Rosa Brooks, the new military blogger for Foreign Policy, shares a fascinating anecdote:
Here’s a small but not atypical example. A few readers may remember the spring 2010 crisis in Kyrgyzstan. Several hundred people were killed by police and ethnically aligned mobs, many more were wounded, and thousands of refugees (mostly from the Uzbek minority population) fled their homes.
Within the White House, these events triggered fears of a possible ethnic cleansing campaign to come, or even genocide. One day, I got a call from a member of the White House’s National Security Staff (NSS). With little preamble, he told me that Centcom needed to “move a surveillance drone over Kyrgyzstan, ASAP, so we can figure out what’s going on there.”
This wasn’t such a crazy idea. Drones and other intelligence, surveillance, and reconnaissance assets have the potential to be powerful tools in human rights monitoring. The ability to watch troops or mobs or refugees move in real time, to see weapons being stockpiled or mass graves being filled, might help us take timely and appropriate action to stop a genocide before it gets off the ground.
There was one enormous problem with my NSS colleague’s request, though: Neither of us had the authority to order Centcom to immediately shift a potentially vital asset from wherever it was currently being used to the skies over Kyrgyzstan.
She goes on to describe how the NSS worried that Osh would turn Kyrgyzstan into another Rwanda and how they didn’t want to “spin up a whole bureaucratic process” to get a drone flying above the city. When Brooks balked at sending a drone over a country where there was no agreement to fly U.S. drones, the NSS staffer tried to pull rank (“I’m calling you from the White House“) and then morals (“The president wants to prevent genocide in Kyrgyzstan”), to no avail.
Brooks means this story as a lesson in civil-military relations, and how the two groups rarely understand how or why the other works. I think there’s an extra lesson there as well. Christian Bleuer saw a rise in anti-Uzbek violence in Kyrgyzstan in the weeks leading up to the June Events, and worried openly about the potential for those smallish brawls to turn into “a more serious ethnic political cleavage.”
Sadly, they did turn into a more serious cleavage. But more than the cleavage resulting in atrocities that the U.S. was unwilling to assist with (the U.S. didn’t even send humanitarian aid for the Uzbek refugees who fled the carnage) is the bias inherent to how the NSS was trying to understand the situation. In 2010 there were U.S. government employees either in or very close to Osh — both at the State Department and with the DOD and various intelligence agencies who run the Manas air base. The NSS didn’t want to contact them to get a handle on what was going on — they wanted a drone.
Drones have a tendency to do this when decision-makers are under a severe time crunch for making a decision: they replace readily available sources and can supplant a strategy. Floating a drone over Osh and Jalal-Abad would not have made it any easier to “prevent a genocide” in southwestern Kyrgyzstan, as it wouldn’t have made interventionist troops any closer or more able to get there given the turmoil also wracking Bishkek at the time.
The desire to generate information is understandable, even laudable, but even there drones would have had limited utility. The fires that raged across Osh (a picture above of one burned out building I saw there, 18 months afterward, is a testament to how awful it was) created clouds of smoke a Predator drone’s main camera couldn’t see through, while the heat they generated would make the infrared camera difficult as well. That doesn’t mean a drone would be blind, but in this case — sending a single drone, re-tasked from monitoring the war it was monitoring — a drone really wouldn’t have told them very much.
One of the remarkable things about the violence in Osh is that the open sources everyone can read — English and Russian press, social media, emails, text messages, and so on — were actually very good at telling the story of the violence as it unfolded in real time. These things created an archive of the massacre as it unfolded, one with far richer data, stories, and media (images, video, audio) than a drone could ever provide.
Yet, the policymaker defaulted to a drone — to the easy-seeming technological solution for gathering information over the messier but far more accurate and comprehensive social and open solution. In the end, the fact that the U.S. never really did anything about the riots, whether monitoring or helping Uzbekistan with the influx of a hundred thousand refugees, doesn’t matter. It was tragic inaction but didn’t affect the outcome. However, the reliance of drones instead of sober consideration — of using a drone as the strategy instead of in service to a strategy, as I’ve noted happens elsewhere — that is what’s so fascinating, and worrying, about Brooks’ account.
Note: Unmentioned in Brooks’ story is seeking permission from the government of Kyrgyzstan to actually fly said drone over Osh. It may not have been possible or feasible given the turmoil occupying Bishkek’s new government at the time. Still, the omission of even discussing local government permission to fly a military asset over their airspace is fascinating: one of the key points undermining the sovereignty argument that Pakistani intellectuals employ to oppose drone strikes there is that the government of Pakistan has given the U.S. permission to fly drones over its airspace. Without that permission, doing so is a violation of sovereignty and if missiles are fired from the drone, it is an act of war. With permission, drones are neither offensive nor anti-sovereignty. Neither the NSS staffer in this story, nor Rosa Brooks herself, seemed to consider Kyrgyzstan’s sovereignty in their argument about drones.