You've said that security can be improved by randomly applying better detection measures. How might this work?
We know that terrorists watch our systems. When a system becomes routine and predictable, they will identify the vulnerability and go after it. The shoe bomber is an example and Abdulmutallab is another. There should be an element of randomness in which some passengers would be subjected to the more rigorous search methods based on the passenger's records and documentation and on behavioral assessments that security professionals make. Observations of people's behavior are just as important as the search procedure itself.
On top of that, you could have a computerized system that selects passengers at random for higher-level screening. This would not be an interrogation or an interview, and it would also apply to frequent travelers. We know that this type of randomness really is an impediment to terrorist planning, but it's also important from the standpoint of civil liberties because passengers are not singled out for any particular reason.
What are the chances that we'll see such an approach at airports in the near future?
There's a certain amount of receptivity to it, certainly on the part of the security screeners. Where it runs into barriers has more to do with public attitude and, to put it bluntly, posturing politicians. Americans prefer their security to be passive, nonintrusive and egalitarian. People feel more comfortable when exactly the same procedures are applied to every single person, which is of course the dumbest way we can do it. The minute you try to segregate people, it immediately raises allegations that this is somehow based on ethnic or racial profiling or there are other nefarious criteria that are being used to make these decisions.
With regard to the attitudes of politicians, sometimes people in Washington pretend that what we have at airports around the country is 100 percent prevention. The fact is there is no such thing as 100 percent prevention.
How much progress has been made with regard to air travel security since 9/11?
If you take the long view on this, security has had an effect on the number of hijacking attempts and airline sabotage attempts. There is no question that, as a consequence of screening measures and other factors, the number of attempts by terrorists has significantly declined over the years. In the 1970s and 1980s we were looking at a terrorist hijacking or an attempted terrorist sabotage of an aircraft something close to once a month. If you look at the post-9/11 environment, clearly there have been plots and failed attempts, but we're looking at one of these incidents per year. We have obliged our adversaries to make smaller devices and use exotic substances to try to conceal them, and that renders them far less reliable. Yes, Richard Reid made it onto a plane with his shoe bomb, and Abdulmutallab made it on with his underpants bomb, but the bombs didn't work. So we've decreased the number of attempts and increased their operational difficulty. That is a positive result, but we're running into this dilemma now as the devices get smaller and concealment gets better—the terrorists will work this out. The challenge is how do we deal with that in our society in a way that is acceptable to society?