We Are All Juan Williams (歧視是不理智，不公正，但完全正常的)

We Are All Juan Williams http://www.slate.com/id/2272262
Associating minorities with crime is irrational, unjust, and completely normal.
By Shankar VedantamPosted Monday, Oct. 25, 2010, at 1:27 PM ET


Timothy McVeigh and Mohammed AttaJuan

Williams told Bill O'Reilly that he gets nervous at airports when he sees Muslims. For this, Williams has been roundly denounced as a bigot. But Williams' association between innocent Muslims and the perpetrators of the 9-11 attacks was less about bigotry—at least, bigotry conventionally defined—than about his mind working normally. To live in America in the post-9/11 age and not have at least some associations between Muslims and terrorism means something is wrong with you.

I am not suggesting that associating ordinary Muslims with terrorists is either rational or right. It's neither. But the association arises via a normal aspect of brain functioning, which is precisely why so many people entertain such beliefs—and why those beliefs have proved so resistant to challenge.
Print This ArticlePRINTDiscuss in the FrayDISCUSSEmail to a FriendE-MAILGet Slate RSS FeedsRSSShare This ArticleRECOMMEND...Single PageSINGLE PAGE
Yahoo! Buzz
Facebook FacebookDigg DiggReddit RedditStumble UponStumbleUponCLOSE

The left is wrong to wish the association away only by pointing out how unfair it is, because that denies the reality of how our minds work. The right is wrong to believe the association must be accurate merely because it is widespread.

It may be helpful to see how the bias works in a less incendiary context than racial profiling and terrorism. Let's say you have an upset stomach on the morning before you take a flight. (This example comes straight from my book.) You don't make much of it because you know that things that occur together are not necessarily related. Now let's say you have a second upset stomach the following week, right before you have to take another flight. Your conscious mind might still say, "coincidences happen," but now a part of your mind starts to think that flying brings on upset stomachs.
Advertisement

What's interesting about this is not that our unconscious minds confuse correlation with causation—our conscious minds do that all the time, too. What's interesting is that the events our unconscious minds link together are only the unusual ones. On both days you had upset stomachs, a newspaper may have been delivered to your door, but you don't associate newspaper delivery with your upset stomach.

These automatic associations make evolutionary sense. If one of our ancestors was wandering in a desert and came by a snake curled up next to the only tree on the landscape, her mind would connect not just that tree with that snake, but all trees with snakes. Illusory correlations are all about seeking out group patterns based on rare individual incidents: all trees and snakes and all flights with stomach upsets, rather than that one tree and that one snake, or that one flight and that one stomach upset. Scientists say correlation isn't causation, but, from an evolutionary point of view, if the snake-tree link is wrong, all that would happen is our ancestor would avoid all trees in the future. If the link was real and she failed to see it, she could get herself killed. Our ancestors constantly drew conclusions about their environment based on limited evidence. Waiting for causative evidence could have proved costly, whereas extrapolating causation from correlation was less costly.

The terrorist attacks of 9/11 were unusual. (Even if you take all the terrorist attacks in the world, they are still unusual.) In seeking explanations for those events, our minds are drawn to other unusual things linked to them—especially at the group level. When members of a minority group are associated with a series of unusual incidents, our minds inflate the connection between the two. As with the snake and the tree, our association is not limited to the particular members of the minority group but all members of the group.

Juan Williams pointed out on Fox that we do not associate Timothy McVeigh and the rude people who protest about homosexuality at military funerals with Christianity. But he didn't understand why our minds fail to make that connection. Illusory correlations disproportionately afflict minorities because, in making associations, we mainly link unlikely events. Whites and Christians are not minorities; they are like the newspaper delivered to our front door every day. We do not associate McVeigh with Christians any more than we associate our upset stomach with the newspaper.

Muslims are only the latest victim of illusory correlations in the United States. African-Americans have long suffered the same bias when it comes to crime. In every country on earth, you can find minority groups that get tagged with various pathologies for no better reason than that the pathologies are unusual and the minorities are minorities.

Whenever people who strongly believe in illusory correlations are challenged about their beliefs, they invariably find ways to make their behavior seem conscious and rational. Those who would explicitly link all Muslims with terrorism might point to evidence showing that some Muslims say they want to wage a war against the West, that a large preponderance of terrorist attacks today are carried out by Muslims, and so on. This is similar to our longstanding national narrative about blacks and crime.

But even if blacks and whites do not commit crimes at the same rate, and even if Muslims are overrepresented among today's terrorists, our mental associations between these groups and heinous events are made disproportionately large by the unconscious bias that causes us to form links between unusual events and minorities.

The researchers Franklin D. Gilliam Jr., Shanto Iyengar, Adam Simon, and Oliver Wright once conducted a simple experiment that demonstrated how illusory correlations work: They showed volunteers a television news program that featured a violent crime. Some volunteers were shown a white suspect, while others were shown a black suspect, but everything else about the program remained identical. The volunteers who saw the black face were more likely to blame blacks as a whole for rising crime than the volunteers who saw the white suspect were to blame whites for rising crime. (The volunteers in the white scenario blamed that individual suspect for the crime.) The bias showed up among white as well as black volunteers.

People in Thailand will associate white American tourists with pedophilia even though many more acts of pedophilia are committed by Thais. But white Americans are a minority in Thailand, as are acts of pedophilia. So you will hear Thai people shout until they are blue in the face about individual anecdotes showing white Americans who are pedophiles. (The same is true of gay men and pedophilia in the United States.)

When it comes to our associations between Muslims and terrorism, commentators on the left are being wishful when they imagine we can rid our minds of false associations merely by holding consciously egalitarian views. Commentators on the right are doing something much more dangerous, however: They are rationalizing and justifying a mental process that is fundamentally not rational and deeply unjust.
Advertisement

If you know there are 1 billion Muslims on our planet (low estimate) and you've heard of 1,000 incidents where Muslims carried out terrorist attacks (an exaggerated number), and terrorist sympathies were (improbably) distributed evenly across the world, the odds that a particular Muslim is a terrorist are about 1 in a million. A rational Bill O'Reilly should be much more exercised about asteroids striking Earth, or dying from dog bites, than about Muslims being terrorists.

The fact that so many of us subscribe to illusory correlations can be blamed on our unconscious minds. The fact so few of us challenge our unconscious minds? That's on us.