) a shift from shame to guilt. Indeed, full empathy and guilt may be two sides of the same coin. Both are the consequences of a mental model that is used to simulate how another person thinks or feels (an imaginary witness to a wrongful act, a person in distress) and to ensure correct behavior by inducing the appropriate feelings (anguish, pity).

Finally, full empathy and guilt are most adaptive where kinship ties are relatively weak and where rules of correct behavior require a leveling of the playing field between kin and non-kin. This has long been the case in Northwest Europe. There seems to be a longstanding pattern of weak kinship ties west of a line running from Trieste to St. Petersburg, as shown by several culture traits that are rare or absent elsewhere:

- relatively late marriage for men and women

- many people who never marry

- neolocality (children leave the family household to form new households)

- high circulation of non-kin among different households (typically young people sent out as servants) (Hajnal, 1965)

Commonly called the Western European Marriage Pattern, this geographic zone of relatively weak kinship was thought to have arisen after the Black Death of the 14th century. There is now good evidence for its existence before the Black Death and fragmentary evidence going back to 9th century France and even earlier (Hallam, 1985; Seccombe, 1992, p. 94). Historian Alan Macfarlane likewise sees an English tendency toward weaker kinship ties before the 13th century and even during Anglo-Saxon times (Macfarlane, 2012; Macfarlane, 1992, pp. 173-174).

This weak kinship zone may have arisen in prehistory along the coasts of the North Sea and the Baltic, which were once home to a unique Mesolithic culture (Price, 1991). An abundance of marine resources enabled hunter-fisher-gatherers to achieve high population densities by congregating each year in large coastal agglomerations for fishing, sealing, and shellfish collecting. Population densities were comparable in fact to those of farming societies, but unlike the latter there was much "churning" because these agglomerations formed and reformed on a yearly basis. Kinship obligations would have been insufficient to resolve disputes peaceably, to manage shared resources, and to ensure respect for social rules. Initially, peer pressure was probably used to get people to see things from the other person's perspective. Over time, however, the pressure of natural selection would have favored individuals who more readily felt this equivalence of perspectives, the result being a progressive hardwiring of compassion and shame and their gradual transformation into empathy and guilt (Frost, 2013a; Frost, 2013b).

Empathy and guilt are brutally effective ways to enforce social rules. If one disobeys these internal overseers, the result is self-punishment that passes through three stages: anguish, depression and, ultimately, suicidal ideation.

People suffering from depression are looking at both others and themselves with suspicion, often believing whatever they have was obtained by cheating, and that it is more than they deserve. Depressives, burdened by moralistic standards, are harsh evaluators of both themselves and others. The self-punishment meted out by depressives is a common if disturbing symptom; while thinking 'I deserve this', they may engage in altruistic punishment turned upon the self. Just as altruistic punishers experience a neuronally based reward from punishing defectors, despite material costs, depressed patients report a sense of relief upon inflicting self-punishment. Patients who are 'cutters', describe relief from tension after cutting and depressives with suicidal ideation may describe the relief they felt when on the verge of attempting a suicidal action. (O'Connor et al., 2007, p. 67)

This pathology is progressively less common in populations farther south and east, not so much because each stage is less common but rather because depression is much less likely to result from empathic guilt and much less likely to lead to suicide (Stompe et al., 2001). This 3-stage sequence does not seem to be a human universal, at least not to the same extent as in Northwest Europeans, a reality that Frantz Fanon noted when describing clinical depression in Algerians: 

French psychiatrists in Algeria found themselves faced with a difficult problem. When treating a melancholic patient, they were accustomed to being afraid of suicide. The melancholic Algerian kills, however. This disease of the moral conscience that is always accompanied by self-accusation and self-destructive tendencies assumes hetero-destructive forms in the Algerian. The melancholic Algerian does not commit suicide. He kills. (Fanon, 1970, pp. 219-220)



Benedict, R. (1946 [2005]). The Chrysanthemum and the Sword. Patterns of Japanese Culture, First Mariner Books. 

Chakrabarti, B. and S. Baron-Cohen. (2013). Understanding the genetics of empathy and the autistic spectrum, in S. Baron-Cohen, H. Tager-Flusberg, M. Lombardo. (eds). Understanding Other Minds: Perspectives from Developmental Social Neuroscience, Oxford: Oxford University Press.  

Creighton, M.R. (1990). Revisiting shame and guilt cultures: A forty-year pilgrimage, Ethos, 18, 279-307.

Davis, M.H., C. Luce, and S.J. Kraus. (1994). The heritability of characteristics associated with dispositional empathy, Journal of Personality, 62, 369-391.

Fanon, F. (1970). Les damnés de la terre, Paris: Maspero. 

Frost, P. (2013a). The origins of Northwest European guilt culture, Evo and Proud, December 7 

Frost, P. (2013b). Origins of Northwest European guilt culture, Part II, Evo and Proud, December 14  

Frost, P. (2014). A pathway to pro-social behavior, Evo and Proud, May 10.

Gillberg, C. (2007). Non-autism childhood personality disorders, in: T.F.D. Farrow and P.W.R. Woodruff (eds). Empathy in Mental Illness, (pp. 111-125). Cambridge: Cambridge University Press.

Hajnal, J. (1965). European marriage pattern in historical perspective. In D.V. Glass and D.E.C. Eversley (eds). Population in History, Arnold, London.

Hallam, H.E. (1985). Age at first marriage and age at death in the Lincolnshire Fenland, 1252-1478, Population Studies, 39, 55-69. 

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

Hermann, E. (2011). Empathy, ethnicity, and the self among the Barnabans in Fiji, in D.W. Hollan, C. J. Throop (eds).The Anthropology of Empathy: Experiencing the Lives of Others in Pacific Societies, (pp. 25-42), New York: Berghahn.

Lepowsky, M. (2011). The boundaries of personhood, the problem of empathy, and "the native's point of view" in the outer islands, in D.W. Hollan, C. J. Throop (eds).The Anthropology of Empathy: Experiencing the Lives of Others in Pacific Societies, (pp. 43-68), New York: Berghahn. 

Macfarlane, A. (1992). On individualism, Proceedings of the British Academy, 82, 171-199.  

Macfarlane, A. (2012). The invention of the modern world. Chapter 8: Family, friendship and population, The Fortnightly Review, Spring-Summer serial

O'Connor, L.E., J.W. Berry, T. Lewis, K. Mulherin, and P.S. Crisostomo. (2007). Empathy and depression: the moral system in overdrive, in: T.F.D. Farrow and P.W.R. Woodruff (eds). Empathy in Mental Illness, (pp. 49-75). Cambridge: Cambridge University Press.

Price, T.D. (1991). The Mesolithic of Northern Europe, Annual Review of Anthropology, 20, 211-233.

Seccombe, W. (1992). A Millennium of Family Change. Feudalism to Capitalism in Northwestern Europe, London: Verso.

Stompe, T., G. Ortwein-Swoboda, H.R. Chaudhry, A. Friedmann, T. Wenzel, and H. Schanda. (2001). Guilt and depression: a cross-cultural comparative study, Psychopathology, 34, 289-298.

Saturday, June 21, 2014

Negotiating the gap. Four academics and the dilemma of human biodiversity

I’ve published a second article in Open Behavioral Genetics: “Negotiating the gap. Four academics and the dilemma of human biodiversity.” You can read it as a PDF here or as a caliber ebook here. The foreword is reproduced below. Comments are welcome.


Twenty-five years ago I met a professor from the medical faculty who had decided to go into anthropology. He was excited by the concept of gene-culture co-evolution and wanted to get in on the action. But he stressed the need for "prudence." He would first earn his credentials as an anthropologist before tackling this sensitive subject, and he would do so gradually and prudently.

He was already a man of a certain age, and I wondered whether he would have time for all of this, but I said nothing. He knew better than me how to plan his life. And his proposal for research on gene-culture co-evolution had been thoroughly worked out. This was no back-of-the-envelope thing.

Over the next quarter-century he carried out fieldwork and published journal articles, but he never touched the subject that had inspired his move to anthropology. Did he change his mind? I suspect the reason was less thought out. Once you begin your research from a certain angle, it is hard to break away and approach it from a totally different angle—you would have to find new sources of funding and make friends with new people. You would also lose friends. So you take the easy way out, for the time being. And you wait for the right moment, which never comes.

Charles Darwin himself had fallen into that trap. When a non-biologist anonymously wrote and marketed a book about evolution, Vestiges of the Natural History of Creation, the resulting controversy impressed on Darwin the need to become a reputable biologist before writing on the topic. So he bided his time and published, published, published … on other topics in biology. One day, however, fate forced his hand. Another biologist sent him a manuscript that set out the very theory that Darwin had kept under wraps for so long. The rest is history.

You may be thinking: "That was Darwin, and this is me. And my situation is different, very different. And this is a completely different issue. It's really important for me to wait until the time is right!"

I hear you. Maybe your situation is different. And who am I to judge?


This essay presents four academics—Richard Dawkins, Claude Lévi-Strauss, John Tooby, and Leda Cosmides—and how they negotiated the gap between personal conviction and mainstream discourse. All four came to the conclusion that human populations differ not only anatomically but also in various mental and behavioral predispositions. These differences are statistical and often apparent only between large groups of people. But even a weak statistical difference can affect how a society will develop and organize itself. Human biodiversity is therefore a reality, and one we ignore at our peril.

Yet most academics do ignore it, their ignorance being either real or feigned. It is easy to forgive the truly ignorant. But what about the ones who know better?  What's their excuse? "I don't have tenure yet." "I'm not well enough known yet." "I don't have enough clout yet." Some will just say: "Please come into my office. Others may hear us talking in the corridor."

And so, among those who do know better, the common response is ... no response. But what else is there to do? How does one go about saying something that is offensive to most people? Is it better to do it gradually? Or all at once? Or is it better to say nothing at all and wait for someone else to speak out?

There are no easy answers, and that may be part of the problem. Too many people are looking for answers that are easy—that cost little in terms of reputation, career prospects, or acceptance at the next cocktail party. Why not instead assume that everything worthwhile has a cost and then look for ways to minimize the cost?

Once you accept that rule of life, everything will fall into place. This intellectual maturity became a source of strength for one of the above academics, Claude Lévi-Strauss, who had to face bitter criticism for what he said. There was an énorme scandale. People were upset and shocked. Yet he carried on as if nothing terrible had happened. Was he so fascinated by his ideas that he simply ignored what others might think? Perhaps. More likely than not, he pondered his dilemma, weighed the pros and cons, and decided that the only sensible thing was to speak out. 

How will you decide? Will you speak out or remain silent?


Frost, P. (2014). Negotiating the gap. Four academics and the dilemma of human biodiversity, Open Behavioral Genetics, June 20  

Saturday, June 14, 2014

A hair-color allele of Neanderthal origin?

Taiwanese aboriginal children, Bunun village (source: Jeremy Kemp). 60-70% of Taiwanese aborigines have a loss-of-function allele at the main hair color gene, MC1R, yet their hair is as black as humans with the original “African” allele. This seems to be a general pattern in Asians. They have fewer MC1R alleles than do Europeans, and the ones they have produce the same hair color.


When I first wrote about the puzzle of European hair and eye color, a common explanation was Neanderthal admixture. Modern humans intermixed with Neanderthals in Europe, and one legacy of this intermixture is the high prevalence of non-black hair and non-brown eyes we see in present-day Europeans.

I was skeptical. Scientists had already retrieved mtDNA from the remains of Neanderthals and early modern humans, and there was no discernible genetic continuity between the two. Neanderthal admixture seemed minor and could hardly account for the high proportion of Europeans who deviate from the species norm of black hair and brown eyes (Frost, 2006).

With the sequencing of the Neanderthal genome, it became apparent that some admixture had taken place, but only on the order of 1 to 4% in modern Eurasians. Neanderthals did resemble modern humans in having the same main gene for hair color, i.e., MC1R, but the Neanderthal allele at that gene was unlike any allele in modern humans (Lalueza-Fox et al., 2007). Moreover, there was no evidence of the polymorphism that exists for European hair color. The same allele was present in the two Neanderthal individuals that had been sampled.

That seemed to be the end of the story. A new twist, however, has been added by a recent paper. Ding et al. (2014) have found that one of the MC1R alleles in modern humans (Val92Met) appears to be of Neanderthal origin:

In this paper, we present evidences of Neanderthal introgression encompassing the MSH receptor gene MC1R. Furthermore, our evidences support that the derived allele at the functional variant Val92Met of MC1R (i.e., rs2228479*A) is likely of two origins: the vast majority of haplotypes carrying this allele in the human gene pool is resulted from Neanderthal introgression, while one haplotype (NA19084_a) carrying this allele may be from a recurrent mutation in the AMH linage, double recombination, or biased gene conversion.

This finding is consistent with the theory, first advanced by Gregory Cochran, that archaic admixture made it easier for modern humans to adapt to new environments. To be sure, Val92Met is only one of eleven derived MCIR alleles that exist in modern humans. But Ding et al. (2014) also believe that some of these other MC1R alleles are mosaics of Neanderthal and non-Neanderthal segments. So Neanderthal admixture may have helped European hair color to diversify by providing raw material for selection to act on.

A silent allele or a silenced allele?

By itself, Neanderthal admixture cannot explain the unusual diversity of hair color in present-day Europeans. It simply provided some of the raw material for this evolutionary development, and in most cases this raw material had to undergo further changes, through mutation and recombination, before it could become useful.

Indeed, despite being a loss-of-function allele, Val92Met seems to produce the same black hair as the original "African" allele. This may be seen in its geographic distribution: ~5% in Europeans, ~30% in continental East Asians, and 60-70% in Taiwanese aborigines (Ding et al., 2014). It has also been reported in South Asians, Papua-New Guineans, Japanese, and Inuit (Harding et al., 2000). Ding et al. (2014) state that this allele is associated with red hair, but the study they cite found only one individual with Val92Met among the 21 redheads examined (Valverde et al., 1995). This proportion is almost identical to the allele's incidence among Europeans in general. More likely than not, that single individual owed her red hair to an allele somewhere else on her genome.

Hair color is much less diverse in Asians, and this is reflected in lower MC1R diversity. Whereas Europeans have eleven MC1R alleles, Asians have only five, and all five produce the same black hair color (Harding et al., 2000). In short, Asians have fewer alleles and proportionately fewer of these differ phenotypically from the ancestral African allele. It looks as if something downstream prevents these alleles from affecting hair color.

As I've argued elsewhere, Europe's diverse palette of hair and eye colors is due to unusual evolutionary circumstances, i.e., intense sexual selection of women within an ecozone (continental steppe-tundra of the last ice age) where almost all food was obtained through long-distance hunting. The consequently higher death rate and lower polygyny rate among hunters dried up the pool of men available for mating and increased competition by women for mates. Women were more strongly selected for eye-catching traits, particularly bright or novel hues, thus creating an increasingly diverse palette of hair and eye colors (Frost, 2006; Frost, 2014).

This ecozone was more suitable for continuous human settlement in Europe than in northern Asia, where it was farther north and farther removed from the moderating influence of the Atlantic. A site in central Siberia from the last ice age has yielded human DNA that shows strong affinities with present-day Europeans and Amerindians, but much less affinity with present-day northern Asians, who seem to be largely the product of repeopling from the south near the end of the last ice age (Maanasa et al., 2014). Europeans have thus better preserved the legacy of this episode of intense sexual selection.

Perhaps the story ends there. Present-day Asians have preserved less of that MC1R diversity and what they have preserved has less functional significance. Or perhaps that diversity was initially functional and then gradually ceased to be functional … because of some other selection pressure? Perhaps, at the end of the last ice age, there was some non-black hair among northern Asians, though much less than among Europeans. Being less common and thus less normal, and no longer favored by intense sexual selection, there may have been stronger social selection to eliminate deviant hair colors.

A similar kind of social selection might explain why red hair is less common than blond hair among Europeans, i.e., stigmatization of redheads that was ultimately due to a mental association between red hair and menstrual blood (Frost, 2012).

There may be a story behind these "silent alleles." 


Ding, Q., Y. Hu, S. Xu, C. Wang, H. Li, R. Zhang, S. Yan, J. Wang, and L. Jin (2014). Neanderthal origin of the haplotypes carrying the functional variant Val92Met in the MC1R in modern humans, Molecular Biology and Evolution, published online June 10, 2014

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

Frost, P. (2012). Why are redheads less common than blondes? Evo and Proud, March 10  

Frost, P. (2014). The puzzle of European hair, eye, and skin color, Advances in Anthropology, 4, 78-88.

Harding, R.M., Healy, E., Ray, A.J., Ellis, N.S., Flanagan, N., Todd, C., Dixon, C., Sajantila, A., Jackson, I.J., Birch-Machin, M.A., and Rees, J.L. (2000). Evidence for variable selective pressures at MC1R. American Journal of Human Genetics, 66, 1351-1361. 

Lalueza-Fox, C., H. Römpler, D. Caramelli, C. Stäubert, G. Catalano, D. Hughes, N. Rohland, E. Pilli, L. Longo, S. Condemi, M. de la Rasilla, J. Fortea, A. Rosas, M. Stoneking, T. Schöneberg, J. Bertranpetit,  and M. Hofreiter. (2007). A melanocortin 1 receptor allele suggests varying pigmentation among Neanderthals, Science, 318 (5855), 1453-1455.  

Maanasa, R., Skoglund, P., Graf, K.E., Metspalu, M., Albrechtsen, A., Moltke, I., Rasmussen, S., Stafford Jr, T.W., Orlando, L., Metspalu, E., Karmin, M., Tambets, K., Roots, S., Mägi, R., Campos, P.F., Balanovska, E., Balanovsky, O., Khusnutdinova, E., Litvinov, S., Osipova, L.P., Fedorova, S.A., Voevoda, M.I., DeGiorgio, M., Sicheritz-Ponten, T., Brunak, S., Demeshchenko, S., Kivisild, T., Villems, R., Nielsen, R., Jakobsson, M., and Willerslev, E. (2014). Upper Palaeolithic Siberian genome reveals dual ancestry of Native Americans. Nature, 505, 87-91.  

Valverde, P., E. Healy, I. Jackson, J.L. Rees, and J. Thody. (2005). Variants of the melanocyte-stimulating hormone receptor gene are associated with red hair and fair skin in humans, Nature Genetics, 11, 328-330.

Saturday, June 7, 2014

A new allele for blond hair

Geographic prevalence of the new allele for blond hair (Guenther et al., 2014). Just one of many alleles that create the European palette of hair and eye colors.

There is a widespread belief that whatever made Europeans fair-skinned also gave them their unique palette of hair and eye colors. In reality, fair skin has only a weak genetic linkage with either non-black hair or non-brown eyes:

If we were to take all the human beings in the world who have dark brown eyes and black or dark brown hair, we would not only have the vast majority of the human species, but would have a group which shows virtually the complete range of human skin color, from black to almost completely depigmented. (Brues, 1975)

Hair color and eye color are largely uncoupled from skin color in our species—or rather in Europeans, to be more precise. It looks as if the pressure of selection has moved European color traits in different directions, on the one hand making the skin progressively lighter, and on the other making the hair and the eyes more and more diversely hued.

There have thus been different changes at different genes. European skin lightened mainly through the appearance of new alleles at three genes: SLC45A2, SLC24A5, and TYRP1. European hair color diversified through a proliferation of new alleles at MC1R. European eye color diversified through a proliferation of new alleles in the HERC2-OCA2 region and elsewhere.

In a recent study, Guenther et al. (2014) have shown that new alleles for European hair color proliferated not only at MC1R but also at other genes. In particular, blond hair seems to be due to an allele near KITLG, although other loci in this region seem involved as well. Thus, the diverse palette of European hair colors is not a side effect of a single genetic change. It is due to a proliferation of new alleles at many different places on the human genome. The only common element seems to be a selection pressure for more hair colors.

Guenther et al. (2014) likewise note that changes to hair color and eye color became uncoupled from changes to skin color in ancestral Europeans:

Some human pigmentation variants alter general aspects of pigment biosynthesis, producing changes in all melanocytes, and therefore have pleiotropic effects on hair, skin and eye color. However, it is well known that hair and eye color can also vary independently, producing common human phenotypes such as light-haired individuals with brown eyes or brown-haired individuals with blue eyes. The rs12821256 variant alters an enhancer that is active specifically in the hair follicle environment, providing a simple genetic explanation for previous observations that this SNP is associated with changes in hair pigmentation but not eye pigmentation in northern Europeans.

Why this uncoupling of European hair, eye, and skin color? These three color traits have probably undergone divergent selection pressures. Moreover, in all three cases the selection pressure seems to have acted primarily on women and only secondarily on men—an indication of some form of sexual selection where women were the sex in excess supply on the mate market (Frost, 2014). Because skin color is sexually dimorphic in all humans, with women being the "fair sex," this dimorphism may have biased sexual selection in the direction of increasingly lighter skin.  

Hair and eye color, however, had no preexistent sexual dimorphism. So sexual selection was driven simply by a desire for bright or novel colors. In other words, whenever a new color arose through mutation, it would have initially benefited from sexual selection. As it became more and more common in the population, this selection pressure would have gradually diminished until the new color had become as prevalent as other existing hues. As a result, the palette of hair and eye colors would have steadily grown larger and larger (Frost, 2006; Frost, 2014).


Brues, A.M. (1975). Rethinking human pigmentation, American Journal of Physical Anthropology, 43, 387-391.

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

Frost, P. (2014). The puzzle of European hair, eye, and skin color, Advances in Anthropology, 4, 78-88.  

Guenther, C.A., B. Tasic, L. Luo, M.A. Bedell, and D.M. Kingsley. (2014). A molecular basis for classic blond hair color in Europeans, Nature Genetics, advance online