War Stories

Smart Bombs, Dumb Targets

Did overconfidence in precision targeting cause civilian deaths in Iraq?

The capture of Saddam Hussein dramatically illustrates an ancient, but often forgotten, principle of warfare: It’s easy to kill people but very hard to kill a person.

Saddam’s nabbing took thousands of Army and special-ops forces, interrogating hundreds of loyalists, tracking dozens of leads, and, in the end, one soldier spotting an out-of-place thread of fabric on what turned out to be a secret cover, then lifting the lid to find the Butcher of Baghdad himself hiding in the bottom of a hole.

It was the work of armed troops on the ground, close-up, ultimately at arm’s length. It could not have been accomplished by pilots dropping smart bombs from two miles high in the sky.

Last spring, while the “shock and awe” phase of the war was still in progress, U.S. targeteers tried to get Saddam and his cronies the easy way—or what seemed to be the easy way—with B-1 bombers, F-117 attack planes, and satellite-guided air-to-surface munitions.

A new report by Human Rights Watch reveals, for the first time, just how persistently they tried to kill individual Iraqi leaders in this manner—and how consistently they failed.

Over the course of the war, U.S. air forces mounted 50 so-called “decapitation strikes.” The bombs accidentally killed several dozen civilians who happened to be near the explosions, but they killed none of the Iraqi leaders they were intended to strike.

The problem was not with the bombs. The bombs were as accurate as advertised; they hit precisely where they were aimed. The problem was with the pre-strike intelligence; the Iraqi leaders—the bombs’ targets—turned out to be someplace else.

The best-publicized decapitation strike took place on March 20, the first night of the war, when the CIA got word that Saddam Hussein was in a bunker under a particular house in Baghdad. Two F-117s, quickly dispatched to the area, dropped four 2,000-pound EGBU-27 earth-penetrating bombs on the site. The results: One Iraqi civilian was killed, 14 were wounded; Saddam was nowhere to be found. A postwar inspection revealed there never was a bunker.

In another high-profile strike, on April 7, a B-1B dropped four 2,000-pound JDAM bombs on a restaurant where Saddam and his two sons were believed to be meeting. Shortly afterward, Vice President Dick Cheney said, “I think we did get Saddam Hussein. He was seen being dug out of the rubble and wasn’t able to breathe.” Someone might have been dug out the rubble; 18 people were killed, but they did not include the Husseins.

A similar strike was intended to kill Lt. Gen. Ali Hassan al-Majid, aka “Chemical Ali.” He escaped, if he was ever there; 17 civilians were killed.

All these leaders—and many others—were killed or captured later. But none of them were taken by aerial bombs.

Why the intelligence was so wrong—why the targets weren’t there when the bombs struck—makes for a flabbergasting tale. According to the report (which was assembled by Marc Garlasco, a former analyst with the Defense Intelligence Agency whose specialty was the Iraqi military), U.S. spy agencies determined the location of these Iraqi leaders by intercepting their telephone conversations. Like many people in the Middle East, the leaders were using Thuraya satellite phones. Thuraya sat phones have an internal global positioning satellite chip. Intercept the phone call, you also find out the GPS coordinates of the phone.

However, it turns out there were three problems with this method. First, there is a lag between the time when the call is intercepted and the time it takes for a bombardier to fly to the area and drop his bombs on the target. The airstrike on the Baghdad restaurant took place 45 minutes after the intelligence intercept. This was heralded as amazingly fast work, but it was also plenty of time for father and sons to leave the scene and travel miles away—if they had been there to begin with.

Second, it has widely been known for a long time that the U.S. National Security Agency tracked al-Qaida terrorists through their Thuraya sat phones. It is quite conceivable—Garlasco thinks it very likely—that the Iraqi leaders were using deception tactics, making calls from a location, then quickly leaving, so that the American planes would waste bombs, kill civilians (and thus make enemies of the victims’ relatives), and look incompetent.

Third and most astonishing, the GPS signal beamed by Thuraya sat phones is accurate only to within a radius of 100 meters. The bombs that were dropped on these targets are accurate to within 10 meters. In other words, even if the caller had still been on the scene when the bombers arrived (unlikely enough), the bombs would very likely have missed because the target could have been as far as a football fieldaway from where the phone’s signal indicated.

The accuracy of the bombs encouraged a false confidence in the officers who planned the strikes. Had the bombs been considered so inaccurate that they would land, on average, 100 meters from a target (a pretty dumb bomb), no U.S. commander would have dropped them on a residential area in the hopes of killing an Iraqi leader who was thought to be somewhere in the neighborhood. Yet relying on such inaccurate intelligence had the same effect.

This information about the Thuraya’s GPS chips is neither new nor classified. The manufacturer’s manual—cited by the Human Rights Watch report—clearly states its technical features, including the error radius of its location-tracker.

An intriguing question, which someone in Congress should pose to a relevant witness: Did the officers who selected the targets know that the intelligence was so imprecise? Certainly the NSA knew the specifications of the Thuraya’s GPS chip. But were these data—and their implications—passed on to the military commanders?

There is a still broader inference to be drawn from the Human Rights Watch report, concerning the whole policy of “preventive warfare” and the risks it carries. Over the past year or so, many have cautioned that such a policy requires extraordinarily reliable intelligence. If we are going to attack nations not to repel aggression nor even to pre-empt an imminent threat, but rather to prevent a threat from arising, then we should have, at the very least, extremely high confidence in the evidence. And the case of the decapitation strikes—with their 0-for-50 success record—indicates that smart bombs don’t ensure smart attacks. That being so, they still less ensure a smart war.