You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

Three Deep Flaws in Two New Human-Rights Reports on U.S. Drone Strikes


Two leading human rights NGOs, Amnesty International and Human Rights Watch, released separate reports on U.S. drone warfare this week. They were focused on different places geographically—Amnesty’s report, “Will I be next? U.S. Drone Strikes in Pakistan,” focuses on Waziristan and the tribal regions of Pakistan that border Afghanistan, while Human Rights Watch’s offering, “Between a Drone and Al Qaeda: The Civilian Cost of U.S. Targeted Killings in Yemen,” looks at six drone strikes (one in 2009, three in 2012, and two in 2013) in Yemen. Following closely on the heels of the Interim Report of UN Special Rapporteur Ben Emmerson on U.S. drone strikes and targeted killing, presented a week earlier at the U.N., these reports represent a determined challenge to the Obama administration’s drone warfare policies, on the grounds of factual claims about the actual civilian harms of the strikes in both Pakistan and Yemen, claims that the alleged “blowback” and other second-order effects of the strikes outweigh any putative benefits, and claims that the framework of the strikes, as well as specific strikes themselves, violate international law.

On the factual claims of civilian casualties, the reports suffer from the limitations of their groups’ ability to gain independent access to the areas at issue, though both groups have clearly done some significant on-the-ground reporting. That said, it is equally difficult for outside commentators—such as ourselves—to say anything authoritative either, either in endorsement of their findings or in critique of them. The facts the groups allege are horrific, including drone strikes gone horribly awry killing numerous innocents, and they warrant further investigation.

They also warrant caution. Neither of these reports even purports to examine a representative sample of drone strikes. Rather, the two groups went looking for drones strikes that had killed civilians—and they found some. Given that President Obama has said publicly that “it is a hard fact that U.S. strikes have resulted in civilian casualties,” this is hardly surprising.

Amnesty’s report is candid on this point. In the introduction, the authors note that the document “is not a comprehensive survey of U.S. drone strikes in Pakistan” but “a qualitative assessment based on detailed field research into nine of the 45 reported strikes” in North Waziristan between January 2012 and August 2013. The report “highlights incidents in which men, women and children appear to have been unlawfully killed or injured.” At one level, this focus makes sense. Human rights groups report on human rights violations, so one can’t fault the groups for shining a light on those cases that seem most problematic. But as a reader, one needs to be careful not to generalize from findings like these. Even if we take all of the facts both groups allege at face value, neither report really changes our understanding of either the likelihood of civilian casualties, the number of them, or the utility of drones as a tool in overseas counterterrorism. Examining the costs is a worthy endeavor, but the much more important question is the scope and magnitude of those costs—and on that point neither report adds much to the existing conversation.

What’s more, military analyst David Axe has raised questions about whether the Amnesty report is really describing drone strikes by the United States in all of its accounts:

It’s not at all clear that pilotless warplanes were truly responsible for all the attacks Amnesty studied. The Pentagon and CIA declined to discuss the drone campaign with the rights group.

Some of the eyewitness accounts in the new report are inconsistent with known drone tactics and the well-understood limitations of unmanned aircraft in general. The attackers could have been manned warplanes, and Pakistani rather than American.

Even assuming the facts are as bad as the groups contend, both reports—in slightly different ways—seem to overstep analytically what the facts they report will actually support. Sometimes, the overreach is subtle and factual. Sometimes, it’s legal—and pretty blatant. Three of these overreaches warrant specific mention.

First, Human Rights Watch builds much of its analysis around the proposition that the drone strikes it examined “did not adhere to policies for targeted killings that President Barack Obama disclosed in a speech in May 2013.” Yet as the group acknowledges, all of the strikes examined in the report predate Obama’s speech at the National Defense University, some by several years. Human Rights Watch mentions this fact a few times but it does not seem to have assimilated it. This speech announced changes in policy with respect to drone strikes. Yet throughout the report, Human Rights Watch holds the administration accountable for not complying with policies it had not yet adopted, and it sometimes seems to treat violations of those policies as somehow indicative of violations of violations of international law. 

Second, the reports—particularly the Amnesty report—have a way of conflating legitimate targeting which may produce civilian collateral damage with horrible errors that simply should not happen. The most glaring example of this is Amnesty’s treatment of the June 4, 2012 strike that killed Abu Yahya Al-Libi, a senior Al Qaeda leader. According to the Amnesty report, an initial drone strike killed five people and injured four others (the report does not say whether any were civilians). A group of 12 people, including both local residents and foreigners “whom villagers said were Arabs and Central Asians who were likely to be members of al-Qa’ida” showed up “to assist victims.” Al-Libi was “overseeing the rescue efforts” and was killed in the second strike, along with between 9 and 15 other people, including six local tribesman who “as far as Amnesty International could determine, had come only to assist victims.” In other words, six tribesman were killed working alongside a group of Al Qaeda operatives under a senior Al Qaeda official were killed.

Amnesty considers this strike a potential “war crime” both because it constituted an attack on civilian rescuers and, quite amazingly, because Al-Libi may not have been directly participating in hostilities at the time of the strike. Amnesty’s position, in short, is that it may be a war crime to target a senior Al Qaeda leader when he’s doing something other than plotting attacks—if, that is, it’s lawful to target him at all. There are many serious issues these reports raise; this kind of overreach undermines them all.

Third, both Amnesty and Human Rights Watch seem to be alleging war crimes while not quite admitting that’s what they are doing. Human Rights Watch does this subtly—by laying out standards of law the United States does not accept, reporting facts that seem to violate those standards, and then demanding “prompt, thorough, and impartial investigations into all cases where targeted strikes may have resulted in unlawful killings” with “criminal prosecutions as appropriate.” Amnesty is more blatant about it. The group coyly declares at the outset that it is “unable to reach firm conclusions” about its case studies and their “status under international law,” going so far only as to say it is “seriously concerned” that the strikes “may constitute extrajudicial executions or war crimes.” But draw firm conclusions it then does. For example, on page 23, it declares flatly that its evidence “indicates that Mamana Bibi was unlawfully killed”—leaving open only the question of whether the illegality was a war crime or an extrajudicially execution.

It is also worth being cautious about the groups’ blithe claims that drones strikes are turning the populations of these countries en masse against the United States and its counterterrorism efforts. At least with regards to Pakistan—the subject of Amnesty’s report—it bears noting that there are voices pushing back on the question of how the local population, of Pakistan in general and the tribal regions where the strikes take place in particular, view drone strikes. It is not entirely consistent with the views of Amnesty’s report; something, in fact, of the opposite. The Economist magazine reported in its October 19, 2013 issue that a “surprising number of Pakistanis are in favor of drone strikes.” The article notes that while opinion polls find Pakistanis widely opposed to U.S. drone strokes, there is pushback both within the country and within the tribal regions themselves:

[W]hen Sofia Khan, a school administrator from Islamabad, travelled with hundreds of anti-drone campaigners to a ramshackle town bordering the restive Federally Administered Tribal Areas (FATA) last October she was stunned by what some tribesmen there had to say. One man from South Waziristan heatedly told her that he and his family approved of the remote-controlled aircraft and wanted more of them patrolling the skies above his home. Access to the tribal regions is very difficult for foreign journalists; but several specialists and researchers on the region, who did not want to be identified, say there is at least a sizeable minority in FATA who share that view.

Surveys are also notoriously difficult to carry out in FATA. A 2009 poll in three of the tribal agencies found 52% of respondents believed drone strikes were accurate and 60% said they weakened militant groups. Other surveys have found much lower percentages in favour. But interviews by The Economist with twenty residents of the tribal areas confirmed that many see individual drone strikes as preferable to the artillery barrages of the Pakistani military. They also insisted that the drones do not kill many civilians—a view starkly at odds with mainstream Pakistani opinion. “No one dares tell the real picture,” says an elder from North Waziristan. “Drone attacks are killing the militants who are killing innocent people.”

The reason for this, it appears, is something that Amnesty’s report elides in its background section about the violence in the border regions. Amnesty’s report is addressed to U.S. drone strikes, and the violence of U.S. drone strikes. And without question, civilians get killed on occasion in drone strikes—though the Bureau of Investigative Journalism, a British NGO, suggests a “fall in civilian casualties” in Pakistan with “most news sources claiming no civilians killed this year despite 22 known strikes.” Amnesty’s report describes many violent actors in the tribal regions, including the Taliban, foreign Al Qaeda fighters, and the Pakistani military doing battle with its own Taliban insurgent groups. It describes all this violence, yet makes its focus U.S. drone strikes, despite the fact that the U.S. drone strikes are the most precise and probably the least harmful to civilians of all of the various forms of violence and fighting in the region.

One would be hard-pressed reading Amnesty’s report to understand that the biggest and least discriminate forms of violent force in the tribal regions comes from the Pakistani military, which has long engaged in bombardment and shelling against villages as it engages with its Taliban insurgents. Compared to that violence, at least some villagers seem to be saying, drones are less feared and far less harmful. Some observers in urban Pakistan have said so; in 2010, the Economist notes, “a group of politicians and NGOs published a ‘Peshawar Declaration’ in support of drones. Life soon became difficult for the signatories. … Many commentators admit to approving of drones in the absence of government moves to clear terrorist sanctuaries. But they dare not say so in print.” 

The villagers who support drone strikes are not crazy. An end to American drone strikes would not mean an end to the violence in the tribal regions, nor would it mean an end to government attacks against Taliban militants. Quite the contrary. It would probably mean an intensification, if anything, as the Pakistani government was forced to fall back on its usual tools: artillery and air strikes. There is precedent for this. A few years ago, it sought to reach an accommodation with the Taliban, and saw the insurgents take the Swat Valley and threaten a move toward the capital. The Pakistani army pushed back with the tools it had available: massive artillery barrages that leveled villages, left thousands dead, and left nearly a million people homeless. If we're going to talk about "blowback" from drones, then we had better talk about "blowback" from "no drones" too.

The Amnesty report acknowledges the existence of the army’s violence, and the Taliban insurgency. But by myopically focusing on American drone strikes, it makes no comparisons between these. It fails to acknowledge the judgment that at least some of the villagers in the Economist story appear to have made: that American drones are what they claim to be, and they are attractive compared to the realistic alternative to them. And for some people, at least, the drones are killing their enemy: Taliban insurgents and militants who have moved to take over their villages.