A More Granular Look at Death by Drone


Is the CIA’s targeted killing campaign the “most precise and effective application of firepower in the history of armed conflict”? Retired Gen. Michael Hayden, former director of both the CIA and the NSA, argued exactly that in The New York Times earlier this year. While Hayden might be right, he elides a number of significant issues that demand consideration. For instance, the targeted killing program may be considerably less precise than claimed. Further, Hayden does not discuss the implications of unmanned aerial vehicles proliferating across the globe. In the end, we side with Hayden on the efficacy of the targeted killing program. Yet we also argue for a more nuanced program that acknowledges the inherent ambiguities of remote targeting and provides more transparency about the costs involved, even if these costs do not outweigh the benefits of the program.

Hayden places a great deal of emphasis on the precision of the drone warfare program, opening with a long fictional vignette that illustrates the decision calculus made by drone operators, the intelligence apparatus that feeds them, and the ultimate authorizing authorities. Yet even in this handcrafted story, an interesting ambiguity emerges. In the vignette, a drone operator drops a small laser-guided GBU-12 bomb on al-Qaeda planners (actually launched from a manned aircraft, but laser-guided by the unarmed drone). Further guidance is then given to the drone operator: “After the GBU hits, if military-age males come out … kill them.” Of course, in this vignette, we don’t know who these military-age males are. Given the presence of al-Qaeda operatives, it’s certainly possible or even likely that they are terrorists as well. However, they could just as easily be innocent domestic servants, non-affiliated family members kept onsite as human shields, merchants, or a host of other possibilities. That Hayden chooses to use — in his opening defense — an example that is open to multiple interpretations is illustrative.

Maybe — just maybe —targeted killing isn’t quite as precise as he makes out. While this may be the case, it’s worth considering if the issue is about the platform (i.e., unmanned vs. manned aircraft) that is hosting the lethal weapons, the decision-making calculus (i.e., intelligence analyst to commander to drone pilot vs intelligence analyst to commander to pilot), or how we define what is acceptable collateral damage? It stands to reason that if the same desired effect is achieved through the same decision-making processes — including checks and balances — then how the lethal blow is delivered should be irrelevant. Whether the lethal projectile is hosted by a manned aircraft or an unmanned aircraft should not be the point of contention. The issue is really about collateral damage, and if the threshold for “acceptable” collateral damage changes as the rules of engagement change.

Hayden also doesn’t engage the “radicalization” argument, articulated prominently by retired Gen. Stanley McCrystal, former Defense Intelligence Agency director Gen. Michael Flynn, and former White House counterterrorism “czar” Richard Clarke. All of these leaders have argued that collateral damage from drone strikes so angers populations on the ground that it leads to further radicalization and provides a recruiting base for violent groups, rendering the drone program a net loss for U.S. interests.

While this position bestows a certain psychological satisfaction, it does suffer from a dearth of evidence. Its arguments tend to be visceral and not empirical. Studies on the effects of civilian casualties are rare and proffer ambiguous findings. We know of no studies that considered drone strikes or even airpower more broadly in isolation from other causes of civilian casualties. This doesn’t mean that the position is wrong. Indeed, the argument draws in some ways upon the writings of early airpower theorists, who maintained that aerial bombing imposed a unique psychological effect upon the receiving population. But in the absence of hard evidence (and one hesitates to think how that research design might be constructed), unknown possible recruits cannot be weighed against known terrorist casualties. Nonetheless, this argument is made by serious people and must be at least acknowledged at some level.

There’s one more lacuna in Hayden’s argument. Hayden simply doesn’t address the global proliferation of unmanned aerial systems technologies including the electronic interception capabilities and “big data” analysis that makes precision targeting possible. The Chinese are manufacturing and marketing near clones of the Predator drone that offer a significant percentage of its capability. This then begs the question, when the Chinese use their lethal drones on Uighur insurgents, or Kurdish separatists become targets of Turkish armed drones, will the U.S. judgments on collateral damage change?

Since Hayden’s initial interview, there’s been a volley of responses and counter-responses to his defense of the targeted killing program. The most recent lobs have come about transparency — by the government and by media. While transparency of mission is important, a core issue remains obfuscated — defining collateral damage and what is an acceptable threshold in today’s “war on terror.”

Again, at the end of the day, we end up very near Hayden. The targeted killing program remains a useful means of fighting worldwide Islamic terrorism. It would be irresponsible to simply shut the program down in the face of these concerns. Sweeping generalizations about the correlation between collateral damage from drone strikes contributing to the growing radicalization of at-risk, vulnerable populations is inferential at best. Insurgent and civilian populations do not react the same to drone strikes in Afghanistan as they do in Iraq; it would be foolish to paint a picture that suggests that blowback following civilian casualties is the same in all circumstances. Semantics and optics do matter. Caution is needed when discussing how precise weapons are, just as when discussing what the collateral effects of these weapons are alleged to be.

We applaud the administration’s decision to release the drone program “playbook,” though the manner in which the data is released will determine its utility. Will it be broken down by time, and will covert and overt programs be conflated? It’s distinctly possible that all the data could be released, but with insufficient parsing to make it of any real utility in assessing and refining policy. We need to discuss acceptable collateral damage thresholds. Increasingly common “gray zone” contests will only further blur the lines between military and civilian personnel and activities.

Drone warfare is here to stay. Even were the United States to unilaterally (and unwisely) renounce this policy, other nation-states are sure to adopt a program that has so clearly demonstrated its efficacy. However, maintaining that the program is cost-free can lead to a skewed decision calculus.


Melissa Hersh is Principal of Hersh Consulting. She regularly advises governments, IGOs and Fortune 500 companies on the full risk spectrum across a variety of aerospace and defense, homeland security, energy, mining and global health issues.

Douglas Ollivant (@DouglasOllivant), a retired Army infantry officer, is an ASU Senior Fellow in the Future of War project at New America, a managing partner of Mantid International, and a national security contributor to Al Jazeera America.


Photo credit: Staff Sgt. Vernon Young Jr., U.S. Air Force