Means-testing the Drone War

by Joshua Foust on 3/10/2010 · 12 comments

The London Times runs a story claiming that the U.S. drone war in Northwest Pakistan is creating fear and paranoia:

The effects of the campaign, however, are beginning to veer dramatically off course as the strikes intensify, according to tribesmen. “Before the drone attacks began the Taleban weren’t so obvious among us and the militancy wasn’t as strong,” Amir said. “But now every home in North Waziristan seems to have one or two Taleban living in it. The youth are joining them. Feelings against the US and Government are rising because of the attacks. Al-Qaeda has been badly affected by drones — but it has benefited too.”

“One cannot deny the effect of the drones in taking out senior leadership, the militancy’s centre of gravity,” a Pakistani army officer admitted. “It has had a huge impact. But at the same time it has become a huge motivation to fight against the Government and the army because of the perception that it is a breach of sovereignty and is killing civilians. All combined, it creates a very negative impact.”

Well, let’s just color me skeptical that any grassroots opposition to the drone strikes is over sovereignty—even people out in the bush know those things don’t happen without Islamabad’s approval. Rather, the innocent victims of the strikes—even NAF’s well-regarded drone data indicates an appalling amount of collateral damage—and their seeming capriciousness, since no one really knows which agency has what standards for firing a missile, would seem a bigger factor.

But that story gets at a very important point, one I see painfully rarely in the DOD: unintended consequences. In the Times, they mention three major consequences of the drone strikes: post-traumatic stress, revenge “spy” killings by militants, and a diffusion of militants from camps into populated areas (one man says there is one or two Taliban in every house in his village).

Unfortunately, we cannot meaningfully change the stress and paranoia that comes from the strikes—if not drones, then bombs and acid-washed girls. But the other two we should have been able to anticipate, or at least mitigate. It is a standard tactic for many violent resistance movements to behave violently toward those they perceive as collaborators—especially when it’s a fairly simple causal chain between “informing an American” and missiles raining down from the sky. It is inevitable they’ll accuse the wrong person and murder capriciously as a result (and to be fair, the militants do not have a monopoly on collective punishment). Similarly, it would make sense, if we began targeting only large gatherings of militants, that they would stop gathering in large numbers and diffuse—this is almost doctrinaire insurgency.

Now, the discussion over drones and whether they’re an effective tactic or strategy in the GWoT rarely moves beyond theoretical constructs—we have polls of questionable methodology indicating the people in the FATA prefer drones to terrorists (just how does one get a statistically relevant sample there?), and we have Kilcullen and Exum making a mostly theoretical argument against their use (it not only clashes with the previous poll, but their casualty count seems to have been way off). But what of concrete testing?

This is, obviously, pretty difficult. In any complex system you cannot draw simple correlates. But we can speak to high-level leadership as one example. In 2004, Nek Muhammad Wazir has been leading one of Waziristan’s many cyclical insurgencies, when, right when he was sitting down at the negotiating table, a U.S. missile fizzed out of the sky and killed him. Wazir’s successor, Baitullah Mehsud, spent the next five years wreaking absolute havoc in the region—and whenever a cease-fire would be signed, it was rarely the militants who broke but, but America (or Pakistan, operating under pressure from America) trying to lob drone missiles at another figure they were able to identify.

The end result of this incessant drone war against militant leadership is that the leadership itself is far more radical and far less willing to negotiate an end to their insurgency than they were in 2004. While the drones could be called a stunning success in going after al Qaeda, they’ve also been used for years to go after the Pakistani Taliban—and in both cases the men who replaced the dead commanders were more vicious and less amenable to overtures from governments to discuss an end to the violence.

While a (very) brief look at the leadership of these organizations cannot really say much about their success or failure in aggregate, it can highlight some of the second order consequences of a somewhat overly narrow focus on degrading leadership. Successful though it may be—and if NAF is to be believed, then a large majority of drone targets are actual bad guys—the drone war still carries with it serious consequences. Even within the insurgency in Northwest Pakistan, we cannot conclusively say that drones have had a major effect on operations, considering how much worse the area has gotten as strike frequency increased (we cannot draw anything more than a correlation on this front). Al Qaeda’s expeditionary reach may have been curtailed, but it seems to have been at the cost of vast swaths of Pakistan… and even Afghanistan. Have we been shooting ourselves in the foot?

Subscribe to receive updates from Registan

This post was written by...

– author of 1848 posts on

Joshua Foust is a Fellow at the American Security Project and the author of Afghanistan Journal: Selections from His research focuses primarily on Central and South Asia. Joshua is a correspondent for The Atlantic and a columnist for PBS Need to Know. Joshua appears regularly on the BBC World News, Aljazeera, and international public radio. Joshua's writing has appeared in the Columbia Journalism Review, Foreign Policy’s AfPak Channel, the New York Times, Reuters, and the Christian Science Monitor. Follow him on twitter: @joshuafoust

For information on reproducing this article, see our Terms of Use


AJK March 10, 2010 at 8:22 pm

Question here: how much of the FATA insurgency do you think is personality-based? That is, if the insurgency is at small-group level where the individuals know their leader, then how much do the leaders’ personalities drive their actions?

I ask this because if the US/ISAF/whoever has a very good read on some individual leaders and who may step up if the leaders are killed, then they could conceivably drive the insurgency in a direction by picking and choosing who to kill.

This relies on an intensive HUMINT element that I’m pretty sure doesn’t exist in reality, but it would be another way to fight the insurgency. If you can get the opposition to aggregate under a precise strategy and then sweep out that strategy from under them, you can make serious headway.

Of course, that last sentence really describes how the insurgency has gotten the US to rely on drone strikes that may not be all that effective, too.

anan March 11, 2010 at 12:43 pm

AJK, I wouldn’t assume that HUMINT doesn’t exist. HUMINT has lead to a lot of successful drone strikes in Afghanistan. Suspect much of the credit for this goes to the NDS.

Perhaps the more interesting question is why the NDS wants certain Taliban leaders killed more than other ones.

Andy March 10, 2010 at 11:06 pm


The drone “war” cannot be considered solely on it’s own, but must be compared to alternative courses of action. Choosing the best action (or no action at all) requires weighing alternatives and each alternative will carry risks, unintended consequences and downsides. Drone strikes are no different. How do drone strikes compare to alternatives? That’s is the question you need to answer.

dcrowe March 11, 2010 at 2:12 am

Both Joshua’s post and Andy’s comment get at something I noticed when reading Bergen and Tiedemann’s report. They spend a lot of time trashing the consequences of the drone strikes (including an often-overlooked drawback: the drones also destroy all the potentially intelligence-rich “litter” around the target) but then just wind up with a conclusion that the drone strikes are essential, w/o explaining the cost/benefit of any alternatives. The “but we don’t have other ideas” is actually a prime example of the phenomenon Andy comments about. If we don’t have any other ideas, we’re not comparing it to anything.

DePetris March 11, 2010 at 12:28 am

There is no question that the U.S. drone program over Pakistan’s tribal region is an extremely controversial issue. One of the very reasons why Pakistanis despite the United States so much (estimates are up to 80 percent hold unfavorable views of the U.S.) is due the types of drone attacks that occur on a regular basis. Women and children are sometimes caught in the crossfire, even if a few Al’Qaeda operatives or Taliban leaders are killed in the process.

But the reality remains that the United States doesn’t really have any alternatives. Legally, they are not allowed to enter Pakistan without Islamabad’s permission, which has been repeatedly denied. A U.S. military invasion of Waziristan (where the Haqqani network is located) is out of the question, both because the Pakistanis would not allow it and because Washington does not have the resources available to launch such an operation. I suppose the soldiers leaving Iraq could be used, but again, the Pakistani Government is in no such mood to simply grant America’s wishes.

And what about relying on the Pakistani Military to do the job themselves? This route has worked well over the past few weeks, evident in the capture of some top Afghan Taliban commanders. But if we have learned anything from Pakistan’s behavior, it is that Islamabad either will not or cannot sustain this. The Haqqani network is Pakistan’s number one strategic asset in Afghanistan once the Americans leave…a militant bulwark against Indian influence right next door. So naturally, Pakistan will think twice before launching an assault on a future ally.

So when taking all of this into consideration, using drones to find and kill terrorists is the only policy the Pentagon has at the moment. I highly doubt the drone program would be used to such an extent if other options (like ground operations) were on the table. It’s not a matter of being the best of bad choices. Rather, it’s a matter of being the only viable choice outside of a new U.S. military invasion.

dcrowe March 11, 2010 at 2:23 am

“Only viable choice” for what desired end? The report from NAF says drone strikes:

– Kill 1 civilian for every 3 suspected militant.
– Hasn’t prevented the Taliban from killing large numbers of civilians.
– Aren’t inflicting losses on the Taliban that they can’t absorb
– Haven’t prevented AQ from training new recruits
– “Don’t seem to have had any great effect on the Taliban’s ability to mount operations in Pakistan or Afghanistan or to deter potential Western recruits”
– “May have ultimately increased the appeal of the Taliban and al Qaeda, undermining the Pakistani state”

Considering these drawbacks, toward what desired policy are we actually advancing via the drone strikes?

dcrowe March 11, 2010 at 10:57 am

Ugh. I meant 1 civilian for every 2 suspected militants.

Dafydd March 11, 2010 at 8:13 am

“Well, let’s just color me skeptical that any grassroots opposition to the drone strikes is over sovereignty”


just because Islamabad has a policy of surreptitiously approving of dron strikes while publicly condemning them, and everyone (especially those in the bush) knows it, doesn’t mean these people don’t vehemently object to that policy.

After all, some of these people scarcely recognise the right of the Pakistani army to patrol around there. Why wouldn’t they see US drone strikes as impinging on their sovereignty?

Joshua Foust March 11, 2010 at 9:21 pm

Dafydd, I really only know how to reference this:

The majority of the respondents (13 of 15) did not fully see the drone attacks as a violation of the sovereignty of Pakistan. Their argument is very simple: the state of Pakistan has already surrendered FATA to the militants, therefore, Pakistan has no reason to object to the drone attacks. Pakistan will have this right only if can retake the areas from the militants.

Now, it’s a survey, and I questioned the other survey above that indicated drones might not be the up-front PR disaster they appear to be. But it is interesting to see that reaction, yes?

Dafydd March 12, 2010 at 10:41 am

Sure, it is interesting. And the ‘Pakistan has surrendered sovereignty’ argument is not totally without weight.

Nevertheless I believe many Pakistanis view drone attacks from any outside power on FATA as a violation of sovereignty that should not be tolerated by their government.

It is quite possible, and quite probable, objections are stronger in areas less affected by militancy.

Abdullah March 15, 2010 at 4:31 pm
Curious to know if you all think this info is accurate? What would be your comments on the sources as well as the numbers?

Bender March 15, 2010 at 8:24 pm

Read this study, not very representative in terms of numbers but seems to suggest just the opposite:
Pakistani sovereignty is not an issue because “Pakistan has already surrendered FATA to the militants”.
The precision of drone strikes seem to be appreciated but the side effect is that Taliban are killing civilians if they as much as suspect they are involved in intel ops.

Previous post:

Next post: