Saturday, July 27, 2013

Skin color and the menstrual cycle

The female torso visibly reddens towards the end of the menstrual cycle. Do men unconsciously pick up on this visual cue? (Figure from Edwards and Duntley, 1949)

Women vary in skin color over the menstrual cycle. From mid-cycle on, their skin steadily reddens because of an increase in blood flow that peaks in the day or two preceding menstruation.This cyclical “blushing” mainly affects the torso:

[…] these cyclic variations in blood flow were observed over the entire trunk and at least the upper parts of the limbs. There is suggestive evidence that the face and the hands and feet may share in these changes. (Edwards and Duntley, 1949)

Facial skin, especially around the eyes, shows a similar cyclical change, according to two questionnaire surveys: 

[…] about half the women questioned had some increase in skin pigmentation, which was noted in every case in the latter days of the menstrual cycle and in some cases during menstruation also. The others showed no skin changes whatsoever.

[…] The site most commonly showing pigmentation changes was the skin around the eyes. Next most frequently affected were the areola of the nipple and the perioral skin. The forehead, axilla, and abdomen were affected in less than one-third of the "positive" subjects. (McGuiness, 1961) 

In answer to the questionnaire, 18 women (62%) consistently noticed darkening of the peri-ocular skin towards the end of the menstrual cycle, i.e. immediately prior to the onset of menstruation; of these, three also noticed darkening of the nipple areolae, two the forehead skin and one the peri-oral skin. (Snell and Turner, 1966)

Snell and Turner (1966) confirmed these observations by measuring the percentage of light reflected by facial skin, although the cyclical variation was rather small. Unlike the torso, the face doesn’t redden towards the end of the cycle. Instead, it becomes browner through increased melanocyte activity:

The results from the skin reflectance readings did not show any great changes. The readings obtained from the cheek and lower eyelid indicated that the melanin content of the skin in these regions tended to rise in the later part of the cycle in many of the women. 

[…] The melanocytes of the anterior abdominal wall skin over the linea alba showed no changing pattern of activity at different phases of the menstrual cycle.

[…] It was concluded that a proportion of normal women, especially dark-skinned brunettes, have darkening of the facial skin during the later days of the menstrual cycle and this mainly involves the peri-ocular skin. (Snell and Turner, 1966) 

Does this cyclical variation provide men with a means to assess female fertility? An unconscious means, to be sure. Pierre van den Berghe thought so, but I ignored his gentle prodding and avoided the subject, all the more so because a search of the ethnographic literature failed to turn up any awareness in any human society of this cyclical change. In contrast, many societies have been keenly aware that women are fairer-skinned and men darker-skinned, often to the point of making this sex difference an artistic convention (van den Berghe and Frost, 1986; Frost, 1988; Tegner, 1992).

One research team has tried to find out whether men pick up on this cyclical variation:

Here, in an initial pilot study, we test the hypothesis that changes in female facial skin coloration across the menstrual cycle could be one of the signals that men have adapted to in order to assess female fertility. Spectrophotometric measurements of the facial skin color of normally ovulating Caucasian women (aged 24–29 years) were collected in the late follicular and midluteal phase of their menstrual cycle. Facial images were also taken in both sessions and judged for attractiveness and health by a panel of German men (aged 16–37 years). In line with Roberts et al. (2004), our results show that men perceive women in the late follicular phase to be significantly more attractive and healthier than those in the midluteal phase. However, we did not detect any significant differences in objective measurements of skin color between the two phases. (Samson et al., 2011)

This study suffers from a few flaws. The authors measured skin reflectance on the forehead and the cheeks, yet these body sites are less involved in darkening and lightening of female skin over the menstrual cycle. It would have been better to measure skin reflectance around the eyes (although premenstrual peri-ocular darkening might have likewise been absent in the fair-skinned German participants). Better yet, this study should have focused not on the face but on the torso, since that body region is the one most affected by this cyclical variation. We should also keep in mind that men unconsciously use two different aspects of female pigmentation for gender recognition. One is the lighter skin of a woman’s face. The other is the higher contrast between facial skin color and eye/lip color (Russell, 2009; Russell, 2010; Porcheron et al., 2013; see also Dupuis-Roy et al., 2009). This contrast effect might be weakened by the premenstrual darkening of skin around a woman’s eyes.

A more recent study has corroborated that men prefer faces of ovulating women to those of premenstrual women (Bobst and Lobmaier, 2012). It concludes that subtle changes in face shape are responsible, although changes in skin color cannot be ruled out. In fact, if men can respond to such subtle changes in face shape, they should also be able to respond to changes in facial color that are no less subtle.

It may be worthwhile to take another look through the ethnographic literature. One of my wonderful commenters has pointed me to an article by a sociologist of Zambian origin, Mwizenge S. Tembo:

It is [with] the frequent circulation of the hormones, the increased flow of blood during ovulation, and especially during pregnancy that women in Sub-Saharan Africa may acquire a characteristic mild to quite remarkable red-orangeish glow to their skin. In fact an obvious tell-tell sign of being pregnant among married women and also among young girls who may have had sex out of wed lock, even when the pregnancy is not even physically visible, is the characteristic lightening of the skin-tone whether the woman is light or very dark. Among the Tumbuka, Chewa, Nsenga, and Ngoni people of Eastern Zambia, several terms are used to describe the state of being pregnant. “Ali ndi pakati” means that “the woman is in between”. Because the majority of women in African societies prior to modern medicine had very high deaths and faced danger during the birthing process, the woman was said to be literally “between life and death” or “living with uncertainty”. “Ali ndi mimba” means “the woman has a stomach” referring to the obvious bulging stomach of a pregnant woman. The most relevant term to this discussion is “ali ndi pathupi” which means “the woman has a body” (Salaun, 1969; Price, 1970) which refers to the characteristic light skin tone or the visible obvious glow the woman assumes when she gets pregnant. 

This light reddish skin is considered desirable by African men and may serve an adaptive purpose: “Among many other possible explanations, the most compelling may be that the lighter skin, even among the darkest of indigenous Africans, may have been a normal and natural biological marker and signal that the woman was very fertile.” Tembo is a fan of evolutionary psychology and may be indirectly echoing a meme that began with me and Pierre van den Berghe. Nonetheless, there may indeed be more awareness of this menstrual change in skin color than I had thought, particularly in settings where most variation in skin color is intra-ethnic.


When all is said and done, this research topic may still be ‘a bridge too far.’ Admittedly, a researcher should have little trouble finding out whether the premenstrual darkening of the eye area is a sexual turnoff for men. I’m sure it is—many women certainly seem to think so. But how would one determine whether this male response is hardwired or not? By measuring it as a function of testosterone levels? Finally, would such a hardwired mental algorithm shed light on other feelings towards skin color?

It might be more interesting to investigate how men respond to the premenstrual reddening of the female torso region. Recruitment of female participants would nonetheless be much more difficult, as would be the task of getting approval from the research ethics committee. There’s also the little matter that this premenstrual ‘blushing’ is visible only in light-skinned women. Conversely, premenstrual darkening of the eye area is visible mainly in darker-skinned women. 


Bobst, C., and J.S. Lobmaier. (2012). Men's preference for the ovulating female is triggered by subtle face shape differences, Hormones and Behavior, 62, 413-417.

Dupuis-Roy, N., I. Fortin, D. Fiset, and F. Gosselin. (2009). Uncovering gender discrimination cues in a realistic setting, Journal of Vision, 9(2), 10, 1–8.

Edwards, E.A. and S.Q. Duntley. (1949), Cutaneous vascular changes in women in reference to the menstrual cycle and ovariectomy, American Journal of Obstetrics and Gynecology, 57, 501-509.

Frost, P. (1988). Human skin color: a possible relationship between its sexual dimorphism and its social perception, Perspectives in Biology and Medicine, 32, 38-58.

McGuiness, B.W. (1961). Skin pigmentation and the menstrual cycle, British Medical Journal, 2, 563.

Porcheron, A., E. Mauger, and R. Russell (2013). Aspects of facial contrast decrease with age and are cues for age perception. PLoS ONE 8(3): e57985

Russell, R. (2010). Why cosmetics work. In Adams, R., Ambady, N., Nakayama, K., & Shimojo, S. (eds.) The Science of Social Vision. New York: Oxford.

Russell, R. ( 2009). A sex difference in facial contrast and its exaggeration by cosmetics, Perception, 38, 1211-1219

Samson, N., B. Fink, and P. Matts. (2011). Does a woman’s skin color indicate her fertility level? Preliminary findings, Swiss Journal of Psychology/Schweizerische Zeitschrift für Psychologie/Revue Suisse de Psychologie, 70(4), 99-202.

Snell, R.S. and R. Turner. (1966). Skin pigmentation in relation to the menstrual cycle, Journal of Investigative Dermatology, 47, 147-155.

Tegner, E. (1992). Sex differences in skin pigmentation illustrated in art, The American Journal of Dermatopathology, 14, 283-287.

Tembo, M.S. (2010). The Rediscovery of the Beautiful Woman in African Societies. Eurocentric Destruction of Indigenous Conceptions: the Secret Rediscovery of the Beautiful Woman in African Societies.
van den Berghe, P. L. and 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.

Vargas-Guadarrama, L. (1971). Pigmentation cutanée et cycle menstruel, Paris, Université Paris VII, Thèse de doctorat.

Saturday, July 20, 2013

Trading in fair-skinned women. Did it happen elsewhere?

Austronesian woman (Roekiah Soeara, 1942, Indonesian actress - source). Austronesian and Papuan peoples intermixed in coastal Papua-New Guinea and on the islands to the east. This intermixture seems to have been mainly due to Austronesian women joining polygynous Papuan households. Did this happen through peaceful exchange (brides for land?) or through raiding and kidnapping? 

In all human populations, the sexes differ somewhat in skin color, women looking paler and men browner and ruddier. This sex difference is mirrored by a cross-cultural tendency to make lighter skin a female norm, which women often accentuate by various means (e.g., staying out of the sun, wearing sun-protective clothing, applying white facial powders or skin-bleaching preparations). Traditionally, this norm was said to be ‘white’ in Europe and East Asia, ‘golden’ in South-East Asia, and ‘red’ in sub-Saharan Africa (van den Berghe and Frost, 1986).

Why are women lighter-colored than men? Some ethologists have argued that light skin is one of several infant traits that the adult female body has adopted to calm aggressive impulses in men and induce caring behavior. This visual stimulus would thus influence male sexual response without being erogenous in and of itself. Whatever the ultimate cause, traditional social environments have tended to make women’s lighter skin a criterion of mate choice, often a leading one (Frost, 2011).

Evidently, skin color varies not only between men and women but also between different human populations. What happens when people become aware of the second kind of skin-color variation? Specifically, what happens to the feelings associated with the first kind? How are they transposed into this new social context?

One result may be a form of trade: women from lighter-skinned populations will become objects of commerce for sale to men in darker-skinned populations. This was the case between the 8th and 19th centuries, when women were exported from Europe to the Muslim world, i.e., Spain, North Africa, the Middle East, and South Asia (see previous post).

Has this kind of trade developed elsewhere? To be economically viable, it should meet certain conditions:

1. The two populations differ enough in skin color to make the trade worthwhile.

2. There is enough supply, i.e., the women are obtained for trade on a large enough scale through local wars or slave raiding.

3. There is enough demand, i.e., the male clients are polygynous enough and wealthy enough.

These conditions came together with the rise of the Muslim world to geopolitical dominance in the 8th century. Elsewhere, and at other times, the conditions were less optimal. People usually had little contact with other people whose skin color substantially differed from their own. In pre-Columbian America, for instance, it’s difficult to see how such trading could have developed, given the slight differences in skin color among different Amerindian groups.

Nonetheless, there are a few intriguing examples, albeit on a small scale:

Sub-Saharan Africa

Skin color does visibly differ among the various peoples of sub-Saharan Africa, to a greater degree in fact than what many non-Africans might think. These differences seem to interact with notions of sexual beauty, as Lugira (1970) writes about pre-colonial Uganda:

The Ganda concept of skin pigmentation considers light coloured complexions to be differing shades of white.  A dark brown skin colour is said to be — eruyeru, that is, somewhat white. A really brown‑reddish‑yellow person is said to be mweru = white, which in comparison would be considered to be blonde; and this in the Ganda aesthetic language is considered as red = myufu, the most perfect skin pigmentation.

As a result, the lighter skin of some groups could become a casus belli:

[…] The Nnyambo people were the handsome looking (brown‑red) inhabitants of the south of Buganda, in the Ziba countries and Kalagwe. The Nnyambo women were one of the reasons that induced Suna II to wage war with Kiziba after which he suffered from small pox and died (Lugira, 1970)

South-East Asia

An incipient trade of this sort existed in 19th- and 20th-century Thailand, where some Chinese merchants would offer their daughters to Thai rulers in exchange for protection and influence:

'Chinese of wealth', wrote the American missionary N. A. McDonald in 1884, 'often become favorites with the rulers and receive titles of nobility, and these noblemen in return present their daughters to Their Majesties.'

[…] William Skinner noted that Chinese women were 'prized for their light skin color'. Here, skin colour was valued as a form of feminine beauty and a sign of 'Chineseness'. (Jiemin, 2003)

This kind of exchange was consistent with indigenous Thai notions of female beauty, as shown by a recent study:

Young women in all four regions of Thailand considered ‘bright face skin’ and ‘white-pink (body) skin’ as the next most important physical appearance characteristics. Women in the North region were most concerned about having bright face skin, perhaps because they already tend to have lighter body skin which is viewed as desirable. Women in South region were most concerned with body skin color, which may be because they tend to have darker skin color. (Rongmuang et al., 2011)

Papua New Guinea / Melanesia

Finally, lighter-skinned women may have been objects of exchange in coastal Papua New Guinea and on the islands to the east (New Britain, New Ireland, the Solomons, Fiji). This area was a zone of intermixture between two streams of settlement: Papuans with dark brown if not black skin and Austronesians with light brown skin. Interestingly, this intermixture mainly took the form of Papuan men pairing with Austronesian women, as shown by comparison of paternally-transmitted Y chromosomes and maternally-transmitted mtDNA:

[…] This genetic admixture was most likely male biased involving mostly Austronesian women and, over time, mostly New Guinean men, resulting in a higher proportion of Melanesian than Asian Y-chromosome together with a higher proportion of Asian than Melanesian mtDNAs as observed in contemporary Polynesians (Mona et al. 2007; see also Kayser et al. 2006)

[…] mtDNA and NRY analyses indicate that this admixture was sex biased (Melton et al. 1995; Kayser et al. 2000; Su et al. 2000; Hurles et al. 2002): about 94% of Polynesian mtDNAs are of Asian ancestry, whereas about 66% of Polynesian Y chromosomes are of Near Oceania ancestry (Kayser et al. 2006). Although the mtDNA support for this sex-biased admixture hypothesis has recently been questioned (Soares et al. 2011), genome-wide SNP data do indicate significantly more Asian versus New Guinea ancestry for the X chromosome of Polynesians than for the autosomes (Wollstein et al. 2010), in agreement with the sex-biased admixture scenario. In addition, Papuan-speaking groups in New Guinea show higher frequencies of Asian mtDNA haplogroups than of Asian NRY haplogroups (Kayser, Choi, et al. 2008).

[…] Overall, the mtDNA haplogroups in the Solomons are predominantly of Asian [i.e., Austronesian] origin, whereas the NRY haplogroups are predominantly of NO [Near Oceania, i.e., Papuan] origin. (Delfin et al, 2012)

Since Papuans are much more polygynous than Austronesians, and in the past more patrilocal, this intermixture was probably due to Austronesian women traveling over some distance and joining polygynous Papuan households.1 How and why is anyone’s guess. Peaceful exchange? “Give us some of your land and we’ll give you some of our women?” Or was it raiding and kidnapping?

In either case, the two groups were probably keenly aware that one of them was lighter-skinned and the other darker-skinned. Even today, after millennia of intermixture, color consciousness remains strong in this contact zone, as noted by a study of the Eastern Solomons:

The Lau were conscious of skin color; some parents, particularly mothers, tried to dissuade their sons from marrying much better-educated girls from the Western Solomons because of their dark skins. Although color consciousness is decreasing, until quite recently clans and persons of higher status have been generally lighter, and marriages have been preferentially within clan or class status levels. (Baldwin and Damon, 1973)

According to an origin myth from New Britain, these differences in skin color arose through the marriage choices of two brothers:

To-Kabinana said to To-Karvuvu, “Do you get two light-coloured coco-nuts. One of them you must hide, then bring the other to me.” To-Karvuvu, however, did not obey, but got one light and one dark nut, and having hidden the latter, he brought the light-coloured one to his brother, who tied it to the stern of his canoe, and seating himself in the bow, paddled out to sea. He paid no attention to the noise that the nut made as it struck against the sides of his canoe nor did he look around. Soon the coco-nut turned into a handsome woman, who sat on the stern of the canoe and steered, while To-Kabinana paddled. When he came back to land, his brother was enamoured of the woman and wished to take her as his wife, but To-Kabinana refused his request and said that they would now make another woman. Accordingly, To-Karvuvu brought the other coco-nut, but when his brother saw that it was dark-coloured, he upbraided To-Karvuvu and said: “You are indeed a stupid fellow. You have brought misery upon our mortal race.  From now on, we shall be divided into two classes, into you and us.” Then they tied the coco-nut to the stern of the canoe, and paddling away as before, the nut turned into a black-skinned woman; but when they had returned to shore, To-Kabinana said: “Alas, you have only ruined our mortal race.  If all of us were only light of skin, we should not die.  Now, however, this dark-skinned woman will produce one group, and the light-skinned woman another, and the light-skinned men shall marry the dark-skinned women, and the dark-skinned men shall marry the light-skinned women.” And so, To-Kabinana divided mankind into two classes. (Gray, 1916, p. 108)


Light-skinned European women became objects of commerce because of an unusual set of circumstances, essentially the relative dominance of the Muslim world and, correspondingly, the relative weakness of the European world.

Circumstances may come and go, but basic notions of human beauty change less quickly. In the near future, a similar situation may develop in response to the impoverishment of common people in Europe and North America and the growing affluence of elites in the Third World, particularly in resource-rich countries.


1. The literature also puts forward the reverse scenario as a possible explanation, i.e., Papuan men marrying into Austronesian communities. In this second scenario, Papuan men would have had to renounce not only patrilocality but also polygyny and low paternal investment. This is a more radical behavioral change than the one associated with Austronesian women marrying into Papuan communities. Elsewhere in the world, we have two other examples of a high-polygyny population coming into contact with a low-polygyny one: Bantu and Khoisans in southern Africa and Bantu and Pygmies in central Africa. In both cases, intermixture has almost wholly involved women moving from the low-polygyny population to the high-polygyny one. There has been little if any movement of men in the other direction.


The Spanish online journal La Tercera Cultura has recently translated and published one of my posts: “Cómo se llegó a pacificar Europa.”


Baldwin, J.C. and A. Damon. (1973). Some genetic traits in Solomon Island populations, American Journal of Physical Anthropology, 39, 195-201.

Delfin, D., S. Myles, Y. Choi, D. Hughes, R. Illek, M. van Oven, B. Pakendorf, M. Kayser, and M. Stoneking. (2012). Bridging Near and Remote Oceania: mtDNA and NRY Variation in the Solomon Islands, Molecular Biology and Evolution, 29(2), 545–564.

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. (2010). Femmes claires, hommes foncés. Les racines oubliées du colorisme, Quebec City : Presses Universitaires de Laval.

Gray, L.H. (1916). The Mythology of All Races, Vol. 9 Oceanic, Boston: Marshall Jones.

Jiemin, B. (2003). The Gendered Biopolitics of Marriage and Immigration: A Study of Pre-1949 Chinese Immigrants in Thailand, Journal of Southeast Asian Studies, 34(1), 127-151.

Kayser M, S. Brauer, R. Cordaux R, et al. (2006). Melanesian and Asian origins of Polynesians: mtDNA and Y chromosome gradients across the Pacific, Molecular Biology and Evolution, 23, 2234-2244.

Lugira, A.M. (1970). Ganda Art, Kampala: Osasa pub.

Mona, S., M. Tommaseo-Ponzetta, S. Brauer, H. Sudoyo, S. Marzuki, and M. Kayser. (2007). Patterns of Y-Chromosome Diversity Intersect with the Trans-New Guinea Hypothesis, Molecular Biology and Evolution, 24 (11), 2546-2555.

Rongmuang, D., B.J. McElmurry, L.L. McCreary, C.G. Park, A. Miller, and C. Corte. (2011). Regional Differences in Physical Appearance Identity among Young Adult Women in Thailand, Western Journal of Nursing Research, 33(1), 106-120.

Van den Berghe, P. L. and 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.

Saturday, July 13, 2013

The other slave trade

Although the slave raiders usually left infants behind, older girls and boys could be taken, if suitable for immediate sale (Selling a child-slave, painting by Vasily Vereshchagin – source).

Europe used to export slaves to the non-European world. Such a statement would astonish most people today, even among the university-educated. Surely, those slaves were few in number, certainly fewer than the African slaves taken across the Atlantic.  And surely all of that happened long before the Atlantic slave trade. 

Well, no and no. The numbers were huge. At the height of that trade, over 10,000 Eastern Europeans were enslaved each year between 1500 and 1650 for export to North Africa, the Middle East, and South Asia … a total of 1.5 million. By comparison, the Americas received fewer than 300,000 African slaves before 1600 and another 1.5 million between 1600 and 1700 (Fisher, 1972; Kolodziejczyk, 2006). Western Europeans were likewise enslaved and taken abroad, mainly to North Africa. How many? More than 1 million between 1530 and 1780 (Davis, 2004).

And, yes, those European slaves were going across the Mediterranean while African slaves were going across the Atlantic. Officially, the “harvest” ended with the Treaty of Carlowitz (1699), which called on the Ottoman Empire to stop all slave raiding (Abou-el-Haj, 1969). Unofficially, it did not end on a large scale until Russia annexed the Khanate of the Crimea in 1783—a quarter-century before the abolition of the slave trade in the British Empire. Fair-skinned women were thereafter exported on a smaller scale until the late 19th century, mainly from the Caucasus.  

Differences from the African Slave Trade

The white slave trade differed from its black counterpart in a few key ways. In Africa, a slave-trader typically purchased prisoners of war who had already lost their freedom through local conflicts. In Europe, he played a more active role.  

This was the case with the Crimean Tatars who lived under Ottoman protection in the Black Sea region. Beginning in the mid-15th century, they would fan out each year on raids into what is now Ukraine and southern Russia. These raids served no military purpose, being driven by the profits to be made in the slave trade:

[…] most of these raids do not appear to have had any military purpose and, moreover, had little or no relationship to Ottoman policy. They were an integral part of the Crimean economy, a "harvesting of the steppe" as the Tatars explained it. (Fisher, 1973)

In a royal [Polish] document dated 1555 we read: “There are many Turks who send Tatars supplied with their horses and armour into our domains, and later share the profits in the fields”, this last expression referring to the fact that the division took place far from the eyes of the Ottoman police and customs officers who might have viewed negatively the breaking of the peace treaty or the failure to pay taxes due by those involved in the slave trade.  

Notwithstanding such efforts to escape the tax duties, the Ottoman state was among the principal share holders in the Black Sea slave trade. According to Pretwicz, the sultan's income from the slave trade in Akkerman and Ocakiv (Turkish Özü ) amounted to a few 100,000 akçe a year. Strikingly similar are the numbers for Caffa established by Inalcik on the basis of Ottoman tax registers. The slave tax collected in Caffa amounted to 620,000 akçe in 1520 and 650,000 akçe in 1529.26 The same author estimates the total state revenue from the slave trade as approaching 100,000 gold florins (i.e. circa 6,000,000 akçe) in the mid 16th century. (Kolodziejczyk, 2006)

The white slave trade was different in a second way. Most black slaves were destined for physical labor on plantations. There was thus a stronger preference for men over women. In contrast, white slaves were used more for domestic service, particularly concubinage and marriage. There was thus a stronger preference for women, as reflected in the sex ratio of the slave population: black slaves were predominantly male, and white slaves predominantly female. Furthermore, while blacks of both sexes sold for the same price, Russian and Circassian women fetched 50% more than men of the same nationality. (Verlinden, 1977, pp. 211, 224, 306, 315, 330-331, 460, 517; see also Frost, 1990). This price differential continued until the end of white slavery. A mid-19th century report from Turkey states that a “trained, strong, black slave” would cost 4,000 to 5,000 piasters, whereas “white slave girls of special beauty” were worth 50,000 piasters or more (Lewis, 1990, p. 13).  


Slave trading existed in many parts of the world and during many historical periods. Trading in fair-skinned women, however, was much more limited in space and time. There is no evidence of it during Roman times, at least not on a large scale. If a Roman notable wanted a bride with milk-white skin, he would look among the families in his entourage and not among the slaves at the local market. After all, a native-born woman of good family would bring a dowry and valuable family connections.

All of this changed in the 7th century with the dramatic expansion of the Arab world into the Middle East and thence into North Africa and Spain. The new elites were darker in skin tone and, also, more polygynous. It was these two factors that would fuel demand for fair-skinned brides and concubines. 

A third factor was of course the relative weakness of European societies, particularly during the Dark Ages that followed the collapse of the Roman Empire. With the gradual strengthening of European states, this trade increasingly took the form of hit-and-run raids that focused on poorly defended areas, such as the plains north of the Black Sea. This raiding would finally end only with European annexation of those “states” that earned most of their income from the slave trade, such as the Khanate of the Crimea and the Beyliks of North Africa.

Would this trade have continued if Europe had remained weak? Probably. Would it have eventually become more humane and sustainable? Doubtful. Though often described as “harvesting,” there never was any effort to make it sustainable. A Tatar raid typically left behind the old and the very young, as a Polish report noted: “In the fields and forests they [i.e. the Tatars] left behind over 200 poor children whom they could not take along since everyone preferred to take horses and oxen rather than children” (Kolodziejczyk, 2006).

The result was widespread depopulation of much of Ukraine and southern Russia, which in turn forced the Tatars to raid farther and farther afield, even as far as present-day Poland. Demographic wastage was considerable: 

The Crimean Tatar society was based on raiding the neighbouring Slavic and Caucasian sedentary societies and selling the captives into the slave markets of Eurasia. Approximately 75 percent of the Crimean population consisted of slaves or freedmen, and much of the free population was highly predatory, engaged either in the gathering of slaves or in the selling of them. It is known that for every slave the Crimeans sold in the market, they killed outright several other people during their raids, and a couple more died on the way to the slave market. (Britannica, 2013)

There was no resource management, only resource depletion (Wikipedia, 2013). As Kolodziejczyk (2006) notes:

We should not close our eyes to the consequences of depopulation, affecting large Slavic territories in Eastern Europe. If an "alternative" history of Ukraine were imaginable, perhaps the country's historical development would have looked different had it not been for the slave trade.

Ukraine is considered to be part of ‘Old Europe’ yet the plains north of the Black Sea were finally opened for settlement at about the same time as the plains of the United States and Canada. 


Abou-el-Haj, R.A. (1969). The Formal Closure of the Ottoman Frontier in Europe: 1699-1703, Journal of the American Oriental Society, 89(3), 467-475. 

Britannica. (2012). “Slavery” in Encyclopedia Britannica’s Guide to Black History. 

Davis, R. (2004). Christian Slaves, Muslim Masters: White Slavery in the Mediterranean, the Barbary Coast, and Italy, 1500-1800, Palgrave-Macmillan.

Fisher, A.W. (1973). Azov in the sixteenth and seventeenth centuries, Jahrbücher für Geschichte Osteuropas, Neue Folge, 21(2), 161-174. 

Fisher, A. (1972). Muscovy and the Black Sea slave trade, Canadian American Slavic Studies, 6, 575-594.

Frost, P. (1990). Fair women, dark men: the forgotten roots of colour prejudice, History of European Ideas, 12, 669-679.

Kolodziejczyk, D. (2006). Slave hunting and slave redemption as a business enterprise: The northern Black Sea region in the sixteenth to seventeenth centuries, Oriente Moderno, 86, 1, The Ottomans and Trade, pp. 149-159.

Lewis, B. (1990). Race and Slavery in the Middle East, New York: Oxford University Press. 

Verlinden, C. (1977). L’Esclavage dans l’Europe médiévale, vol. II, Ghent.

Wikipedia (2013). Crimean-Nogai Raids,

Saturday, July 6, 2013

White skin privilege

A new arrival (painting by Giuilo Rosati - source). The privilege of white skin …

Earlier this year, fashion model Cameron Russell condemned the unbearable whiteness of her industry:

[…] I won a genetic lottery, and I am a recipient of a legacy. For the past few centuries, we have defined beauty not just as health and youth and symmetry that we’re biologically programmed to admire, but also as tall, slender figures with femininity and white skin. This is a legacy that was built for me, and that I’ve been cashing in on. (Russell, 2013)

Yes, Ms. Russell did win a genetic lottery, being certainly more attractive than average. But she also mentioned a second unearned windfall: a beauty privilege due to the “legacy” of the past few centuries, when Europeans lorded over the world. Without that legacy, she would presumably be a very ordinary woman, perhaps even ugly.

This presumption can be tested. There was a time, not so long ago, when Europeans were weaklings on the world scene, when large parts of their continent were ruled by other peoples, and when the center of geopolitical power lay in the Middle East. In such a context, women like Cameron Russell would have had much less beauty privilege to cash in on.

In reality, they had plenty, and not just in Europe. ‘White slavery’ today means the international trafficking of women for prostitution. Back then, it meant the provisioning of the Muslim world with European concubines, who were valued for their white skin (Lewis, 1990, pp. 11-13, 56, 72). This trade was considerable in Muslim Spain:

The same convoys of booty also included women, these Frankish women who were all the more sought after in Cordova because they were blond and fair-skinned. It was among them, as among the captive women from Gascony, that the Umayyad princes chose their most pampered concubines and who, once they became mothers, were themselves raised to the rank of veritable princesses, of proven sultanesses (umm walad) who were influential and quick to enter, with the assistance of Slav eunuchs, into secret and complicated palace intrigues. But the Frankish women did not populate only the caliph's harems; the dignitaries of the khassa and the rich burghers of the cities also procured them at lavish prices, like, in the modern period, the Circassian women who have so curiously tinted the upper classes of oriental Muslim society. (Lévi-Provençal, 1953, p. 179)

Such women came from places that were poorer and less advanced than the Muslim world. Neither they nor their future masters knew what white skin would signify over a half-millennium later. Indeed, no one foresaw the rise of Europe to geopolitical preeminence, certainly not this 11th-century Muslim author:

For those who live furthest to the north between the last of the seven climates and the limits of the inhabited world, the excessive distance of the sun in relation to the zenith line makes the air cold and the atmosphere thick. Their temperaments are therefore frigid, their humors raw, their bellies gross, their color pale, their hair long and lank. Thus they lack keenness of understanding and clarity of intelligence, and are overcome by ignorance and dullness, lack of discernment, and stupidity. Such are the Slavs, the Bulgars, and their neighbors. (Lewis, 1990, p. 47)

Nonetheless, their women were considered strikingly attractive, even to the point of being simply called ‘beautiful girls.’ An 8th-century Arab musician wrote: “They used not to train beautiful slave girls to sing, but they used only to train yellow and black girls. The first to teach valuable girls to sing was my father” (Lewis, 1990, p. 56).

What gave rise to this desire for light-skinned foreign women? It seems that fair skin has long been key to Arab notions of female beauty:

Praise of a girl's looks is traditionally couched in such terms as: Her face is like the full moon, her mouth is an almond, her nose a cardamon, she is plump, and dimpled etc. [...] The highest praise is perhaps that she is as white as snow — strange praise indeed to come from a people very few of whom had ever seen snow. (Haim 1978, p. 88)

[The moon] is the most common image used to represent female beauty. When attempting to draw the attention of a beautiful girl on the street, a young man may call out, “’Es ya qamar?” (roughly, “What’s happening, O moon?”). Two important components of the image, brightness (fairness of skin) and roundness (of face), convey the popular conception of beauty in Palestinian and Arab culture. (Muhawi and Kanaana 1989, p. 60, cf. also 122, 181)

Not just in Arab societies …

In general, traditional human societies share a belief that women should be fairer-skinned than men (van den Berghe and Frost, 1986). This cultural norm runs parallel to a physical norm, i.e., in all human populations, women are less pigmented than men from puberty onward. Both melanin and cutaneous blood are involved, with the result that women look paler and men browner and ruddier. Women also display a sharper contrast between facial skin color and eye/lip color. These visual cues are subconsciously used by the human mind to determine whether an individual is a man or a women (Dupuis-Roy et al., 2009; Frost, 2011; Russell, 2010; Russell, 2003; Russell and Sinha, 2007; Tarr et al., 2001).

In addition to aiding sex recognition, these visual cues may also trigger feelings that in one way or another depend on the sex of the person being observed. Since lighter skin is specific not only to women but also to infants, some authors view it as one of several features (smooth, pliable skin, high-pitched voice, small nose and chin, etc.) that the adult female body has borrowed for the purpose of calming aggressive impulses in the adult male and inducing feelings of care (Frost, 2010, pp. 134-135). Such feelings may feed into male eroticism but are not erotic per se. Desire for darker female skin is attested as an alternate, though secondary mode of sexual arousal, even in contexts where exotic otherness seems to play no role, such as premodern European peasant societies, specifically within a context of passionate but short-lived relationships (Frost, 2010, pp. 90-91). This alternate eroticism, previously repressed, has become popular in the Western world since the 1920s with the growing acceptance of tanned skin as a female fashion accessory (Frost, 2010, pp. 91-103).

Men thus seem to be innately oriented toward paler female skin, if only as part of a mechanism for sex recognition. This orientation can, but does not always, translate into erotic attraction and mate choice. One notable exception is the modern Western world, where tanned female skin has become increasingly popular. Another seems to be the high-polygyny region of sub-Saharan Africa and Papua-New Guinea, where attitudes toward female skin color tend to be ambivalent (Frost, 2010, pp. 83-97). First, the relative scarcity of female mates ensures that all available women have takers. Second, due to the higher polygyny rate, fathers invest less in their offspring and mothers invest more. Darker women may thus benefit from a perception that they are better at hoe farming and providing for their children. Ardener (1954) makes this point with regard to the Ibo of Nigeria:

In the choice of a wife, yellow-skinned girls are regarded as beauties, and, other things being equal, they command higher bride prices.  On the other hand it is generally held, especially by dark-complexioned persons, that yellow-skinned people are not as strong as the dark and do not live as long.  A 'black' girl is said to be a harder worker. […] A Mission headmaster was of the opinion that the preference for yellow girls was greater nowadays than in his youth.  He thought that the reason for this was that people formerly looked for strength rather than beauty and tended to marry black girls.


There is a widespread belief, particularly among proponents of whiteness studies, that notions of beauty are determined by power relationships. The strong and mighty are inevitably ‘beautiful.’ This belief is so entrenched that little concern is shown for counterfactual evidence, such as the medieval trade in fair-skinned women for clients in North Africa, the Middle East, and South Asia.

This trade existed for two reasons. On the one hand, European states were too weak to stop it. On the other, European women were considered beautiful by people in geopolitically stronger states to the south and east.  Again, this pattern is inconsistent with the belief that power relationships determine notions of beauty.


Ardener, E.W. (1954). Some Ibo attitudes to skin pigmentation, Man, 54, 71-73.

Dupuis-Roy, N., I. Fortin, D. Fiset, et al., et al. (2009). Uncovering gender discrimination cues in a realistic setting, Journal of Vision, 9(2), art. 10, 1–8.

Haim, S.G. (1978). Love in an Arab Climate, Encounter, 50, 86‑91.

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. (2010). Femmes claires, hommes foncés. Les racines oubliées du colorisme, Quebec City : Presses Universitaires de Laval.

Lévi-Provençal, É. (1953). Histoire de l’Espagne musulmane, tome III, Paris: G.P. Maisonneuve.

Lewis, B. (1990). Race and Slavery in the Middle East, New York: Oxford University Press.

Muhawi, I. and S. Kanaana. (1989). Speak, Bird, Speak Again, Berkeley: University of California Press.

Russell, C. (2013). Model Cameron Russell: I get what I don’t deserve, February 18, CNN Edition International

Russell, R. (2010). Why cosmetics work, in R. B. Adams, N. Ambady, K. Nakayama et al. (eds.) The Science of Social Vision. New York, Oxford.

Russell, R. (2003). Sex, beauty, and the relative luminance of facial features, Perception, 32, 1093-1107.

Russell, R. and P. Sinha. (2007). Real-world face recognition: The importance of surface reflectance properties, Perception, 36, 1368-1374.

Tarr, M.J., D. Kersten, Y. Cheng et al. (2001). It’s Pat! Sexing faces using only red and green, Journal of Vision, 1(3), 337, 337a.

Van den Berghe, P. L. and 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.