Harvard Finds a Cure

MITHRIDATES
So Eric Holder thinks we're a nation of cowards when it comes to race and George Bush doesn't care about Black puppets, but where do the rest of us stand? To find out I took one of Project Implicit's association tests. And like the vast majority of people who take the test I was found to have a preference for white people. Could it be true? After my liberal upbringing, living in a predominantly black neighborhood for several years, and voting for a black President? Hell, I even picked a black dog.

So are we all racists deep down? Maybe so. Who knows, really? But here are a few alternative explanations. Before I give anything away, take the test here. There are tons of them, but take the one on race. It's definitely worth doing.

More...

First explanation: The test takes the form of a game in which we try to sort black, white, good, and bad into the right categories. It takes some getting used to and, not surprisingly, I think I got a lot better at it after a few minutes of playing. The first time I played they gave me good/black and bad/white and I didn't do very well; the next time they gave me good/white and bad/black and I did better. They concluded that I was better at assigning good to white than good to black and that I have an implicit preference for white people.

But hold on a second. Maybe I'm not racist and just got better at the game. What if they gave everyone the test in the same order. Could they be that stupid? So I did the whole thing again and they gave them to me in the reverse order. They still concluded I had a preference, but only slight this time. At least they appear to be randomizing the order. This may lead to valid conclusions about the group, but not any one individual.

But it does raise an issue about the "cure" as described in wired.

After being trained to distinguish between similar black male faces, Caucasian test subjects showed greater racial tolerance on a test designed to to measure unconscious bias.

Hmmmm. Or maybe they just got better at the game like I did . . . Sounds a bit fishy.

Second explanation: The way I took the test was to think "black-good" and "white-bad" and sort as best I could. This reminded me of a speaker we had at my lefty-minded, New England prep school who told us about self-esteem among black kids by noting how the Thesaurus was filled with negative words related to "black" and positive words related to "white". As in black mark, Black Monday, black magic; and white knight, white wedding, whitewash. Or note the antonyms for "white" on thesaurus.com: black, dark, dirty.

This is a big problem with the researchers' conclusion. They take the synonyms of "black" — bleak, atrocious, horrible, sinister, nasty, foul, threatening — and test to see how good we are at associating them with "white". All our lives we've seen and heard negative associations with the word black and positive ones with the word white. It would only be natural for someone who didn't care one way or the other about black people or white people to perform the way most people did on the test. This could certainly be problematic for the self-esteem problem described above, but it certainly causes us to doubt the assumption of preference for white people over black people.

Third explanation: Well, let's just hope it's the first or second . . .