Colbert King's op-ed today—laying out the case against the use of racial profiling against "young Muslim male"—hits a bunch of the really crucial notes, I think, but it's possible to expand this argument a bit. First, a few misconceptions to clear up. Racial profiling probably isn't motivated by bigotry. Many minority police officers support the practice, after all, because they see it, sensibly enough, as a statistical tool. Moreover, for some intents and purposes, the statistical tool can work. Racial profiling can, in theory, offer a more cost-effective way to attack a certain problem. If, say, most people committing X action are young Colombian men, and if a reliable way to identify young Colombian men exists, then racial profiling will reduce X action. Clear enough, and you can see why so many people find the concept so attractive. Nevertheless, it's still an awful idea and ought to be abolished.

One of the big worries is that defenders of racial profiling want the practice to evade the "strict scrutiny" that generally gets applied to various forms of racial discrimination. Police departments and other proponents argue that, in racial profiling, they aren't taking action based solely on race—which is true—but rather using race as just one of a variety of factors to identify suspects. As such, they say, "strict scrutiny" shouldn't apply. Indeed, many mayors and governors will denounce racial profiling when "race is the only factor," but approve of other types, as if this makes all the difference in the world. It doesn't. In the real world, most discrimination doesn't take race as the only factor. Let's say I preferred to associate only with white people (or hire only white people, or admit only white people to my grad school program), but I would make exceptions for blacks and Hispanics who attended Ivy League universities. Clearly I'm discriminating by race, although race isn't the only factor in my decision. Point is, when we start approving of those types of racial discrimination in which race is just "one of many factors," we start heading down a troubling slippery slope.

That's the conceptual problem. The practical problems with racial profiling are more straightforward. For one, it antagonizes the group of people being profiled. One might argue that in the case of anti-terrorism profiling, the targets here are relatively small in number. (Not that many people ride public transit or fly on airplanes, after all.) That seems plainly false. Even a "Muslim-looking male" who never boarded an airplane would still know full well that if he wanted to do so, he would likely be stopped, and that in itself could cause resentment. Nor is getting pulled out of a line at an airport or subway station, as a result of racial profiling, just a "minor inconvenience," since the person being profiled knows all the while that it's not just this one time he's being yanked aside; rather, he's likely to have go through this process many more times in the future.

Now, as it happens in the case of, say, young Muslim men in America, I'm not sure if the resentment that would flare up as a result of racial profiling would necessarily "create" new terrorists. Maybe not. (That still wouldn't make it right, just less dangerous.) But it might piss off people who would otherwise help in a terrorism investigation, or who might call the police on a tip, or whatnot. Significant? It could turn out that way.

By the way, how many people would be affected if the police start profiling "young Muslim men"? A lot. A whole lot. As King notes, "Muslim-looking" men encompasses a wide, wide swath of races and nationalities, from Nigerians to Iranians to Indonesians. (Those three don't look anything alike.) Then you have your Central Asian Muslims, who often resemble the Chinese more than they do Mohammed Atta, not to mention your Chechens and other Caucasian Muslims, who can often pass for white. And so on. A lot of different people are getting profiled and antagonized here. Then we have to deal with the fact that many of the people we might think are Muslims probably aren't; around three-quarters of all Arab-Americans follow Christianity, for instance.

Meanwhile, even if racial profiling isn't motivated by bigotry, over time the practice would very likely create racial tension, or bigotry, among law enforcement officers, who, after all, would be out there day after day looking suspiciously at every Arab or North African they see. (Not to mention that they would likely have, over time, many a tense confrontation with "Muslim-looking men" who resent being targeted.) Pretending that these security personnel could continue to operate in a race-neutral manner day after day seems extremely naïve to me. Meanwhile, the practice would encourage civilians to view anyone they considered a "young Muslim male" suspiciously, which would further inflame racial tensions. How could it not? To top it all off, as King notes, police officers engaged in racial profiling will be far more likely to overlook white terrorists, who are as old as the republic itself, and include such stalwarts as Eric Rudolph and the dude pictured to the right. A police officer focusing hard on what that swarthy fellow standing in line is up to will almost certainly miss suspicious behavior by the white dude with the crew-cut and bulky backpack. Nor, for that matter, does it seem like it would be terribly difficult for a "young Muslim male" to pass himself off as, say, Hispanic or Greek (or white, if from the Caucus region) or whatever if one really wanted to pull off some bombing or other.

But maybe not. If cops were to use racial profiling against what they thought were young Muslim men, it might well very reduce terrorist incidents. Who knows? Nevertheless, even if that was the case, the loss of this statistical tool would have to be the price we pay for racial equality. As with all things, trade-offs sometimes become necessary. If preserving racial harmony means that the DHS needs to spend an extra couple billion dollars on some other, less cost-effective, security measure, well, fine. That sounds like a worthwhile trade-off to me. Now sure, one can imagine any number of "ticking-bomb" scenarios to question these principles—say that we had impeccable intelligence that a group of four Arab men were planning to bomb the New York subway tomorrow, but didn't know who they were; what then?—but clever hypotheticals like these don't disprove the general rule here.