Colourism: the sweet juice of racism

One of my favourite authors of all time is a Canadian named Lawrence Hill. He, like me, is of mixed racial heritage. He, like me, struggled with crafting an identity in an era where ‘biracial’ or ‘mixed’ wasn’t really an option. I found his writing a major source of both inspiration and comfort in my teenage years where the race question loomed largest in my life (at least, compared to now). If you haven’t read any of his stuff, I highly recommend you put The Book of Negroes or Any Known Blood on your reading list.

In one of his books Black Berry, Sweet Juice, he riffs on an old racist adage: “the blacker the berry, the sweeter the juice.” The implication is that dark-skinned women are more sexually attractive. Hill wryly completes the rhyme: “the blacker the berry, the sweeter the juice. But if you get too black, it ain’t no use”, aptly noting that life is much kinder to light-skinned black folks than dark-skinned ones. This is a fact that is well-known within the black community, as both an internal conflict and an external one:

For the first time, a study indicates that dark-skinned African Americans face a distinct disadvantage when applying for jobs, even if they have resumes superior to lighter-skinned black applicants. Matthew Harrison, a doctoral student at the University of Georgia, presented his research last month at the 66th annual meeting of the Academy of Management in Atlanta. Along with his faculty supervisor, Kecia Thomas, a professor of applied psychology and acting director of UGA’s Institute for African American Studies, Harrison undertook the first significant study of “colorism” in the American workplace.

“The findings in this study are, tragically, not too surprising,” said Harrison. “We found that a light-skinned black male can have only a bachelor’s degree and typical work experience and still be preferred over a dark-skinned black male with an MBA and past managerial positions, simply because expectations of the light-skinned black male are much higher, and he doesn’t appear as ‘menacing’ as the darker-skinned male applicant.”

Hiring is one of those weird things where a whole host of non-overlapping bits of information are used to accomplish something quite difficult – evaluate the worth of another human being. Aside from the things that are on paper – education, experience, list of skills – we are also called upon to see how well someone would ‘fit’ within our environment. Personal details – sense of humour, eye contact, level of comfort – things that may have no bearing whatsoever on job performance, become part of the package of details used when deciding to whom a position will be offered.

Like all decisions that are made when instinct and ‘gut feeling’ are brought to bear, a whole host of cognitive biases are allowed to run wild. In this case, a person’s level of “comfort” with an applicant appears to be closely related to skin colour. As we demonize black people along racial lines, we begin to demonize blackness itself. This is really no surprise, and is spoken to explicitly in this morning’s post – we have an association between positive characteristics and white skin. While the reasons for this are multitude and would require a PhD-level dissertation to fully explore, I hope that you will simply accept my assertion that ‘whiteness’ is considered, in general, to be superior to ‘blackness’. Not by all individuals, not through active considered bigotry, but through our society’s subtle indoctrination that has a history stretching back centuries.

The consequence of such demonization shows up in the results of this study. Applicants to jobs can find themselves discriminated against, not just for being black, but for being too black. Despite our repeated assertions that we live in a “post-racial” society, we have just found new ways to express our racism. As a result, the overlapping cognitive processes of “I am afraid of black people” and “I want to hire someone who makes me feel comfortable” strike a bias against dark-skinned candidates – even more so in some cases than our desire to hire the most qualified person for the position.

The perverse upshot of this study is that it can help us tease out anti-black racism without interference from racial compensation. I’d imagine that most people presented with white and black faces will recognize that racism might play a role in the decision-making process. They are therefore likely to actively make the attempt to avoid being racist. This might mean looking more explicitly at qualifications to mitigate the effect of race, or overcompensating by picking a black candidate “just to be fair”. Since colourism is not quite as well understood, and all of the faces in front of the study participants were ostensibly ‘black’, compensation for racism will be less likely to occur.

Because of my mixed racial heritage, I benefit from the effects of colourism. I can’t deny this, nor do I wish to. What knowledge of this bias allows me to do, however, is act as a “foot in the door” – being a ‘non-scary’ black guy who can help break the cognitive link between blackness and its associated negative baggage. This doesn’t excuse me from still having a responsibility to ensure fairness in hiring whenever I can, or make it “okay” that colourism exists. It simply means that I am in a position to exploit this bias to achieve the breakdown of even these subconscious kinds of racism.

Once again, the way to diminish the power of these kinds of bias is to first learn that they exist, and then to train our brains to notice them. Finding methods of in-depth research to pinpoint and measure these biases makes this all the more possible. But if we don’t even try, then it ain’t no use.

Like this article? Follow me on Twitter!


  1. says

    Did the study rule out the other factors? The eye contact, the confidence, and all that jazz? I ask because, as a honky who has often had to go through the interview process, I don’t believe I’ve ever considered race. The closest I have come is not giving a job to someone because the job required strong communication skills, and the applicant (though excellent) had a very thick Eastern European accent. If two black people came in for interviews, I would highly doubt that I would choose the less black looking for that reason. If the other candidate was stronger, I’d be an idiot not to hire them.

    But I don’t obviously speak for everyone, and it’s entirely possible that I’m speaking as someone who has never been in that situation and wouldn’t recognize that I was doing it if I was.

    While it is nowhere near the same thing, I regularly lose out on jobs because of my appearance. I work as a Business Analyst and Computer Programmer, and I’m also a great big monster with a shaved head and tattoos. I’m not pulling the “Oh, what about me!” card there, I’m saying that losing out on a job you are more qualified for because of appearance sucks hard, and I’d like to see that end.

  2. Crommunist says

    I would highly doubt that I would choose the less black looking for that reason. If the other candidate was stronger, I’d be an idiot not to hire them.

    I’d be willing to bet that every single participant in the study would have said the exact same thing. Racism is not always (and in fact is quite rarely) so overt as we like to think it is. Oftentimes these kinds of subtle cognitive biases aren’t available to our conscious mind – that’s the whole point. Our understanding of racism, what it is and where it comes from, needs to evolve. The way to reduce the impact is to be aware that we are all susceptible to these kinds of biases, and that we need to take deliberate care not to let them influence our behaviour, rather than just saying “I am not a bad person so I wouldn’t do that.”

    Did the study rule out the other factors? The eye contact, the confidence, and all that jazz?

    I couldn’t find the study (it was a master’s thesis, may never have been published), but the article says that the applicants were hypothetical, with pictures attached to their résumés. Eye contact and confidence are therefore not applicable to this particular study. Even if they were, I’d wonder why there would be a bias in terms of eye contact/confidence that favoured light-skinned people over dark-skinned ones.

  3. Enkidum says

    I’d be extremely surprised if this basic finding isn’t true. This isn’t the conscious “you” saying “I hate n*ggers”, or even “Black people make me a tiny bit uncomfortable”. “You”, in the sense of the “you” that has a conscious awareness of yourself and your actions, doesn’t really enter into the picture (or at least not in most of these kinds of studies).

    Most of these implicit bias studies are based on a kind of priming. To take a far less controversial example, if I ask you to use a sentence with the word “bank”, you will probably say something about that place that you associate with money and mortgages and so forth. But, if slightly beforehand, I talked about rivers, streams, water, and so forth, and then asked you to produce a sentence with the word “bank”, you would be more likely to use the less common meaning, and talk about river banks.

    You would have absolutely no awareness of this manipulation – from your conscious perspective it would just seem to be the first sentence that popped into your head. But we can look at the data from such experiments and see that there is a massive effect of such manipulations – in essence, you are using the “river bank” form of “bank” because we primed you with the earlier associated words. There are a vast number of extremely well-replicated effects similar to this.

    In the ugly race cases, much the same thing is going on (or at least that seems to be a big part of it). Whether you like it or not, the images and stories you have been presented with since childhood are skewed in subtle ways, and one of the ways they are skewed is “white=good, black=bad”. (Like Cromm says, this is hugely oversimplified, but just take that for the sake of argument.)

    So this means that you have all sorts of positive associations with whites that you don’t have with blacks (or you have to a lesser extent with blacks). Again, you have no control over this. This is just a statistical learning mechanism, completely unconscious (well, I’m theorizing a bit here, but I think there’s good reason to think it works this way). It’s picked up on the associations that you’ve been presented with throughout your life (white=good, black=bad) and will automatically spit out “bad” when you present it with “black”, because it’s a good learning mechanism. (Actually, probably more accurately, it will spit out fewer good things when presented with black than it would when presented with white.) Does that make sense?

    Cromm – someone else mentioned the Harvard implicit associations group on your other article today, which is my main source of knowledge about the race-specific stuff. There’s some other stuff on the “own-race” effect in visual search I can look up for you too. And if you’re interested in the effects that always testing middle class white kids has on psychology, you should check out Joe Henrich’s paper on WEIRD people – #8 on this page:,

  4. says

    I get all that. I’m just saying that in my own experience, I don’t believe race or skin tone (or shade of skin tone) has ever or would ever come into it. I recognize that there is the potential for unconscious action taking place, but I don’t think I have those unconscious parts of me. I grew up in a middle class community in Calgary, and there were no black people around. When I did meet black people, such as the one black family at my church, they were fantastic people who I totally enjoyed. I wouldn’t say that I’ve never heard a bad thing said about black people, but I’ve never consciously thought of that as an issue of race so much as assholes, and there are assholes of every hue.

    As far as the issue of resumes and photos, obviously that rules out the majority of non-verbal queues. The only way that bias could enter into that would be if the images were leading in some way, such as all of the lighter skinned photos wearing suits and all of the darker skinned photos wearing casual clothing. Since that’s probably not the case, it seems like the findings are valid. Interesting stuff.

  5. Retired Prodigy Bill says

    Ah, Enkidum, you beat me with the IAT reference!

    I think it is fairly obvious that this is a true thing. What we call “ourselves,” our consciousness, is the result of dynamic, fluid and recursive filters over which we have little awareness and no control, other than through the self-reflexivity portion of the process. To racial bias you can add height and weight bias, speech bias, intelligence bias, tatoo bias, clothes bias, all sorts of things. Becoming more aware of our biases is the only way to alter them. Whenever I travel back to the USA (I currently live in Mississippi) I’m struck by a) the lovely, heterogeneous composition of the citizens and b) the variety of personal appearance and accessories that would prevent someone from being hired in Mississippi.

    Since moving to Mississippi three years ago I’ve apparently gone, according to my IAT results, from a slight bias for European Americans to a slight bias for African Americans, so environment seems to count.

  6. Charles Sullivan says

    I wonder if native-born Africans in the US experience the same forms of discrimination in hiring based on relative darkness of skin tone.

    I suspect that it may not be the case (or is less significant), and this itself may reveal something about American racist attitudes.

  7. audiolight says

    but the article says that the applicants were hypothetical, with pictures attached to their résumés.

    I have seen other similar psychology studies done before (matching theoretical descriptions with people’s supposed pictures), and while I agree this study shows obvious racial bias in participants, I am skeptical that this result can be directly applied to ‘real world hiring practices’.

    Participants in the study that Harrison, himself an African-American, directed for his master’s thesis included 240 undergraduate students at the University of Georgia, some of whom participated in the study voluntarily, while others got class credit for their involvement. While there was a disproportionate number of females in the study (72 percent), this was due to the high percentage of women majoring in psychology at UGA and was adjusted for in reporting the research.

    In practice, the study would need to be more careful to set up (and control for) real world circumstances before it could conclude more than cognitive racial bias in undergraduates. Undergradates are not HR people, hiring managers or job interviewers, nor have they been instructed on what factors they can look for and what they can’t when reviewing resumés. They likely have not yet taken an upper-year course on “Personnel Selection” in the Psychology or Business disciplines (since many study participants were PSYCH 101 students ‘getting credit for their course’). In fact, the study up-front admits that there was a disproportionate amount of females in the study, when statistics still currently show a disproportionate amount of male managers in North America (who actually make the hiring decisions!).

    Understandably, university studies are typically limited this dimension (can you imagine the study budget if you actually went out into industry and monitored everyone’s hiring practices? Wow!) But, I think we should be slightly cautious about sample representation and bias when making our conclusions on this study.

  8. Crommunist says

    You’re probably right to be skeptical about how well this finding maps onto real-world hiring practices, but that’s more a question of magnitude than it is anything else. What the study does establish is that colourism is a real, measurable phenomenon. It was also in Georgia, which has a very different history of racism than, say, New Hampshire (or Toronto). No one study is capable of definitively answering this kind of question, but it’s certainly a compelling piece of evidence that there is something going on here.

  9. Nice Ogress says

    I haven’t any doubt at all that color bias is real. Of course it’s real.

    I’ve gotten the handshake for being the ‘nicest’ applicant in a hiring drive, and it was pretty obvious at the time that I wasn’t the most qualified, best dressed, politest, or whatever-have-you. I was the ‘nicest’ because I was young and white.

    A lot of folks in HR trust their ‘gut feelings’ about hires, and unconscious bias is just that – unconscious. They tell themselves, ‘oh, she’s got a nice face’, or ‘she looks honest’, or ‘she seems like she’ll fit in’, but the subconscious trigger they’ve hit on is whiteness. This person is similar to me. Therefore, she is good.

    I’ve often joked that I have a ‘baby face’, and I do. My cheeks are round and my eyes are wide and slightly downturned, which makes me look ever-so-slightly sad all the time. Strangely, though, I’ve noticed that darker-skinned folks with these same physiological features don’t seem to share the same advantages I do. GOSH I WONDER WHY THAT IS. TRULY IT IS A MYSTERY.

    I’ve actually been strongly advocating for more diversity in my workplace for the last three years. Fortunately my bosses are sensible folks, so it’s not onerous or dangerous (particularly) for me to do so. I don’t do this out of ‘white guilt’ or whatever, I do it because I can.

  10. audiolight says

    What the study does establish is that colourism is a real, measurable phenomenon.

    Well, the measurable phenomenon of (unconscious/uncontrollable) cognitive racial bias has already been established by many other psychology studies previous to this one. But yes, this study certainly adds to the body of evidence for the cognitive racial bias phenomenon.

    No one study is capable of definitively answering this kind of question

    For sure, but there are certain study constructions that could make the conclusions of the study more directly applicable. University undergraduate studies are always “easy” to do (they’re cheap and local), but they unfortunately lead sample biases that researchers need to be aware of when drafting their study conclusions. (Although, I do also understand that we’re currently talking about someone’s Master’s thesis!)

    I’d love to see a study done with certified PHRs/CHRPs and see if the racial bias effect is still *as strong* as we might expect it to be, when compared to the general population.

  11. says

    I think your point about excavating and being aware of our own unconscious biases is crucially important. I’m beginning to wonder if Humanist communities should engage in sustained anti-bias education to try to get at this.

    They just released a similar study which seemed to demonstrate prejudice against gay applicants too, particularly when hiring for posts which are seen to require what have traditionally been considered more “masculine” characteristics (assertiveness etc.). Here it is:

  12. A. Noyd says

    Retired Prodigy Bill (#3)

    To racial bias you can add height and weight bias, speech bias, intelligence bias, tatoo bias, clothes bias, all sorts of things.

    Or combinations where one bias is masking a more controversial one! Some random white woman I encountered outside a gas station somewhere between Oregon and Tennessee felt she needed to share her opinion of how impractical and silly sagging was. There were a couple of young black fellows nearby sporting the fashion, which apparently unsettled her enough that she felt compelled to attempt to establish special white-people solidarity with the nearest available white person. My honest reply was that I don’t consider sagging any more impractical or silly than high heels, and that it’s refreshing to see an inanely restrictive fashion adopted by men for once. It was funny-sad watching her try, in the face of massive solidarity fail, to defend the now-apparent double standard without owning up to the racism that lurked below it.

  13. says

    I stumbled on your blog whilst researching on colourism. I can assure you that it is real and certainly practised unashamedly in Africa. Methinks that this ‘Priming’ is done from the cradle and is seen to be acceptable. Hell we even have songs written about ‘Yellow-yellow ‘ girls an apparent reference to light skinned ladies who are considered to be more attractive


Leave a Reply

Your email address will not be published. Required fields are marked *