Popular Posts

Thursday, March 11, 2010

The Blonde Jihadist

The indictment of Colleen LaRose, a 46 year old white woman who was charged with providing material support to terrorists has raised, once again, some interesting questions about racial or ethnic profiling. LaRose, aka "Jihad Jane", converted to Islam "several years ago" according to law enforcement officials and was involved in trying to recruit jihadists to kill Lars Vilks, a cartoonist whose "depictions of the Prophet Muhammad incited protests by Muslims".

Rachel Maddow at MSNBC has pointed to the case of LaRose as a clear example of why racial profiling for law enforcement purposes is pointless and "stupid". Maddow's point, presumably, is that profiling doesn't work, because when a 46 year old white woman with blonde hair can end up "Jihad Jane", it's clear that not just young men of Arab descent are engaged in terrorist planning and activities. Maddow, in other words, makes what I'll call the Effectiveness Claim: profiling doesn't work because people outside the profile engage in criminal or terrorist activities. This makes the profiling strategy bad, because when looking for people who fit the profile, actual criminals or terrorists could slip right by. Fair enough.

Whatever the merits of the Effectiveness Claim, however, it's doubtful that Jihad Jane provides much confirmation of it. For one, LaRose is an outlier, an abnormality, which is why she ended up on the Rachel Maddow in the first place (if "she" had been a "he", from Saudi Arabia, say goodbye to MSNBC coverage). Outliers do not disprove statistical norms. Not every person who is a terrorist will fit the profile, as even profilers agree. The rationale behind profiling is that most of the people who engage in some behavior will fit a profile, or even that many will. The point is that the distribution of factors like ethnicity, age, and gender in the population of terrorists are not simply random. So, given the Effectiveness Claim, the LaRose case says nothing, since we would expect outliers in any realistic sample of terrorists.

Second, the LaRose case is troubling for the Effectiveness Claim because, recall, LaRose is a self-admitted Muslim. In this sense, then, she fits a profile of jihadist terrorist types rather than helps disconfirm one. She is, in other words, another example of a Muslim engaging in jihadist terrorist planning or activities. This of course doesn't show that all Muslims are terrorists (certainly not true), or even that most or many Muslims are terrorists. But it does show that one more Muslim is a terrorist, and it does not show that non-Muslims are terrorists, too.

So, the LaRose case does little, actually, for proponents of the Effectiveness Claim, like (apparently) Maddow. What Maddow and others who oppose profiling might want to argue instead is something I'll call the Moral Claim, which says that because not all terrorists are members of a particular religion or ethnicity or gender, even if many are, it is morally unacceptable to profile people based on those factors. It's much harder to argue with the Moral Claim, particularly when one recognizes the frustration and humiliation that wrongly targeted people must feel when they find themselves victims of profiling.

Read story here.


Rebecca said...

A question about profiling has always stumped me: How many points on a profile does a person need to hit in order to actually fit a profile? Muslim, Arab nationality, male, age 17-40, arguably can hit a "profile." But, how many of those in that profile are actually a terrorist given the odds? Your Moral Claim makes sense, in many ways if looking at probability of "terrorists fitting a profile." So, it follows to ask, is profiling an effective tool?
I guess it may be the equivalent of asking "how many licks does it take to get to the center of a Tootsie pop." Becky

Erik J. Larson said...

Becky, not sure on either score (number of points on a profile, or licks to get to the center of a Tootsie pop). I suppose in the latter case it would depend on the amount of saliva present at the time of the licks, and --- just kidding.)

On a serious note I'm abivalent about profiling. It seems wrong to grab people out of a line because they look "like terrorists", of course, and in this sense I think we can all agree that such policies would not constitute a good society. Sounds positively Orwellian, actually.

On the other hand, if a woman reports that she was attacked in the park, but can provide no physical description, the police are not likely to start looking for another woman. Why? Because men mostly attack women in parks. To make this even more clear, imagine that the woman described some actions consistent with an attempted sexual assault, like grabbing at body parts and so on. In this case even if the woman could provide no physical description, it would be utterly absurd for the police to spend time and resources questioning other women, though it's of course possible that the assailant was another woman.

So, profiling in this sense is part of police work. And note here that the profiling I've described is NOT just based on behavior (acting suspicious and so on), but involves including prior knowledge about the likelihoods of men or women attacking women in parks.

All of this is to say that I think the issue is very complicated, and I'm not sure what the just society should do in all cases.