Abstract: Probably not.
Psychology tends to take a rather gloomy view on racism: that we automatically categorise people on the basis of race, and there’s little that can be done about that. So I was interested to read this article by Vaughan Bell, reporting on some research from ten years ago which suggests that categorising people based on race can actually be unlearned really, really super-quickly. Could it really be true? Is overcoming racism that easy?
The paper, “Can race be erased? Coalitional computation and social categorization” by Kurzban, Tooby and Cosmides, published in PNAS (hur hur) certainly seems equally optimistic about the swiftness of unlearning racism. However, they’re being a little bit too optimistic.
The way they studied categorising people on the basis of race was by using a memory confusion protocol. Participants (most of whom were white, hispanic or Asian-American; no data on the relative ratios here, and the authors folded hispanic and white into the same category) saw a sequence of sentences with a photo of who had said it. From what the people said, it was apparent who was in a group with whom, and that there were two groups at odds with one another. After this, participants were shown the sentences, and had to remember who had said it. Errors, according to this paradigm, are telling about how participants are categorising the speakers, and they would be more likely to make within-category errors than between categories, i.e. they would mix up two people from the same group rather than two people from different groups. So looking at errors suggests how participants are grouping people together.
The photos used were of two black men and two white men per group, with the exception of a condition where the authors tested the effects of gender, so photographs of women were also included. What is known that in one experiment the people depicted in the photos all wore the same colour baseball jersey, while in the second experiment people in each rival group wore a different coloured jersey.
You can see where this is going, can’t you?
Ultimately, what was found was that in the first experiment, participants made more errors in categorising on the basis of race. In the second experiment, participants used the colour of the jersey rather than skin colour as a method of classification. In other experiments, perception of sex differences didn’t go away: the authors had hypothesised that this was more deeply encoded than race.
Well, that seems like pretty strong evidence yes? Not really. First and foremost, it is crucial to remember that this was a laboratory study, with variables strictly controlled. It probably is more generalisable than it was when the study was conducted–ten years ago, there wasn’t much communication taking place as intergroup conflicts between people represented only by short sentences and a picture of themselves, but these days there’s Twitter. However, we don’t tend to fall into these neat little groups which are equally racially balanced (and gender-balanced).
The implicatations of this research are explored more fully in a study from 2009, from Bavel and Cunningham. In this study, participants were assigned to a group which didn’t really exist–in their group, were twelve photographs of men, six black and six white. They were also told about the existence of another group, which featured a similarly mixed group. After learning the faces of the people in their own group and the outgroup, participants were tested on automatic responses to liking or disliking these faces. There was also a control group, which were not assigned membership to either group, but participated in the learning and evaluation task.
The results are kind of complicated, but ultimately, people are still kind of racist towards black people in the outgroup, while not racist towards white people. Basically, they were more likely to negatively evaluate black faces in the outgroup, while more likely to positively evaluate black faces in the ingroup, while positive evaluations of white faces held steady throughout. Also, across all conditions, black faces were still less liked than white faces.
Admittedly, this study has a lot of holes again. Once again, we’re looking at a lab study, and there are problems with the sample: there is no report of the race of the participants, which would be useful to know, and the majority of participants were women, while all of the stimulus faces were male faces.
At any rate, what this body of research is showing is not that it is easy to suddenly not see race but, rather, that we dislike black people less if we feel like we’re on the same side as them. Meanwhile, the other study shows we’re less likely to mix them up if they’re wearing their group affiliation really visibly. If anything, it somewhat exposes the underlying racism of the phrase “I don’t see colour”, so beloved by some white folk.
This is one of those instances where once again, it is crucial to be aware of the biases that our naughty little brains throw at us, and consciously strive to overcome them and work against them. That, and it’s important to remember that racism probably won’t magically end if everybody wears bright baseball jerseys to show off what team they’re on.