Schneier on Security

Richard Steven Hack • April 24, 2007 2:53 PM

Racial profiling to stop terrorists works as long as the terrorist groups are dumb enough to send obvious foreign nationals to pull off operations.

As soon as some operations fail because of that, they will learn and racial profiling will stop working.

It's that simple. It's not hard to use front people and patsies to pull off terrorist operations. It's been done by other groups and will be done by all groups - at least those that have an international character - sooner or later.

A similar problem arises with stupid projects like searching people at random in the New York Subway. With the fact that ninety percent of people are carrying some kind of bag these days, how do you pick who to search? Cops know what suspicious people look like - or they think they do. Has that ever been tested scientifically to see whether they really do or that it's just "I don't like that guy's face - or race".

And what are the odds that any given search will actually turn up someone with a bomb, given the number of people involved and the number of cops involved?

A more interesting question: how many "false flag" operations can you detect with such profiling? Can you tell an Israeli from an Arab with 100% proficiency?

The Israeli Mossad was caught running a "false flag" operation in the Palestinian area by posing as "Al Qaeda" recruiters trying to recruit Palestinians. That operation was exposed by Palestinian security.

How many others weren't?

In the intelligence community, you never know who is working for whom. Therefore, why assume that any given terrorist operation is performed by any given group, absent some sort of authenticated claim of responsibility (and how do you authenticate a claim?), let alone any given ethnic group?

Another interesting bit related to this is that in all the tests where airport security is found wanting, I think we can assume that the testers were the same ethnicity as the security people. It would be interesting to run the same tests with people of varying ethnicity. This would provide some evidence that ethnic profiling has any validity at all.

Another problem with such tests is that a real terrorist is operating under emotional stress - and a tester is presumably operating with less emotional stress. How much does emotional stress influence the outcome of such tests?

People should read Richard Marcinko's Red Cell SEAL Team exploits. These guys went everywhere and busted into everything which is supposedly secure - Camp David, Air Force One's hangar, Navy nuclear weapons lockers, Naval Intelligence HQ, nuclear subs at Groton. They pretty much proved that you can get in anywhere with weapons and explosives if you want to - and if you look like Americans, anyway.

Worst comes to worst, I'm sure the Arab terrorist groups will start hiring Western terrorist mercenaries and criminals to pull off their operations if they have to, even without ideological reasons. As an example, there are still theories that McVeigh and his accomplice had Arab terrorist connections, and that they were either accomplices to Arab terrorists or patsies for same.