Friday, September 14, 2007

When did Europeans become 'white'?

It is often assumed that Europeans have always looked much like they do now. Even Neanderthals are often depicted as white folk who need a shave and a haircut. Yet, clearly, Europeans have not always been European. At some point in time, their ancestors came from somewhere else and looked like people from somewhere else.

The current thinking is that modern humans arrived in Europe about 35,000 years ago by way of the Middle East and ultimately from Africa. When did these proto-Europeans begin to look like their present-day descendants? Probably long after. The current phenotype seems to reflect later in situ changes to morphology and coloration that are still far from uniform among Europeans.

In a study of prehistoric European skeletons, Holliday (1997) found that early Upper Paleolithic skeletons had ‘tropical’ body proportions and clustered with recent Africans. Late Upper Paleolithic and Mesolithic skeletons clustered with recent Europeans. The shift to a more European phenotype is placed by Holliday (1997) at around 20,000 BP— during the last ice age and well after the arrival of modern humans.

Even later are the changes to European skin color, eye color, and hair color. Whitening of the skin, through allelic changes at the AIM1 gene, is dated to about 11,000 years ago (Soejima et al., 2005). Allelic changes at other skin color loci are similarly dated to the late Upper Paleolithic or even the Holocene (Voight et al., 2006). No less recent is diversification of eye color alleles at the OCA2 gene (Voight et al., 2006). Diversification of hair color alleles at the MC1R gene has yet to be reliably dated but is likely contemporaneous. This whitening of the skin and diversification of eye and hair color clearly constitute an in situ evolution within Europe, being most prominent within a zone centered on the East Baltic and covering the north and the east. Within this zone, skin is unusually white, almost at the physiological limit of depigmentation, eyes are not only brown but also blue, gray, hazel or green, and hair is not only black but also brown, flaxen, golden or red. As one moves further east and south, these color traits disappear and merge into the standard human pattern of dark skin, brown eyes, and black hair.

If I had unlimited funding, I would like to retrieve nuclear DNA from European skeletal remains that date from the arrival of modern humans (c. 35,000 BP) to the near present (c. 5,000 BP). I would then study changes in genes for skin color, eye color, hair color, hair length and form, and facial morphology. My hunch is that all of these changes took place within a narrow time-frame, most likely between the glacial maximum (c. 20,000-15,000 BP, when Europe was largely depopulated) and the end of the last ice age (c. 10,000 BP).

These changes were driven not so much by adaptation to the natural environment as by intense female-female competition for mates. Men were in short supply among early Europeans, especially among those who pursued reindeer herds across the northern and eastern plains. This was because of 1) very long hunting distances, which greatly increased death rates among young men; and 2) limited opportunities for food gathering, which made women dependent on men for provisioning and thus ruled out polygyny for all but the ablest of hunters. With too many women competing for too few men, conditions were optimal for sexual selection of women. Men were able to translate their subtlest preferences into mate choice. Intense sexual selection is particularly indicated by a shift to brightly colored traits, especially color polymorphisms—such as those of eye and hair color (Frost, 2006).

How did Europeans look previously? They probably looked distinctly non-European. Holliday, as already discussed, noted the ‘tropical’ and even African appearance of early modern humans in Europe. Other anthropologists have noted the same, particularly in relation to a pair of skeletons discovered in 1901 at Grimaldi, northern Italy. The skeletons were initially dated to the beginnings of modern human occupation in Europe, c. 30,000 BP. Associated artifacts have since been radiocarbon dated to 14,000-19,000 BP but may come from more recent layers of occupation (Bisson et al., 1996).

These skeletons exhibit an array of dental and morphological characteristics normally found in sub-Saharan Africans. As Boule and Vallois (1957, pp. 285) report:

When we compare the dimensions of the bones of their limbs, we see that the leg was very long in proportion to the thigh, the forearm very long in proportion to the whole arm; and that the lower limb was exceedingly long relative to the upper limb. Now these proportions reproduce, but in greatly exaggerated degree, the characters presented by the modern Negro. …

The skulls likewise look non-European. The face is wide but not high. The nose is broad and flat. The upper jaw projects forward whereas the chin is weakly developed. The well-preserved dentition is not at all European. Among currently living populations, the ones who most closely resemble the Grimaldi humans seem to be the Khoisan peoples of southern Africa. Boule and Vallois (1957, pp. 290-291) write:

For our part, we have been greatly struck by the resemblances these Grimaldi Negroids bear to the group of South African tribes, the Bushmen and the Hottentots. Comparisons which we have been able to make with the material at our disposal, in particular with the skeleton of the Hottentot Venus, have led us to note, for instance, the same dolichocephalic character, the same prognathism, the same flattening of the nose, the same development of the breadth of the face, the same form of jaw, and the same great size of teeth. The only differences are to be found in the stature and perhaps in the height of the skull.

We know less about their soft-tissue characteristics. Alongside the skeletons were a number of female statuettes with big breasts, protruding bellies, full hips, and large buttocks. The hair seems to be short and matted (Boule & Vallois, 1957, p. 311).

These Grimaldi humans may have been ancestral to later European populations:

Verneau has investigated the survivals of the Grimaldi race at different prehistoric periods. He has first of all compared this type with the Cro-Magnon, which succeeded it in place. ‘At first sight’, he says, ‘the two races appear to differ greatly from each other; but on examining them in detail, we see that there is no reason why they should not have had some ties of kinship.’ Verneau even declares that the Grimaldi Negroids ‘may have been the ancestors of the hunters of the Reindeer Age’. (Boule & Vallois, 1957, p. 291).

Interestingly, Grimaldi-like humans are reported to have persisted in parts of Europe as late as the Neolithic:

Verneau likewise discovered, in both prehistoric and modern races, survivals or reappearances of the Grimaldi types.

‘In Brittany, as well as in Switzerland and in the north of Italy, there lived in the Polished Stone period, in the Bronze Age and during the early Iron Age, a certain number of individuals who differed in certain characters from their contemporaries’, in particular in the dolichocephalic character of their skull, in possessing a prognathism that was sometimes extreme, and a large grooved nose. This is a matter of partial atavism which in certain cases, as in the Neolithic Breton skull from Conguel, may attain to complete atavism. Two Neolithic individuals from Chamblandes in Switzerland are Negroid not only as regards their skulls but also in the proportions of their limbs. Several Ligurian and Lombard tombs of the Metal Ages have also yielded evidences of a Negroid element.

Since the publication of Verneau’s memoir, discoveries of other Negroid skeletons in Neolithic levels in Illyria and the Balkans have been announced. The prehistoric statues, dating from the Copper Age, from Sultan Selo in Bulgaria are also thought to portray Negroids. In 1928 René Bailly found in one of the caverns of Moniat, near Dinant in Belgium, a human skeleton of whose age it is difficult to be certain, but which seems definitely prehistoric. It is remarkable for its Negroid characters, which give it a resemblance to the skeletons from both Grimaldi and Asselar.

It is not only in prehistoric times that the Grimaldi race seems to have made its influence felt. Verneau has been able to see, now in modern skulls and now in living subjects, in the Italian areas of Piedmont, Lombardy, Emilia, Tuscany, and the Rhone Valley, numerous characters of the old fossil race
(Boule & Vallois, pp. 291-292).

Although the concept of atavism or ‘throwback’ is no longer widely accepted, there may have been some human groups in Europe that still looked African long after most had moved away from this phenotype. Indeed, if sexual selection were the cause, the phenotypic transformation should have occurred unevenly, beginning among populations on the former steppe-tundra of northern and eastern Europe, and then percolating outward through gene flow. In some peripheral regions, the transformation may still have been incomplete at the dawn of history.

Observations similar to those of Boule and Vallois have appeared elsewhere in the literature. Angel (1972) noted that 14% of skeletal samples from early Neolithic Greece displayed apparently Negroid traits, in contrast to later periods.

To be sure, there is a lot of pooh-poohing in the literature about the Grimaldi skeletons. Some say that the skeletal restoration must have been defective, or that the pressure of overlying layers had distorted the skull and the jaw, or that the apparently Negroid traits are of the sort that occur sporadically in Europeans.

For Carleton Coon (1962, p. 577), Europeans were ‘Caucasoid’ throughout the entire Upper Paleolithic:

There was, in fact, only one Upper Paleolithic European race. It was Caucasoid and it inhabits Europe today. We know this not only from skeletons but also from the representations of the human body in Upper Paleolithic art.

With reference to the Grimaldi skeletons, specifically their dentition, Coon (1962, p. 584) states:

These are dental characteristics of the Negro, but not exclusively. They are also seen on a number of teeth from Krapina and on those of Neanderthals, and are also present, as we have just mentioned, in the Mount Carmel population. An upper canine from the Magdalenian maxilla of Farincourt has the same features. The Grimaldi child was no more Negroid than the Palestinians of Skhul and many living Europeans of the Mediterranean region.

These statements are true, more or less. The dentition of sub-Saharan Africans does conserve many archaic characteristics that are absent in other modern humans but are present in Neanderthals, in the Skhul-Qafzeh hominins, and in other hominids (Irish, 1998). But the Grimaldi skeletons are clearly modern human. And while it is true that individual Negroid traits occur sporadically in living Europeans, it would be unusual, very unusual, for all of them to co-occur in a single living European. Finally, as Coon himself points out, other European skeletons from the Upper Paleolithic also show these traits.

Carleton Coon believed in the multiregional model of human evolution. He felt that European modern humans evolved out of European Neanderthals. So they could not have come from elsewhere. The Grimaldi skeletons, and others like them, must be an aberration … unless one accepts the ‘Out-of-Africa’ model. From the standpoint of this second model, we have one more piece in a puzzle linking modern Europeans to a demographic expansion that began to spread out of Africa some 50,000 years ago and that reached their continent about 35,000 years ago.

According to the Out-of-Africa model, the first modern humans in Europe could not have looked ‘white’. Indeed, they probably did not for at least the next 15,000 years. Their physical appearance seems to have changed, and radically so, within a later and relatively narrow time-frame, probably the second half of the last ice age.

The cause? It does not seem to have been natural selection, i.e., gradual adaptation to the ecological conditions of Europe. As I have discussed elsewhere, the cause is more consistent with an intensification of sexual selection of women, as a result of the unusually strong female-female competition for mates that prevailed among herd-hunting peoples in ice-age Europe. Since most genes are not sex-linked, this selection spilled over on to members of both sexes, thus modifying the appearance of the entire population (Frost, 2006).


Angel, J.L. (1972). Review of Blacks in Antiquity, American Anthropologist, 74, 159-160.

Bisson, M.S., Tisnerat, N., & White, R. (1996). Radiocarbon dates from the Upper Paleolithic of the Barma Grande. Current Anthropology, 37, 156–162.

Boule, M. & Vallois, H.V. (1957). Fossil Men. New York: Dryden Press.

Coon, C.S. (1962). The Origin of Races. New York: Alfred A. Knopf.

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

Holliday, T.W. (1997). Body proportions in Late Pleistocene Europe and modern human origins, Journal of Human Evolution, 32, 423-447.

Irish, J.D. (1998). Ancestral dental traits in recent Sub-Saharan Africans and the origins of modern humans. Journal of Human Evolution, 34, 81-98.

Soejima, M., Tachida, H., Ishida, T., Sano, A., & Koda, Y. (2005). Evidence for recent positive selection at the human AIM1 locus in a European population. Molecular Biology and Evolution, 23, 179-188.

Voight, B.F., Kudaravalli, S, Wen, X, Pritchard, J.K. (2006). A map of recent positive selection in the human genome. PLoS Biology, 4(3), e72 doi:10.1371/journal.pbio.0040072


Anonymous said...

Dear Pr. Frost,

the ancestors of mongoloid people were also game-hunters during the Ice Age, however they were not selected for light hair and eyes..?

Anonymous said...

The key factor in raising male mortality is not game hunting per se, but game hunting over very long distances, as in hunting of migratory herds over continental steppe-tundra. This environment did exist in Ice-age Asia but was much colder and drier, being closer to the Arctic Circle and further from the Atlantic Ocean.

In Europe, the Scandinavian icecap pushed the steppe-tundra zone much further south, thereby raising its bioproductivity and potential for human settlement.

In Asia, the steppe-tundra zone could not continuously sustain a substantial human population upon which sexual selection could operate. This was so throughout most of the ice age.

Anonymous said...

Here is a speculative thought. This hypothesis may help explain the dark skins of the Tasmanians. They lived isolated in Tasmania without seacraft capable of traversing Bass Straight (between Tasmania and Australia). Tasmanian latitudes south are equivalent to the latitudes north of modern New York or Florence. In their island isolation they probably would not have had any opportunity to engage in the long distance game hunting of people on the northern steppes. Without this selective pressure they maintained their dark skins in a relatively high latitude environment contra the more common climatic / environmental adaptation explanation for light skin colours.

mathilda said...

I've seen two ginger variants on MCR1 dated to about 30k and 80k. Is this wrong? I've stashed the article with the dates somwhere.

If it were correct, wouldn't being ginger have had some effect on skin tone? I've never seen a dark skinned ginger mixed race kid.

mathilda said...

"The hair seems to be short and matted (Boule & Vallois, 1957, p. 311)"

I thought they are now believed to be wearing hats, as they show some other woven clothing on them too. The Venus of Willendorf, in a bobble hat. LOL.

Anonymous said...

Re - "diversity..of hair colour..being most prominent in a zone centred on the East Baltic"

I think your theory is the only explaination that makes sense so this is like carping, but I must demur if the quote means red or golden hair and the accompanying subtle skin tones are more common in the Baltic. Red hair is most common in the Scottish Borders.

The low regard in which it is held seems to have made Coon and co. uninterested in making maps on its occurrence, it exists nonetheless.
On golden hair I have more support to point out that it is not centred on the Baltic, Coon mentions it in the West of Ireland (Arran Isles especialy) where red hair is also common.
The population of the west coast of Ireland is beleived to be relatively Paleolithic (judging by their appearance and the Y chromosome data). Could red hair(with light skin) have been the first sexually selected change to non-whiteness. I beleive red hair is older than other hair colours

Anonymous said...

I enjoyed your article. However, my years of experience with men in general (I am a 60-year old woman) and my knowledge of history, lead me to theorize a different process.
While gentlemen, and indeed all men, may prefer blondes, they seem to mate with women rather indiscriminately. The Mesolithic hunters were not likely monogamous. When they came back from the hunt bearing bacon, they may have used it to lure the more fair-skinned women. However, that would not likely have kept the darker-skinned women from mating with them, too, even if only for scraps, in the hope of fair treatment, or for the dark women's own sexual desires. Black and Native American women during the slavery years in the United States could be called to witness.
Now, it may be that the fair ladies received more of the food, putting their offspring at an unfair advantage for attaining reproductive age. Or it may be that the more fair-skinned offspring fared better in the gloomy Ice Age because their skin was able to make more Vitamin D. It even may be that fair skin provided protective coloration or a hunting advantage in a land of snow and ice. Or the genes for fair coloration may have been dragged along because they assist the genes for some other natural advantage. Any of those cases would be an example of natural selection.

Unknown said...

I am still trying to understand how an african people could become fairskinned; being of european extraction on all sides of my ancestors, I find your article somewhat helpful. I do know that we are very dependent on Vitamin D. But where did CroMagnon man come from? any explanation? were they "white"? and how did they become tall? So many questions, so little time. Thanks!

Unknown said...

Your theory is wrong firstly evolution doesn't take ten to twenty thousand years but millinia and also all these skeletons show clearly that black people were the first to inhabit Europe before the whites invaded Europe from Central Asia. Red hair and blue, green eyes are synonymus with albinism, end of story.

Anonymous said...

"Your theory is wrong firstly evolution doesn't take ten to twenty thousand years but millinia and also all these skeletons show clearly that black people were the first to inhabit Europe before the whites invaded Europe from Central Asia. Red hair and blue, green eyes are synonymus with albinism, end of story."

You are ridiculous. And completely wrong. "Black" people did not inhabit Europe in any record. WHG were slightly darker toned than modern Europeans and they had blue eyes. That doesn't make them "black" or "sub-Saharan". We see direct lineage of living Europeans to WHG. There is no WHG related found in sub-Saharan Africa. Dark skin does not automatically mean sub-Saharan African. That is absurd. Southern Asians are not sub-Saharan and they certainly can have black skin.

The alleles for skin lightening and hair lightening have no connection to albinism or Neanderthal heritage. The lightening alleles that exist developed outside of Africa. Skin darkening alleles developed within Africa, which is why latter sub-Saharan African populations can be far darker than older lineages such as the Khoisan. The Khoisan can not be considered "black" by any genetic measure. They defy racial classification and it's likely all humans stem from proto Khoi-San.

Red hair alleles developed in the Levant and blond hair alleles developed in Asia, in at least 3 different locations. The WHG blue eyes are not found in Africa anywhere, but they are found in latter lineages in the Levant and North Africa. There were many obvious back migration to Africa throughout antiquity and prehistory. That is how V88 arrived in Africa and how North Africa became populated with skin and eye lightening alleles. Which again have no connection to albinism, which is a specific genetic deviation, separate from the usual lightening alleles.

Anonymous said...

This entire post and source material are unverified, non peer reviewed and contradict all known academic work in this area. This is some clear Uthman dan Fodio Institute rubbish. Anyone and everyone related to that faux "institute" are crackpot theorists pushing agenda and propaganda narratives.

Anonymous said...

And there are "dark" skin, "ginger" people. The skin tone is not VERY dark, but certainly a light brown.

Red hair alleles are in no way related to albinism or Neanderthal heritage. Blue eyes come from multiple sources, including WHG and CSP.

There were numerous back migrations to Africa, which brought Neanderthal dna and multiple haplotypes as well as alleles for lightening, red hair, blue and green eyes, which can largely be seen in the North African population as it is only partially sub-Saharan in origin and mostly in the last 1300 years or so. Prior to that, going all the way back to the Neolithic at least, North Africa was largely populated by a light skinned people who developed into latter North African lineages from Mauritania to Libya, as well as settling the Canary Islands.

You should ignore any crack pot theorists, serving up nonsense propaganda, from the Uthman Dan Folio Institute, Clyde Winters, Erhan Elhaik and various other heterodox faux academics pushing pre-designed narratives and dismissing actual findings in the fields involved.