Evolutionary Psychology

Evolutionary Psychology is moving to SAGE. The new address is evp.sagepub.com. Submissions here.
Robert Kurzban

The Evolutionary Psychology Blog

By Robert Kurzban

Robert Kurzban is an Associate Professor at the University of Pennsylvania and author of Why Everyone (Else) Is A Hypocrite. Follow him on Twitter: @rkurzban

Stopping Stereotyping and Prejudice

Published 18 May, 2011

Recently, Satoshi Kanazawa posted an entry at Psychology Today entitled “Why Are Black Women Rated Less Physically Attractive Than Other Women, But Black Men Are Rated Better Looking Than Other Men.” (There is a  screen shot at scribd.) The post no longer appears at Psychology Today. (One blogger at PT says that it was removed by editors at PT, but I have not been able to confirm that. I can’t decipher the cryptic quote at the end of the article on HuffPost.)

The post can be seen at scribd, and most readers are probably familiar with this by now, so I’ll just summarize. Kanazawa reports data from Add Health, a large survey of adolescents. As part of the survey, interviewers rate the attractiveness of respondents on a scale from 1 to 5. Kanazawa shows data indicating that Black women are rated as less attractive by the Add Health interviewers than members of other ethnicities (White, Asian, Native American).

Kanazawa writes that “Black women are objectively less physically attractive than other women.” I think a lot has turned on the word “objectively.” He used it, I think, to distinguish the data he’s discussing from the self-report data. (This point is made at evolvify.) I can, however, see how this would be read by many as making a claim about the women, as opposed to ratings of the women.

My point here isn’t going to be about the data, but rather about the issue of stereotypes and prejudice. In terms of the data, I will make one remark. Rhodes et al. (2005) argued that if people prefer faces that constitute an average of the faces that they experience, then, as they put it, faces “should be more attractive when their component faces come from a familiar, own-race population.” They indeed showed some evidence for an “own race” effect. If raters in Add Health were Caucasian men, then this might help to explain observed differences. Now, this idea doesn’t by itself explain all the data, since this line of argument predicts that ratings across all non-white groups would be lower, and it doesn’t explain ratings of males. However, it does suggest that experience with faces influences what one finds attractive, which in turn suggests that the explanation might be located in the particular pattern of experiences of the raters, most of whom might have been White males. (I don’t know the Add Health interviewer demographics.) This is not to say that there are not commonalities in what people find attractive. (See Rhodes, 2006.) Not knowing about the people doing the rating in this case does seem to make the data difficult to interpret, but it also seems to me that it is defensible to argue that there is at least something to be explained here. (Unless, of course, there are some questions which simply must not be asked. I would be interested in a defense of such a claim.)

But let me say something about stereotypes. Humans are social creatures who coordinate activities and cooperate in substantial numbers. Possibly as a consequence of this, even when there is no objective existence of groups, we hallucinate that they exist, and we attach real meaning to them (see, e.g., Kurzban et al., 2001). Kanazawa writes, for instance, that “There are many biological and genetic differences between the races.” My reading of the data is quite different, and in my view (Cosmides et al., 2003), “race” is a way to group people that exists in the heads of humans; this in part explains why the categories we use change over time.

And, further, sometimes people’s views of groups are laden with emotion, and this is where I want to direct my main remarks. Kanazawa’s post has led to something of a firestorm; for instance, responding to this post, PZ Myers says that Kanazawa is among the reasons that he “detests” evolutionary psychology.

Myers, just like others, cannot seem to help seeing the social world – indeed, the scientific world – in terms of groups and coalitions, and here his reaction to Kanazawa’s post fuels his anger for an entire category of people who had no role in writing the post in question. I am strongly opposed to this, and I find his remarks offensive. It makes no more sense to hate an entire scientific discipline because of the work of one person than it does to hate a group (racial or otherwise) based on the ideas or the actions of one of its members.

Along similar lines, he says of evolutionary psychologists that “more of them need to stand up and repudiate this bigoted clown…” Holding aside what he means by “standing up” and the name calling, of all the possible categories that Myers could have chosen, why this one? Why not people at LSE? Or social scientists generally? To return to a point I made earlier in the context of the Hauser scandal, the choice of groups illustrates something about the predispositions of the observers more than anything else.

Not only that, even if one were to think it was reasonable and just to tar an entire field with the same brush, it’s something of a perversity to base one’s views of the field on Kanazawa’s work. As Scott Barry Kaufman’s post documents, and others have similarly pointed out, many evolutionary psychologists take issue with fundamental ideas as the heart of Kanazawa’s research. If Myers wants to dismiss the entire field, or at least justify his hatred for it, he needs to engage the principles that underlie the field as a whole. If he wants to do this, I look forward to his explanation for why human behavior (but not, presumably, any other species’ behavior) should be studied scrupulously free of the theory of evolution, or whatever remedy it is that he has in mind to relieve his anger.

These remarks are not, of course, intended to be specific to Myers, but apply to anyone who is similarly inclined to generalize from one person to a group of people.

I am, to be clear, against racism in all forms. And, I agree that bloggers bear an additional burden when discussing emotionally charged issues.

I look forward to a day when people are evaluated not by the categories to which they belong, but on the content of their blog posts.

I hope that day comes soon.


Cosmides, L., Tooby, J. & Kurzban, R. (2003). Perceptions of race. Trends in Cognitive Sciences, 7(4), 173-179.

Kurzban, R., Tooby, J., & Cosmides, L. (2001).  Can race be erased? Coalitional computation and social categorization. Proceedings of the National Academy of Sciences, 98(26), 15387-15392.

Rhodes, G. (2006) The evolutionary psychology of facial beauty. Annual Review of. Psychology, 57, 199-226.

Rhodes, G., Lee, K., Palermo, R., Weiss, M., Yoshikawa, S., Clissa, P., Jeffery, L. (2005). Attractiveness of own race, other-race, and mixed-race faces. Perception, 34, 319–340.

  • http://www.ToddKShackelford.com Todd Shackelford

    This is a wonderful comment, Rob–measured, thoughtful, and insightful.

    • Rob Kurzban

      Thanks, Todd. Much appreciated.

  • http://twitter.com/daiyami daiyami

    A sidenote, tangential to your point: I’m afraid your information is incorrect. Kanazawa originally titled the article “Why Are Black Women Less Attractive…” The “Rated As” was added after the firestorm broke. Kanazawa’s choice of title, as well as the use of “objectively” would also indicate he was making a claim about the women, not about the ratings in the women.

    • Robert Kurzban

      That is a good point. My belief in the third paragraph would seem to be incorrect. Apologies for the error, and thanks for pointing this out.

  • http://ionian-enchantment.blogspot.com Michael Meadon

    I’m wondering Rob, do you agree Kanazawa’s data is preposterously, pathetically, inadequate to provide even slight support for his hypothesis? If he had a large number of raters (>50, including those of different races, sexes) and the finding still emerged, okay, then perhaps we’d have something to talk about.

    • Crys T

      >50 is a “large” number of respondents? I don’t have the actual random sample equation to hand, but I’ve been using one of the online calculators, and even with error set at 5% and 90% confidence, a target population of only 100 would require a sample size of at least 70.

    • Robert Kurzban

      Right, I should be explicit. In the context of the strong claim, which is one about “objective” attractiveness, I think the only way to interpret such a claim is that there is a universal preference. That is, independent of the features one encounters, such and such a set of features will always be less preferred. If that is the claim, it seems to me that one would want to sample across experiences. (In this case, “race” could be a proxy for that.) That is: if there is something like a same-race effect, then you would need to sample across experiences for the sort of claim I take him to be making. So I agree with you that given he has no data about the raters, the universal claim he wants to make cannot be sustained. Does that make sense?

      • http://ionian-enchantment.blogspot.com Michael Meadon

        Yeah, makes sense.

        @Crys T: Sure. I was thumb sucking an example number. If he had 50 raters, I’d certainly take the results more seriously, but I would have to modulate that by looking at a sample formula, of course.

    • http://facelab.org Lisa DeBruine

      Modern research on attractiveness shows us that attraction is influenced by a very complex interaction among physical (e.g., facial symmetry, skin smoothness) and transient (e.g., expression, gaze direction) characteristics of the observed, characteristics of the observer (prior visual experience, own rank in the current environment) and characteristics of the interaction (e.g., judging in the context of a short-term or long-term sexual relationship, as a potential ally). There is no was to measure “objective attractiveness”, as attractiveness isn’t a measurable physical characteristic. Therefore, the question of “which race is more attractive” is nonsense from the start.

      See FaceLab.org for a very large list of downloadable scientific studies on facial attractiveness.

      • Alex

        I am not an expert in this area, but that argument seems flawed to me. Simply because something is complicated and multiply determined does not mean that it is impossible to determine some “objective” value. To see why, we could similarly argue that one’s ability to hit a baseball is a complex interaction between one’s ability to hit a ball, the ability of the pitcher, whether it is raining, whether the pitcher is left or right handed, whether there are runners in scoring position, whether one is on a hot streak or in a big slump, who one is hitting against (an opposing pitcher, in batting practice, hitting at a son’s little league game), ect. I don’t think it’s silly to talk about objective hitting ability though, if what one means is that, across many contexts, this person is more likely to get a hit than others. Similarly, to argue that one is objectively attractive means they are more likely to get hit (on) than others.

  • DiscoveredJoys

    I gave up reading PZ’s blog regularly some time ago. Although I valued his science paper summaries and broadly agree with his social attitudes, I found his blogging on certain hot button issues (race, abortion, oil spills, politicians) intemperate and lacking in self awareness.

    I know people (including me!) have hot button issues, but it seems to me that a scientific blog, reporting on science, should try for a more dispassionate approach.

    “Comment is free but facts are sacred” as CP Scott said.

  • Tim Bates

    Excellent post Rob: The way in which minds bind the evaluation “bad” onto an idea when that idea is held by a member of an out-group is so powerful in understanding behaviour, but is not given remotely the weight that it needs. Scientifically, or in public and private life.

    @ Michael Meadon: Not sure what “preposterously, pathetically” small is in technical terms. Those terms are more in keeping with out-group demonisation than hypothesis evaluation.

    Inter-rater reliability for ratings of attractiveness is very high, so with sample numbers like Add Health has (thousands), small numbers of raters are not an issue for reliability. The question is about a confound: For instance of same-different race for raters and ratees. That can be addressed by experimentation, so go ahead and test: That’s the best answer to questions.

    • Marco DG

      100% agree.

    • http://ionian-enchantment.blogspot.com Michael Meadon

      You’re right, of course. But I have a long history of taking on Kanazawa’s silliness, and my frustrations sometimes show through.

      (And I don’t grant the point that ridicule is always inappropriate in scientific discussions – especially pseudo or fringe science. “One horse-laugh is worth ten thousand syllogisms.” ~ H. L. Menchen.)

  • Jesse Marczyk

    I also gave up on reading PZ’s blog a while back, largely for the reasons you outlined above; the commenter quality was also sub-par, generally speaking.

    I do find it ironic (perhaps that’s not the correct word, since it’s totally expected at this point) that someone – like PZ – can condemn the practices of another – stereotyping – while doing the same thing themselves. I suppose so long as it’s the proper out-group being targeted, it’s no big deal.

  • Bjarne Holmes

    A very well written, logical, and excellent set of points Rob. Thank you!! think your analysis is spot-on and works very well in a world of scientists and academia. However, the problem is that the statements on Kanazawa’s blog operate and exist in the context on a larger world, where they can be and will inevitably be used to fuel misunderstanding etc. However silly it may be to lump groups together (and I agree), the fact is that humans do so (a lot) and both scientists and in this case more importantly the lay-public will do so. In as much, in my opinion, critique is the most powerful when it comes from those who are judged as being closest to the in-group that exists (perhaps arbitrarily) in peoples minds. In other words, in the case of Kanazawa, it is more powerful and meaningful to most people to see and read critiques of fellow researchers’ work or statements in the same grouped area (in-group), than from others. So, most people will for instance listen closer when a EP scientist speaks out against comments by Kanazawa than say when a cultural critic does so or a relationship researcher like myself (with little EP research experience) does. Like it or not, we live in a world of in-group and out-group identities and we all contribute to this quite actively and use it to form our own identities. The reputation of our in-groups (however arbitrarily they may be constructed) is ours to define and defend if we so wish. In other words, EP as a concept exists and hence anyone doing work in the area will get lumped together to some degree. It is the responsibility of those lumped in the in-group to critique itself – not to stifle or shutdown anyone, but to help clarify and define the in-group to the rest of the world (concepts evolve and change constantly). Scientists are not alone responsible for what happens with results once they have been written about, but that doesn’t mean either that we should wash our hands entirely of how our work is interpreted and perhaps misused by the “masses”. Let’s face it, EP is not alone in this, but there are ideas that have been lumped with EP (wrongly) that create a problem for EP and that scientists working within EP are most suited and best armed to tackle. And example would be the very annoying idea I get from some undergrad males on campus… they have heard/read (and clearly not understood) some basic EP ideas and have said things to me stipulating that rape of females is “natural” and that if it happens to occur on a date, then they shouldn’t be faulted as “my biology means I can’t help myself”. Clearly, they know and understand nothing of EP, but clearing up and setting the record straight on how EP is misunderstood and misused not only is most powerful when it comes from researchers within the “in-group” of EP. However, I think it would also serve EP a hell of a lot of good in terms of making sure the concept of EP is understood correctly and not misused.

  • http://vertebratesocialbehavior.blogspot.com Clara B. Jones

    1. IMHO, the critiques of Lisa’s position are on target. I would add that, if she were correct, we’d have to throw out many of the studies on perception and on “beliefs, attitudes, and values”.
    2. I don’t think your view that race is a social construct can easily be defended. On my blog, I discuss a range of research showing associations between markers and human characteristics and/or distributions. For example, networks of individuals are, ceteris paribus, composed of individuals with similar traits, including similar genetic markers (as every psychologist knows, “similarity” is a litany of social psychology). Another body of work evaluating single nucleotide polymorphisms (SNPs) clearly demonstrates spatial assortment by individuals with similar markers. Such assortment is often regional (geographically).
    3. It would be interesting to unpack the study in an attempt to determine, as one example, what features account for differential perceptions/attitudes/opinions (?) x race x sex, etc. If there are real patterns in the study (a comment above mentions “consistencies” across individuals which is a better way to think of it), then one wants to get some handle on what is going on and, for some investigators, it might be important to get a handle on why–including possible adaptive significance(s) of the observations/phenomena.

  • Marco DG

    Regardless of the quality of Kanazawa’s post and analysis, here’s a related finding that may be of interest to some: the cross-cultural preference for comparatively darker skin in males and lighter skin in females, coupled with the tendency for males to be darker than females within the same population.

    This effect may indeed contribute to explain a pattern such as that reported by Kanazawa (especially if raters are males). Of course, better data would be needed to exclude and/or control for more obvious sources of rater bias, as already noted by many commenters.

    Some links to pages with references:


  • http://vertebratesocialbehavior.blogspot.com Clara B. Jones

    Again, following my previous comments & the above post, it would be very interesting to unpack what’s going on in these studies–for instance, what stimuli and/or stimulus patterns are the operative ones. This type of study might seem more difficult than it actually is; I vaguely recall that research has been conducted identifying what features of faces lead to judgments about them, including recognition. Unless I am incorrect, and my memory is fuzzy here, this work was done by one of the social psychologists in the psych dept @Harvard.

  • http://www.realadultsex.com figleaf

    Left unmentioned so far is his serenely confident but argumentatively gratuitous shot that while “black women are on average much heavier than non-black women” that’s not why black women are uglier. Next he blithly asserts that blacks on average are stupider (have lower intelligence) than all non-blacks… but that’s not why, quoth he, black women are uglier. Oh no, because, see, even though black men are just as stupid as black women they’re still significantly more attractive than non-black men. (Or, one supposes from his amended version, stupidity not withstanding black men are rated more attractive. Which I guess is supposed to be less racist.)

    Eh. Maybe they’re not gratuitous as structural arguments: he may have brought it up by way of eliminating his favorite racist stereotypes as factors: “well, you’d think black women were uglier because blacks are fatter and stupider but no, even filtering out their fatness and stupidity black women are still ugly.

    Oh, and this is lovely! He says that black male attractiveness eliminates as a reason the “fact” that since blacks “have existed much longer in human evolutionary history, blacks have more mutations in their genomes than other races.” And, you see, purer races prefer lower “mutation loads.” But once again, despite those preferences (and, don’t forget, men’s seed-spreading willingness to screw anything that moves… er… to make lower genetic “investments”) and all those icky mutations make black men “if anything, more attractive.”

    (Speaking of “objectivity,” one can imagine that were Kanazawa of blacks heritage he’d have as easily concluded not that blacks have more “mutations” but that they have more robust genetic diversity whereas “inbred” races have an “evolved” aversion to replenishing their degenerate gene pools. He could say the same numbers back that up too! But I digress.)

    (Also speaking of “objectivity,” one can imagine that black people have more “mutations” because, as you say Rob, “black” is only a race in the sense that “black” people have darker skin, with the result that while “black” people descended from populations recently indigenous to north Africa, south Africa, central, east, and west Frica, south Asia, the Pacific Islands, Australia, parts of India, and so on are, yeah, a $@^%@ of a lot “older” and racially “mutated” since some of them are likely more genetically similar to what ever relatively genetic monoculture Mr. Kanazawa calls homeland than they are to each other.)

    But nope, nope. Instead he says “mutations” don’t make black women uglier either. In fact “the only thing I can think of that might potentially explain the lower average level of physical attractiveness among black women is testosterone.” Yup, that’s probably the only other thing that could possibly explain the difference.

    It’s also the point at which he stops being a racist asshole using raw statistics and becomes a… racist homophobe!

    Testosterone, he claims, makes everybody look more manly. And blacks women have more testosterone. Which makes them look more manly. And so by inference that makes anyone who’s attracted to blacks women Teh Gay. And, as we all know, Teh Gay is an evolutionary dead end. So all right-minded, offspring-maximizing men recognize that blacks women are ugly: QED.

    Nearly all the preceding crap is just Kanazawa being an unencumbered racist, none of which in isolation should reflect badly on his professed field of evolutionary psychology.

    The trouble starts when he stopped waiving around statistics from standard KKK tracts and started pulling out EP tools to explain that attractiveness is a measure of genetic and developmental health. Because those are in fact EP assertions that Kanazawa has used repeatedly in past columns — without challenge from other EPs — to confirm other but heretofore milder racist, sexist, and homophobic points. Which puts one in the sticky position of wondering when and where the line should be drawn.

    So while I agree, one hundred percent, that it’s wrong to brand all of evolutionary psychology on the basis of its best-known, most widely followed, and most widely-recognized, and most widely-quoted practitioner. But whether it’s fair or not, unless all evolutionary psychologists want to be tarred with the same brush then, yeah, “more of them need to stand up and repudiate this bigoted clown…” Because like it or not, until they do he’s their Carl Sagan.


  • http://www.realadultsex.com figleaf

    And for the record, unlike the broad group of scientists in evolutionary psychology, who really can’t do much more than lament or denounce Kanazawa (or possibly decline to invite him to their conferences or publish him in their journals) the fact of his continuing employment at the London School of Economics implies their direct endorsement of his views. So yeah, since he’s their direct employee they should be denouncing him too.

    As for whether or why social scientists in general shouldn’t denounce Kanazawa, I believe most of them already do. In fact I’m given to understand that PZ Meyers, himself a social scientist, is the author of the term “bigoted clown.” Which seems like a pretty firm denunciation.


    • Jesse Marczyk

      As for whether or why social scientists in general shouldn’t denounce Kanazawa, I believe most of them already do

      Just a thought, and I’m going out on a limb here, but I’ll bet that PZ didn’t even bother to check beforehand what the evolutionary psychology community at large actually thinks about Kanazawa before he started mouthing off about what we do or don’t collectively think.

      I say that because I have never once met someone who has spoken well of Kanazawa. By in large, he’s been ignored in my social circle when not outright belittled by people familiar with things like this.

      The trouble starts when he stopped waiving around statistics from standard KKK tracts and started pulling out EP tools to explain that attractiveness is a measure of genetic and developmental health. Because those are in fact EP assertions that Kanazawa has used repeatedly in past columns — without challenge from other EPs — to confirm other but heretofore milder racist, sexist, and homophobic points.

      I think you have this completely backwards; the problem is the racism and bad conclusions, not the evolutionary psychology theories about developmental stability.

  • Nandalal Rasiah

    well, the unhappy upshot of this is that the central route to persuasion, regarding lay people and the worth of EP, is getting progressively narrower. The most sober non-EP discussion of the Kanazawa debacle is to be found at Ta-Nehisi Coate’s blog at The Atlantic and his call for EP education (something PZ is obviously too sciency to do) began with a quote from Chris Ryan(that did not end in, “but bonobos” thankfully:”
    “Anyone who takes evolutionary psychology seriously has to overcome the fact that many of the most prominent voices in the field don’t. One gets the sense, reading their sweeping, provocative proclamations that they are more salesmen than scientists. The more shock and schlock they can pack in to whatever they’re selling, the better. “

  • http://vertebratesocialbehavior.blogspot.com Clara B. Jones

    Putting aside criticisms of this paper’s author and the controversies surrounding it, a significant literature exists on aspects of “attractiveness”. Perhaps, using the same or some variant of this paper’s basic question (Does perceived attractiveness vary by race?) combined with a sound research design, the study might be repeated and the findings further investigated. In my opinion, the question is non-trivial, though, of course, may be culture-bound within and between populations.

  • http://vertebratesocialbehavior.blogspot.com Clara B. Jones

    This article not unrelated to the topic being discussed (link from column in today’s NYTimes by David Brooks):


  • http://www.cultures.org Bruce Lepper

    I can confirm from personal experience the cultural element in this.
    After spending two years in Africa, when I came back to Europe white women seemed thin-lipped, skinny-faced and unattractive. After a while the effect wore off and they became attractive again…
    This indicates that at least some aspects of attractiveness are not only different between populations and/or environments, but can be unwittingly “learned” and “unlearned” in a relatively short time.
    Any suggestion that there is “objective beauty” outside of a cultural context is thus, at best, inaccurate. That’s not to say that there aren’t universally-appreciated attractions such as symmetrical features and signs of good health.
    As already noted, no information is given about Kanazawa’s “objective” interviewers. In any event, they probably all live in a predominantly non-black culture where the ideal feminine beauty is well established and incessantly displayed. The short answer to the first part of the question he asks himself is probably: “Because you’re in America”.
    My personal experience also indicates that adding more raters from different races is probably not going to change much. Raters of any race residing in different racial populations would be more pertinent.

  • Peter Koppejan


    1. Nice logo (“say no to racism”). That will make your point clear to even the dimmest members of the pitchfork crowd that is now descending on Kanazawa. After all, he called you a friend, so better be safe than sorry.

    2. You assert that people (like Kanazawa) “hallucinate” about non-existent groups i.e. races.
    Races are simply genetically clustered groups of humans. The fact that different people make different classifications or that boundaries are fluid doesn’t mean the clusters are non-existent. It always amazes me that so many academics refuse to see what any 5-year old can see.

    3. Of course when Kanazawa writes that “black women are less attractive” he means ON AVERAGE. The PC crowd is trying to turn this into a verdict about ALL black women. That is a strawman and you know it.

    4. While I am not convinced by Kanazawa’s data per se, anyone with 2 braincells knows that at least in western society, black women are considered pretty much the least attractive on average.
    – interracial couplings between black women and white or asian men are VERY uncommon.
    – rape of black women by white or asian men is VERY uncommon.
    – black men who are succesfull (an thus have options) go disproportionally for blond bombshells.
    – blacks always complain about the lack of black (super)models. given the huge presence of liberals and gays in the fashion industry, this rules out racial bias.
    And that’s why the “sistas” are mad now; Kanazawa articulated something they knew pretty much already, but he didn’t offer any condemnation of society, like a good anti-racist boy would do.

    4. People should be treated as individuals, not as group members. Exactly! Please tell that to the modern day Inquisition that frame discussion about group differences as an insult to all individual group members.

    5. What are you going to say to people that WANT to be treated as group members. Many of the howling black americans who say they’re offended and hurt by Kanazawa’s piece have a strong racial identity, for example watching Black Entertainment Television. When was the last time you went to a “White Bookstore”?

  • Peter Koppejan

    Look at this:

    “Men don’t write black women back. Or rather, they write them back far less often than they should. Black women reply the most, yet get by far the fewest replies. Essentially every race—including other blacks—singles them out for the cold shoulder.”

    Again, this proves what everybody knew all along.

Copyright 2011 Robert Kurzban, all rights reserved.

Opinions expressed in this blog do not reflect the opinions of the editorial staff of the journal.