Is everyone racist?

Nicholas Kristof asks tactfully whether everyone is a little bit racist, but since the “little bit” includes what might lead one to kill a person of another race, I’m leaving out the little bit.

The column is a fairly short but useful explanation of implicit bias. There are some of the usual links – e.g., to the implicit bias test – but one that I hadn’t realized is online. That’s the shooter task or “the police officer’s dilemma”. You are asked to decide whether someone holding something has a gun; if so, you press “j” for “shoot” and otherwise “f”. All the people one sees are men. The test is quite famous and I believe people, both white and not, are more inclined to shoot unarmed people of color than unarmed white people.

I failed miserably because for most of the time I couldn’t react in under a second, which is required. If police are making these split second decisions, I think we should try hard to stop that practice. I quickly found myself trying to create heuristics, ways of reacting before I had a full identification of what was being held. Pretty quickly I was prepared to shoot anyone who had something narrow and straight protruding from a hand. And that’s one of the kinds of mistakes police report themselves as making.

So I don’t know how the test fares as a test of racism, but I do think it should make us all very worried about armed policemen encouraged to shoot very quickly.

6 thoughts on “Is everyone racist?

  1. I got to 500, but it just wouldn’t stop :)

    I don’t get how Bayes’ theorem is relevant. Implicit biases aren’t evidentially based, afaik. Indeed, the typically cause evidence fails: You *don’t* update because of implicit biases or you over update. So it’s not just racist priors, but the fact that nothing like Bayes’ theorem is actually being applied.

  2. It’s worth noting that there’s a long tradition in social justice movements that would reply to the author’s question with something like: No, because racism is a property of social systems, not of persons. Implicit bias tests measure individual racial prejudices, not racism. Of course, those prejudices are importantly related to racism, and a person with racial prejudices might be more likely to do and say things that advance racism.

    I think when it comes to the case of the cops, it would be really awesome to have the cops work on their racial prejudices. Maybe the IAT test can help advance that. But in the USA, even if we had a country full of cops that aren’t confusing wallets for guns, we’d still have a racist criminal justice system.

Comments are closed.