Profiling

Having just returned from a month of overseas travel, and plenty of long lines at airports from Los Angeles to Heathrow to Nairobi, the issue of airport security was never all that far from view. Add to that the Christmas day attempt by Umar Farouk Abdulmutallab to blow up the plane he was on by igniting first his loins then himself then the aircraft, a ballsy attempt if nothing else, and you can’t help wondering about the issue of profiling.  

Profiling is the shorthand name for paying more attention to some passengers than others when it comes to the amount of scrutiny and level of searches and questioning each passenger will receive. Profiling is the idea that some categories of people are more of a threat than others, statistically speaking, and that best outcomes follow from concentrating more resources on those most likely to pose a threat. 

For instance, based on all you know about the world, who do you think is more likely to try to blow up a plane? Is it a) a young, muslim male or b) an octogenarian, Danish female? If you think the answer is a), then you need a reason for arguing that both should have the same odds of being subject to extra scrutiny when they pass through security at airports. You need an argument for ignoring what you know is a statistical truth.  

The arguments most civil libertarians give, when trying to argue against profiling, is one of these. Some claim that profiling will itself turn into a sort of a recruiting tool for terrorism, as those subject to it will become bitter at regularly being singled out. And that may well be true. But even if it is true, the gaps in this argument are huge. We are being asked to imagine that there will be a causal chain of events moving from someone being regularly stopped at Sydney airport for an extra few minutes of searches to his (and odds are it is a ‘his’) deciding to join al-Qaeda and seeking to become a suicide bomber.  

In any cost-benefit analysis I can imagine, that miniscule possibility will be outweighed by the significant benefits of a far more efficient use of security staff’s time.  

The other main anti-profiling argument is this, and it’s the one Shami Chakrabarti, the director of the United Kingdom’s civil rights group Liberty, likes to use. After alluding to the first argument above, Ms. Chakrabarti also points out that terrorist groups aim to recruit from a wide variety of backgrounds.  They will try to find suicide bombers who don’t fit the profile.  

And surely she is correct about that. But so what? Profiling, when done properly, is not just about who is young, or male, or holds a particular set of religious beliefs. It’s about all the things that go into making a solid generalisation.  

Go back to our friend the underwear bomber, Mr. Abdulmutallab. And leave aside the fact his dad had tried to warn the American embassy in his country, and that his name was on a British government watch list. 

The fact is that our underwear bomber had bought a one-way ticket with cash; that he had not checked in any luggage for a journey taking him near-on half-way around the world; that his ultimate destination was the United States, target number one for al-Qaeda; and that he came from a half muslim country not entirely free of fundamentalists. 

Profiling, as I said, is about playing the odds.  Of course lots and lots of people might fit into all these categories and be perfectly innocent. Indeed we would expect that going in. After all, we live in a world where tens of thousands of people, probably more, are in a commercial flight up in the air at any time of the day.  By contrast, suicide bombers on flights are incredibly few and far between.  

Worse, nothing any government can do will ensure the fanatical terrorist never succeeds. Even the most up-to-date body scanners and best X-ray machines fail sometimes. This is about making things more difficult for the fanatic, in a world where nothing can make it impossible for him. It’s about weighing up benefits and costs and opting for a course of action that is likely to produce the best net overall outcome. 

Now that means that we can all accept that terrorist groups will want to expand the scope of whom they recruit and nevertheless see the force in allocating limited search time in part based on statistical generalities.

After all, we are not talking about a world in which some people will skip merrily through security with no checks at all. Every single traveller, and all of his or her bags, will be checked. All of them will go through the detectors and scans. What we are talking about is who, normally, will get the extra few minutes of searching and questioning.

Advocates of profiling say it should be whomever falls into the groups whose conduct (such as buying a one-way ticket and having no checked luggage) or whose characteristics (say, being young, male and from Saudi Arabia) makes them statistically more likely to be a threat.  

It seems to me that people who reject this approach either prefer to wallow in pieties rather than confront uncomfortable facts, or they simply don’t care that the facts in question are true because they see the world in moral abstractions such as ‘discrimination is always evil’ or what have you.  

Readers attracted to the latter outlook might like to remember that when insurance companies charge young male drivers more than young female ones, or when they base premiums on the fact women tend to outlive men, they are profiling. They are making a decision based on statistics, on what is likely (though not always) to be true based on past experience.  

It makes sense to me.

James Allan is Garrick Professor of Law at the University of Queensland.