The critical error behind almost all reasoning about “racial profiling” for terrorists

Proponents of racial (ethnic, religious) profiling for terrorists inevitably begin by citing the same demographic comparison. In the words of notorious racist Ann Coulter,

“The perpetrators [of terrorist attacks on Americans since 9/11] have all had the same eye color, hair color, skin color and half of them have been named Muhammad…This is not racial profiling; it’s a description of the suspect.”

Cheekiness aside, the argument is simple: A terrorist is more likely to come from x-racial-ethnic-religious group rather than groups y or z; therefore, we are warranted in searching for terrorists among persons who are (or appear to be) from that group particularly. This might entail singling out men fitting this description for baggage or ID checks in subways, or funneling them through a separate check-in line at airports.

For the sake of argument, let us assume that Coulter’s data on terrorists is correct—that most terrorists come from this Arab/Muslim sector (though in fact they don’t). Let us also assume there are no logistical or moral obstacles to profiling (also dubious).

My objection, rather, concerns the form of the argument. In short, Coulter’s numbers could be 100% sound, and it still wouldn’t make racial profiling an effective technique for finding terrorists.

* * *

The problem is the profiling argument gives way too much weight to the (statistical) relationship of ethnic groups to one another—that is, to the percentage of terrorists within one group as compared to the percentage in each of the others. I contend this relation—the very lynchpin of the profiling argument—has precisely zero relevance to the business of hunting terrorists. If we’re thinking properly, what should matter instead is the statistical relationship of terrorists to their own racial groups.

To put this visually (proportions not accurate):

The ‘profilers’ only care about how the beige sliver on the left compares in size to the sliver on the right: If the right is bigger, they say, we should search A for terrorists, rather than C. But this comparison is irrelevant. What matters is not which sliver is larger but how large—in absolute terms—the larger slice actually is.

Put another way, the most this kind of comparison by itself could tell us is that terrorists are unevenly distributed among ethnic groups. What it cannot tell is whether terrorists are represented in any one ethnic group highly enough to make profiling a viable technique for finding them. Just as only knowing your meal is bigger than your neighbor’s will not tell you whether it is big enough to satisfy you. (For both meals could still be very small.)

Testing profiler logic: Two analogies

Some everyday examples of similar reasoning should further clarify.

Analogy #1: Children from “broken homes” are by some percentage more likely to become serial killers later in life than children from two-parent homes. This hardly means that shaking down a bunch of people with divorced parents would be a wise deployment of police resources. Even if 100% of serial killers had divorced parents, those products of divorce who actually commit serial murder are such a tiny minority among all children of divorce that the strategy would still be suspect.

Clearly, the key statistic isn’t (i) how many offenders come from divorce, but (ii) how many people who come from divorce actually turn out to be offenders. Focusing on the first stat only tells us that one set of odds is greater than some other set of odds; it tells us nothing about how good the second set of odds—the one we are banking on—actually is. We may as well say that buying two Powerball tickets has a greater chance of winning millions than buying just one; that’s true, but it doesn’t make buying one or buying two very likely to yield a winner. Likelier is still miles away from likely.

* * *

Analogy #2: Imagine we have a haystack which has some probability of containing a needle. (Let’s say, there is some probability that one of the straws is a needle.) Let this equal the probability that a given, random person fitting the “terrorist profile” is a terrorist; that is, drawing a random straw is as likely to yield a needle as detaining a random “Muslim” is likely to yield a terrorist. Let us assume this method of finding needles is ineffective, counterproductive, even (somehow) immoral; also, that we have some far better method of finding needles in haystacks—using magnets, X-ray, floating the straw on water so the needle sinks, etc. We still want to root out needles, but have long abandoned the strategy of drawing random straws.

Now, imagine we discover that all along there has been a second haystack nearby which has an even lower probability of yielding a needle than our haystack. Perhaps we discover several more, each with some probability of success lower than the original, but still greater than zero.

It has become clear that a needle is more likely to come from the first haystack than from any of the others. Still, it would be irrational in the highest to conclude that we should, on this basis, resume our random straw-draws. The simple fact that a less promising haystack exists does not magically make checking this stack a good idea, if it wasn’t a good idea before. If an activity’s probability of success is extremely low, it isn’t made better just because there exist other activities whose probability of success is even lower. (That’d be crazy, right?)

Conclusion

Profiling advocates are confusing better odds with decent odds. The simple fact that terrorists are more likely to come from Arab/Muslim men than from some other group doesn’t mean that the likelihood of randomly finding terrorists among Arab/Muslim men is very good at all. And that is the real question.

So: Just how good is that that likelihood? I haven’t exactly crunched the numbers; you can do the math if you like. (The burden of proof is on the profilers anyhow.) But there are millions of persons in the world who fit Coulter’s “profile,” and the number of these who commit terrorist acts against Americans is, in relative terms, very nearly zero. Even fewer do so in those stereotypical ways that profiling would address. Fewer yet operate in the U.S., where ours laws can actually penetrate.

Clearly, we are dealing with numbers akin to those “children of divorce” who commit serial killing. It is quite likely that if we incarcerated every other Muslim male in the world, it would register nothing in practical terms to diminish the odds of the next terror attack. Yes, we can theoretically halve a .000003% chance of something. Getting married later in life will halve one’s chances of committing suicide someday. Hell, there is shit you could do right now to seriously diminish your chances of being brainwashed by a cult or eaten by a mountain lion. I mean, Who gives a shit? Differences of this infinitesimal grade do not drive anybody’s consideration of anything in the real world; far less should they drive national security policy.

___________________________________________________________________________________________

Bonus: The above aside, profiling is still a shitty idea

This is not to mention that radical Islamists come in all “colors” and (duh) will easily work around any profile we make.

In addition, ethnic profiling is counterproductive; it alienates the very communities which are most critical for intelligence on (and testimony against) the potential attackers that move and live among and gain cover from them. As social psychologist Tom R. Tyler masterfully argues in his seminal Why People Obey the Law, individuals who feel they are singled out unfairly by law enforcement tend to avoid contact with the latter as much as possible; at the same time, they internalize this treatment, minimizing their stake in the system, giving them less incentive not to offend.

Finally, as with finding needles in haystacks, we have alternative strategies for fighting terrorism that are superior to profiling. Granted, jihadists will cite a number of gripes against the US if you ask them. Some of these concern cultural factors like women’s liberation and sexy music and movies. But according to the evidence, these aren’t the “root” reasons they turn to terror. As I’ve argued elsewhere, the violence is above all a response to US foreign policy in and toward Muslim countries and populations. Bin Laden, for instance, clarified his grievances across many years in interviews with Robert Fisk.

Michael Scheuer’s Imperial Hubris: Why the West is Losing the War on Terror summarizes nicely the main Islamist concerns; to paraphrase:

• U.S. support for Israel’s occupation of Muslim Palestine
• U.S. and other Western troops in every state of the Arabian peninsula
• U.S. support for Russia, India, China, Phillipines and Uzbekistan against their Muslim populations and militants
• U.S. pressure on Arab energy producers to keep oil prices low
• U.S. military and economic sanctions on Muslim nations (sometimes through the U.N.): Syria, Libya, Iraq, Sudan, Pakistan, Iran, Indonesia, Somalia
• U.S. support for apostate, corrupt, and despotic Muslim governments (often a vehicle for the above concerns)
• And now, via the WOT: U.S. occupation of Iraq and Afghanistan; incarceration without trial of thousands of Muslims suspected of being mujahideen; pressure on Muslim governments to track, control and limit Muslim’s donations to charitable organizations; pressure on these governments to tailor school curricula to give a more pro-Western brand of Islam

Thankfully for us, these concerns are quite reasonable, technically solvable, and are morally “overdetermined”—that is, they should be met for a host of reasons even aside from fighting terror.