How accurate are US drones?

Drone strikes in Pakistan are never publicly acknowleged by the US government [Al Jazeera]

The announcement that a botched US drone strike killed 23 Afghan civilians in Uruzgan province is another blow to Nato’s promise to protect Afghan civilians – one of several fatal mistakes in Afghanistan this year.

But there was at least a measure of accountability involved: The US military investigated the incident and (publicly) disciplined the soldiers responsible.

No such accountability exists on the other side of the Afghanistan-Pakistan border, where the drone strike program is run by the CIA. The strikes – successful or not – are never publicly acknowledged by the US government. Mistakes are never admitted, death tolls never confirmed.

The Pakistan drone strikes almost certainly kill civilians, but exactly how many is the subject of much debate. Pakistani analysts claim the strikes overwhelmingly miss their targets: A study published in April 2009 claimed that 687 civilians had been killed, along with just 14 al-Qaeda members, a 50-to-1 ratio. A similar report, published in January 2010 in the Pakistani newspaper Dawn, cited more than 700 civilian casualties.

Those numbers are generally dismissed in the US, where the most oft-cited number comes from the New America Foundation, which estimates that between 290 and 387 civilians have been killed by 134 reported drone strikes – approximately 30 per cent of the total reported fatalities.

A forthcoming study, conducted by University of Massachusetts professor Brian Glyn Williams, reaches a different conclusion. Williams finds just 44 confirmed civilian deaths – 3.5 per cent of the total – with another 240 unknown victims. (The rest of the victims are non-civilian fighters, Williams concludes.)

Interactive map of drone strikes in Pakistan (courtesy New America Foundation)

Williams doesn’t disagree on the total number of people killed – just on the alleged affiliation of those victims.

Both surveys are based on English-language media reports, which the authors readily acknowledge are flawed: Journalists have little access to the tribal regions where the drone strikes are carried out, forcing them to rely on sometimes unreliable local stringers, or on reports from the very groups targeted by the bombings.

Perception versus reality

The back-and-forth over civilian casualties masks a bigger reality about the drone programme, though: It may not matter whether or not Williams’ more optimistic figures are accurate, because they will do little to change Pakistani public opinion.

Polling data is difficult to come by, but the drone strikes are widely believed to be unpopular in Pakistan. There are a few contrary data points, like a study from a Pakistani think tank – the Aryana Institute for Regional Research and Advocacy – which found mixed support for the drones: 52 per cent of respondents said they believe the strikes are accurate.

US studies estimate between 3 and 30 percent of people killed by drones are civilians [AFP]

But another survey from the tribal areas found 80 per cent of respondents do not support the drone strikes, according to Mosharraf Zaidi, a Pakistani journalist. In an op-ed earlier this month for The News, a Pakistani newspaper, Zaidi described villagers in Pakistan’s tribal agencies who were “traumatised” by the constant threat of drone strikes (on the one hand) and Taliban attacks (on the other).

And an Al Jazeera-Gallup poll conducted in 2009 found just 9 per cent of Pakistanis favor the attacks.

That unpopularity stems, in large part, from a belief that drones kill sizable numbers of civilians. Andrew Exum, a fellow at the Washington-based Center for a New American Security – and a critic of the drone strike programme – argued on Saturday that such a perception is ultimately more important than research studies about civilian casualties.

“I care, in other words, less about reality as defined by verifiable facts and figures and more about reality as it is interpreted in Pakistan and within Pakistani diaspora communities,” he wrote.

As Zaidi has pointed out, not a single study shows drone strikes to be perfectly accurate; thus the Pakistani public retains the (legitimate) belief that the strikes kill civilians.

What is more, because the drone programme is officially secret, there is little that policymakers can do to shift Pakistani public opinion. Even if the reports of civilian casualties are inflated, the US government cannot offer evidence to correct those reports – because doing so would reveal the programme’s existence.

Peter Bergen and Katherine Tiedemann, the analysts behind the New America drones study, argued for a more transparent programmein an op-ed last month in the New York Times.

“Should the American government’s claims about the small number of civilian deaths be verified, some of the Pakistani hostility toward the United States might dissipate. This would be much easier if the now-classified videotapes of drone strikes were made available to independent researchers,” they wrote.

The Obama administration has so far faced little pressure to make the drone programme more transparent. That may change in the coming days: Philip Alston, the United Nations special rapporteur on extrajudicial executions, issued a report on Thursday questioning the legality of CIA-directed drone strikes. Alston’s report concluded that the military, not the CIA, should run the programme, as a way of ensuring some measure of accountability.

Al Jazeera is not responsible for the content of external websites.