Understanding the Strategic and Tactical Considerations of Drone Strikes
Joshua Foust
January 2013

In Brief:
• To be effective for the long run there must be a broader strategic framework placing drones in service of more fundamental goals. • Drone strikes can only be understood by combining two different levels of analysis: the immediate, tactical effects, and the long-term strategic consequences. • These two levels of analysis contain a mixture of data; focusing on what the data say and don’t say is vital for informed debate. • From a tactical level, drones have been incredibly successful at killing high-level terrorist leaders everywhere they’re used. • From a strategic level, drones present long-term challenges that are not yet fully accounted for in U.S. policy decisions.

The United States first used an armed drone to conduct a targeted killing on November 17, 2001. It was a strike on a housing compound in Kabul, Afghanistan against Mohammed Atef, an Egyptian al Qaeda militant who planned the 1998 embassy bombings in Nairobi, Kenya, and Dar es-Salaam, Tanzania.1 The Atef strike marked a sea change in how the U.S. would conceive of the use of force: up to that point, officials had been reluctant to use manned aircraft to conduct strikes.2 A new era in warfare had begun.

Joshua Foust is a Fellow at the American Security Project 1100 New York Avenue, NW Suite 710W Washington, DC


Over the next year, drones carried out two more strikes – one in Eastern Afghanistan at al Qaeda’s Zhawar Kili encampment,3 and one against Ali Qaed Senyan al-Harthi, a Yemeni member of al Qaeda who planned the October 2000 attack on the USS Cole.4 The U.S. military carried out the strike against Mohammed Atef; the Central Intelligence Agency (CIA) carried out the Zhawar Kili and al-Harthi strikes. The first barely elicited a shrug at the time – it was one of many precision air strikes by the U.S. military against al Qaeda leaders during the start of Operation Enduring Freedom. The latter two raised some eyebrows: it wasn’t the military wasn’t conducting violent operations but the CIA, and in the case of al-Harthi it was killing outside of a warzone. In the years since, however, the U.S. government has used armed drones for precision strikes at an increasing rate. The CIA in particular adopted the platform as a reliable method for limited strikes against high-value targets in otherwise inaccessible places. Drone strikes expanded from Afghanistan into neighboring Pakistan, where Pakistani officials quietly accepted their use.5 As the pace of drone strikes increased – slowly at first under President Bush and then rapidly under President Obama – public debate about lethal strikes intensified. Yet, despite the growing public discussion about lethal drone strikes, there remains little data to support many claims made by critics and supporters alike. Without sound data about drone strikes, there is no basis for evaluation. The public needs facts to understand the policy. Collecting data on drone strikes, however, is not simple. The program is classified to various degrees within the U.S. government: drone strikes in Pakistan, which are carried out by the CIA, are so secret the government cannot officially acknowledge them.6 In Yemen and Somalia, the U.S. military, through the Joint Special Operations Command, or JSOC, can also carry out drone strikes which are classified differently. When U.S. officials discuss drone strikes publicly,7 they are careful not to mention Pakistan; rather, they mention drone strikes elsewhere (most often Yemen).

An Air Force Airman 1st Class coordinates air cover during an operation in the Sroghar Mountains of Afghanistan. U.S. Air Force

Different aspects of the program have been leaked to reporters.8 However, there is no way to confirm that these accounts9 accurately reflect the actual process by which drone strikes are decided and carried out. And because the U.S. government is not forthcoming about the drone strikes it carries out, researchers are left to gather as much information as possible from other sources. This severely limits any analysis of the program. Many areas where drone strikes occur, such as the Federally Administered Tribal Areas (FATA) of Pakistan, are off-limits to foreigners. Additionally, most of the publicly available data about drone strikes is assembled from government sources, either in the U.S. or in the host country, which introduces unavoidable bias in

reporting. As a result, there are almost no reliable data in the public about the targets and effects of drone strikes, making firm conclusions about them extremely difficult. Despite the fundamental weaknesses of all data about drone strikes, it is possible to indirectly measure some effects they have. This paper is intended to begin a new public discussion about how we can better understand the use of drone strikes and their effects. It looks at drones from two perspectives: the strategic level, where broad social and political activity takes place, and the tactical level, where individuals experience conflict and drones. In doing so, we aim to tighten the focus on data in the public debate – both what we can know with the right research and what we cannot reasonably know. There are some consequences of drone strikes that we can infer from the information we have on hand, and there are some that we cannot reasonably infer. In discussing this at two levels, we also demonstrate that drones are neither inherently “good” nor “bad” – it is the policy to use them that really matters.

Measuring Drone Effects: Direct versus Indirect Metrics
Because of the highly classified nature of drone programs, much of the data about them is difficult to collect directly. From a tactical perspective, the total number of drone strikes, casualties from individual strikes, and other data about the intended targets of these strikes cannot be confirmed by outside researchers. At the strategic level, the social or political effects of strikes are difficult to measure as well, since other factors external to drone strikes can also influence public opinion or the behavior of foreign leadership. Nevertheless, there are indirect metrics that can indicate what effects drone strikes have. A tactical-level indirect measure of drone strike effects is the flow of Internally Displaced People (IDP): do drone strikes create the mass movement of civilians fleeing conflict? One example of an indirect strategic metric is the political rhetoric about them over time: do drones drive anti-Americanism or political dysfunction in the host government? In an ideal world, this combination of direct and indirect metrics would create an expansive, comprehensive picture of how effective drone strikes are in achieving counterterrorism objectives. However, because many of those metrics are not or cannot be measured, we’re left with a muddled, vague understanding of what effects drone strikes really have. For the sake of simplicity, we include both direct and indirect measurements in our list of metrics below.



Tactical Metrics of Drone Strikes
1. Total number of strikes, by region
Put simply: the public does not know how many drone strikes the U.S. government has carried out in Pakistan, Yemen, and Somalia. No government involved with drones strikes – the U.S., Pakistan, Yemen, or Somalia – has provided official tallies of the number of drone strikes, or even the locations where the strikes occur. Without data on the number and location of strikes researchers must rely on indirect measurements of drone strikes, most often inferring strike data from officials speaking to journalists. Relying on media reports as an indirect measure poses serious challenges to conducting a rigorous study of drone strikes. The most basic problem with media reports is that they are unverifiable. In Yemen, civilians near air strike events tend to describe everything as a drone regardless of the type aircraft carrying out the strike –clouding any effort to determine which strike was the result of drones.10 Relying on civilians to identify when drones are used in an air strike is problematic – even in the U.S., drones are often misidentified and spark public concern.11 Because few claims can be verified, it is nearly impossible to reconcile different versions of strike events in media reports.
An MQ-1 Predator takes off from Creech Air Force Base, Nevada. U.S. Air Force

In Pakistan, too, strikes can be misreported. Many incidents are misreported by news agencies relying on different sources within the governments of Pakistan and the U.S. (if they are reported at all).12 One research group suggested that the U.S. government underreports the number of drone strikes (as in: it misreports drone strikes as coming from other types of aircraft). 13 Yet U.S. military officials have suggested that some reported drone strikes were actually the result of manned aircraft, such as the 2004 strike that killed Pashtun militant Nek Muhammad.14 It is unclear why the military would be privy so such information, since the drones program in Pakistan is run by the CIA and not the U.S. military. Regardless, outside researchers possess little capacity to verify or dispute incidents of misreporting. Pakistan may have the capacity to assist in the verification of some strikes. Military troops are on the ground near certain strike zones, and might be able to assess the damage and report back who was killed and injured. But few military officials ever comment on strikes, since they are coordinated by the Inter-Services Intelligence agency (ISI) and not the Pakistani military.15


Military officials of both the U.S. and Pakistan have a limited capacity to comment on drone strikes because intelligence services run the program. While researchers can make broad guesses about how many drone strikes have occurred in any individual combat zone, it is important to remember that these figures are approximate at best. Without official confirmation of drone strikes, determining the full extent of the drone campaign in any target area is difficult.

2. Casualties from drone strikes
Determining the casualties caused by drone strikes – how many people were killed or injured – can help determine the scope and collateral damage of the campaign and thus help gauge public reaction to future strikes. But because the regions where drone strikes occur are so dangerous as to be inaccessible to journalists and outside researchers, collecting this data on casualties is extremely difficult. Journalists often rely on government sources for their data on drone casualties. This reliance poses serious drawbacks. Depending on the perspective of the media outlet, reports on the same strike event can vary significantly. American media reports are likely to differ substantially from those of Pakistani counterparts solely due to different national points of view, to say nothing of any one outlet willfully manipulating facts for political and social gain. The vast majority of media accounts about drone strikes are sourced from the government, often in the intelligence services. There is no way to confirm or refute the accounts in these reports. In Pakistan researchers cannot directly access drone-affected areas because the government denies them access.16 Without being able to directly visit these sites, researchers are reduced to informed guesswork about the nature and characteristics of any individual strike. While residents of the FATA can leave the area and head for other parts of Pakistan to talk with researchers, there is no way to verify their stories, which leads to confirmation bias. 17 Without direct access to the FATA, researchers are not able to determine whether the stories told outside of it are representative or not.
Members of the Pakistan military and civilians from multinational relief agencies carry an injured woman to a U.S. Army UH-60 Blackhawk helicopter for a medical evacuation in the remote Pakistani village of Rawalakot in the aftermath of an earthquake in 2005. U.S. Navy

Researching drones and their effects in Yemen and Somalia is difficult for similar reasons. Neither the Yemeni nor Somali government officially forbids foreign researchers to visit drone-affected areas, but the poor security environment of these areas limits access. Researchers also currently lack the means to account for injuries resulting from drone strikes. The countries in question do not keep detailed records about births, deaths, or burials – key data that could allow an analysis of the records to determine how people died.18 The lack of data about drone casualties severely limits any effort to understand their effects (this will be discussed further in point 5).



3. Identity of intended drone strike targets
The identity of the targets fired upon by drones is arguably one of the most contentious aspects of the drones program. U.S. officials have suggested that they do not always know the identity of the people they fire upon – the so-called “signature strikes”19 or Terrorist Attack Disruption Strikes (TADS). U.S. officials have also openly discussed the creation of a secret “kill list” that contains the names of people they are pondering striking. Each Tuesday, President Obama sits down with his counterterrorism staff to examine the list, and there he decides who will be killed that week. 20 More recently, U.S. officials have discussed making the kill list a permanent feature of counterterrorism policy and adding other options besides killing to the menu of choices they face when deciding to take action against the names on it (the so-called “disposition matrix”).21 This list of whom the government is targeting is highly classified – revealing the names on the list could, according to officials, possibly reveal how they acquire information on terrorist groups, which would pose a steep risk to future intelligence-gathering. But the list’s classification presents a Catch-22: outside researchers are not privy to the list, so the identities and intended targets of drone strikes cannot be confirmed. In some high-profile cases, such as al Qaeda figureheads like Abdel Rehman al-Hussainan,22 AQAP leaders like Nadir Haider Nasser al Shaddadi,23 or senior militants like Baitullah Mehsud,24 officials will reveal who their intended target was. But naming the intended target of a drone does not always assuage skepticism of the drone strikes. A November strike against ‘Adnan al-Qadhi, a high profile figure in Yemeni politics long suspected of ties to al Qaeda, raised many questions among his community about why he was targeted.25 Yemen analyst Gregory D. Johnsen wrote that al-Qadhi was not actively plotting against the U.S. and wondered what criteria lead to his targeted killing.26 Simply naming the targets of drone strikes may not be enough to assuage public skepticism of the drones program.

UH-60 Black Hawk helicopter depart from Combat Outpost Bari Alai north-east of Jalalabad. U.S. Army

In many cases the identity of the targets is left unclear – “suspected militants” or “operatives” are common terms used to describe drone targets.27 These strikes are less remarked upon in the media discussion about drones, but are the focus of greater concern by several journalists and human rights advocates who argue that it is immoral for the government to kill people it cannot positively identify. Without more specificity about who the intended targets of drones are, it is impossible for outside researchers either to verify or refute the U.S. government’s claims to zero28 (or near zero)29 civilian casualties. Some researchers who rely on media reports have attempted to piece together a picture of how many civilians are killed compared to militants. The New America Foundation (NAF) compiles one set of data about civilian casualties,30 and the Bureau of Investigative Reporting (TBIJ) compiles different data.31 Both dispute the others’ analysis.32

Because both NAF and TBIJ use unreliable sources to build their databases (like media reports), there is no way to mediate their conflicting claims. There aren’t good sources of data to confirm their divergent assessments of alleged civilian casualties of drone strikes. No one in the public sphere is privy to the intelligence or decision-making process that went into any single decision to launch a drone strike. Without increased transparency from the U.S. government about its targeting process, its targeting lists, and the intended targets of its drone strikes, it is unlikely outside researchers will be able to independently verify or dispute the effects of individual drone strikes.

4. The behavior and movement of Internally Displaced Persons (IDPs)
In the three undeclared combat zones where drone strikes occur (Pakistan, Somalia, and Yemen) civil conflict has displaced large numbers of people.33 The behavior of locals in all three countries suggests that when locals fear for their safety, they flee the area. However, no international agency has identified any significant movement of displaced people from drones themselves. The lack of IDPs could mean that people are too besieged to leave; it could also mean that they do not view drones as a threat on par with local militants or military ground forces. In Pakistan, there are clear patterns to IDP flows. In 2007, after a tense standoff, Pakistani security forces stormed a radical mosque in Islamabad, killing 50 militants.34 It was a watershed for many Islamists in Pakistan. One, Maulana Fazlullah, used the siege as a reason to join forces with the Tehrik-e Taliban Pakistan (TTP), the Pakistani Taliban then led by Baitullah Mehsud.35 Fazlullah leads a banned Pakistani Islamist fundamentalist movement called the Tehreek-e-Nafaz-e-Shariat-eMohammadi (TNSM). By the time of its alliance with the Pakistani Taliban, the TNSM had raised fears among analysts as “one of the most dangerous religious militant groups in Pakistan.”36 Over the next several months, the TNSM occupied large areas of the Swat Valley, plunging a former tourist haven into violence. In response to this dramatic rise in violence, thousands of Swatis fled the area.37
U.S. Army soldiers and Pakistan army personnel form an honor guard carrying the flags of both countries during a transfer of authority ceremony in Muzaffarabad, Pakistan. U.S. Air Force

The Pakistani military launched an offensive to “clear” out the Swat Valley, and by the end of 2008 an estimated 250,000 Pakistani civilians had fled the fighting.38 As fighting between the Pakistani military and the Pakistani Taliban continued over the next year, more than a million people fled



the area.39 Despite the many challenges of data collection in northwest Pakistan, the international community was nevertheless able to measure and report on the civilians fleeing the Swat conflicts. Swat is not alone in Pakistan. Just this year, even the threat of a Pakistani military offensive into Waziristan, prompted thousands of civilians to flee the area in anticipation of the fighting.40 In contrast, there have been no reported mass movements of IDPs fleeing drone strikes in Pakistan. Civilians in Yemen follow a similar pattern of behavior. In 2011, when AQAP and its political wing, Ansar al-Shariah, occupied several towns in the south of the country, civilians fled the area en masse.41 When the Yemeni military forced them out a year later after heavy fighting, residents had little to return to; the area was devastated.42 In contrast, there are no known displacements of civilians from drone strikes in the country. Somalia also adheres to this pattern: the reported IDP flows have been in response to fighting with al-Shabaab militants as well as the dire humanitarian situation in much of the country.43 There are no reported movements of IDPs in reaction to drone strikes.

5. Forensic examinations of alleged drone strike victims
No drones report published to date has included an expert who can analyze the blast effects of the weapons drones fire. Similarly, no drones report published to date has included medical experts who could accurately diagnose the injuries attributed to drone strikes. Though advocacy groups and some reporters publish images and even video of disfigured civilians,44 there is no way to determine conclusively if a drone actually caused those injuries. Forensic examination of drone strike locations and purported victims is important to determine the effects these weapons have on communities, as well as to distinguish between drone strikes and other forms of air strike. Armed drones can fire a variety of missiles, all of which have relatively small explosives on them. The common MQ-1 drone can fire missiles as small as the 44-pound AGM-175 Griffin45 or as big as the 100-pound AGM-114 Hellfire anti-tank missile. The larger MQ-9 Reaper can carry the Hellfire missile as well as the 500-pound GBU-12 Paveway laser-guided bomb and the 500-pound GBU-38 JDAM. Each of these weapons types are different in size and type of warhead, and thus will create different blast patterns, explosive effects, and personal injuries when they’re used. The Hellfire, for example, can be fitted with either a metal slug that can kill a single person in place or with a small thermobaric warhead that can incinerate people while leaving a structure mostly unharmed.46 The Hellfire only has about 20 pounds of explosive, which means it wouldn’t leave an enormous crater upon impact. The Griffin has a warhead of only 13 pounds, which is why military and intelligence planners are interested


in using it – such a small amount of explosive promises to greatly curtail the collateral damage of drone strikes. In contrast, the Paveway and JDAM are much larger – each weighs 500 pounds and would create a considerably larger explosion than the Griffin or the Hellfire. These weapons are not often carried on military drones, and they are almost never carried on CIA drones.47 Both types of bombs are carried far more often on conventional aircraft. Thus, when assessing the damage allegedly caused by drone strikes, understanding the type of weapon used by drones can help differentiate legitimate claims to injury, and illegitimate claims. If purported victims are claiming injuries that are inconsistent with the weapons drones fire, then they should not be given credence – a determination only a weapons forensics expert can make. Groups like the United Nations Assistance Mission in Afghanistan (UNAMA) possess the capacity to forensically investigate violent incidents.48 They even have an office in Peshawar, Pakistan, near to the tribal areas where strikes occur. Yet UNAMA have not been permitted to send its experts into the tribal areas to investigate alleged strikes.

An MQ-9 Reaper on approach for landing. U.S. Air Force

In addition, there has been no comprehensive record-keeping of burials, funerals, or other measurable consequence of the victims of drone strikes in any country. Without this key data, it is impossible to verify most allegations about the nature, extent, and frequency of injuries and deaths resulting from drone strikes. The covert nature of the drone program in Pakistan creates a dilemma for the U.S. The Pakistani media (and the Pakistani government) are not similarly constrained, and have successfully blamed the U.S. for injuries without properly investigating their real cause. Classification issues mean the U.S. government cannot publicly acknowledge or respond to such claims. Because the government cannot officially acknowledge that the program even exists, it cannot launch public investigations into any claims of collateral injury or death.

While the above tactical metrics can suggest what the immediate effects of any individual drone strike (including whether it targeted the right person) are, those measures do not indicate whether the campaign as a whole is effective or not. Thus, there must be a broader consideration of the strategic effects of these strikes: not just about an individual strike but about the total effects of these strikes over time. Five of those considerations are listed below.



Strategic Metrics of Drone Strikes
1. The changing behavior of targeted groups
One direct way to measure the effects of drone strikes is to see how drones prompt terrorist groups to modify their behaviors in response. It is reasonable to assume that as drones strike at terrorist groups, those groups will modify their behavior to try to avoid being targeted in the future. However, the classified nature of much reporting on the makeup and behavior of terrorist groups can limit how much these changes can be measured. In general, drone strikes have resulted in three broad changes to terrorist group behaviors: rejecting technology, going into hiding, and violently attacking those suspected of participating in the targeting process. Pir Zubair Shah, a Pakistani journalist, recounted in a recent article how the Pakistani Taliban had reacted to drone strikes: “The Taliban and al Qaeda had stopped using electronic devices, they told me. They would no longer gather in huge numbers, even in mosques to pray, and spent their nights outside for safety, a life that was wearing thin.”49 Other journalists have described how the militants in Pakistan hate and fear drones. “The strikes fueled a vicious paranoia among the Taliban,” wrote David Rohde, who was imprisoned by militants for seven months in 2008. “For months, our guards told us of civilians being rounded up, accused of working as American spies and hung in local markets.”50 Reports of violently “purging” suspected informants for the drone program are routine.51 In Yemen, militants have adopted similarly brutal methods for responding to suspected informants. In one especially horrific instance in the summer of 2012, al Qaeda in the Arabian Peninsula videotaped their crucifixion of a Yemeni man on a lamppost, accusing him of helping the U.S. drone campaign.52 They did this while fleeing the town of Ja’ar in the face of an onslaught by the Yemeni military that was supported by U.S.administered drone strikes.

Soldiers from the 1st Battalion, 10th Special Forces Group (Airborne) teach mounted infantry tactics to soldiers from the Malian Army in Timbuktu, Mali. U.S. Air Force

Other Yemeni groups have adopted complicated methods for hiding their training facilities from drone reconnaissance. In at least one instance, Yemeni militants set up a “pop-up” training camp, where they drove into the desert in pickup trucks, set up some targets, fired weapons at those targets, then packed up and left within a matter of hours. The idea was to set up a training session so quickly that they could avoid being targeted for a strike.53


In Somalia, drone strikes are reported to have “limited al Shabaab’s mobility,” according to a local analyst.54 Since the militants are afraid to move around in large groups and operate openly, their ability to contest control of some areas is limited. In addition, al Shabaab militants have openly bragged of executing their own cohorts on suspicion of spying for the west.55 Academic studies are not in consensus about the overall effect drone strikes have on militant behavior. Examining Pakistan, one study found a strong correlation between drone strikes and a temporary disruption in terrorist activity, specifically suicide attacks involving improvised explosive devices.56 In a comparative study between Yemen and Pakistan, however, another researcher found that over time drone strikes correlate with more attacks against local government forces, though in Pakistan, the number of militant attacks appears to have held steady.57 A third study suggests that “failed” drone strikes, defined as strikes that miss their intended target, increase militancy in Pakistan, but successful strikes serve as a temporary deterrent.58 Other studies have suggested that drone strikes, particularly in Pakistan, do not noticeably affect the propaganda output of al Qaeda and associated groups, leading the researchers to question what other effects drones may have on group dynamics.59 Taken together, these anecdotes and studies suggest some types of drone strikes are likely to temporarily disrupt militant groups. However, they also suggest that researchers understand very little about how these groups are coping with drones, and that groups may potentially become more violent than they were before the drone strikes.

2. Changes in targeted group size or recruitment
If drones are effective at degrading terrorist groups, then one can indirectly measure their effects through how targeted groups react to them. Estimates of terrorist group sizes should reflect the impact of successful drone strikes. However, official estimates of terrorist group sizes are an imprecise science, and reporting agencies often have agendas for either inflating or minimizing their estimates for political purposes. In Pakistan, any estimation of the size of terrorist groups is complicated by the rich ecosystem of terror groups that operate there. In a 2008 testimony before the South Asia Subcommittee of the House Foreign Affairs Committee, Ashley Tellis broke Pakistani terror groups into five broad categories: sectarian, anti-Indian, Afghan Taliban, and al Qaeda & affiliates.60 Of these groups, drones are known to have targeted al Qaeda and affiliated groups (the primary target of drone strikes in Pakistan), some Afghan Taliban groups (such as the Haqqani Network in North Waziristan), and possibly some sectarian groups (such as Lashkar-e-Taiba, along the border with Afghanistan). As the pace of drone strikes increased in Pakistan in 2009, Canadian journalist Sharmeen ObaidChinoy reported a rise in Pakistani militants recruiting children to carry out suicide attacks.61 Other reporting from Pakistan suggests that militants recruit as often from affluent and well-educated areas of Pakistan as they do from the tribal areas.62



However even before the use of drones in Pakistan, reports suggested that al Qaeda and affiliated movements already had active recruitment efforts in these same areas.63 U.S. officials have reported that the flow of western recruits to terror groups in Pakistan has increased, but hard numbers are difficult to come by.64 Subsequent research in Pakistan suggests that other causes for recruitment, like childhood education, also plays a strong role in radicalization and eventual recruitment.65 Taken as a whole, it is difficult to single out drones as a major cause for either increasing or decreasing the size of terrorist groups in Pakistan. Too many factors go into the decision to join one of the many disparate groups to single out drone strikes. Other studies comparing Somalia and Yemen suggest structural factors, like failed institutions, porous borders, and a lack of central control by the government; all contribute substantially to the growth of terrorist groups.66 Yemen presents an especially puzzling story. Former Ambassador Barbara Bodine says that when she arrived in Yemen in 1997, “there was already an Al Qaeda presence.”67 Despite the 2002 drone strike in Yemen, al Qaeda did not become a significant presence again in the country until 2006, when 23 al Qaeda inmates escaped from prison.68 Those escapees rebuilt al Qaeda in Yemen, eventually rebranding it AQAP.

A U.S. Army Sergeant trains Ethiopian National Defense Force soldiers. U.S. Air Force

Since then, it’s been difficult to get a solid grasp of how AQAP has changed. The group grew from the initial group of 23 escaped militants to upwards of 300 by the end of 2009.69 By mid-2012, U.S. officials believed AQAP had expanded to “more than 1,000 members” with connections to al Qaeda leadership in Pakistan.70 Between 2009 and 2012, the U.S. dramatically increased the use of drones to carry out air strikes in Yemen. The correlation between the rise of AQAP and the rise of drone strikes has led some analysts to conclude that drones are driving AQAP’s expansion,71 but such analysis fails to account for the group’s rise before drone strikes began in 2010. The data about how terrorist groups change over time in Pakistan, Yemen, and Somalia make it difficult to single out drones as a cause for any increases or decreases in recruitment.


3. Political and/or social effects in the host country
Based on anecdotal evidence, drone strikes seem to cause a rise in anti-American sentiment in every country they are used. Appearances can be deceiving, however. The reality is that it is extremely difficult to single out drone strikes as a unique or even primary cause of anti-Americanism. Nevertheless, the political and social effects of drone strikes can provide an indirect means of measuring their success. In Pakistan, opposition to the drone strikes has become an issue of national politics, with an entire political party organized around ending their use.72 In Yemen, there is a growing body of anecdotal evidence that drone strikes are driving anti-Americanism and opposition to the central government.73 More broadly, airstrikes in Afghanistan – some of which come from drones – have stirred protests from both the government and from the population.74 However, these anecdotes are not conclusive about the negative effects of lethal drone strikes. In Pakistan, people protest a variety of U.S. policies they dislike, including the war in Afghanistan.75 On drones in particular, Pakistanis express a mixture of opinions. Sometimes, shocking incidents like the Raymond Davis incident can galvanize public reaction. In January of 2011, the Pakistani police caught American Raymond Davis after he killed two assailants in Lahore, Pakistan. The U.S. insisted he had been granted diplomatic immunity, which Pakistan rejected. News of his arrest quickly spread along with public disagreements from both governments over his diplomatic status. Pakistani officials began to leak sensitive information about his employment, alleging that he was an operative of the CIA76 collecting targeting information on local terrorist groups for eventual drone strikes.77
A U.S. Navy chaplain consults with local nun over U.S.provided donations to a Catholic school in Djibouti. U.S. Air Force

The revelations about Davis’ employment sparked public outcry78 and resulted in anti-American riots after his release from Pakistani custody: many Pakistanis expressed outrage not only that he wasn’t tried for killing two men, but that he was collecting drone intelligence in the first place.79 In contrast, reports from the tribal areas also suggest there is some guarded support from the people who live there, who seem to prefer the drone strikes over other alternatives like conventional military campaigns.80 The contrast between the two views – urban Pakistanis in the Punjab seem to hate drones, while rural Pakistanis in the tribal areas seem to guardedly support them – suggests a more complicated view of Pakistani reactions to the strikes than a simple pro- or anti-American narrative.



Yemen presents a similarly muddled case. When journalists have gained limited access to droneaffected areas of the country, they have reported deep anger at the government in Sana’a as well as the United States.81 Yet other researchers interviewing community elders from non-urban areas report the opposite: those communities seem to guardedly support drone strikes as a least-bad option for countering al Qaeda groups.82 Both stories could conceivably be right, but without further data, it is difficult to say conclusively what can be drawn from either account – there just aren’t enough data to definitively identify drones as the cause of anti-Americanism absent other factors (such as U.S. support for the Yemeni government). Yemen faces many more political challenges than the challenge of countering AQAP. The Houthi rebellion in the north and the Southern secessionist movement in the south both pose substantial challenges to the central government. 2012 saw a historic change of power, when Ali Abdullah Saleh abdicated the presidency and Abd Rabbuh Mansur Hadi replaced him.83 But Hadi has faced the same restive population, unresolved political challenges, and threat from AQAP that his predecessor did84. It is difficult to say that drone strikes have improved or hurt his standing or the efficacy of his government.

4. The enablement of local government forces
Outside of declared combat zones, drones are most often employed as the last resort option for striking terrorists. In Pakistan, Yemen, and Somalia, drones are used for lethal strikes because there aren’t always suitable alternatives to accomplishing some counterterrorism goals. However, the data show that drones don’t have permanent effects – local forces are needed to build any long lasting outcome from drone strikes. The way drone strikes enable local forces to accomplish counterterrorism goals is a direct measure of their effectiveness. “Local forces” does not necessarily mean the national army. In many disconnected peripheral communities in Pakistan, Yemen, and Somalia, a national force might be seen as an outside occupier and inspire resentment. Local representatives of the government, however, represent a theoretical solution to the problem: they are still responsible to the capital, but come from local areas and thus would be more responsive to local concerns. Creating a local force is not easy, however, and the history of trying to (especially by outsiders, like in Afghanistan) is fraught with challenges and failure. Even so, local forces are the only feasible way to build upon the political and security space drone strikes provide. In Pakistan, data suggest drone strikes have a temporary effect on militancy and violence but may actually lead to more violence over time (see point #1 above). Even so, if local forces could move into drone strike areas, then the longer term increases in violence could, conceivably, be thwarted. In general, Pakistan presents a unique case for drone use. Drone strikes take place in the tribal areas of the northwest, a legal wasteland that offers no means for law enforcement except for collective punishment. For decades, residents of the tribal areas could not vote. When Islamabad granted them


the right to vote in 1996, political parties were still forbidden in the area, so politicians could not build bases of support to run for office. In the vacuum that resulted, Islamists build support through mosques and madrassas and displaced civic and tribal leadership.85 The residents of the tribal areas do not have normal political rights, and the Pakistani state does not exercise normal sovereignty over the area. So there are no real local forces to build upon the space that drones provide. The Pakistani Army is not suited to be a local force: drones strike in predominantly Pashtun areas, and the army is not mostly Pashtun. While ethnicity is not always destiny, surveys suggest that locals do not find the Pakistani Army very trustworthy.86 Additionally, news of Pakistani Army offensives tends to create thousands of IDPs fleeing the area, suggesting that they are not an appropriate choice for “filling the gap” left by drone strikes. Given the tribal areas’ complicated history with central control, it’s unclear what could, conceivably, be a viable replacement.
U.S. Air Force Brigadier General Michael Callan visits graduating Senegalese infantry trainees. U.S. Air Force

In other words, drones in Pakistan are, at best, a temporary tactic for managing part of the terror problem; there is no viable replacement that offers a better long-term outcome just yet.

Yemen presents a different case. There, drones have been used in a far more restrained fashion than in Pakistan and have enjoyed open support by the government. More important, Yemen does not have territories that are legally separate from the rest of the country, the way Pakistan’s tribal areas are. In mid-2011, AQAP occupied several towns in the southern province of Abyan. Zinjibar, the provincial capital, and Ja’ar, a smaller city, were occupied by AQAP’s political front organization called Ansar al-Shariah.87 A year later, the new Hadi government launched a massive operation to clear Ansar alShariah out of the area. U.S. drones joined U.S. and Yemeni aircraft to bombard AQAP emplacements and chase them out of town.88 Yemen did not only use air strikes, however. Ground troops managed to seize control of both Ja’ar89 and Zinjibar.90 Those ground forces had extensive help from local forces (such as popular committees, many of which were composed of tribal leaders). It remains unclear whether this victory is meaningful for the long run – there are reports of AQAP fighters simply hiding to avoid reprisal by U.S. and Yemeni forces91 – but it did represent an important shift from how drone strikes are used in Pakistan. Using local forces to fill in the gaps from drone strikes is an important component to making any campaign effective. Drones have played an even smaller role in Somalia than in Yemen. There are no data to suggest that they’ve played a decisive role in Somali and African Union forces establishing control of Mogadishu


over the last year. While drones can be effective at temporarily managing a terrorist challenge, the data are clear that at a bare minimum, actual territory must be controlled by ground forces. Otherwise, any benefit or edge that drone strikes might give vanishes and the targeted militants simply return to committing violence.

5. End State Considerations
Ultimately, the question that must be answered when evaluating drone strikes is “what is the end state?” Drones have some discrete and measurable effects, but what purpose are the strikes meant to serve? The stated U.S. policy is to destroy, degrade, and defeat al Qaeda. But determining what that looks like is no simple task. While drones can be effective at destroying parts of al Qaeda and thus degrading its capacity to launch attacks, they are also insufficient on their own for accomplishing the broader goals of U.S. counterterrorism policy. Most academic studies agree that targeted killing conducted by armed drones may be effective as part of a broader strategy.92 Drones, however, have limits. Where drone strikes are found to have a measurable effect, it tends to be temporary. Successful strikes correlate in some circumstances with a temporary reduction in the incidence and intensity of terrorist violence,93 but may also correlate with long-term increases in retaliatory attacks against local government and persistent instability.94 This suggests that while drones can manage the terrorist problem for a short time, they are not necessarily contributing to a long term reduction of the threat. The long term reduction of threat is absent in most discussions of the drone program. Drones have killed many al Qaeda terrorists, but the threat appears to be migrating elsewhere and taking on new forms.95 So what is that end state drones are meant to accomplish, and can we measure whether that end state is being reached? This is where understanding the difference between strategic and tactical effects matters greatly. Paradoxically, the strikes are easier to measure and understand strategically than they are tactically: the broad, social effects that accompany their use are plainly visible, but the minute details of any individual strike (or even series of strikes) are difficult to determine. Strategic effects cannot be constructed meaningfully by aggregating the many poorly-reported tactical effects in the media and in various drone databases. Rather, developing a global perspective matters: are drones making the problem of terrorism worse, or not? From a tactical perspective, drones have been a stunning success by most measures. They kill dozens of al Qaeda-linked terrorists and put no Americans at risk in the process. There are many indications that non-combatant civilians have also been killed in these strikes, but few policymakers think the numbers are high enough to justify a major rethink of the program. From a strategic perspective, however, the effects of the drone strikes are less clear. It’s true that lots of terrorists have been killed and the capacity for al Qaeda and its branches to directly attack the


US has been curtailed. But that success has come at a steep price. The strikes have poisoned the bilateral relationship with Pakistan – even though the government privately consents to strikes, its public statements have used them to deflect criticism onto the United States. Pakistan is hardly more secure now than when drone strikes began in 2004, and militants have recently begun attacking Pakistan’s nuclear weapons facilities.96 As a consequence, anti-Americanism is higher than ever while the Pakistani government is as dysfunctional as ever. In Yemen, too, drone strikes have killed many leaders of AQAP while also driving anti-Americanism and possibly aiding in militant recruitment. In Yemen a narrow view suggests drone strikes have been stunningly effective: a drone strike has killed AQAP figures and contributed to a successful Yemeni offensive to retake towns that had been occupied by AQAP. But the vast majority of reporting suggests that these same strikes also drive Yemenis to embrace al Qaeda.97 Drones are a minority of other U.S. spending in the country,98 and it’s far from clear that that spending is effective at the broader U.S. goals of promoting stability and the rule of law.99 This suggests that many studies about drones that focus only on casualties and strikes miss the larger effects and consequences of lethal drones. The few studies that do try to account for larger social and political effects of drone strikes focus only on Pakistan and are limited by selection bias issues.100 Partly, this is a result of the poor quality of available data. But partly, it is also because of the narrow scope of such research. As it’s currently constructed, drone strikes are a narrow solution to a narrowly defined problem; they do not address the larger, systemic and structural issues that create environments welcoming to terrorists. The mission of policymakers responsible for the drone strikes is to make those strikes as effective as possible for killing terrorist leaders while minimizing blowback, including from civilian casualties; they are not charged with safeguarding the long-term political or social health of the countries where strikes occur.

U.S. Marine Corps CH-46 Sea Knight helicopter landing in Somalia. U.S. Department of Defense

This conclusion suggests a change in focus is needed to make drone strikes more effective. They need to be considered beyond the narrow issue of how many they kill and instead viewed from a broader perspective. Though U.S. officials always insist in public statements that they are focused on a broader strategy,101 the programs in place do not always effect such a consideration. While the amount of overt aid to Yemen has been almost evenly split between civilian and military/counterterrorism programs (the U.S. has also sent an unknown amount of covert aid to the Yemeni government), the last three years have seen the rapid growth of an entrenched al Qaeda group, the collapse of the previous regime, and the growth of widespread anti-Americanism. 102 The aims might be broader than the simple elimination of terrorists, but the policies enacting the current strategy are not achieving those aims. Developing ideas and considerations for that strategy will be covered in the next essay in this series.


Where to Go
Policymakers now face a critical choice: how will drones be used in the future? Many questions remain about the current use of drones for striking at terrorists. They should be addressed before the program is expanded to new areas of conflict. For starters, the U.S. government needs to provide better data about how and why it conducts drone strikes to bolster public confidence in the program. While the openness with which officials have discussed the highly classified program is noteworthy, such talk has not provided data to substantiate the broad claims officials have made – such as extraordinarily low civilian casualties.103 The scattered reporting about drone strikes disputes many of these claims. Official openness about drone strikes would enhance the public debate and build support for the program. However, increased government transparency is not a simple prospect. The different organizations that operate drones do so under different parts of U.S. law. The CIA’s activities take place under one section of national security law (Title 50), while the military functions under another (Title 10).104 Additionally, most of the drone strikes in Yemen and Somalia are considered clandestine, which means they’re secret but the government can officially acknowledge. Drone strikes in Pakistan, on the other hand, are covert, which means the government cannot even acknowledge their existence. In addition, each organization has dramatically different reporting requirements – the CIA reports to a small group of cleared officials in the administration and Congress, while the military has broader reporting requirements throughout its chain of command. Resolving the different legal authorizations for drone strikes, and unifying reporting requirements, is a vital step toward getting government data about drone strikes. Drone strikes will be most effective if combined with broader policies to handle their many effects. Developing those policies will be the subject of future research at ASP.

Joshua Foust is a Fellow at the American Security Project


Further Reading:
PAPER: The Strategic Context of Lethal Drones The US and its UAVs: The Financial Cost versus Strategic Value of Drones FACT SHEET: Fact Sheet U.S.C. Title 10, Title 22, and Title 50 FACT SHEET: Contextualizing the Drones Debate Annotated Drones Bibliography FACT SHEET: Yemen FACT SHEET: Mali Endnotes
1. 2. 3. 4. 5. 6. 7. 8. 9. Khaled Dawoud, “Mohammed Atef,” The Guardian, 18 November 2001, Richard Clarke, “Give drones a medal,” New York Daily News, 02 December 2012, Ann Scott Tyson, “Does bin Laden matter anymore?” The Christian Science Monitor, 01 March 2002, http://www.csmonitor. com/2002/0301/p01s02-usmi.html Tony Karon, “Yemen Strike Opens New Chapter in War on Terror,” Time, 05 November 2002,,8599,387571,00.html Peter Bergen and Katherine Tiedmann. “Washington’s Phantom War.” Foreign Affairs. July/August 2011. articles/67939/peter-bergen-and-katherine-tiedemann/washingtons-phantom-war According to an off-the-record interview with a senior U.S. official involved in the drones program in Pakistan. John Brennan, “The Ethics and Efficacy of the President’s Counterterrorism Strategy,” The Woodrow Wilson International Center for Scholars, 30 April 2012, Jo Becker and Scott Shane, “Secret ‘Kill List’ Proves a Test of Obama’s Principles and Will,” The New York Times, 29 May 2012, http:// Greg Miller, “Plan for hunting terrorists signals U.S. intends to keep adding names to kill lists,” The Washington Post, 23 October 2012, 2012/10/23/4789b2ae-18b3-11e2-a55c-39408fbe6a4b_story.html

10. According to an off-the-record discussion on 7 November 2012 with a journalist based in Yemen who covers drone strikes. 11. John Henrehan, “Military drone mistaken for ‘UFO’ along DC highways,” Fox 5, 14 June 2012, 12. Joshua Foust, “Phantom Militias,” The Columbia Journalism Review, 3 November 2008,


AMERICAN SECURITY PROJECT tom_militias.php 13. International Human Rights and Conflict Resolution Clinic, Stanford University and Global Justice Clinic, New York University School of Law, Living Under Drones: Death, Injury, and Trauma to Civilians from US Drone Practices in Pakistan, September 2012,, pp. 32-35 14. According to an off-the-record interview with a senior U.S. military official who was involved in the drones program in Pakistan, the 2004 drone strike that killed Nek Muhammed in South Waziristan was actually a missile strike by an F-16 fighter jet. The author was unable to confirm or refute the claim. 15. Pir Zubair Shah, “My Drone War,” Foreign Policy, March/April 2012, war?page=full 16. Daily Times, “Ministry issues guidelines for foreigners visiting Peshawar, FATA,” 25 June 2012, page=2012%5C06%5C25%5Cstory_25-6-2012_pg7_18 17. Reporting from advocacy groups in particular fall prey to this form of bias: in seeking out only aggrieved victims of drone strikes, they exclude potential access to both victims and non-victims who don’t share that view. 18. For details, see the remarks of C. Christine Fair at the ASP Event, “U.S. Drones Policy: Strategic Frameworks and Measuring Effects,” 20 August 2012, 19. Greg Miller, “CIA seeks new authority to expand Yemen drone campaign,” Washington Post, 18 April 2012, http://www.washingtonpost. com/world/national-security/cia-seeks-new-authority-to-expand-yemen-drone-campaign/2012/04/18/gIQAsaumRT_story.html?hpid=z1 20. Jo Becker and Scott Shane, “Secret ‘Kill List’ Proves a Test of Obama’s Principles and Will,” The New York Times, 29 May 2012, http:// 21. Greg Miller, “Plan for hunting terrorists signals U.S. intends to keep adding names to kill lists,” Washington Post, 23 October 2012, http:// 0/23/4789b2ae-18b3-11e2-a55c-39408fbe6a4b_story.html 22. Ismail Khan, “U.S. Drone Strike Kills a Commander for Al Qaeda in Pakistan,” The New York Times, 9 December 2012, 23. Mohammed Mukhashaf, “Drone kills nine near formerly rebel-occupied Yemen town: sources,”Reuters, 18 October 2012, http://www. 24. Declan Walsh, “Pakistan’s top Taliban leader Baitullah Mehsud killed in US drone attack,” The Guardian, 7 August 2009, http://www. 25. Adam Baron, “Family, neighbors of Yemeni killed by U.S. drone wonder why he wasn’t taken alive,” McClatchy, 28 November 2012, http:// 26. Gregory D. Johnsen, “Why Did the US kill ‘Adnan al-Qadhi?,” Waq al-Waq, 13 December 2012, 27. See, for example, Rasool Dawar, ”US drones kill 3 suspected militants in Pakistan,” Associated Press, 9 December 2012, com/us-drones-kill-3-suspected-militants-pakistan-081710082.html 28. Scott Shane, “C.I.A. Is Disputed on Civilian Toll in Drone Strikes,” The New York Times, 11 August 2011, http://www.nytimes. com/2011/08/12/world/asia/12drones.html?pagewanted=all 29. Mark Landler, “Civilian Deaths Due to Drones Are Not Many, Obama Says,” The New York Times, 30 January 2012, http://www.nytimes. com/2012/01/31/world/middleeast/civilian-deaths-due-to-drones-are-few-obama-says.html 30. See, for example, 31. See, for example, 32. See, for example, Chris Woods, “Analysis: CNN expert’s civilian drone death numbers don’t add up,” The Bureau of Investigative Journalism, 17 July 2012,; and Meg Braun, “Counting civilian casualties in CIA’s drone war,” Foreign Policy’s AfPak Channel, 2 November 2012, 33. United Nations High Commission on Refugees. “Country Profiles”. 34. BBC World News, “Pakistani soldiers storm mosque,” 10 July 2007, 35. Hameedullah Khan, “Swat cleric ends peace deal,” Dawn, 22 September 2007, 20

36. Hassan Abbas, “The Black-Turbaned Brigade: The Rise of TNSM in Pakistan,” Jamestown Terrorism Monitor, 30 November 2006, http:// 37. C. Christine Fair, “Pakistan Loses Swat to Local Taliban,” Jamestown Terrorism Focus, 14 November 2007, single/?no_cache=1&tx_ttnews%5Btt_news%5D=4537 38. Jane Perlez and Pir Zubair Shah, “Confronting Taliban, Pakistan Finds Itself at War,” The New York Times, 2 October 2008, http://www. 39. Muhammad Imran, “Majority of Pakistan Swat IDPs back home,” Ground Report, 28 August 2009, 40. AFP, “Thousands flee Waziristan in fear of military offensive,” 26 August 2012, 41. Faisal Darem, “Residents flee Zinjibar after al-Qaeda takeover,” al-Shorfa, 31 May 2011, 42. Ahmed Dawood, “Ansar al-sharia militants withdraw from Zinjibar and Jaar to port of Shuqra,” Yemen Times, 14 June 2012, http://www. 43. UNHCR, “Almost 320,000 civilians flee Somalia this year, including 20,000 to Yemen,” 21 October 2011, http://www.unhcr. org/4ea185356.html 44. Spencer Ackerman, “Rare Photographs Show Ground Zero of the Drone War,” Wired Danger Room Blog, 12 December 2011, http:// 45. William Matthews, “Smaller, Lighter, Cheaper,” Defense News, 31 May 2010, 46. Michael Smith, “Army ‘vacuum’ missile hits Taliban,” The Times, 22 June 2008, 47. According to an off-the-record interview with a former intelligence official, 22 October 2012. 48. See, for example, here: 49. Pir Zubair Shah, “My Drone War,” Foreign Policy, March/April 2012, war?page=full 50. David Rohde, “The drone war,” Reuters Magazine, 26 January 2012, 51. Associated Press, “Thousands Cheer as Pakistani Militants Decapitate, Shoot Afghans Accused of Spying for U.S.,” Fox News, 27 June 2008,,2933,372883,00.html 52. Iona Craig, “Little sympathy for crucified ‘spy’ after Islamists flee town,” The Times, 20 June 2012, world/middleeast/article3450564.ece 53. For details, see the remarks of Aaron Zelin at the ASP Event, “U.S. Drones Policy: Strategic Frameworks and Measuring Effects,” 20 August 2012, 54. Armin Rosen, “How Africa’s Most Threatening Terrorist Group Lost Control of Somalia,” The Atlantic, 21 September 2012, http://www. 55. Zoe Flood and Abdiaziz Abdinuur, “Al-Shabaab ‘execute British and American spies,’” The Telegraph, 23 July 2012, http://www.telegraph. 56. See, for example, Patrick Johnson and Anoop Sarbahi, “The Impact of U.S. Drone Strikes on Terrorism in Pakistan,” Working paper, February 25, 2012, 57. Luke Olney, “Lethal Targeting Abroad: Exploring Long-Term Effectiveness of Armed Drone Strikes in Overseas Contingency Operations,” Graduate School of Arts and Science of Georgetown University, Masters Degree Thesis, April 14, 2011, 58. David Jaeger and Zahra Siddique, “Afghanistan and Pakistan? On the Dynamics of Violence between the United States and the Taliban,” Institute for the Study of Labor Discussion Paper Series, December 2011, 59. Megan Smith and James Igoe Walsh, “Do Drone Strikes Degrade al Qaeda? Evidence form Propaganda Output,” Forthcoming, Terrorism and Political Violence. January 27, 2012, 21


60. Ashley Tellis, “U.S.-Pakistan Relations: Assassination, Instability, and the Future of U.S. Policy,” 16 January 2008, 61. Sharmeen Obaid-Chinoy, “Pakistan’s Taliban Generation,” CBC, 27 June 2009, 62. Zia Ur Rehman, “Taliban Recruiting and Fundraising in Karachi,” CTC Sentinel, 24 July 2012, 63. C. Christine Fair, “Militant Recruitment in Pakistan: Implications for Al Qaeda and Other Organizations,” Studies in Conflict & Terrorism, 27:489-504, 2004, 64. Craig Whitlock, “Flow of terrorist recruits increasing,” Washington Post, 19 October 2009, 65. C. Christine Fair, “Militant Recruitment in Pakistan: A New Look at the Militancy-Madrasah Connection,” Asia Policy, 4 (July 2007), 66. Chantel Carter, “Pre-Incident Indicators of Terrorist Attacks: Weak Economies and Fragile Political Infrastructures Bring Rise to Terrorist Organizations and Global Networks,” Global Security Studies, Fall 2012, Vol. 3, Iss. 4, Final.pdf 67. Azmat Khan, “Understanding Yemen’s Al Qaeda Threat,” PBS Frontline, 29 May 2012, 68. Mark Trevelyan, “Jailbreak in Yemen stirs concern abroad,” Reuters, 10 February 2006, articles/2006/02/10/jailbreak_in_yemen_stirs_concern_abroad/ 69. Stefano Ambrogi, “Yemen says may harbor up to 300 Qaeda suspects,” Reuters, 29 December 2009, 70. Pam Benson, “New terrorist plot to attack plane foiled,” CNN, 8 May 2012, 71. Michael Kelley, “Obama’s Drone War In Yemen May Be Al Qaeda’s Best Recruitment Tool Ever,” Business Insider, 30 May 2012, http:// 72. Masood Salman. “In Pakistan, Drone Protest Takes Detour For Safety,” The New York Times, 08 Oct. 2012, http://www.nytimes. com/2012/10/08/world/asia/in-pakistan-drone-protest-led-by-imran-khan-takes-detour-for-safety.html 73. Andrew Hammond, “Yemeni debate over drones emerges after Saleh’s fall,” Reuters, 17 October 2012, 74. Laura King and Hashmat Baktash, “Afghan leader says NATO airstrikes on homes ‘absolutely banned,’” The Los Angeles Time World Now blog, 12 June 2012, 75. AFP, “Thousands rally against US, NATO in NW Pakistan,” 18 December 2008, 76. Declan Walsh, “American who sparked diplomatic crisis over Lahore shooting was CIA spy,” The Guardian, 20 February 2011, http://www. 77. Rob Crilly, “Detained US official ‘in telephone contact with Islamic terror group,’” The Telegraph, 10 February 2011, 78. M Ilyas Khan, “Behind the scenes of Pakistan Raymond Davis ‘spy’ saga,” BBC World News, 01 March 2011, world-south-asia-12598684 79. Carlotta Gall and Mark Mazetti, “Hushed Deal Frees C.I.A. Contractor in Pakistan,” The New York Times, 16 March 2011, http://www. 80. Pir Zubair Shah, “My Drone War,” Foreign Policy, March/April 2012, war?page=full 81. Adam Baron, “Family, neighbors of Yemeni killed by U.S. drone wonder why he wasn’t taken alive,” McClatchy, 28 November 2012, http://


82. Christopher Switft, “The Drone Blowback Fallacy,” Foreign Affairs, 01 July 2012, 83. Laura Kasinof, “Yemen Leader Leaves for Medical Care in New York,” The New York Times, 22 January 2012, http://www.nytimes. com/2012/01/23/world/middleeast/yemeni-president-leaves-heading-for-us.html 84. Laura Kasinof, “For Yemen’s New President, a Battle for Control and a Tug of War With the Past,” The New York Times, 13 June 2012, 85. C. Christine Fair, Nicholas Howenstein, and J. Alexander Their, “Troubles on the Pakistan-Afghanistan Border,” U.S. Institute of Peace, December 2006, 86. Naveed Ahmad Shinwari, “Understanding FATA: 2011 – Attitudes Towards Governance, Religion & Society in Pakistan’s Federally Administered Tribal Areas,” Produced with the support of Community Appraisal & Motivation Programme. 2012, http://understandingfata. org/ 87. Jeb Boone, “Southern Yemen steeped in conflict,” GlobalPost, 19 June 2011, 88. Bobby Ghosh, “The End Of al-Qaeda?” Time, 17 September 2012,,9171,2123810,00. html?pcd=pw-w 89. Mohammed Mukhashaf and Sami Aboudi, “Yemeni army says retakes militant stronghold of Jaar,” Reuters, 12 June 2012, http://www. 90. Fawaz al-Haidari, “Yemen army retakes Qaeda bastions,” AFP, 12 June 2012, 91. Adam Baron, “A Hollow Victory,” Foreign Policy, 2 July 2012, 92. Carvin, Stephanie. “The Trouble with Targeted Killing,” Security Studies, vol. 21:3, pp. 529-555. August 2012. 93. Johnston, Patrick B. and Anoop Sarbahi. “The Impact of U.S. Drone Strikes on Terrorism in Pakistan,” Working paper, February 25, 2012. 94. Olney, Luke A. “Lethal Targeting Abroad: Exploring Long-Term Effectiveness of Armed Drone Strikes in Overseas Contingency Operations,” Graduate School of Arts and Science of Georgetown University. Masters Degree Thesis, April 14, 2011. 95. Joshua Foust, “How Strong Is al Qaeda Today, Really?” The Atlantic, 01 May 2012, 96. Declan Walsh, “Militants Attack Pakistani Air Force Base,” The New York Times, 17 August 2012, world/asia/pakistani-air-force-base-with-nuclear-ties-is-attacked.html 97. Andrew Hammond, “Yemeni debate over drones emerges after Saleh’s fall,” Reuters, 17 October 2012, 98. Munoz, Carlos, “Details emerge on US counterterrorism support for Yemen,” The Hill Defcon Blog, July 16, 2012, blogs/defcon-hill/policy-and-strategy/238091-details-emerge-on-us-counterterrorism-support-for-yemen 99. Angus McDowall. “Donors pledge $6.4 billion in aid to Yemen,” Reuters, September 4, 2012. us-yemen-aid-idUSBRE8830FR20120904 100. Joshua Foust, “Targeted Killing, Pro and Con: What to Make of U.S. Drone Strikes in Pakistan,” The Atlantic, 26 September 2012, http:// 101. John O. Brennan, “U.S. Policy Toward Yemen,” PBS Newshour, 08 August 2012, 102. Government Accountability Office, “Uncertain Political and Security Situation Challenges U.S. Efforts to Implement a Comprehensive Strategy in Yemen,” February 2012, 103. Scott Shane, “C.I.A. Is Disputed on Civilian Toll in Drone Strikes,” The New York Times, 11 August 2011, http://www.nytimes. com/2011/08/12/world/asia/12drones.html?pagewanted=all 104. For a description of these different types of authorities, see Ashley Boyle, “Fact Sheet: U.S.C. Title 10, Title 22, and Title 50,” American Security Project, August 2012,


Building a New American Arsenal The American Security Project (ASP) is a nonpartisan initiative to educate the American public about the changing nature of national security in the 21st century. Gone are the days when a nation’s strength could be measured by bombers and battleships. Security in this new era requires a New American Arsenal harnessing all of America’s strengths: the force of our diplomacy; the might of our military; the vigor of our economy; and the power of our ideals. We believe that America must lead other nations in the pursuit of our common goals and shared security. We must confront international challenges with all the tools at our disposal. We must address emerging problems before they become security crises. And to do this, we must forge a new bipartisan consensus at home. ASP brings together prominent American leaders, current and former members of Congress, retired military officers, and former government officials. Staff direct research on a broad range of issues and engages and empowers the American public by taking its findings directly to them. We live in a time when the threats to our security are as complex and diverse as terrorism, the spread of weapons of mass destruction, climate change, failed and failing states, disease, and pandemics. The same-old solutions and partisan bickering won’t do. America needs an honest dialogue about security that is as robust as it is realistic. ASP exists to promote that dialogue, to forge consensus, and to spur constructive action so that America meets the challenges to its security while seizing the opportunities the new century offers.