A U.S. Army soldier from the 1-320th Alpha Battery, 2nd Brigade of the 101st Airborne Division points his rifle toward COB Nolan during a patrol in the volatile Arghandab Valley in Kandahar province, Afghanistan, on July 20, 2010
Having taken over command of the U.S. mission in Afghanistan, General David Petraeus is thinking of easing restrictions imposed by his predecessor on when and how American troops can attack suspected enemy fighters. But a new, independent study from the nonprofit National Bureau of Economic Research (NBER) finds that the restraints put in place by the recently fired General Stanley McChrystal actually resulted in a decrease in insurgent attacks.
Soldiers in Afghanistan have grumbled for months that curbs on their firepower ordered by McChrystal last summer have emboldened the Taliban and put U.S. troops at greater risk. Those rules of engagement bar allied air and mortar strikes on houses unless allied troops are in immediate danger. "His rules of engagement put soldiers' lives in even greater danger," a soldier privately told Rolling Stone in the article that got McChrystal fired last month. "Every real soldier will tell you the same thing." Petraeus, in his Senate confirmation hearing after being tapped to succeed McChrystal, told the Armed Services Committee he is "keenly aware of concerns by some of our troopers on the ground about the application of our rules of engagement" and that he plans to "look very hard at this issue."
While U.S. officials expect no wholesale changes in the rules of engagement, they suggest Petraeus may undo some of the red tape surrounding their enforcement, streamlining them and making them easier to understand and obey. "As you and our Afghan partners on the ground get into tough situations, we must employ all assets to ensure your safety," he said as he assumed command earlier this month.
The 69-page NBER study analyzed 4,000 civilian casualties and 25,000 clashes between U.S.-led forces and insurgents over the 15 months ending April 1. The pattern it detected was that civilian casualties caused by U.S. and NATO actions "are associated with a substantively and statistically large increase in attacks" by the Taliban and other Afghan militants. The average civilian-casualty event caused by the International Security Assistance Force (ISAF) killed or wounded two Afghans. According to the analysis, districts in which such an episode occurred averaged six more violent clashes over the following six weeks than did otherwise similar areas that were spared civilian casualties. "The data are consistent with the claim that civilian casualties are affecting future violence through increased recruitment into insurgent groups after a civilian-casualty incident," the Air Force–funded report says. "Local exposure to violence from ISAF appears to be the primary driver of this effect."
There were nearly 11 civilian casualties per day in Afghanistan over the 15 months measured in the study; 10% of them were women or children. While the insurgents caused more than 80% of the civilian casualties — primarily with roadside bombs — each side killed roughly the same number of women and children. "This means," the report notes, "that as a proportion of all civilian casualties, ISAF kills or injures many more women and/or children."
The study also found that U.S.-led attacks in Iraq did not seem to generate follow-on attacks by Iraqi insurgents. "In Afghanistan, we find strong evidence of a revenge effect," it says. "In Iraq, we find no such effect." That, the report's four authors theorize, is because Afghanistan's militants are rural and centrally commanded, while Iraq's are largely urban and operate under a more autonomous command structure. "The dynamics of the two wars are fundamentally different, and so one should be cautious about relying too much on the Iraq experience to guide policy in Afghanistan," says Princeton University's Jacob Shapiro, one of the report's authors. Joining him in the research for the NBER report, which was based in Cambridge, Mass., were Luke Condra and Joseph Felter of Stanford University and Radha Iyengar of the London School of Economics.
One observation in the study was of particular interest to the Air Force after last year's complaints that ill-aimed air strikes were killing too many Afghan civilians. According to the NBER, allied air strikes accounted for just 6% of the civilian casualties among Afghan women and children. That's well below the 16% killed or injured in traffic accidents between ISAF and Afghan vehicles.