Indiscriminate

Another blog post cobbled together from Facebook posts. I should get a way to edit my research code on my iPhone…


Last week, a student came to talk to me about postgraduate research and I suggested that she speak to my colleague as her previous experience was aligned with his current research. I mentioned this to the colleague in question and he asked me wether she “looked Nigerian”.  I said I honestly couldn’t say, as I don’t know what a Nigerian is supposed to look like, to which he replied “oh, do all black people look the same to you?”


This exchanged bothered me enough to tweet about it, but I’ve been unpacking why it bothered me for a while. The thing is, I can distinguish between black people easily, it’s just that I don’t think I could classify a stack of pictures of any people (of any colour) into country groups with any accuracy.  I’ve been thinking that I would have remarked on a French accent, because I can recognize it. I’ve been looking at photos of Nigerians online, and I’ve even googled around for West African facial features, but to be honest I would say many of my South African students share those same features, which as far as I can glean are wide face, wide nose and dark skin.



I would say context adds a lot of clues. If you already know someone is from India, you can go into nuances, but how easy is it to take a picture from a stack and say with certainty that a person is from India instead of Pakistan – or South Africa for that matter? Is it easy to tell someone is from West African descent, or is it a stereotype (and one I am unfamiliar with). I know what the stereotypical Dutchman looks like, although I have only met one person from Holland that actually looked like that (tall, blond, biggish teeth). I just didn’t know the Nigerian stereotype, and found its very hard to get that info from looking at Nigerians.

There’s a more subtle point here about the utility of such a distinction. In fact, I find it vaguely racist (or at least discriminatory) to categorize beyond pure description in identifying a student. It’s one thing to refer to skin colour when describing someone’s appearance, as this is directly descriptive. It’s another thing to add references to race when talking about a person’s academic performance, as it’s a clue that you feel that race is related to academic performance.


Many people I spoke to about this repeated the tired old line that stereotypes are based on a grain of truth. Unfortunately, when it comes to perceptions of national character in personality, this paper shows that everyone agrees on the typical character of a German, but Germans don’t test the same as these stereotypes. And if you’re thinking “But they didn’t test the real Germans”, look up the No True Scotsman fallacy. Of course, you may also be thinking that casual classifications like these are not harmful, but you may be interested in the idea of stereotype threat. By the way, if you think you’re impervious to the group effect, have a look at the out-group homogeneity effect, which makes groups you are not a part of look a lot more homogeneous than groups you are a part of. Basically, this explains why you don’t here South Africans asking “did she look like a South African?”


I think it serves everyone best to be vigilant in the use of possibly biased language. I try to refrain from gendered language except where it applies to a known individual. Similarly, I try to avoid mentions of race or other irrelivant information when describing someone’s competence. I know people who seem obliged to precede positive reports of minorities by a description like “oh, that black girl who just started working here is doing so well”, while the “default” white male’s competence is a given. This makes me cringe. I know a couple of my preconceptions end up slipping into my language as well, but at least I am (partly) aware of the issues. If you clicked on the three links in this story, you may be more aware now, too.