An Army That Learns From Its Mistakes

Acknowledging and studying civilian casualties would strengthen us in Iraq.

By Eric Umansky
Eric Umansky writes for Slate.

Los Angeles Times

May 17, 2005

Trying to flee fighting in western Iraq last Tuesday, one family in a taxi tried to speed through a Marine checkpoint. There are few details available about what happened next. The Los Angeles Times says only that the car was fired upon, the driver killed and a mother and daughter wounded. It quotes one Marine saying, "We were just sick to death when that lady got out of the car with her baby."

It's tempting to dismiss what happened that night as just something that happens in war. But it shouldn't be shrugged off that way. To see why, one has to look at the results of the Army's investigation into the March 4 attack on the car carrying former Italian hostage Giuliana Sgrena — one of the few checkpoint incidents investigated.

The report enraged Italians for its conclusion that the GIs involved did nothing wrong. But while clearing the soldiers, the investigation ends up implicating the military's policies themselves. The report is a case study in how the United States has failed to adopt what could be a relatively painless, and moral, counterinsurgency tactic.

The unit involved in the Sgrena shooting, part of the Army's 3rd Infantry Division, was manning a temporary "blocking position," meant to turn back all traffic. There were no written rules for that kind of operation. Instead, the unit learned on the job from the soldiers it was replacing, which turns out to have been a problem because the departing unit's tactics were outdated and dangerous. Rather than using signs, the units waved down drivers with a spotlight and laser pointer.

The night of the shooting, about two dozen other cars were stopped, but not without problems. "Many of the vehicles screeched their tires when stopping," says the Army report, adding, "While effective for accomplishing the mission, the spotlight and laser pointer may not be the best system from a civilian point of view."

There were other established checkpoint procedures, but they too were problematic. For instance, if drivers don't see or ignore the laser pointer and lights, soldiers were advised to fire warning shots. Of course, some people then assume they're being shot at, which seems to be what happened with the Italian driver of the Sgrena vehicle, who told soldiers he "panicked and started speeding."

The final element of the procedures calls for the GI holding the spotlight to drop it, shoulder his rifle and fire at the engine block. Again, the Army's report says that's exactly what happened. And again, as the report later acknowledges, it's a flawed tactic. It's not easy to hit an engine block in a moving car, especially when you have to drop a spotlight, pick up your gun and fire within seconds.

At the end of the report is a series of achingly common-sensical recommendations, including: Use warning signs with international symbols; reconsider requiring a soldier holding a spotlight to also be a gunner; and start a public awareness campaign.

Checkpoint killings, unfortunately, are nothing new, nor are the recommendations to ameliorate them. In late 2003, Human Rights Watch documented 11 civilian deaths at checkpoints in Iraq. Among the report's suggestions: "Initiate a public service campaign to inform Iraqi civilians about proper behavior at checkpoints."

Individual units that inadvertently kill civilians are required to file incident reports. But nobody seems to add them up. "We don't compile those," said military spokesman 1st Lt. Ryan Fitzgerald. Perhaps the Pentagon doesn't add the numbers because it fears bad publicity. Maybe it's bureaucratic inertia. Whatever, it's shortsighted.

There is some kind of actual guerrilla infrastructure that can be battled, but its most dangerous asset isn't a physical weapon; it's the insurgents' mind-set. Working to minimize civilian casualties may or may not dent the guerrillas' motivation, but there's little cost in trying.

GIs don't want to kill civilians, nor obviously does the military as a whole. But what's missing is a militarywide mechanism for learning from the casualties that do occur. The military has something called, self-explanatorily, the Center for Army Lessons Learned. Imagine if keeping civilian casualties to a minimum were part of the lesson plan.

Humans make mistakes. The point is not to assign individual blame. The issue is whether institutions have the structures in place to learn from them.

Before she was killed last month by a suicide bomber in Iraq, activist Marla Ruzicka was pushing a similar idea. She wanted a government office created to catalog, and ultimately offer compensation for, civilian casualties. Ruzicka's vision was driven by morals: We should make amends to those we harm. It's fortunate that it also has the possibility of being a counterinsurgency tool. If only the military would recognize that and take the step that's a prerequisite for learning from such casualties: acknowledge them.