Droning on

The Washington Post reports today that the CIA wants to expand drone attacks in Yemen:

Securing permission to use these “signature strikes” would allow the agency to hit targets based solely on intelligence indicating patterns of suspicious behavior, such as imagery showing militants gathering at known al-Qaeda compounds or unloading explosives.

The United States has long used this less discriminating approach in Pakistan, where I am told we killed a lot of tall guys in long white robes before finding Osama bin Ladin holed up in his Abbottabad villa.

This is not an easy policy choice, but the right course is to err on the side of caution.  The Post article emphasizes the risk of drone strikes putting the U.S. on the government side in Yemen’s wars with several groups of insurgents.  I don’t see that as the main issue. After all, we recognize and support the government in Sanaa, even if we don’t intend to get involved in Yemen’s internecine battles.  None of the insurgents are going to think we are not on the government’s side.

The Post also emphasizes that the drone strikes have killed a lot of the “right” people, more than are killed in strikes based on specific intelligence about their whereabouts.  That, too, is not pertinent to the decision-making.  We’d kill a lot of the “right” people by mowing down whole villages too, but it wouldn’t be morally correct or wise.

The issue is the impact of the strikes.  Do they work, or do they not?  Do they reduce risks to the United States or American forces?  A recent quantitative analysis of the drone strikes in the Federally Administered Tribal Areas (FATA) of Pakistan suggests that they have reduced militant violence, BUT:

Finally, it is important to reiterate that any reduction in terrorist activity associated with the drone campaign appears modest in scope. Although a decline in violence in FATA in 2010 coincided with the peak of the drone campaign, FATA militants remain active and violence remains high. To the extent drone strikes ”work,” their effectiveness is more likely to lie in disrupting militant operations at the tactical level than as a silver bullet that will reverse the course of the war and singlehandedly defeat al Qaeda.

Others find more ambiguous results.

The bigger question is the impact on the population in areas where drones strike, and on the broader political context . This is where things get dicey, in particular if you hit the wrong people (no matter how many “right” ones you kill).  Joshua Foust warned about these consequences earlier in the year.  While some Pakistanis and Yemenis may celebrate the deaths of particular militants, there will always be collateral damage, the more so if the “rules of engagement” are loosened.  It is difficult to imagine that most Pakistanis and Yemenis will welcome the deaths of innocent countrymen in U.S. drone attacks intended to protect Americans.  How long will their governments put up with us?  We’ve seen in President Karzai the negative consequences of too many mistaken strikes (not only by drones) and night raids.

General Petraeus, whom I know and respect, needs to repeat the question he asked himself about detention facilities when he took over in Iraq and later in Afghanistan:  are we creating more terrorists than we are taking out of circulation?  No RAND study will likely answer this question.  We’ll have to rely on good judgment, which is in short supply as Washington gears up for its quadrennial blood-letting between Democrats and Republicans.  There isn’t much mileage for an American politician in not doing the max to get the terrorists in Yemen, but restraint might in the end save more American lives.