A law professor stops—and frisks—the myth that racial profiling works.
Given Attorney General John Ashcroft’s recently announced plan to “question” thousands of young men about terrorism based on little more than their demographic profile, a searching debate on the merits of racial profiling would seem more urgent than ever. Harris (Law/Univ. of Toledo), writing before September 11, offers a conclusion that, if true, could have a profound effect on civil liberties in our suddenly anxious world: Racial profiling is not just bad; it doesn’t work. Not many people, of course, are willing to defend racial profiling as a stand-alone good, but the constant refrain from the practice’s supporters, especially those in law enforcement, has been that it’s necessary because racial minorities commit more crimes than other members of society. Harris convincingly explains what is known as the lamppost phenomenon: If law enforcement agencies look for violations amid a particular group, they are bound to find them at higher rates. When he turns to actual legal cases, his reportage is less convincing. Where he truly falls short, however, is where he promises most: his attempt to explain why racial profiling is ineffective. He contends that in jurisdictions where racial profiling is used, “hit rates”—the rates at which violations are found—are the same, or lower, for targeted groups as for other groups. Because hit rates are the same, profiling must be unjustified. But this analysis ignores the decreasing marginal probability of finding violations. A supporter of racial profiling could therefore reply that it should be used to the extent that hit rates are equal across racial and ethnic groups, even if that means stopping more minority drivers. As for Al Qaeda, Harris draws a line between preemptive and investigative profiling, and it’s not clear what his position would be on the current controversy.
Provocative, but not quite a direct hit.