Charles V. Pena
Our America Initiative Advisory Council Member of Defense & Foreign Policy
The recent hostage crisis in Sydney, Australia that resulted in the deaths of two innocent people at the hands of an Iranian-born would-be jihandist (who was also killed) is a stark reminder that terrorism is still a very real threat. But as this New York Times article also points out, the anti-terrorism measures that Australia enacted were largely ineffective. The immediate reaction by Australian authorities has been what you would expect. How could this happen? How can we prevent it from happening again? The same questions that the U.S. asked in the aftermath of 9-11.
And reality should temper our expectations about what is reasonable in terms of security. All too often we are willing to adopt a "better safe than sorry" attitude and accept security measures (many of which invade our privacy or, worse, erode our constitutional freedoms) even if they may not be very effective. Here's an example. Suppose instead of metal detectors or body scanners at airports there were terrorist scanners that were 90 percent accurate (which means they are also wrong 10 percent of the time) identifying someone as being a terrorist. Let’s also suppose that 3,000 people need to pass through a scanner and that one of them is a would-be terrorist. Thirty people pass through the scanner without any problem and the next person fails. How sure are you that the person is actually a terrorist?
A. 90 percent
B. 10 percent
C. 0.3 percent
The answer is actually C, or less than one-half of 1 percent. How is that so? If 3,000 people are tested, and the test is 90 percent accurate, it is also 10 percent wrong. So it will probably identify 301 terrorists – 300 by mistake and 1 correctly. The problem is that you won’t know from the result which is the real terrorist. So the chance that passenger 31 walking through the detector is actually a terrorist is 1 in 301 or 0.3 percent.
The problem with “better safe than sorry” is that it focuses almost exclusively on the consequences of “what if” and largely ignores the costs and consequences of “what is.” Yet the latter may be more real than the former.