The Ineffectiveness of Profiling

If you had the choice between Airplane A, where all the passengers from countries considered “high risk” were systematically searched, and Airplane B where they weren’t, which would you chose?

This is the question I was asked following my column on Tuesday, in which I wondered about the legitimacy of the profiling directives imposed by the Obama administration on travelers from 14 countries considered “high risk.”

The question, as it is formulated, suggests that the correct answer is A. I would have believed that, before digging further into the subject and realizing that common sense, in this case, is misleading.

Whether we want it or not, in this time of terrorist threats, profiling is a necessary evil in airports. But there are several types of profiling. And it’s wrong to believe that all are equally effective.

In a piece published in The New York Times, security expert Bruce Schneier, author of the book “Beyond Fear: Thinking Sensibly About Security in an Uncertain World,” distinguishes two types of profiling. They are behavioral profiling, as its name suggests, consists of paying attention to the way a person behaves, and automatic profiling, which consists of paying particular attention to the name and the nationality of the person.

According to Schneier, the first method – recommended by Canadian authorities – can be effective, though it’s hard to put into practice correctly. As for the second – officially adopted by the United States as of Monday – it’s not only ineffective, but actually puts our security in even further danger.

Terrorists don’t all have the same profile and the same mustache. They don’t all come from the same country, don’t all have the same skin color. There’s no software that can pick them out from a crowd of travelers. American conservatives are calling – without jest – for a separate security line for all passengers named Abdul or Mohammed. But that wouldn’t have stopped Richard Reid, the terrorist with the booby-trapped shoe, a Brit with Jamaican origins. Nor would it have stopped Jose Padilla, the American from Puerto Rico who became an Islamic radical, accused of plotting a “dirty bomb” attack. And that wouldn’t have stopped a Timothy McVeigh, either, completely white and American.

In the search for a simple solution, some propose imposing stricter searches only on Muslims. This miracle solution, as it is seen in the eyes of some on the American right, is considered by numerous experts to be as ineffective as it is uselessly discriminatory. Not least because, in practice, it would be a real headache, as the ex-FBI agent and terrorism expert Michael German underlined in his own New York Times piece. Would we rely on a person’s name? On his appearance? On her Arab-American origins? The vast majority of American Arabs, he reminds us, are not only devoid of any sympathy for the terrorists, but also follow the Christian faith.

Such religious profiling would require us to disproportionately target blacks, Asians, whites, and Latin Americans, notes the terrorism expert. As an exercise, it’s unjust, tedious, and not very promising. It would be much more useful to devote additional resources to intelligence services than to get involved with such an operation. That’s not even counting that avoiding such profiling would prevent the terrorists from exploiting its injustice, clothing it in the rhetoric of victimization.

While racial (or national) profiling is a measure that may calm those who scream about lax security, statistical studies show that it’s no more effective than random searches. Worse yet, observes Bruce Schneier, it creates two paths through security: One where we’re very attentive to the slightest detail, and another where we’re much less so. This ultimately invites the terrorists to find a way to take the less-watched path and to recruit, as they do already, combatants who would evade the profiling. On the list of the alleged Al-Qaeda criminals most wanted by the FBI there are already people like Adam Gadahn, a young American born in Oregon who was raised in a Christian family. Automatic profiling would be totally ineffective in a case like his.

So, to answer the question: Airplane B is a better bet than airplane A. And if the passengers on Airplane A could be subjected to a smart behavioral profiling, that would even better. As long as we’re constraining liberties, we might as well do it effectively.