You are here
Satoshi Kanazawa on the physical attractiveness of blacks
In May 2011, Satoshi Kanazawa wrote an article for Psychology Today online titled “Why Black Women Are Less Physically Attractive Than Other Women,” which the editor(s) quickly changed to “Why Are Black Women Rated Less Physically Attractive Than Other Women, But Black Men Are Rated Better Looking Than Other Men?”. The article was yanked by the magazine following a racism outcry; copies survive in some places (e.g., scribd, pdf). Kanazawa’s university, London School of Economics (LSE), asked him to refrain from writing in non-peer reviewed publications pending an investigation of his research, many of his “peers” went into overdrive attacking him, he was fired from blogging at Psychology Today, and many tried to have him fired as LSE professor.
A discussion of matters that Kanazawa addressed in his deleted article has come up at this site repeatedly and created numerous problems for me. Thus they should be addressed properly.
Kanazawa’s main findings
Kanazawa discussed the attractiveness ratings of a very large, reasonably representative sample of Americans, each of whom had been rated 3 times by 3 observers over a period of 7 years (three per participant is small, but the raters, of whom there were hundreds, agreed somewhat on average, and the participants were many). The ratings were part of a general longitudinal study (participants followed over a couple of years)—the National Longitudinal Study of Adolescent Health (ADD Health)—which assessed thousands of variables. The important findings are summarized below:
Results from factor analysis. Pink represents females, blue males. Black females were rated least attractive. There was a non-significant trend for white females to be rated most attractive. White males were rated more attractive than black or American-Indian males. There was a non-significant trend for white males to be rated more attractive than Asian males.
Critics of Kanazawa’s article
Some critics invoked bias against blacks on the part of the interviewers; some noted that the ADD Health study does not describe the interviewers and hence there are unknown bias variables. However, we can infer a few things about the interviewers in general. The study was clearly designed to serve the needs of academic sociologists and psychologists, people predominantly leaning toward the political/academic left in the period during which this study has been carried out. Therefore, it is unlikely that those more likely to be prejudiced against blacks were recruited in large numbers; the bias of the interviewers should, on average, be in favor of blacks. Note the proportions of women assigned to different categories of attractiveness:
Since the number of ways of developing a specified face/body shape are fewer than the number of ways in which these goals are not met, the size of the [very] unattractive-to-average group should be larger than the average-to-[very] attractive group. But the interviewers were far more likely to rate participants attractive than unattractive. This is a strong bias in favor of participants, including blacks. In short, indications are that the interviewers’ bias was, on average, in favor of blacks.
Most of the criticism of Kanazawa’s article is useless and of no scientific merit. Some of the seemingly scientific criticism is along the lines of “all the variables and confounds that were unaccounted for (also this),” which is basically useless. Most of these are irrelevant to the big picture, just as physics does an excellent job of describing the gist of physical phenomena in spite of ignoring many real world issues when it brings in massless strings, rigid rods, frictionless surfaces, black bodies, etc. (see also some parallels between physics and population genetics; population genetics is certainly relevant to ethnicity). Since Kanazawa left it at testosterone being likely responsible for the lower attractiveness of black women, some have cited literature showing that the difference is not in the direction Kanazawa thinks it is, but these critics leave the larger issue unaddressed, namely that different ethnic groups are not perceived as equally attractive, and if testosterone does not explain this, then there are other factors that should be discussed.
The only noteworthy scientific criticism I have come across is Wicherts and Kaufman’s reanalysis of the ADD Health data. Wicherts and Kaufman started with ANOVAs (analysis of variance), which is basically meaningless as the kind of data involved (attractiveness assessed in terms of very unattractive, attractive, average… ) does not warrant the use of this analysis.
The ratings of attractiveness applied to 4 cohorts (same group at different ages), with average ages 15.9 (wave 1), 16.5 (wave 2), 22.1 (wave 3) and 28.9 (wave 4). Kanazawa did not address wave 4 (the data were out a month before his publication, and were restricted to academics satisfying certain criteria). Wicherts and Kaufman did not address males, consistent with numerous critics finding racist Kanazawa’s claim of black women being less attractive but not his claim that black men are more attractive. Wicherts and Kaufman reported that the ADD Health study shows only teenage black females, not adult black women, rated less attractive, as group differences were not reported for wave 4, when all were adults. They then ignored the finding for teenage girls, pretending that only adults count (most females, particularly black, who are 17-19 look adult).
There are two interesting findings in the re-analysis by Wicherts and Kaufman that they do not comment on. The attractiveness ratings of the participants agree somewhat between waves 1 and 2, but they agree less between waves 2 and 3 and least between waves 3 and 4. Similarly, the interviewers agree among themselves as to the attractiveness of the participants, somewhat, for wave 1, agree less for wave 2, even less for wave 3 and least for wave 4. What happened? Here is one possibility. That Wicherts and Kaufman ignored the findings for teenage girls is consistent with politically correct beliefs regarding attractiveness within and between populations being more applicable to adults. So if people like Wicherts and Kaufman are the interviewers, and there is every reason to believe that the majority bias among the interviewers is similar, then, in the absence of official guidelines regarding how much one should adjust own bias or inflate ratings, disagreement among raters and inconsistency of attractiveness ratings should be greater when the perceived need for such adjustment/inflation is greater (older samples). In wave 3, whereas the average age of the participants was 22.1, some of them were still in their late teens, whereas in wave 4 all were adult. Therefore, we have a potential explanation of the two interesting finds in the re-analysis by Wicherts and Kaufman.
How relevant are the ADD Health data?
A major problem with the attractiveness ratings in the dataset is that they cannot be interpreted in terms of how one ethnic group compares to another. Chances are that the raters judged the attractiveness of a participant based on their estimation of the norms within the participant’s ethnic group or best-fit ethnic group.
This brings us to another issue. It is unlikely that Kanazawa was playing with ADD Health data and discovered that black females were rated less attractive. Like the typical person he should have been aware of how the attractiveness of blacks is viewed by most others and chances are that he stumbled across data or was probing a dataset that he could use to shed light on this topic with some academic rigor.
Common observations about the physical attractiveness of blacks
We can address people’s views of what they find more attractive and the underlying reasons why. In others words, we can address the issue of “why one group is found/rated less attractive than another” as opposed to “why one group is less attractive than another” [not discussed in this article].
Kanazawa’s dataset focuses on American blacks (African-Americans).
The majority of the international public views blacks—particularly the sub-Saharan African and Australian aboriginal varieties—as the least attractive people, on average. Data on this are difficult to find in science journals because, depending on the journal, the topic is either of limited importance, too much trouble to print or taboo. On the other hand, it is hardly necessary to spend time looking for publications on this topic as there are plenty of easily accessible real-life observations/data.
An excellent example is fashion modeling, where the models are selected based on physical appearance alone. High-fashion models are overwhelmingly white and, regardless of where they come from, overwhelmingly Northern European types. Some black women that end up as high-ranked high-fashion models are put there to avoid trouble with the civil rights people. An example is the following Vogue edition:
Vogue May 2007 edition cover showing women claimed to be top-ranked high-fashion models. Based on the amount of work that the featured models were getting, no one would dispute that the white models deserved their ranking or placement, but the African-American model, Chanel Iman, was getting nowhere close to the amount of work the others were getting to deserve a spot in the group. Strictly speaking, there are two oddities among the white models: Doutzen kroes and Caroline Trentini, but what makes them odd is unrelated to their ranking/placement; Kroes was not exactly built for high-fashion modeling like most other top high-fashion models, Trentini is unusually feminine for a high-fashion model. On the other hand, nothing except avoiding the attention of the Politically Correct police could explain the inclusion of Chanel Iman.
Some black women cannot make it in high-fashion modeling even after crying about racism:
Kema Rajandran, a black woman of the south Indian variety was told by an Australian modeling agency that she is unlikely to get much work. Rajandran was livid and went public about discrimination. It did not help her get any high-profile fashion modeling jobs. One year before this incident, the Australian government had encouraged the use of diverse models. Some modeling agencies made an effort to recruit ethnic models (e.g., Malaysians), who ended up being sent back from photo shoots because nobody wanted them.
Some models of the sub-Saharan African variety are put there for reasons similar to the occasional display of bizarre clothing:
Overall, there is just not much interest in using blacks—regardless of whether they are sub-Saharan Africans, American blacks, Australo-Melanesian aboriginals, Sri Lankans and south India people, southeast Asian Negritoes, etc.—as high-fashion models, which has nothing to do with difficulties in obtaining sufficient numbers of tall (except the negritoes and other small-statured blacks) and very thin black women. This could be blamed upon the prejudices of fashion designers, but if we consider the online world, many websites devoted to the fashion world that allow reader commentary, such as the fashion spot, make it clear that most people have no visual interest in women of sub-Saharan African varieties, and among those with a visual interest in them, the interested are disproportionately ethnics of the sub-Saharan African types—in contrast to the international interest in white models—and some of this interest expresses itself in the form of co-ethnics lamenting the underrepresentation of black models.
In other areas where the preferences of the public can be evaluated, non-looks factors increasingly come into play, and the assessment is neither strictly about looks nor about representative samples of each ethnic group [whereas those selected for fashion modeling are not representative of their ethnic group, the selection is about looks alone].
One example comprises of strippers, where the selection is limited to those willing to strip before strangers. In this case, anyone who thinks that in a multiethnic society, such as the United States, men prefer black, Asian or Latina strippers as much as white female strippers is either deluded or clueless.
One could also consider nude models, where African-American, sub-Saharan African and other dark women are again among the least favored, and a similar statement applies to pornography, but non-looks factors come into play here, such as the availability of women willing to pose nude or participate in pornography, the enthusiasm with which porn performers act, the willingness of porn performers to participate in more disinhibited activity, etc. Nevertheless, if one were to carefully separate the non-looks factors in these scenarios, public preferences regarding physical appearance will undoubtedly remain a major factor. In the case of men, African-American porn stars are apparently more widely cast than their female counterparts, but there is no indication that this is related to perceived physical attractiveness; some pornographers use them to promote the myth of larger genital endowment among black men, a baseless claim [see Orakwe and Ebuh (2007) (discussion); analysis of the claims of Jean Philippe Rushton; British reference range vs. Nigerians]. These pornographers are forced to recycle a small number of black male performers, for want of finding enough well-endowed performers, which would not be the case if blacks had significantly larger phalluses compared to other non-Asian ethnic groups contributing to male performers.
Other areas are even less relevant as non-looks factors play a more important role. One example is dating in America, where African-American women are the least favored, but personality and other non-looks variables are a factor apart from physical appearance. African-American men, in contrast, are not the worst off when it comes to success at dating, for which there are at least two reasons: some of their competitors are from ethnic minorities that are culturally more alien to the majority, and—what is presumably the factor having the greatest weight—judging by a good number of women that African-American men are seen with, whom men of other ethnic backgrounds would be reluctant to date, including African-American women, they are less discriminating. Whereas some competitors of African-American men are ethnic men with less masculine builds on average (Asians), this is unlikely to carry much weight as women will emphasize this factor if men are otherwise well-matched on desirable characteristics, and only few will place major emphasis on a more masculine build (compared to Asians) for various reasons, including better prospects of success in violent acts against strangers.
The reasons why: face shape and skin color
The reasons behind the attractiveness evaluations/ratings of blacks are not difficult to figure out. A major reason is skin color. Most humans in most populations prefer light skin, e.g.:
In the Human Relations Area Files (HRAF) database, skin color preferences are reported for 51 cultures. 30 prefer lighter-skinned women, 14 prefer both men and women with lighter skin, 3 prefer lighter-skinned men and 4 prefer darker skin.
A second major reason is overall face shape. Humans generally prefer more overall derived faces, whereas sub-Saharan Africans and aboriginals in Australia, Melanesia and Southeast Asia have the most overall ancestral faces.
The skin color factor is acknowledged by many African-Americans and many of their dark counterparts, by their behaviors. A few examples:
Accomplished African-American men disproportionately end up with lighter-skinned women. In American samples, anywhere from a substantial minority to a majority of African-American men and African-American women prefer lighter skin. In sub-Saharan Africa, many populations prefer lighter to darker skin. Lots of dark people do not fancy spending much time in the sun to avoid getting darker. Some dark Africans go to extremes trying to lighten their skin with hydroquinone and other bleaching agents. African-American children prefer to play with white dolls than African dolls, which was initially proposed to be a consequence of lower self-esteem, but research showed that African-Americans have higher self-esteem than American whites.
In multicultural societies, blacks are particularly sensitive about the darkest skins, getting offended even if whites put black paint on themselves (blackface) to celebrate their holidays such as St. Nicholas day in the Netherlands, where black paint is put on to represent Zwarte Piet (Black Pete), with the act having no racist connotations whatsoever.
Some blacks get offended when reminded of how dark they are. A dramatic example comprises of former high-profile fashion model Naomi Campbell. She burst with anger when an advertisement compared her to milk chocolate. The milk chocolate in this case happened to be lighter than Ms. Campbell, and one would think this is a compliment, but she reacted as if her skin had been compared to charcoal.
In the ADD Health study itself, lighter-skinned African-American women were rated more attractive:
Since the attractiveness ratings of the African-American women are with respect to skin colors, which include light brown and white, their blackness is an inappropriate concept and one must refer to their Negroidness to avoid confusion.
We have a proportion-Negroid sequence: (full Negro) 1,... 15/16, 7/8, 3/4, 1/2, 1/4, 1/8, 1/16,.. (no Negro) 0.
Since most variation in skin color is due to additive genes, one can use any of the following approximations:
- An additive gene at a particular locus either results in the production of melanin or not. A large number of such loci will result in a skin color distribution approximating the continuous distribution of skin color observed.
- The additive genes come in variants resulting in different amounts of melanin production. A large number of such variants results in the chi-squared distribution approximating a normal distribution.
The proportion-Negroidness can be roughly partitioned into 5 groups ( ≥ 7/8, 3/4, 1/2, 1/4, ≤ 1/8 ) that correspond to the skins black, dark brown, medium brown, light brown and white, respectively. Assuming that the Negroidal element is entirely the blackest peoples of Africa and the non-Negroid element is entirely European, we can estimate the Negroidness of the African-American women in the ADD Health study as roughly 62%.
On the other hand, genetic analyses of the Negroidal element in African-Americans reveal a very different extent of Negroidness:
- Estimates reveal 10-20% European ancestry, on average, among African-Americans (review: the percentage European contribution to African-Americans by geographic area was calculated as west (19.9), south (24.0), midwest (19.4), southwest (17.0), average (21.9) (not designed to examine Indian ancestry)).
- Another estimate (STRUCTURE): 72.5% Negro, 19.6% European and 8% Asian.
- Two types of estimates for a dataset:
- STRUCTURE: 81% Negro, 16% European, 4% Indian.
- Maximum likelihood: 75% Negro, 17% European, 8% Indian.
- 19% average European ancestry.
Most of the sub-Saharan African ancestry of African-Americans comprises of populations such as the Yoruba, the Mandenka and the Bantu, or the blackest humans, whereas the non-Negroid element mostly comprises of the pale Northern European, a darker Southern European and a darker still American Indian, i.e., a lighter minority Negroid element in the founding populations of African-Americans is countered by a darker-than-pale non-Negroid element, and thus the color assumptions behind the 62% Negroidness calculation should not be severely violated.
In other words, we can state that the representative group of African-American women is not as black as it is Negroid. This inference should be regarded as tentative given the crudeness of the data and calculations, and, if true, is indicative of selection (sexual or natural) for something leading to lighter skin, which may be for less Negroid faces and thus lighter skin indirectly, or for higher intelligence and thus lighter skin indirectly, or specifically for lighter skin, or some combination of these (ADD Health data do not allow one to figure out).
Regarding face shape, the following comparison between former United States President George Bush Jr. and a chimpanzee is well known.
On the other hand, blacks are extremely sensitive about any comparison between them and non-human primates, especially apes, even being offended by police officers eating bananas when supervising gatherings where African-Americans are celebrating Martin Luther King Jr. day. This sensitivity cannot be accounted for in terms of skin color as there are numerous monkeys with very light skin, and some chimpanzees have light skin. This extreme sensitivity on the part of blacks is tacit acknowledgment of where their faces lie on the ancestral-to-derived discriminant. This is explicitly acknowledged by a sub-Saharan African population known as the Woodabe, a face-wise unusual-looking people given where they live, who describe the faces of neighboring sub-Saharan Africans by referencing non-human primates. Even American blacks, as well as other ethnic groups, more quickly identify apes when primed with black faces and take longer to identify apes when primed with white faces, a find independent of prejudice against blacks or motivation to control prejudice, academic proof for what is obvious to all humans regarding Negroid faces, which when explicitly pointed out by a white person gets him slandered as a racist.
In contrast, when black supremacists have compared Northern Europeans to non-human primates, they have invoked monkeys, not apes, and based the comparison on hair colors, not overall face shape.
When colonialism was ending in sub-Saharan Africa, black supremacists, in the process of kicking whites out, promoted propaganda that blacks had bigger brains than whites and hence were more intelligent, and compared Northern Europeans to monkeys because of the variety of non-black hair colors, especially the lighter ones, found among them.
Note also that the most attractive black faces, regarded as such by African-Americans, are disproportionately non-Negroid East African (e.g., fashion model Iman) and thus descended from populations whose skulls are much closer to European norms than to sub-Saharan African ones, or obviously mixed-ancestry faces with or without cosmetic surgery to lessen the Negroid appearance (e.g., Shemar Moore, Vanessa Williams, Halle Berry (white mother and nose job)), or majority-component sub-Saharan African faces with cosmetic surgery (e.g., Tyra Banks and her two nose jobs), or sub-Saharan African faces that are natural outliers for their group by looking more derived and less sub-Saharan African.
Other possible reasons: obesity, testosterone, estrogens, etc.
Kanazawa reported results after adjusting for the prevalence of obesity. In statistical studies, one would need to control for obesity to test whether there are other factors affecting attractiveness, but in the real world no one does this. People who find obesity less attractive are bound to find a more obese population less attractive, on average, assuming other things are similar. So yes, African-American women will be found less attractive in Western nations, on average, with respect to body weight. Similarly, notwithstanding the attractiveness ratings, Kanazawa was able to say that African-American men are more attractive than others when one controls for intelligence, which people do not do in real life. Also, ADD Health data lack IQ scores and hence Kanazawa’s argument for attractiveness ratings after controlling for intelligence is either speculation or an attempt to use a proxy for intelligence, such as educational performance. In addition, lighter-skinned blacks are more intelligent (r = 0.17), which Kanazawa never mentioned.
Not finding obesity a good reason, Kanazawa suggested that higher levels of testosterone are the reason why black women are less attractive than others. This is a weak argument. People derive cues regarding sex hormone levels from physical appearance and behaviors. So Kanazawa clearly referenced secondary sexual characteristics, but why only testosterone? Testosterone has masculinizing effects on secondary sexual characteristics, but women are not just less masculine versions of men; they are also more feminized. Some of the hormones responsible for feminizing secondary sexual characteristics—larger breasts, smaller waists, rounded hips, etc.—are estrogens, which Kanazawa did not consider, and this is important.
For instance, African-American women may have higher testosterone levels than Chinese women, but what if they have higher estrogen levels, too? If men were more inclined to favor Chinese women because of their lesser muscle mass, they should be more inclined to favor African-American women for their larger breasts.
If we proceed further, we note confounding factors. Among the ethnic groups in the study, African-American women have the most bone mass, which one expects to relate to attractiveness. How do we know that this is due to higher testosterone levels? Chances are that testosterone has little to nothing to do with it because African-American men have similar testosterone levels as white American men but still have more bone mass; also see this.
African-American women have less body hair than white women, on average. So should this be considered a more feminine feature or is it an ethnic difference not related to feminization? Indications are that this difference is an ethnic one that is not related to feminization. African-American women certainly do not look less feminine on a number of important counts. When it comes to breast size and backside protuberance, they look more feminine than the others in the study, on average. When it comes to body hairiness, lip thickness and smoothness of the supraorbital region, they look more feminine than white women, on average. When it comes to arm length or muscularity, they look less feminine than the other groups, on average, and so on.
I started looking for data on hormone levels in representative samples of women from various ethnic groups, but most of the data are based on diseased or middle-aged and older women. Nevertheless, I stopped after finding a reasonable dataset to contrast young adult African-American and young adult white women for illustrative purposes; the differences in this study may not be replicated in a a representative sample, but the data are good for what we need to understand.
In this study, the black and white women ranged from 18-36 and had an average age of 26; they had the same height and similar levels of sex hormone-binding globulin. Compared to white women, black women weighed more, had a higher waist-to-hip ratio (larger waist for a given hip circumference), had lower estrogen levels and also lower testosterone levels. Let us assume that the differences are representative of that in the general population. We can then calculate the proportion of instances when a young adult black woman and a young adult white woman are randomly selected from the general population such that the black woman weighs less than the white woman or has higher estrogen levels, etc. Some of the results are listed below.
proportion of cases in random selections
|Black woman weighs less than white woman.||33%|
|Black woman has WHR less than white woman.||39%|
|Black woman has higher estradiol concentration than white woman.||34%|
|Black woman has higher free estradiol concentration than white woman.||38%|
|Black woman has lower testosterone concentration than white woman.||57%|
|Black woman has lower free testosterone concentration than white woman.||59%|
|To calculate these values from the data in the study, note that the difference between two normal distributions is a normal distribution with mean equal to the difference in the means of the distributions and standard deviation equal to the square root of the sum of variances of the distributions, allowing us to calculate the cumulative probability that the difference between the groups is greater than or less than zero.|
In the dataset, the differences between black and white women are less pronounced for testosterone than for free estradiol levels and WHR, and the most striking differences are for total estradiol levels and body weight, but even if we consider body weight, in 1 out of 3 random pickings, a black woman would weigh less than a white woman. Said alternatively, if one were presented body weights, one randomly from a black woman and another randomly from a white woman, and asked to assign the body weight to the right woman, the safe guess would be that the greater body weight belongs to the black woman, but using this rule, one would be correct only 67% of the time. Contrast this with skin color and face shape. If one were to randomly photograph the skin of sub-Saharan African blacks and whites, and have people assign the photographs to the correct ethnic group, the assignment will be achieved with over 99% accuracy. Similarly, if one were to randomly photograph the faces of blacks and whites, assign them the same color on a computer and ask people to assign the pictures to the correct ethnic group, the assignment will again exceed 99% accuracy. Yet, Kanazawa considered testosterone [which is less relevant than body weight in the table] and did not address skin color and face shape in shedding light on why blacks are at the bottom of the attractiveness scale, on average, as far as international public opinion goes.
Skin color and face shape differences in this case are predominantly associated with ethnic differences not involving masculinity-femininity. There are other such ethnic differences that are relevant, one being the tendency of African-American women to pack more mass in their upper thigh and have more protruding buttocks while packing little mass in the lower legs, a scenario that many among the international public will regard as disproportionate mass distribution. But again, some ethnic features favor blacks, an example being that their thicker skin is less susceptible to developing fine wrinkles.
The answer Kanazawa is looking for primarily lies in ethnic differences in physical appearance that have hardly anything to do with masculinity-femininity, the most important ones being skin color and face shape. Other factors, such as how feminine the woman appears, are relevant but apparently far less important than skin color and face shape when it comes to explaining the perceived attractiveness of blacks.
Skin color and face shape preferences: conditioned or intrinsic or a combination?
Many critics of Kanazawa’s article have attempted to dismiss his arguments by mentioning how living in a white-dominated society influences ideas about beauty. This needs to be proven just as an assertion about the role testosterone plays needs to be proven. Beauty preferences or evaluations are not a simple matter of who is dominant or in power:
Consider the following data about the ADD Health participants rating their own attractiveness. If indeed living in a white-dominated society affects people’s perception of attractiveness, for some reason it is having the opposite effect on African-Americans as they rate their own physical attractiveness very highly.
We know that Northern European slaves were admired for their looks in Rome; their downtrodden status did not matter.
Notwithstanding a culture promoting tanning, whites most in need of a tan (Northern Europeans) show, on average, no preference for skin darkening due to melanin (the brown variety of which is largely responsible for ethnic differences).
Skin color is a function of pigments produced by the body [melanin (brown, red), hemoglobin (red), keratin (jaundiced yellow)], the foods one eats [carotenes (yellow-orange)] and injuries/infections/abnormalities. The Scottish sample studied showed no preference for skin browning but a preference for skin pigmented by carotenes.
Controlled studies show that notwithstanding very thin and masculinized female high-fashion models being the top-ranked models and their looks extensively exposed to just about all people in Western societies, most men and most women in the West overwhelmingly optimally prefer a narrow subset of the medically healthy body weight range and above average femininity in the looks of women.
These examples suggest that whereas standards associated with the elite or dominant group or ruling class are expected to have some influence, specific arguments must be proven, i.e., it must be proven that white dominance is the reason for lower ratings of the attractiveness of blacks.
Skin color preference is a complex issue. I can conceive of numerous scenarios where it is intrinsic, conditioned or some combination of the two.
In one example we have a population living in a tropical jungle, heavily dependent on hunting. The whiter the hunter, the more light he reflects and the more likely he is to tip off prey. If we start with variation in skin color preference, those that prefer darker skin will disproportionately end up with hunters better at hunting, and given sufficient time living as hunters in a jungle, this population could become darker and one where the majority prefers dark skin.
In a different scenario, a lighter and more intelligent population migrates and dominates a darker population. There may be some ethnic mixing. The lightest people disproportionately have high status and the darkest disproportionately the lowest status. Lighter skin hence gets associated with higher status and becomes desirable.
In a third scenario, upper class people indulge in less manual labor under the sun in pre-industrial society and thus are less sun tanned. Lighter skin is thus associated with higher socioeconomic status and greater social desirability.
One can conceive of many other scenarios, but the short answer is that I do not know how to explain the [median/modal] international skin color preference for light skin.
But face shape is a different matter. Human preference for more overall derived facial features goes with how humans have deviated from ancestral species, and is by all means intrinsic. To the extent that any social conditioning is suggested, the apparent explanation is that social exposure makes one aware of latent preferences, which, said in a different way, is becoming aware of one’s own preferences.
Whereas Kanazawa was forced to grovel for his “sin,” the truth is undoubtedly more unpalatable to the people he offended, a truth that is so simple that some people will find it remarkable that scientific literature had to be discussed to explain something so obvious.
Since the issue of what harm may result from an article like this is bound to come up, this should be addressed, and is best addressed in the context of evolutionary psychology.
Satoshi Kanazawa and evolutionary psychology
This is a digression from what this site is about, but one good thing came out of the kanazawa debacle. Kanazawa’s article prompted a backlash against evolutionary psychology, in part prompting dozens of evolutionary psychologists to broadly attack Kanazawa.
Here is a short introduction to evolutionary psychology for those unfamiliar with how bizarre this discipline is.
Within the biological sciences nobody disputes that genes affect behavior, including human behavior. Genes reflect one’s inheritance and thus there is no controversy over [the milieu shaping one’s] genetic heritage affecting one’s behavior. So why does one need a special moniker such as “evolutionary psychology” when the topics covered are easily accommodated within behavioral ecology, behavior genetics, psychology or sociology? Academic specialization is an inadequate answer. And no, applying evolutionary biology/principles to psychology does not evolutionary psychology make. The answer lies in the vanguard or core group among evolutionary psychologists that has made the discipline heavily focused on human universals [some differences are explored, but these usually involve gender].
“Human universals” is a curious approach. If one were to compare the spacing of the eyes among Europeans and East Asians, the fact that both populations normally have two eyes each is automatically taken into consideration, whereas to note that both these populations comprise of members with normally two eyes each does not require that any differences involving eye placement be taken into account. Thus there are cases where an examination of differences automatically considers the similarities and leads to a more comprehensive understanding but the reverse is not true. Then, there are numerous examples from, say, medicine, where focusing on the diseased, or the abnormal, or the different helps one better understand the normal or typical. More generally, to focus on differences is to either take into account similarities/universals or end up understanding similarities in a clearer manner. Thirdly, scientific experiments often involve deriving useful insights from differences in outcomes when some variables are manipulated while others are held constant. In short, focusing on differences, between individuals and groups, is clearly the path to understanding human universals pertaining to behavior and the topics covered by psychology and sociology.
So why the focus on human universals? At this point we should go back a few decades. In the mid 1970s, Edward O. Wilson came up with the sociobiology synthesis. Whereas little of it focused on humans, the notion of a biology-based sociology (the study of human social behavior) riled some powerful people. The hallmark of E. O. Wilson’s work was synthesis—the scientist and science at their most brilliant—though his approach was not unique (e.g., the sociological work of Gerhard and Richard Lenski). The backlash against sociobiology was so severe that much as this moniker was appealing, a number of scientists abandoned the term and continued their work under behavioral ecology or other specialties/generic fields. As sociobiology waned, more specifically the use of the term rather than sociobiological research, the discipline of evolutionary psychology started gaining traction from the late 1980s. Critics regarded it as a soft-core or politically correct version of sociobiology.
However, if evolutionary psychology were a more politically correct version of sociobiology, then its core proponents would ignore some politically sensitive issues, obfuscate their work with specialty terms, and distance themselves from political activists/activism, but the core/vanguard has played an active role in spreading disinformation about various group differences, including maintaining paradigms that pre-emptively minimize or dismiss various group differences, which are beyond the scope of this article and this site.
Evolutionary psychology was, and remains, controlled opposition. The most effective way to undermine opposition is to lead it. When the writing is on the wall—e.g., data on genes, hormones affecting behaviors and personality, and experiments on non-human species are coming in—denial and suppression will only carry one so far. Joining in to limit the “damage” is a better strategy, and leading the opposition is the best one. This controlled opposition was not simply the brainchild of a few researchers, such as John Tooby and Leda Cosmides, that was widely adopted and ended up toppling competing approaches by virtue of superior methodolgy and results; there were plenty of behind-the-scenes activities that caused the discipline to flourish:
Aside from a focus on human universals, there are two other major elements of evolutionary psychology. One is the mental modules paradigm. In my estimation, the appropriate response to this is... if there is a neurophysiological explanation, a neural network simulation, there is something; if there is an essay or tome on the modularity of the mind, who gives a rat’s? Note also how rich a “theory” they came up with early on; science does not manifest itself in this manner... in the case of the science of the mind it should be mostly masses of data awaiting more data for synthesis (a theory in science). The third major element of evolutionary psychology does not distinguish it from things that are taken for granted in biology, particularly evolutionary biology, which is what Satoshi Kanazawa heavily draws upon. This is not to say that Kanazawa mostly writes mainstream science or what is passable as science, but Kanazawa is clearly not an evolutionary psychologist by a long shot yet calls himself one because he draws upon elements of evolutionary biology to explain human behavior. One can imagine how much this infuriates the core/vanguard of evolutionary psychology who must keep the dupes leashed and concentrated on human universals. As it is, they have to deal with the likes of Peter Frost, who thinks he is contributing to evolutionary psychology by undermining the strength of human universals, the last thing they need is Kanazawa, who is as clueless about the nature of evolutionary psychology as his critics, such as Stanton Peele, who portray evolutionary psychology or caricature it with a broad brush using the work of Kanazawa.
If Kanazawa were a white male, his arguments would be dismissed as racist and supremacist. But he is Japanese; cannot be described as racist against blacks when a) he reports black women as least attractive but black men as most attractive and b) the Japanese did not go around colonizing black Africans, thus preventing him from being assumed racist a priori; and he has made an argument that notwithstanding what the IQ studies say, Northeast Asians like himself will not be making many outstanding/cutting edge contributions to science. This means that critics will go after the discipline that Kanazawa is practicing, which he, cluelessly, claims to be evolutionary psychology. More Kanazawas and this discipline will see its [very welcome] demise sooner rather than later.
What will the demise of evolutionary psychology mean for the advancement of science? Let us consider some excerpts from Satoshi Kanazawa:
...the pursuit of knowledge as the most important goal of science... I believe that the pursuit of knowledge for its own sake is the only legitimate goal in science (by which I mean basic science, as opposed to applied science like medicine and engineering), and the truth is its only arbiter. Nothing else should matter in science besides the objective, dispassionate, and single-minded pursuit of the truth, and scientists must pursue it no matter what the consequences.
From my purist position, everything scientists say, qua scientists, can only be true or false or somewhere in between. No other criteria besides the truth should matter or be applied in evaluating scientific theories or conclusions. They cannot be “racist” or “sexist” or “reactionary” or “offensive” or any other adjective. Even if they are labeled as such, it doesn’t matter. Calling scientific theories “offensive” is like calling them “obese”; it just doesn’t make sense. Many of my own scientific theories and conclusions are deeply offensive to me, but I suspect they are at least partially true.
It is not my job as a scientist to “use” scientific knowledge in any way to improve the human condition; that’s the job of politicians, policy makers, physicians, and other social engineers.
The only responsibility that scientists have is to the truth, nothing else. Scientists are not responsible for the potential or actual consequences of the knowledge they create. Holding scientists responsible for uses and misuses of their research by others is a sure-fire way to detract them from the single-minded pursuit of the truth, because that would make them pause and entertain other criteria besides the truth. If the truth offends people, it is our job as scientists to offend them.
Obviously, Kanazawa’s attitude is far more likely to result in a scientific synthesis than that of the core of evolutionary psychologists’, which is why evolutionary psychology needs to go, and we will see its demise.
This core will not do research along the lines of the discussion in the article. Is this because they are concerned about any harm that may result to blacks or is it that they are concerned with the harm they may cause to themselves? Let us look at what harm blacks may face.
Many blacks will be infuriated, but given their higher self-esteem, a large proportion will blame the messenger and their self-esteem would remain unaffected. Some would be offended because something unflattering, which they know to be correct, has been stated, and because they know it is true, their self-esteem will remain unaffected. The minority that is not offended will emerge with self-esteem intact because they know it is true.
The other type of harm that may result is discrimination or increased prejudice. The likelihood of this can be assessed by asking oneself what may happen if the media orchestrate an extensive “black is beautiful campaign”? If the black models they use are East Africans, mulattoes or those with bleached skin or nose jobs, everyone will see what is going on. If they use full Negroes or black aborigines, a successful campaign is no more likely than the possibility of this article increasing discrimination or prejudice against blacks because people know what blacks look like and most share a broadly similar concept of beauty. In short, no harm to blacks will result.
So are they disinclined to do proper research fearing reprisal from blacks? If this were the case why would they be directing animus toward Kanazawa in the manner they did, and in the midst of a push to get Kanazawa fired as professor? Because they must protect their bogus discipline. The sociobiologists who faced character assassination at the hands of the predecessors and ideological kin of the evolutionary psychology core group never responded in kind but always with evidence. Contrast the scientists who came to the defense of Kanazawa (not defending his claim) with some of the evolutionary psychologist core people [and dupes] critiquing him.
This is one of numerous illustrations that the disinclination of the evolutionary psychologist core/vanguard to do proper research or to properly treat some topics is not a result of concern over the type of harm [to blacks] they superficially wish to avoid or fear of reprisals from the group studied [blacks], but concern over a different type of harm to the group that has created and funded the controlled opposition [not blacks].