The Drama of the Drone Papers

Man walks past a graffiti, denouncing strikes by U.S. drones in Yemen, painted on a wall in Sanaa
Graffiti denouncing strikes by U.S. drones in Yemen.Photo: Khaled Abdullah/© Corbis. All Rights Reserved.

There are two newsmaking items from the Drone Papers, the latest batch of national security revelations published by the Intercept last week. The first is that Glenn Greenwald, Jeremy Scahill, et al, have a major source who is not Edward Snowden. The second, more consequential news is that for every intended or authorized target the United States kills in distant drone strikes, it also kills about six other people who are not intended or authorized targets.

We have always known that the drone program involves collateral damage (to use that old military term). The administration has been eager to publicize the extensive intelligence work and moral labor that goes into choosing assassination targets. But all of that deliberation concerns only a small minority of the actual victims. Most of those people the United States kills in the drone programs, it now seems, are not targets carefully selected for assassination, but people who are simply in the vicinity, about whom neither the intelligence analysts nor the drone operators really know anything at all.

These numbers come from a classified Pentagon review of a single program called Operation Haymaker, in which special-operations teams on the ground in northeastern Afghanistan’s Kunar Province worked with intelligence analysts and drone operators around the world to pinpoint targets. According to the Pentagon review, 39 targets were killed. So were 219 other people. The civilian casualties tabulated in these classified documents, according to the Intercept’s calculations, are roughly double what had been acknowledged publicly.

In other operations, at other places or times, these ratios are bound to be a bit different. But Operation Haymaker was a large and well-supported campaign, and from what information we have about the drone program generally, it doesn’t seem to be an outlier. If you are trying to decide what to believe about drone assassinations, these numbers are probably the best place to start.

One question — maybe the most pressing question — is how the public feels about that brutal ratio of one targeted death to five or six unintended. The evidence so far is that the public is more or less okay with it. Though these numbers are more concrete than we’ve had before, and though the Intercept has done admirable work in further detailing how the decision to assassinate is made and put into operation, the basic story — that the aerial assassinations are not really so surgical — is pretty familiar. “Every independent investigation of the strikes has found far more civilian casualties than administration officials admit,” Scott Shane of the Times put it, summing up a decade’s worth of findings this April.

And yet even though there have been some iconic cases — the strike that hit a wedding party in rural Yemen in December 2013, for instance, killing a dozen people but not its intended target — there has not been much of a political movement against the program. It is hard to imagine the question of Afghan civilian casualties becoming an issue in this presidential election. So far our ongoing military involvements overseas have had virtually no profile there at all.

Scahill’s team handles one category of revelations in the Drone Papers with particular elegance: the chaos and uncertainty of the intelligence that undergirds the identification of a target. The system relies very heavily on “signal intelligence” — the streams of metadata that are attached to cell phones — and often, according to the Intercept’s reporting, that intelligence is just wrong. “It’s stunning the number of instances when selectors are misattributed to certain people,” says the Source, as the person involved in the program, who leaked the Drone Papers, is referred to throughout. “And it isn’t until several months or even years later that you all of a sudden realize that the entire time you thought you were going after this hot target, you wind up realizing it was his mother’s phone the whole time.”

The preference of the Source and for the Pentagon review group seems to be for more human intelligence to supplement and confirm the digital signals. But an approach that leaned more heavily on human intelligence would inevitably require more humans, and if the public has any clearly expressed preferences in this long half-war against terror, it is that we should continue to engage it while involving as few Americans as possible. The preference for signal intelligence over human intelligence may lead to all sorts of lethal bungling. But there is a political context. The decision isn’t arbitrary.

Because they traffic in official secrets, stories like this are inevitably about two things at once. They are about the program they seek to expose, and in that sense the Drone Papers are remarkable, even heroic documents. But they are also about the bureaucracy of secrecy that surrounds them. The absurdity and paranoia of that bureaucracy (and the single-mindedness and triumphalism of those who oppose it) has a way of making a political melodrama out of each leak, as if the truth held in secret must differ vastly from what the public understands.

That isn’t the case with the Drone Papers, but they matter in a lower amplitude way. One problem of governmental secrecy is the atmosphere of vagueness it creates, of plausible moral deniability. What is happening in secret turns out to be very much like what we expected all along. But now we know just how many civilians are being killed. The question is whether that clarity will make any political difference at all.