Illustration: Brian Stauffer

Civilized society is at far greater risk from natural disasters and industrial accidents than from terrorism, yet we behave as if it were the other way around. We spend ­billions on unproven technical remedies for imagined terrorist threats while skimping on known methods of mitigating the effects of hurricanes, floods, and toxic-waste spills, which occur with remarkable regularity and predictability. We should concentrate instead on defending against these more frequent and disastrous threats. By thus identifying our worst vulnerabilities and reducing them, we would reduce the size of the terrorists’ targets as well.

Our present efforts depend on bureaucratic organizations, but we should not expect too much from them. Two decades ago, in a study of industrial accidents, I sought to show that any organizational solution to a human problem that is complex enough to be interesting will necessarily be imperfect. Stated that way, my assertion might have seemed unobjectionable. Yet I noticed that people would always ask why no one had prevented a particular problem—typically the last one in a chain, hence the “proximate” cause of the disaster. For example, in the explosion of the space shuttle Challenger, in 1986, the blame was placed on those who failed to take into account the chilled O-ring that was the proximate cause of the explosion. Similarly, the company that neglected to trim tree branches in Ohio shouldered the blame for the shorted power lines that started a cascade of failures that plunged much of the eastern United States and Canada into darkness in the summer of 2003.

It was in part to address this mental illusion of operator error, as I regard it, that I coined the deliberately paradoxical phrase that became the title of my book: “normal accidents.” I showed how in complexly interactive systems whose elements are “tightly coupled,” great accidents are inevitable. In air travel, electric power distribution, petroleum refining, and nuclear reactors, thousands of things must go right to avoid disaster. One day a few inevitable failures will interact in an unexpected way, bringing down the system. We therefore must either give up on the many benefits we derive from complex, tightly coupled systems whose failures might be catastrophic or put up with occasional disasters. It is a question of coldly weighing the alternatives.

Terrorist acts differ from normal accidents in that they are caused deliberately. Terrorists need not wait for blind chance to turn up hidden weaknesses but can seek them out with energy and insight. Moreover, defensive measures taken against normal accidents will be less effective against terrorists who, as strategic adversaries, can evade them.

Any strategy must begin by taking into account both what terrorists can do, in principle, and what they are inclined to do, in fact. It is abundantly clear that they can do us far more harm than they have already done, without incurring costs greater than they have already proved willing to pay. Their failure to do so is astounding, and it merits investigation.

We can put down a portion of terrorists’ inaction to simple incompetence. When al-Qaeda dispatched a man to New York City to reconnoiter the Brooklyn Bridge, it had in mind a harebrained scheme to cut the cables strand by strand with an acetylene torch; the agent prepared for his mission by viewing photos of the bridge taken from the original King Kong movie, in 1933.

In the United States, terrorists’ inaction also reflects a lack of human resources. This is not the case in many European cities, where terrorist cells have been much more common.

Yet incompetence and lack of resources cannot explain the entire pattern, because sometimes the terrorists have shown considerable strategic sophistication. One example was the 9/11 plan of training pilots, hijacking airliners, and turning them into guided missiles; the terrorist network had even thought to assassinate the leader of the Northern Alliance beforehand, eliminating the chance that he would lend support to the gathering U.S. invasion of Afghanistan.

The terrorists have also shown political acumen. Al‑Qaeda appears to put more emphasis on showing ­prowess, making a statement, and recruiting activists than on merely raising the body count. Indeed, there is some evidence that Osama bin Laden and his associates did not intend to kill as many people as they did in the 9/11 attacks. They were also surprised by the result and criticized by some of their cohorts for the resulting invasion of Afghanistan, which deprived them of their main training facilities. A similar calculation of effects may lie behind the evident decision of bin Laden’s operational chief to cancel, in 2003, a chemical attack on the New York subway or the decision to bomb London’s public transit system rather than using the explosives to detonate toxic chemical storage tanks and railroad cars.