Racism and Moronic Science

Satoshi Kanazawa, a British-based evolutionary psychologist, managed to achieve headlines last week by claiming on the web site Psychology Today to have carried out research showing that black women are less attractive than other woman (article now removed but a copy exists here). He of course provoked an immediate backlash from those who had found his “discovery” offensive. Though I immediately doubted his finding, science isn’t served without giving ideas a fair hearing.

So what was Kanazawa’s method? He showed some American men and women a selection of photos of women of various races, and the black women scored lowest. This immediately rings alarm bells for at least two reasons: firstly, how were the subjects selected? More importantly, this took place in America, the country that probably has a greater in-built cultural bias against black people than any other; the reasons for this are complex but primarily descend from the US’s “Jim Crow” racial segregation laws. Although these laws were abolished following the civil rights movement of the 1960s, their social hangover exists to the present day. American society is still very segregated. In a nutshell, any study of this kind conducted only in the United States would be almost guaranteed to “prove” that black women were “less attractive” than whites or Asians. This study is simply an analysis of American attitudes towards black women – and we didn’t need a survey to guess what the results would say.

As if this wasn’t unscientific enough, Kanazawa’s explanation was even more moronic:

Africans on average have higher levels of testosterone than other races…women with higher levels of testosterone have more masculine features and are therefore less physically attractive.

Um right… so he set out to “prove” that black women are less attractive, set up a study that would provide the answer he wanted, and then plucked a random “fact” out of the air to explain it.

I should now point out that in American “one drop” thinking, “black” refers to pretty much anyone who appears to have at least some African heritage; Barack Obama of course is probably 50% African and 50% European, while Tiger Woods is no more than 25% African, yet they’re both referred to as black. It’s likely that many of the “black women” in Kanazawa’s study are actually mixed race rather than African in origin.

So what is Kanazawa up to? Well, he already has form on racial matters, having weighed into the area of race and intelligence previously; this is a favourite one for racists, because IQ tests do tend to give higher scores for whites than blacks. This line of reasoning has been utterly discredited however, as IQ tests have been shown to measure the amount of science-based education a person has received. As more black people get access to good education, black IQ scores have been rising faster than white ones, and the gap is closing; the differences are cultural and social, not genetic – and Kanazawa surely knows that.

So the man is without doubt a racist; his work has also been repeatedly discredited – click here and here for examples.

After this episode, Kanazawa will have little academic credibility left, but will no doubt become a poster boy for American white supremecists, a dying breed who need every bit of credibility they can find.

3 thoughts on “Racism and Moronic Science”

  1. You have completely misunderstood both the article and what is wrong with it!

    Firstly, the data set Kanazawa used was not his own. He used the Add Health data, a highly respected data source. Kanazawa was not claiming to have 'proved' that african-american women are less attractive; he was asking 'why', and thus posited the theory of higher levels of testosterone. So, in that respect, you have misunderstood the article.

    Secondly, Kanazawa is actually incorrect in his reading of the data set. His article is premised on 'sexual' attractiveness (otherwise the sex hormone levels would be irrelevant). However, the only data sets to show a distinction between ethnicities were children. Adult respondents showed no significant distinction in ratings based on ethnicity. This is why you are wrong about what is wrong with the article.

    And finally, one PhD student and one review do not academically 'discredited' make. This is not my field, and I don't know whether the man is generally considered an idiot; but academics are supposed to disagree with each other – sometimes, most viciously. 'Widely discredited' means far more than contentious, a bit foolish, media-hungry or less-than-respected. Your sources don't support the claim that he is widely discredited; in very much the same way that the cited data does not support Kanazawa's claim that it shows african american women are less attractive than other ethnicities.

  2. ebint – agree on your first point: the data were not generated by Kanazawa. So you're right on the semantics: he effectively claims that the data prove black women are less attractive, rather than claiming that HE has proved this.

    "…black women are objectively less physically attractive than other women…"

    This is immaterial, and you're overstating the difference it would make to any of the points I've made. He's misstated the implications of a study, and he's leapt straight to a baffling conclusion with apparently nothing to back up the process in between.

    However I've seen nothing about child vs. adult data you mention – he states/assumes that the data relate to adults throughout. I assume you're familiar with the Add Health data?

    I still stand on the main point I made – that a dataset based on US responses reflect US attitudes to black people, as well as a US definition of what constitutes "black".

    He has made other academically suspect claims, especially regarding the issue of race and intelligence – reading IQ tests as indicative of "genetic" intelligence, as he has done, is utterly discredited, and on its own is enough to disregard his views on race as deeply biased.

Leave a Reply