When Drones Kill, Is It Just?

This article is more than 8 years old.

Drones, or unmanned aerial vehicles, now make up nearly a third of U.S. warplanes. The inventory increased fortyfold in the last eight years, and the Air Force now trains more drone pilots than conventional ones. Drones would appear to be the perfect weapon.

Indeed, if Saint Thomas Aquinas had designed a weapon that would meet the jus in bello requirements — the acceptable conduct for a “just war” — he might have come up with drones. What makes drones close to ideal is how discriminating they are. It is increasingly easy to kill the person you mean to target, with very little “collateral damage.”

And yet, there is something about them that makes some of us uneasy — something horrifying about the idea of killing an enemy half a world away, just by pushing a button, with no physical danger to the operator. It is hard to know whether this unease is moral intuition, or fear about what will happen when the bad guys acquire the same technology and target Americans they consider enemy combatants.

As long as we’re ahead in this technological game, the weapon looks remarkably compliant with the justice-in-war requirements. But it is the nature of weapon systems to proliferate. And ethical questions remain.

A just war must meet two broad requirements: Both the ends (jus ad bellum) and the means (jus in bello) must be just. Drones used to target al Qaeda terrorists would seem to meet both criteria.

But at whom, precisely, do you aim? In the war against al Qaeda, Congress (via the “Authorization to Use Military Force,” passed Sept. 18, 2001) authorized the president to “use all necessary and appropriate force against those nations, organizations, or persons he determines planned, authorized, committed, or aided the terrorist attacks that occurred on September 11, 2001.” The AUMF specified no geographical area in its broad authorization of force, meaning there is a legal basis for attacking al Qaeda, its affiliates and supporters wherever they may be found.

Most drone strikes have taken place in Pakistan. While the Pakistani government publicly denounces the strikes, operators have allegedly relied on Pakistani intelligence, and the drones have been launched from Pakistani soil, often against enemies of the Pakistani state.

But al Qaeda has far-flung branches, franchises and allies. One can easily imagine the use of drones expanding into many new regions. What about the Malian group Ansar Dine, responsible for destroying historic religious sites in Timbuktu and accused of killing at least 82 Malian soldiers? What about the Nigerian group Boko Haram, which has bombed churches and claims responsibility for 620 deaths in 2012 alone? And what if the individuals are American citizens, as was the case for Anwar al-Awlaki and Samir Khan, both killed in a strike in Yemen on Sept. 30, 2011?

Critics worry that the relatively low cost of using these weapons, especially in terms of American lives, may make conflict both more widespread and more likely. Perhaps this is one of the reasons why majorities in 17 out of 20 countries surveyed by Pew disapproved of the strikes. America is an outlier, Pew found. Sixty-two percent of Americans approve of the use of drones.

Another concern: Will the enemy use broad public opposition to drones as a way to mobilize new recruits? This is a question we must always ask ourselves when evaluating counterterrorism policies. Do the benefits of killing known terrorists exceed the costs of increased mobilization? Recent interviews in Yemen, conducted by the Washington Post, suggest that opposition to the strikes is growing. “The drone strikes have not helped either the United States or Yemen,” said Sultan al-Barakani, a top adviser to former president Ali Abdullah Saleh. “Yemen is paying a heavy price, losing its sons. But the Americans are not paying the same price.”

And then there is the question of how good we are at hitting the intended target. Perhaps more than with any other weapon system, the accuracy of a drone is only as good as the intelligence it relies on. In Yemen, political opponents have reportedly been nominating their enemies as legitimate targets, a particularly troubling development.

Further, how good is the technology at pinpointing the target and avoiding those sitting next to him at the dinner table? Early on, while the U.S. government minimized its reporting of civilian casualties, outside observers reported relatively high amounts of “collateral damage.” But the numbers have gone steadily down. According to Scott Shane of The New York Times, even the highest estimate for the percentage of civilian deaths caused by drone strikes in Pakistan is 20 percent, while the Pakistani Army's recent conventional efforts to root out terrorists in the border regions between Afghanistan and Pakistan caused an estimated 46 percent civilian casualties.

Still, given the inherent limits of the technology, is the target important enough that we will go after him despite knowing that his family or neighbors may be at risk? When that happens, we’re told the president himself decides, but there is still something unseemly about a president admitting that he is responsible for a “kill list.”

Although Congress gave the president a very broad mandate to target al Qaeda and its allies anywhere in the world in the immediate aftermath of 9/11, it is now time for Congress to tighten the authorization. It’s also time to begin a national debate to ensure we are using these weapons in ways that are consistent with our national interests, as well as with the criteria for just war. Drones may be an ideal weapon, but one is left with a puzzling discomfort, wishing there were a saint around to query as to why.

This program aired on July 23, 2012. The audio for this program is not available.