What The Intercept Found in “The Drone Papers”—And What I Found In Them

The Intercept yesterday released its latest scoop: a cache of leaked documents on the U.S. drone program, presented as a series of blockbuster stories.

The “Drone Papers,” as The Intercept calls them, consist of two Pentagon slide decks on “support to small footprint counterterrorism operations” in Somalia and Yemen, another slide deck on “Operation Haymaker” (a targeted killing program in Afghanistan), and a “geolocation watchlist.”

On the basis of these documents, The Intercept finds:

• Targeted killings “depend on unreliable intelligence and hurt intelligence gathering.” The program is heavily dependent on SIGINT, which often comes from potentially unreliable “host-nation partners.” This “is a primary factor in the killing of civilians.”

• “The military labels unknown people it kills as ‘enemies killed in action’” (or EKIA). When drone strikes “miss their target and end up killing someone else, they label that person EKIA.”

• “Strikes often kill many more than the intended target.” The Intercept takes this information from slides showing that, in Afghanistan from May to September 2012, JSOC achieved 19 “jackpots”—airstrikes that hit the intended individual—with 136 other individuals labeled as EKIA in those strikes. “This means that almost 9 out of 10 people killed in these strikes were not the intended targets.”

• The leaked slides, which show the chain of responsibility for approving a target and taking a strike, “do… appear to be consistent with Obama’s comment in 2012 that ‘ultimately I’m responsible for the process.’”

• As of 2013, “JSOC had a 60-day window to hit a target” after receiving approval. This two-month window suggests “the administration’s broad interpretation of ‘a continuing, imminent threat.”

As The Intercept has made the documents public, the rest of us have a chance to comb through them and draw our own conclusions—as Adam Klein did yesterday. Let’s just say that I would characterize these facts very differently. Moreover, as Shane Harris discussed on Rational Security, there’s a lot in those documents that was of greater interest to me and him than it was to the Intercept authors.

Here are the salient points I would have made had The Intercept asked me to write their Drone Papers story:

• The drone program is not a panopticon. We already knew this, but it’s worth repeating, if only because of how pervasive the notion of the drone as “all-seeing eye” has become in the public imagination. The documents, however, describe a program plagued by insufficient signals intelligence and “blinking,” a phenomenon in which a drone’s movement causes a gap in the video feed. Think of it as a cyclops with bug caught in his eye. Interestingly, some of the intelligence limitations are caused by the program itself in what seems to be a vicious cycle: it’s impossible to interrogate someone whom you’ve targeted with a drone, and that lack of interrogation then leads to a scarcity of intelligence when identifying and tracking the next target.

• At least in Somalia and Yemen, the geographical areas on which the two “small footprint” slide decks focus, this lack of intelligence has led to numerous cases in which drone operators failed to take a shot because of an inability to acquire or maintain “positive identification” of the targeted individual. This situation might be different in Afghanistan and Iraq, as the slide deck indicates both that these programs operate under a less restrictive targeting process, and that the long flight time of drone operations in Somalia and Yemen creates additional intelligence difficulties.

• As The Intercept notes, the slides’ depiction of procedural mechanisms behind the program appears to confirm President Obama’s insistence on his ultimate personal responsibility for the drone program. Insofar as the target selection process goes, the slides suggest that all decisions must ultimately go through the president, though the formal chain of command does not put him in control of approving individual strikes in the moment. Additionally, the slides indicate that the average approval time for a target was 79 days, or 58 days once an unidentified outlier was removed from the sample. Adam Klein suggests that we might take this as an indication of a “reassuring” procedure ensuring that “these operations go forward only after a deliberate, individualized process.” I’m less inclined than Klein to characterize this information as “reassuring,” but the slow-moving process does seem to indicate that these decisions are, at the very least, not hastily made.

• The procedure of designating unknown male casualties of drone strikes as EKIA confirms Jo Becker and Scott Shane’s 2012 New York Times report on the government’s decision to designate “all military-age males in a strike zone as combatants” unless proven otherwise, a controversial choice that many point to as the sleight-of-hand behind low civilian casualty counts. (Scott Shane noted this on Twitter this morning.)

• The ratio of individuals designated EKIA to “jackpots” warrants a closer look than The Intercept gives it. The Intercept reads this information as indicating that “9 out of 10” individuals killed in strikes are people other than “the intended target,” which implies that the strikes hit entirely unknown individuals 90 percent of the time. This characterization seems misleading. If these are personality strikes, as they seem to be, a large proportion of the individuals labeled as EKIA were killed because of their proximity to the “jackpot”—which will tend to imply affiliation. In the theoretical case of a strike hitting the intended target’s convoy, which kills not only the target himself but some number of other guys as well, the fact that the other people’s specific identities were not known makes them only very marginally less likely to be fellow fighters. In this case, those designated EKIA would likely be totally legitimate targets under the laws of war. This is not to say that, in some cases, individuals have been killed because of proximity to targets despite being, in fact, civilians or to suggest that those individuals would never be falsely labeled as EKIA under the “permissive” policy reported on by Becker and Shane. But to lump these two scenarios together is to obscure an important distinction. Being a non-combatant and an unidentified combatant are very different things, after all.

If this doesn’t sound much like what you read in The Intercept, either tonally or substantively, there’s a reason for that: There actually isn’t all that much new information in these decks. So everyone will tend to see in them what they brought the conversation. They are not apt to change anyone’s mind. So while you can read the documents yourself and draw your own conclusions, I suspect you will find in them what you already believe.