PG NewsPG delivery
Pittsburgh Post-Gazette Home Page
PG News: Nation and World, Region and State, Neighborhoods, Business, Sports, Health and Science, Magazine, Forum
Sports: Headlines, Steelers, Pirates, Penguins, Collegiate, Scholastic
Lifestyle: Columnists, Food, Homes, Restaurants, Gardening, Travel, SEEN, Consumer, Pets
Arts and Entertainment: Movies, TV, Music, Books, Crossword, Lottery
Photo Journal: Post-Gazette photos
AP Wire: News and sports from the Associated Press
Business: Business: Business and Technology News, Personal Business, Consumer, Interact, Stock Quotes, PG Benchmarks, PG on Wheels
Classifieds: Jobs, Real Estate, Automotive, Celebrations and other Post-Gazette Classifieds
Web Extras: Marketplace, Bridal, Headlines by Email, Postcards
Weather: AccuWeather Forecast, Conditions, National Weather, Almanac
Health & Science: Health, Science and Environment
Search: Search by keyword or date
PG Store: Pittsburgh Post-Gazette merchandise
PG Delivery: Home Delivery, Back Copies, Mail Subscriptions


Headlines by E-mail

Headlines Region & State Neighborhoods Business
Sports Health & Science Magazine Forum

Forum: About those 'normal accidents'

Another failure of a missile-defense flight test shows the folly of building trust in the system, says Gordon R. Mitchell

Sunday, July 16, 2000

The failure of yet another National Missile Defense (NMD) flight test last weekend raised fresh doubts about the technical integrity of the Clinton-Gore administrationšs proposed missile shield.

Perhaps most troubling for the beleaguered NMD program was not the fact that the missile-busting technology misfired again, but rather how it failed.

  Gordon R. Mitchell is an assistant professor and director of debate at the University of Pittsburgh. He is the author of "Strategic Deception: Rhetoric, Science and Politics in Missile Defense Advocacy" (Michigan State University Press). His e-mail address is 

Like most rockets, NMD interceptors are made up of several removable parts or "stages." The "kill vehicle" (the part designed to destroy enemy missiles) is propelled into space by two booster stages. As the interceptor nears its target, the boosters are supposed to separate and fall away, leaving the kill vehicle on its own to zero in on the enemy missile.

In the most recent test, the second-stage booster failed to separate from the kill vehicle late into the flight, preventing the interceptor from hitting its target. That such a basic malfunction would stymie the multibillion-dollar NMD system caught the Pentagon brass off guard.

At a press conference held shortly after the test, Lt. Gen. Ronald Kadish (the U.S. missile defense chief) was asked, "Of all the things that could have gone wrong with this flight, was this at the very bottom of your concern list?" Kadish's reply was revealing: "It wasn't even on my list. . . . this was not something we thought would happen."

Similar garden-variety mishaps have plagued other NMD tests. A January 1997 trial run never even got off the ground, when a data-link malfunction prevented the interceptor from leaving the launch pad. In October 1999, engineers loaded a faulty star map into the memory of the kill vehicle's on-board computer, causing it to drift off course during an attempted intercept. In January 2000, an ice cube formed in the coolant liquid circulating around the kill vehicle's infrared seekers, preventing the interceptor from "seeing" its target.

Undersecretary of Defense Jacques Gansler rationalizes these accidents by pointing out that NMD is an "extremely complex system." Other missile defense insiders explain that NMD has a "fragile architecture" prone to malfunction.

Yale sociologist Charles Perrow has analyzed similar "complexly interactive, tightly coupled" systems in the nuclear power, petrochemical and aerospace industries. His conclusions may help explain why a variety of unexpected breakdowns continue to plague the NMD program.

Some highly complex industrial systems have many sophisticated components that all depend on each other's flawless performance. According to Perrow, this interlocking complexity makes it impossible to foresee all the different ways such systems could fail. "The odd term 'normal accident' is meant to signal that, given the system characteristics, multiple and unexpected interactions of failures are inevitable," he explains.

One way to counteract the tendency of a complex system such as NMD to suffer "normal accidents" is to simplify the conditions in which the system is asked to perform. Pentagon missile defense officials did just that by dumbing-down NMD test conditions after the program's first flight test, reducing the number of decoys used in subsequent experiments from nine to three, then from three to one.

This backtracking in the NMD testing program brings to mind the image of an overmatched high-jumper who keeps lowering the bar to salvage some modest success. In the NMD testing regimen, "[s]o far the bar has been set very, very low," comments Lisbeth Gronlund of the Union of Concerned Scientists, "[m]uch, much lower than it should be, which leads me, at least, to conclude that this is really not about assessing the technology."

While this engineering approach might eventually improve NMD's flight test track record, it is unrealistic to expect such experimental success to carry over into reliable combat performance. History shows that missile defense systems developed under tightly scripted testing regimens do not do well when called upon to perform in challenging wartime environments. The Patriot system was a perfect 17-for-17 in tests on the White Sands Missile Range before it botched nearly every attempted intercept of Iraqi Scud missiles in the 1991 Persian Gulf War.

Kadish testified to Congress in February that "we must ensure that the NMD system we will eventually deploy will work with a very high level of confidence." However, it is unlikely that the current NMD engineering effort will ever yield a system that satisfies Kadish's criterion. Indeed, it would be extremely difficult to reach a "very high level" of confidence based on the oversimplified tests now being administered, even if the NMD system did not continue to fail those bare-bones tests in unpredictable ways.

Missile defense proponents dismiss such concerns by insisting that any defense, no matter how leaky, would be better than defenselessness. This argument has strong rhetorical appeal until one considers just how nuclear strategy would play out in a world filled with wildcard missile defense systems.

Secretary of Defense William Cohen is fond of touting NMD as a tool to counter so-called "rogue state blackmail." This blackmail scenario envisages major powers embroiled in a diplomatic or military dispute with a what used to be called a rogue state (now the State Department uses the term "states of concern"). The model is a North Korea, Iran or Iraq - armed with weapons of mass destruction - attempting to exact concessions from wealthy states by threatening a long-range rocket attack.

In such a tragic situation, it is suggested that missile defense would preserve "freedom of movement" to call the bluff. To do this, however, leaders of shielded nations would need to have supreme confidence in their missile defense systems. Such a level of confidence hardly seems warranted in light of the complex NMD system's vulnerability to normal accidents.

An unusually brave politician might be candid about such vulnerability after deploying a missile defense system worth billions of dollars. The danger is that a more pliant leader would allow decision-making in a crisis to be guided by the technological hubris of missile defense contractors with a penchant for false advertising.

Emboldened by misplaced faith in a leaky missile shield, such an overconfident, underinformed leader might be willing to wager American cities for diplomatic advantage in risky nuclear gambles.

One shudders at the prospect that discrepancy between actual and advertised effectiveness of a U.S. NMD system would be brought to light by a real ICBM launch on an American city. Yet this is precisely the scenario that would seem possible if citizens embrace Secretary Cohen's suggestion that missile defense can be a potent and coercive tool to counter rogue-state blackmail. In such situations, rather than pursuing diplomatic alternatives in a stalemated conflict, shielded nations would dig in and dare a "state of concern" to follow through on its promise to launch an ICBM, hoping that NMD's ace-in-the-hole leverage would force the adversary to back down.

But what if the "rogue state" wouldn't fold? Bluffing poker players who are forced to show their poor cards after a called bet often lose the hand. Presidents bluffing about the capability of their missile defense systems could lose entire cities.

bottom navigation bar Terms of Use  Privacy Policy