Jump to ContentJump to Main Navigation
Targeted KillingsLaw and Morality in an Asymmetrical World$

Claire Finkelstein, Jens David Ohlin, and Andrew Altman

Print publication date: 2012

Print ISBN-13: 9780199646470

Published to Oxford Scholarship Online: September 2012

DOI: 10.1093/acprof:oso/9780199646470.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 22 June 2018



Targeted Killings

Gregory S. McNeal

Oxford University Press

Abstract and Keywords

Critics of the U.S. policy of targeted killing by unmanned aerial vehicles (UAVs or drones) generally lack credible information to justify their critiques. In fact, in many instances their claims are easily refuted, calling into question the reliability of their criticism. This chapter highlights some of the most striking examples of inaccurate claims raised by critics of the U.S. policy of drone-based targeted killing. Specifically, it offers a much-needed corrective to clarify the public record or offer empirical nuance where targeted killing critics provide only unsubstantiated and conclusory statements of fact and law. The chapter is organized as follows. Section I discusses the decision protocol used by the U.S. military before launching a drone strike, a process that goes to extraordinary lengths to minimize civilian casualties. Section II addresses the critics' unsubstantiated claims about the legal, diplomatic, and strategic results of drone strikes.

Keywords:   targeted killing, US policy, unmanned aerial vehicles, drone strike

Critics of the U.S. policy of targeted killing by unmanned aerial vehicles (UAVs or drones) generally lack credible information to justify their critiques. In fact, in many instances their claims are easily refuted, calling into question the reliability of their criticism. This chapter highlights some of the most striking examples of inaccurate claims raised by critics of the U.S. policy of drone-based targeted killing. Specifically, this chapter offers a much-needed corrective to clarify the public record or offer empirical nuance where targeted killing critics offer only unsubstantiated and conclusory statements of fact and law.

Section I of this chapter discusses the decision protocol used by the U.S. military before launching a drone strike, a process that goes to extraordinary lengths to minimize civilian casualties. Although this decision protocol was once secret, recent litigation in federal court has resulted in the release of extensive information regarding U.S. targeting practices. An analysis of this information indicates that the U.S. military engages in an unparalleled and rigorous procedure to minimize, if not eliminate entirely, civilian casualties. Although independent empirical evidence regarding civilian casualties is hard to come by, it is certainly the case that statistics proffered by some critics cannot be empirically verified; their skepticism of U.S. government statements is not backed up by anything more substantial than generic suspicion.

Section II of this chapter addresses the critics’ unsubstantiated claims about the legal, diplomatic and strategic results of drone strikes. Although the counter-observations raised in this chapter do not, by themselves, demonstrate that targeted killings are morally or legally justified, they do suggest that some of the moral or legal objections to targeted killings are based on empirical claims that are either dubious, impossible to verify or just plain false. (p.327)

I. Claims about the process of targeted killing

The central claim raised by critics of the U.S. policy is that drone attacks are indiscriminate and result in a high number of civilian casualties. The claims are dramatic. For example, David Kilcullen and Andrew Exum argued in the New York Times that “drone strikes have killed about 14 terrorist leaders. However, according to Pakistani sources, drones also have killed some 700 civilians. This is 50 civilians for every militant killed, a hit rate of 2 per cent—hardly ‘precision’.”1 In a more moderate estimate, the non-profit Bureau of Investigative Journalism claims that since August 2010, “45 or more civilians appear to have died” in CIA drone attacks.2 Before citing contrary claims, it is important tonote the lack of precision in the Bureau's own numbers. They are not certain if 45 civilians—or some higher number—were killed, and, rather than cite an actual death toll, they simply claim that some civilians “appear” to have died. Similarly, the claims made by David Kilcullen and Andrew Exum that drone strikes have a two per cent hit rate have little, if any, evidentiary support. In regard to these figures, Foreign Policy's Christine Fair concludes, “It would be a damning argument—if the data weren't simply bogus.”3 She points out to readers that “The only publicly available civilian casualty figures for drone strikes in Pakistan come from their targets: the Pakistani Taliban, which report the alleged numbers to the Pakistani press, which dutifully publishes the fiction.”4

Contrast this with the definitive claims made by the U.S. government in response to the Bureau's report. The first comes from a senior U.S. counterterrorism official who stated:

There haven't been any noncombatant casualties for about a year, and assertions to the contrary are wrong. The most accurate information on counter-terror operations resides with the United States, and this list is wildly inaccurate. Those operations are designed to protect America and our allies, including Pakistan, from terrorists who continue to seek to kill innocents around the world.5

The second statement comes from John Brennan, Assistant to the President for Homeland Security and Counterterrorism, who said, “In fact I can say that the types of operations . . . that the US has been involved in, in the counter-terrorism (p.328) realm, that nearly for the past year there hasn't been a single collateral death because of the exceptional proficiency, precision of the capabilities that we've been able to develop.”6

Further undermining the claims of the Bureau, another U.S. counterterrorism official stated that:

[o]ne of the loudest voices claiming all these civilian casualties is a Pakistani lawyer who's pushing a lawsuit to stop operations against some of the most dangerous terrorists on the planet . . . His evidence, if you can call it that, comes from a press release. His publicity is designed to put targets on the backs of Americans serving in Pakistan and Afghanistan. His agenda is crystal clear.7

Although drone critics are not required to take government denials at face value, these statements by U.S. officials demonstrate that the empirical claims made by critics are highly contested by those who operate or oversee these programs. If critics wish to discredit these official statements, the critics should marshal supportable evidence for their statistics.

(a) The administrative process of collateral damage estimation and mitigation

In my own field of research I have reported on the administrative process the U.S. military follows to prevent civilian casualties in combat.8 Many critics of the U.S. policy of targeted killing fail to address the scientifically grounded mitigation steps followed by U.S. armed forces. Those mitigation steps are designed to ensure that the probability of collateral damage from a pre-planned operation is below 10 per cent. In practice, the mitigation steps have resulted in a collateral damage rate of less than one per cent. Specifically, in pre-planned operations the U.S. military follows a rigorous collateral damage estimation process based on a progressively refined analysis of intelligence, weapon effects, and other variables.

To gather the evidence to support the information summarized above and explained in greater detail below, I followed proven case study research techniques9 designed to ensure construct validity, external validity and reliability. Construct validity was ensured by using varied sources of evidence, including publicly available government documents, multiple open-ended interviews, and scholarly and press accounts of the collateral damage estimation and mitigation (p.329) process. I began by collecting and reviewing sources in the public record such as court documents and government declarations and then summarized my initial tentative observations. Then I tested these observations using triangulation techniques and multiple data sources to confirm or falsify my observations and test my research methods. Triangulation is the process by which a case study researcher provides confidence that findings are meaningful and reflect scientific truth.10 External validity was achieved by limiting the generalizability of the findings to the U.S. practice of collateral damage estimation and mitigation during pre-planned military operations.

During any military operation, armed forces are required by the laws of war to minimize collateral damage.11 Following international humanitarian law (IHL) concepts of distinction, proportionality and precautions generally satisfies this requirement. In pre-planned operations, or when time and combat circumstances permit, the U.S. military implements its IHL obligations by employing a multi-step process known as the “collateral damage methodology” (CDM).12 This methodology is grounded in scientific evidence derived from research, experiments, history, and battlefield intelligence, and is designed to adapt to time-critical events.13 The CDM takes into account every conventional weapon in the U.S. air to surface and indirect fire inventory and is a tool that assists commanders in mitigating unintended or incidental damage or injury to civilians, property, and the environment.14 The methodology assists commanders in assessing proportionality and in weighing risks to the collateral objects.15 The CDM is based on empirical data, probability, historical observations from the battlefield, and physics-based computerized models for collateral damage estimates.16

If in the targeting process a commander or their subordinates realize that there is a possibility of collateral damage resulting from an operation, they will employ a series of mitigation techniques intended to ensure, with a high degree of certainty, that there will not be an unacceptable probability of damage or injury to collateral concerns. This mitigation process involves a series of steps based on a progressively refined analysis of available intelligence, weapon type and effect, the physical environment, target characteristics and delivery scenarios keyed to risk thresholds established by the Secretary of Defense and the President of the United States.17

Perhaps the simplest way to understand the mitigation process undertaken by U.S. forces is to think of it as a series of tests based on risk to collateral concerns. These (p.330) tests are implemented in five levels (known as CDE levels); at each level, if a commander determines that collateral concerns are not within the collateral hazard area for a given weapon system, an operation can be commenced. However, if collateral concerns are within the collateral hazard area the operation cannot be undertaken without employing the available mitigation techniques at the next higher level of analysis. At the highest level (CDE Level 5) no mitigation technique will prevent civilian casualties and a senior commander, the Secretary of Defense or the President of the United States must make a judgment about proportionality.18

A few examples can illustrate these concepts. In Iraq, as of 2003, high collateral damage targets were defined as those that “if struck, have a ten percent probability of causing collateral damage through blast debris and fragmentation and are estimated to result in significant collateral effects on non-combatant persons or structures, including: (A) Non-combatant casualties estimated at 30 or greater . . . [including those] . . . in close proximity to human shields.”19 Thus, if after mitigation a commander in Iraq expected a pre-planned operation would result in more than 30 noncombatant casualties, the strike would have to be briefed through the chain of command and authorized by the Secretary of Defense. If the collateral damage estimate was less than 30, the target was defined as a low collateral damage target, and, in most circumstances,20 required approval by either the Commander of Multinational Forces Iraq21 or a Division Commander.22

Notably, Iraq in 2003 was not a counterinsurgency operation, whereas current (2011) operations in Afghanistan, relying on counterinsurgency doctrine, employ a noncombatant casualty value (NCV) of 1 for pre-planned operations.23 This (p.331) NCV of 1 reflects the strategic importance of minimizing collateral damage in counterinsurgency operations. Thus in pre-planned operations (for example, circumstances when troops are not in contact, or a target is not time-sensitive) in Afghanistan, if an operation is expected to result in even one civilian casualty, the strike must be approved by the National Command Authority.24

(b) The results of the administrative process

When followed, this administrative process dramatically reduces the collateral damage in U.S. military operations and also ensures high levels of political accountability. According to my research, less than one percent of pre-planned operations that followed the collateral damage estimation process resulted in collateral damage. When collateral damage has occurred, 70 per cent of the time it was due to failed “positive identification” of a target. Twenty-two per cent of the time it was attributable to weapons malfunction, and a mere eight per cent of the time it was attributable to proportionality balancing; that is, a conscious decision that anticipated military advantage outweighed collateral damage. Furthermore, according to public statements made by U.S. government officials as described above, the President or the Secretary of Defense (also known as the National Command Authority) must approve any pre-planned ISAF strike where one civilian casualty or greater is expected, thus ensuring high levels of political accountability.

This is an important point for the debate over CIA targeted killing operations. The military, operating in Afghanistan, follows a rigorous process of collateral damage estimation and mitigation. If, at the end of that process, mitigation is impossible and the estimated number of civilian casualties exceeds pre-determined limits,25 commanders must complete a sensitive target approval and review (STAR) process. This process is for targets whose engagement presents: the potential for damage and/or injury to noncombatant property and persons; potential political consequences; or other significant effects estimated to exceed predetermined criteria, thus presenting an unacceptable strategic risk.26 The National Command Authority must approve STAR targets.27 As discussed above, in Iraq in 2003 the President determined any operation in which 30 or greater civilian casualties were expected represented a significant strategic level event with geopolitical ramifications necessitating approval at the highest levels of government.28 Thus, if after mitigation a commander in Iraq expected a pre-planned operation would result in more than 30 noncombatant casualties, the strike would have to be briefed to—and (p.332) authorized by—the Secretary of Defense. If the collateral damage estimate was less than 30, the target was defined as a low collateral damage target and, in most circumstances,29 still required high-level approval, although that approval would occur within the military chain of command, usually with approval by either the Commander of Multinational Forces Iraq,30 or a Division Commander.31 Notably, the difference in strategy between Iraq in 2003 (major combat operations) versus Afghanistan in 2011 (counterinsurgency operations) was a driving factor in the different NCV (with the Afghanistan NCV being set at 1), meaning high levels of approval and therefore care prior to any air strike.32 This is a critically important point, as this very low NCV reflects the strategic importance of minimizing collateral damage in counterinsurgency operations.

These points regarding the care exercised by the U.S. military in targeted killing operations, and the recognition by the President of the political consequences that flow from civilian casualties, raise questions about the logic behind the arguments made by targeted killing critics. If an operation in Afghanistan is expected to result in even one civilian casualty, the strike must be approved by the National Command Authority.33 In light of this fact, it seems questionable that the CIA would exercise less care in its targeted killing operations just over the border. This is especially curious when one considers that the CENTCOM commander supervises U.S. forces in Afghanistan and has responsibility for operations in Pakistan. That commander personally approves nearly all strikes that are expected to result in one or more civilian casualties in Afghanistan, yet if the critics are to be believed this same commander is sidelined (and apparently silent) when it comes to CIA strikes in Pakistan. Are we to believe that the CENTCOM commander, who reports directly to the President, has no input into the fact that the CIA is allegedly inflicting 50 civilian casualties for every militant killed? Why would the CENTCOM commander be directed by the President to minimize civilian casualties for strategic and political reasons on the Afghanistan side of the border, yet the Director of the CIA, who also reports to the President, would have a free hand to inflict massive civilian casualties on the Pakistani side of the border? Perhaps the more likely conclusion is that the critics lack reliable information. (p.333)

In fact, these issues call into question the claims made by critics, in particular the logic behind their claims. While it is true that the CIA operates covertly and therefore has more leeway in its operations, and it is also true that the campaign in the tribal areas of Pakistan is a counterterrorism rather than a counterinsurgency operation,34 in order to believe the critics, one would nonetheless have to believe in grand inconsistencies in care and outcomes for agencies that both directly report to the President. As I have detailed in my research, the military follows a rigorous targeting process designed for use in most combat operations. That process has been highly successful in minimizing the number of civilian casualties from military airstrikes. Yet the numbers cited by critics, such as the ratio of 50 civilians killed for every strike is beyond even the 30 civilian NCV in Iraq at the height of major combat operations (not counterinsurgency operations). If we are to believe that the CIA substantially departs from the military's collateral damage estimation and mitigation processes, the onus should be on the critics to demonstrate this fact, which they plainly have failed to do. Moreover, the statements of the government with its own access to real-time information about operations must be weighed against those of journalists who cannot travel to remote regions of Pakistan,35 and civilians who are paid “blood money” if they claim to be collateral victims.36 Finally, even if the critics are correct and the CIA operates with fewer targeting restrictions, this would not demonstrate the illegality of targeted killing per se but would only support the prudential conclusion that drone operations should be handled by the U.S. military exclusively or the CIA should adopt the military targeting process.

Not only should the claims of critics be viewed skeptically because they are illogical, they should also be viewed skeptically because they are inconsistent. For example, one of the most prominent critics of U.S. targeted killing policy is law professor Mary Ellen O'Connell. In her essay “Unlawful Killing with Combat Drones: A Case Study of Pakistan 2004–2009,” she criticizes the U.S. policy by frequently relying on media reports (or others who summarize media reports) to bolster her claim that air strikes are causing high numbers of civilian (p.334) deaths.37 However, when General McChrystal ordered restrictions on air strikes and media sources noted a decline in the number of unintended victims, she suddenly began to question the reliability of the media. In her words, “[w]hether the numbers have actually declined is difficult to confirm because the U.S. and Pakistan have succeeded in keeping journalists out of the border region.”38 This passage reveals an inconsistency in O'Connell's claims. If media reports are not to be believed when they report a reduction in civilian casualties, why should they be believed when they report consistent or increasing civilian casualties? O'Connell offers no explanation for her inconsistent analytical techniques.

A similar inconsistency is displayed in O'Connell's criticism of U.S. claims of no or few civilian casualties. She writes that “the U.S. has little reliable on-the-ground information to confirm or discredit computer data,”39 and that “[i]n Afghanistan and Pakistan, the local informants, who also serve as confirming witnesses for the air strikes are notoriously unreliable.”40 If these people are unreliable as informants in support of strikes, why are they reliable as witnesses to collateral damage? How can any critic of targeted killing claim to be better at the task of sorting out reliable from unreliable informants than the CIA and other government officials on the ground in Afghanistan and Pakistan? Writer Jane Mayer, in her frequently cited New Yorker article, makes similar claims without much evidence, short of CIA officers who noted that sometimes informants were unreliable.41 O'Connell favorably cites Mayer, stating that “[i]n Pakistan, the U.S. has had little on-the-ground (p.335) information, and what it has had has not been very reliable.”42 Yet Mayer does not claim that the United States has little on-the-ground information, and even if she did make such a claim, one would want to know how a journalist had enough information about covert CIA assets to render such a conclusion.

This is especially the case when one considers Mayer's claim that one of President Obama's first drone strikes, on January 23, 2009, “targeted the wrong house, hitting the residence of a pro-government tribal leader six miles outside the town of Wana, in South Waziristan. The blast killed the leader's entire family, including three children, one of them five years old.”43 There is no evidence that this information was verified independently in Mayer's article or in O'Connell's favorable citation of it. In fact, both Mayer and O'Connell cite a New America Foundation database that claims the Wana strike killed no Al Qaeda or Taliban leaders but did kill eight other individuals. However, both the New America Foundation and Mayer failed to note a conflicting report in the Times of London. That report stated that “[s]even more died when hours later two missiles hit a house in Wana, in South Waziristan. Local officials said the target in Wana was a guest house owned by a pro-Taleban tribesman. One said that as well as three children, the tribesman's relatives were killed in the blast.”44 Based on these reports, there is a conflict over whether this house was the right house, specifically whether it was the home of a “pro-government tribal leader” or a “pro-Taleban tribesman.” Despite conflicting information from different sources, both authors only cite the evidence that supports their conclusion.

Critics also make bold and unsubstantiated claims about the accuracy of drones. For example, O'Connell writes that

one thing is clear, the use of drones in Pakistan has resulted in a large number of persons being killed along with the intended targets. Several factors suggest why this has been the case. One problem is structural. The remote pilot of a drone is relying on cameras and sensors to transmit the information he or she needs to decide on an attack. The technology is improving, but it is still difficult to be certain about targets.45

However, O'Connell fails to clarify how the reliance on cameras and sensors by drone operators is any different from the reliance on cameras and sensors by a helicopter or F-16 pilot (who will have far less time loitering over the target area compared to a drone). As Predator pilot Lt Colonel Matt J. Martin has noted:

We followed the same rules of engagement and used the same procedures as all other aircraft, manned or unmanned, that employed weapons in support of the fight on the ground. To us the Predator is a longer-duration, lightly armed (and much less (p.336) survivable) version of an F-16—with the benefits of persistence, global distribution of video and data, the ability to leverage the entire intelligence apparatus through ground communications links, and the ability to think clearly at zero knots and one G.46

If anything, the technology employed by drones is an argument in favor of their usage. As a recent article in the Economist points out:

Drones such as the Predator and the Reaper can loiter, maintaining what one former CIA director described as an ‘unblinking stare’ . . . Thanks to the drone's ability to watch and wait, its ‘pilot,’ often thousands of miles away, can patiently choose the best moment to fire its missiles, both increasing the chances of success and minimizing the harm to civilians.47

Moreover, even Mayer conceded the point that drones are probably more accurate than conventional aircraft when she wrote “Predator drones, with their superior surveillance abilities, have a better track record for accuracy than fighter jets, according to intelligence officials.”48 Despite this fact, O'Connell further claims that “[t]he operators never see with their own eyes the persons they have killed.”49 If by this she means that the operators are not standing above the bodies of their targets, looking with the naked eye, her claim is correct. However, this is also true of nearly every air campaign since the Second World War, every naval campaign that used cannons or naval guns aided by a telescope, and every shot fired by a rifle using a scope. As the Economist notes, “There are still plenty of human beings in the operational loop—it takes a team of about 180 to run and service a Predator—and it is clear that the responsibility for the decision to fire a missile rests as much with the pilot in a distant command centre as with a pilot in any cockpit.”50 O'Connell continues, “[i]n the trailer in Nevada, the pilot knows she will not be attacked. She will go home to her family at the end of the day, coach a soccer game, make dinner, and help with homework.” But these same facts could be used as an argument in favor of targeted killings as such a person will be more apt to exercise care.

In fact, it seems more likely that a person who is not dodging anti-aircraft fire and is not exposed to danger will be calm, careful, and deliberative. As Randall Hansen noted in his book about the Allied bombing of Germany in the Second World War, bomb aimers struggled to do their jobs when faced with the mortal fear of anti-aircraft flak;51 the use of remotely piloted aircraft removes that fear. Retired Lieutenant General David Deptula, who oversaw the U.S. Air Force's drone (p.337) program from 2006 to 2010, questions O'Connell's logic, asking, “Are these people arguing that . . . we should only fight if you are exposed to threats and putting your life at risk?”52 Deptula concludes, “That's silly, and I think it's ill-founded.”53 Edward Barrett, director of strategy and research at the U.S. Naval Academy's Stockdale Center for Ethical Leadership expands on this critique. Barrett points out, “A soldier in the situation is scared and possibly hasty in deciding what to do and acting—and possibly even angry, whereas an operator who's not threatened can use tighter rules of engagement and is not going to be fearful and therefore is going have a much cooler head.”54 These factors are especially relevant given the measures taken to enhance accountability in target acquisition, as well as in actually engaging the target. Furthermore, as O'Connell herself notes, “a—1000 people see the video [from the drone]—from—pilots in their trailers in Nevada and New Mexico, to intelligence analysts at Central Command (CENTCOM) headquarters in Florida, to persons in—Japan, to—POTUS (the President of the United States).”55 A pilot subject to that level of scrutiny certainly will be more careful and discriminating than a traditional pilot faced with imminent danger whose gun camera footage will only be reviewed after the fact.

Critics also frequently make arguments that display an utter lack of familiarity with the U.S. military's targeting and strike practices. For example, a recent Nation article defined “bug splat” as “the splotch of blood, bones, and viscera that marks the site of a successful drone strike. To those manning the consoles in Nevada, it signifies ‘suspected militants’ who have just been ‘neutralized’; to those on the ground, in most cases, it represents a family that has been shattered, a home destroyed.”56 In fact, the term “bug splat” does not refer to people, blood, or bones at all; rather, it refers to the predicted blast pattern made by explosives (based on simulations). Contrary to popular conception, bombs do not explode in a perfect circle but are flattened on one side, similar to the shape of a bug that hits a windshield. A “bug splat” refers to the shape of the planning tool used as an overlay to predict a collateral effects radius. This sounds highly technical, but one need not complete military training to know this fact; one only needs to Google the words “bug splat” and “drone” to find a Washington Post article explaining the term.57 (p.338)

The criticisms reveal other problems. O'Connell claims, for example, that “[s]uspected militant leaders wear civilian clothes. Even the sophisticated cameras of a drone cannot reveal with certainty that a suspect being targeted is not a civilian.” This straw man argument displays a lack of familiarity with the targeting cycle followed by the U.S. military.58 The cycle includes a minimum of 24 hours of “cycle of life surveillance” prior to an attack, and according to Mayer “the recent Senate Foreign Relations Committee report [notes that] the U.S. military places no name on its targeting list until there are ‘two verifiable human sources’ and ‘substantial additional evidence’ that the person is an enemy.”59 Of course, if the only information about a suspected militant in civilian clothes came from the camera of an attacking drone, that would be insufficient evidence for a strike. But that hypothetical does not reflect actual U.S. practice. In actual practice, U.S. forces must positively identify that the person they are targeting is who they believe they are. Moreover, since at least June 2008, no ISAF pilot (drone or otherwise) could fire a weapon at their own discretion, because there is no Fire Support Coordination Line (FSCL). What this means is that air power has no battlespace where they can engage targets without clearance from the battlespace owner (the ground commander).60 Thus all strikes in Afghanistan post-June 2008 must be either pre-planned or contemporaneously authorized by the ground commander and directed by a Joint Terminal Attack Controller (JTAC) (a trained officer on the ground directing aircraft fires); this practice is reflected in U.S. military doctrine61 and reflects the fact that in close air support missions the ultimate responsibility for engagements rests with the ground commander. Thus, the only discretion the pilot has in most circumstances is to not fire their weapon (for example, when they see something that the commander does not, such as the presence of civilians or other collateral concerns). This is the reality of air ground operations in Afghanistan, not the fanciful camera-driven, push-button, videogame-style warfare story that is told by O'Connell and Mayer.

Furthermore, it is inexplicable why critics would believe that the image from a drone camera is the only piece of information on which the military or CIA relies. Why would the United States ignore signals and human intelligence ranging from cell phone intercepts to on-the-ground informants? Just because O'Connell's hypothetical supports her conclusion does not mean it is grounded in reality, especially when the U.S. government has released detailed information about its targeting (p.339) cycle.62 She further errs when she writes that “[l]ittle information is available as to whether the U.S. takes any precautions when carrying out drone strikes.” In fact, prior to the publication of her draft chapter the U.S. released information about the precautions taken during drone strikes; for example on March 5, 2003 at 11:55 a.m. CENTCOM publicized a series of briefing slides entitled “Targeting and Collateral Damage.” These slides were presented at a Department of Defense News Briefing where military officials detailed a simplified version of the CDE process I explained in the beginning of this chapter. Moreover, since the publication of her draft chapter, the U.S. released substantial information in the al-Awlaki litigation—a case in which O'Connell herself filed an expert affidavit. In that case, the United States revealed extensive details about its collateral damage estimation and mitigation techniques.63 Although there might be an empirical debate about the true number of casualties caused by any particular drone strike, the targeting process used by the U.S. military is now a matter of public record.

Another questionable claim by O'Connell is that “the U.S. military is no longer training its members in the law of war as it once did.”64 Citing a “former drone commander,” O'Connell noted that he claims “he had never had a single day of training in the law of armed conflict (LOAC) in his 17 years of active duty.” On its face this claim should strike any reader familiar with the U.S. military as questionable. First, Air Force Instruction 51–401 mandates that:

All Air Force military and civilian personnel will comply with the LOAC in the conduct of military operations and related activities in armed conflict, regardless of how such conflicts are characterized. In support of this policy, the Air Force will conduct specialized training programs for military and civilian personnel designed to prevent LOAC violations.65

Furthermore, the Air Force's Air Education Training Command is required to “[d]evelop training plans, and procedures to instruct and train Air Force personnel at the start of their service on the content and requirements of the LOAC. The amount and content of any specialized training and instruction shall be commensurate with each individual's projected duties and responsibilities.”66 Thus, any “drone commander” who had responsibility for targeting operations would have to comply with Air Force doctrine on targeting;67 that doctrine spells out in an appendix the basic principles of the Law of Armed Conflict as it relates to (p.340) targeting.68 Moreover, other sources make clear that, “[a]t the most basic level, the Department of Defense (DoD) has institutionalized law of war training in the services through DoD Directive Number 5100.77 DoD Law of War Program.”69 Additional training protocols exist for the Air Force to ensure familiarity with law of armed conflict rules, including relevant Hague and Geneva conventions.70

Perhaps the anonymous “former commander” interviewed by O'Connell never received the training mandated by the Secretary of the Air Force, but in that unlikely case it is not a result of a policy of neglect on the part of the U.S. government, or as she claims a decision to “no longer train its members in the law of war as it once did.” Rather, there are clear requirements for continued officer training that mandate the very training that the critics claim is absent.

Furthermore, even if the “drone commander's” claim were true, it simply does not have the significance that critics assign to it. A “drone commander” is not in the operational chain of command for UAV operations and does not exercise control (p.341) over targeting decisions. Air assets are attached to the combatant commander, who is most certainly well versed in the law of war.71 For example, in Afghanistan, UAV pilots operate under the direction of CENTCOM, ISAF and subordinate commands; the commanders of those units are graduates of the U.S. Army's Command and General Staff College. Training for those graduates includes:

The 27 credit hour course, Case Studies in the Law of War, focuses on the application of the law of war during high-intensity conflict. The focus is on in-depth examination of the application of international human law through case studies. Specific topics include the principles of the law of war, targeting decisions, projected persons and places on the battlefield, prisoners of war, war crimes prosecution, genocide, the legality of weapons, and training on the law of war. In addition to the readings and case studies, a student briefing on the course subject matter is required. The other 27 credit hour course, Legal Issues in Contingency Operations, covers legal issues commanders and staff officers are likely to face in future contingency operations. A portion of the course specifically deals with cooperation with war crimes tribunals.72

O'Connell's claims about a lack of law of war training is false and her conclusion that this has somehow impacted targeted killing operations by drones is simply wrong.

In the same passage, O'Connell further notes, without citation, that an unnamed “currently serving Army lawyer reported he had had only three days of international law training during his specialized course at the Army Judge Advocate General (JAG) School in Charlottesville, Virginia in 2005.” I looked at the curriculum for the Army JAG School and, contrary to O'Connell's claims, I was able to confirm that the Basic Course requires eight days of instruction in Operational Law.73 Moreover, a JAG officer advising on operational law would likely not come straight from the Basic Course, but would instead have years of experience and education that would also include a supplemental advanced two-week course in (p.342) Operational Law. In addition to this training, all JAGs undergo training with their unit and pre-deployment training to further enhance their skills. Finally, one JAG officer would not be operating alone, especially if he or she were inexperienced; they would instead have an entire technical chain of more experienced JAGs to whom they could turn for guidance and verification of judgments.74 The claims of this unnamed “currently serving Army lawyer” are not a reflection of U.S. military practice, and even if they are, they are not as significant as O'Connell suggests. Finally, O'Connell also states in passing—and again without citation—that the “Naval War College may be trimming its law of armed conflict program.”75 This statement suggests that the Naval War College has lost interest in law of armed conflict training, a surprising point given that in the past two years the Naval War College sponsored major international law conferences focused on the law of armed conflict.76

Taken together, O'Connell uses a series of unsubstantiated suggestions to support her conclusion that “inadequate training may account for the high rate of unintended deaths” by drones.77 This conclusion rests on two false premises: first, that there is a high rate of unintended deaths, a controversial point at best; and second, that there is inadequate training, a point that lacks substantial foundation as the preceding discussion illustrates. The fact that O'Connell fails to substantiate her claims with adequate citations and fails to consider that those claims could be easily refuted, undermines the force of her provocative argument that the United States is engaged in “unlawful killing by Predator drone.”

II. Claims about drones and the targeted killing policy itself

Beyond claims about the process of targeted killing itself, critics also make unsupportable geo-political claims. For example, O'Connell writes (without a supporting citation) that “the U.S. needs Afghanistan's consent to carry out [drone] raids from Afghan territory. Afghanistan has not, apparently, given this consent.” How can she know whether Afghanistan has given its consent? In fact, O'Connell's (p.343) conclusions run counter to that of Amrullah Saleh, head of Afghan intelligence from 2004 to 2010, a position under the direct control of President Karzai. In a Frontline documentary, “Fighting for bin Laden,” Saleh admits to supplying the CIA with on-the-ground intelligence to carry out drone strikes.78 Further, O'Connell maintains that the United States and Pakistan disagree as a matter of policy as to what steps Pakistan should take, and because Pakistan has not given consent to the United States to use drones in Pakistani territory she writes that “The U.S. may not disregard (Pakistani) sovereignty to carry out its own police actions.”79 She also writes that “Pakistan has not expressly invited the United States to assist it in using force,”80 and that “Pakistan has neither requested U.S. assistance in the form of drone attacks nor expressly consented to them.”81 Is she privy to private diplomatic communications between the United States and Pakistan? If such strident claims about Pakistani sovereignty are to be believed, critics ought to provide evidence that Pakistan has not consented to U.S. operations.

In fact, the body of evidence available to the public runs contrary to the notion that the United States is violating Pakistani sovereignty. It has been widely documented that there has always been at least tacit Pakistani approval for the drone program. A Reuters article provides an American diplomat's account of a meeting in Pakistan, which illustrates such approval.82 The diplomat describes a meeting in which Pakistani parliamentarians repeatedly condemned the drone program. “Then, in the middle of the session . . . one of the parliamentarians slipped the American guest, who specializes in the region, a handwritten note.” The note read, “The people in the tribal areas support the drones. They cause very little collateral damage. But we cannot say so publicly for reasons you understand.” The Reuters article also cites several tribal elders from the regions most frequently targeted by the drones.83 One tribal elder commented, “[civilians] want to get rid of the Taliban and if the (Pakistani) army cannot do it now, then it (drone attacks) is fine with them.”84

It is also widely understood that Pakistani military and intelligence services provide essential assistance to the drone program. The same Reuters article also quotes Ikram Sehgal, a Pakistani security expert. Sehgal agrees that, “the intelligence underpinning the drone strikes has improved precisely because of increased Pakistani cooperation.”85 Likewise, terrorism expert Peter Bergen notes how U.S. and Pakistani strategic interests are now more aligned than they were at the outset of operations in Pakistan and that proof of this can be seen in the fact that the (p.344) “United States is [operating] the drones with Pakistani government permission and also Pakistani government intelligence.”86 In fact, easily verifiable facts, such as the existence of U.S. air bases in Pakistan since 2006, undercut O'Connell's claims. For example, The Times reported in February 2009 that “[t]he US was secretly flying unmanned drones from the Shamsi airbase in Pakistan's southwestern province of Baluchistan as early as 2006, according to an image of the base from Google Earth.”87 The article further noted that “Major-General Athar Abbas, Pakistan's chief military spokesman . . . admitted on Tuesday that US forces were using Shamsi, but only for logistics. He also said that the Americans were using another air base in the city of Jacobabad for logistics and military operations.”88 Beyond this article, there are other sources that undermine O'Connell's sovereignty claim, albeit after the date of her most recently updated draft. For example, diplomatic cables published by Wikileaks reveal that:

At the same time the Pakistani public was decrying the CIA's use of drone strikes in their country, Pakistan's top army general was asking a top U.S. official in behind-the-scenes meetings for more drones to help during military operations . . . Referring to the situation in Waziristan, [Pakistani General Ashfaq] Kayani asked if [U.S. Admiral William] Fallon could assist in providing continuous Predator coverage of the conflict area. Fallon regretted that he did not have the assets to support this request.89

Furthermore, as Islamic History Professor Brian Glyn Williams has noted, members of the Pakistani military have made statements lending support to the notion that the United States and Pakistan are working closely on drone strikes. Specifically, “the Pakistani general in command of forces in the embattled North Waziristan tribal agency told reporters that ‘a majority of those eliminated [in drone strikes] are terrorists, including foreign terrorist elements.’ Until this statement, the Pakistani military and government had not confronted the perception created by Pakistani media and anti-U.S. politicians that U.S. drones target, almost exclusively, civilians.”90 Williams concludes that:

There is, however, little precedent for a general of Mehmood's rank speaking out on such a sensitive topic without the approval of his superiors. To do so would be a grave breach of military decorum, if not a breaking of direct orders, and would certainly lead to the end of an offending officer's career. The fact that no one in the Pakistani military or government has rejected Mehmood's statements is indicative. Clearly, there are voices in the Pakistani military who support the drone strikes against an (p.345) enemy that many in Pakistan's military establishment have come to see as the greatest threat to the Pakistani state . . . Those making public statements on the drones, both in Pakistan and in the West, must now take General Mehmood's on-the-ground perspective about the effectiveness of the drone's targeting into consideration.91

Beyond law, O'Connell also questions the strategy of targeted killing. She notes that Baitullah Mehsud, who was killed in a drone strike, was quickly replaced by other leaders. To bolster her criticism she cites David Kilcullen, who believes that “going after leaders with drones is proving counterproductive of peace and stability in Pakistan.”92 However, such a claim fails to recognize that other counterterrorism experts believe that a policy of targeted killing, even if it results in leaders being replaced, still imposes costs on enemies. For example, Boaz Ganor, writing about the effectiveness of targeted killing, notes that:

. . . when the goal is to deter an organization from terrorist activity and to obstruct its actions, it would appear that this goal can be achieved by hurting a senior terrorist in that organization, which will then embark on a ‘power struggle’ among those eager to fulfill his role. The difficulty in finding another leader with professional skills, charisma or other positive characteristics to fill the position left by the dead activist could interfere with the organization's activities. Disrupting the organization's routine is liable to have ongoing consequences, rather than merely a short-term effect. The organization might then have to invest considerable resources—financial resources, manpower, and time—in defense and ongoing protection for its senior officials. In certain cases, the organization's senior activists take preventive steps and long-term security measures, and may even adopt new behavior because of their fear of personal attack.93

This imposition of costs—including the fear of attack—was proven recently when it was revealed that Osama bin Laden had to rely on couriers to supervise Al Qaeda operations and situated himself far from a personal leadership role out of fear that he might be targeted by U.S. air power.94 Many officials believe this phenomenon also extends to significant Taliban leaders captured in large urban centers, such as Mullah Abdul Ghani Baradar.95 When the infamous Taliban commander, best known for rapidly increasing the use of improvised explosive devices (IEDs) in Afghanistan, was captured by Pakistani intelligence officers in Karachi, Baradar became yet another example of top Taliban and Al Qaeda leaders fleeing the battlefield for a more densely populated city center out of fear of being targeted by a drone strike. Michael Scheuer, the former head of the CIA's bin Laden unit, summarized (p.346) Baradar's capture, stating, “The more violence you inflict, the more intelligence you get because people who are being shot at make mistakes.”96

Taken together, the claims of critics regarding the wisdom, efficacy and legality of targeted killing operations that have been surveyed here simply cannot be believed because they are riddled with inaccuracies or amount to nothing more than easily refuted empirical claims. In fact, in many instances critics reveal a fundamental lack of familiarity with basic aspects of military operations, doctrine, or training. Moreover, their geopolitical claims fail to take account of a wealth of contrary information contained in the public record. Such errors render their criticism unreliable at best.

III. Conclusion

This chapter has highlighted the lack of empirical foundation for many of the most common criticisms raised by targeted killing opponents. This chapter demonstrated how many claims made by critics are easily refuted, which calls into question the reliability of their criticism. Evidence in the public record and research I conducted indicates that the U.S. military follows a rigorous collateral damage estimation and mitigation process designed to minimize civilian casualties. The U.S. military takes unparalleled steps to eliminate civilian casualties. While evidence of civilian casualties is hard to come by, it is certainly the case that statistics proffered by some critics cannot be empirically verified; their skepticism of U.S. government statements is not backed up by anything more substantial than generic suspicion. Moreover, critics of targeted killing also lack proof for their claims about the legal, diplomatic and strategic results of drone strikes. Although the counter-observations raised in this chapter do not, by themselves, demonstrate that targeted killings are morally or legally justified, they do suggest that some of the moral or legal objections to targeted killings are based on empirical claims that are either dubious, impossible to verify or just plain false. This chapter provides a necessary corrective to some of the most strident claims raised by critics.


(1) See David Kilcullen and Andrew Exum, “Death from Above, Outrage Down Below,” New York Times, May 16, 2009.

(2) Emphasis added, Chris Woods, “U.S. Claims of ‘No Civilian Deaths’ are Untrue,” The Bureau of Investigative Journalism, July 18, 2011, Covert Drone War; available at 〈http://www.thebureauinvestigates.com/2011/07/18/washingtons-untrue-claims-no-civilian-deaths-in-pakistan-drone-strikes/〉 accessed November 4, 2011.

(3) C. Christine Fair, “Drone Wars,” Foreign Policy, May 28, 2010; available at 〈http://www.foreignpolicy.com/articles/2010/05/28/drone_wars〉 accessed November 4, 2011.

(4) Ibid.

(5) Woods, supra n. 2.

(6) Ibid.

(7) Pam Benson and Elise Labott, “U.S. Denies Report Alleging Drone Strikes Kill 160 Kids in Pakistan,” CNN World, August 12, 2011, available at 〈http://articles.cnn.com/2011-08-12/world/pakistan.us.drone.strikes_1_drone-strikes-drone-campaign-drone-program/3?_s=PM:WORLD〉 accessed November 4, 2011.

(8) Gregory S. McNeal, “Collateral Damage and Accountability,” Pepperdine University School of Law Working Paper (April 22, 2011), available at 〈http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1819583〉 accessed November 4, 2011 (hereafter Collateral Damage and Accountability).

(9) Joe R. Feagin et al., A Case for the Case Study (University of North Carolina Press, 1991); Robert K. Yin, Case Study Research: Design and Methods 2, 2nd edn (Sage, 1994).

(10) Yin, supra n. 9, 91–3.

(11) Laurie Blank and Amos Guiora, “Teaching an Old Dog New Tricks: Operationalizing the Law of Armed Conflict in New Warfare,” 1 Harv. Nat'l Sec. J. 45, 56 (2010).

(12) Hereafter CDM, the methodology or methodology.

(13) McNeal, supra n. 8, 4.

(14) Ibid.

(16) Ibid.

(17) Ibid.

(18) Ibid., citing Declaration of Jonathan Manes, The Joint Targeting Definitions and Process; see also Nasser Al-Aulaqi vs. Obama, 727 F. Supp. 2d 1, 46–7, 51–2 (D.D.C. 2010) (the person responsible for making this decision is prescribed by the Rules of Engagement).

(19) McNeal, supra n. 8, 19, citing “The Dissident Voice,” U.S. Iraq Rules of Engagement Leaked, available at 〈http:// dissidentvoice.org/2008/02/us-iraq-rules-of-engagement-leaked-raises-question-about-rumsfeld-authorizing-war-crimes/〉 (linking to MNFI Rules of Engagement published by WikiLeaks and quoting the relevant sections of the ROE regarding High Collateral Damage targets).

(20) McNeal, supra n. 8, 18. (The Rules of Engagement in Iraq made complex and significant distinctions between types of targets and approval authority. These distinctions ranged from facilities with significant cultural or political value, to individuals such as former regime members to members of specified terrorist groups.)

(21) Ibid. at 19, noting this will usually be an officer holding the rank of General (4-star).

(22) Ibid. at 19, noting this will usually be an officer holding the rank of Major General (2-star). The Rules of Engagement may allow for delegation of this authority, depending on the weapons used and circumstances.

(23) Ibid. at 19. See also, Pamela Constable, “NATO Hopes to Undercut Taliban With Surge of Projects,” Washington Post, September 27, 2008, A12 (quoting Brig. Gen. Richard Blanchette, chief spokesman for NATO forces stating “—[i]f there is the likelihood of even one civilian casualty, [NATO] will not strike, not even if we think Osama bin Laden is down there.” This quote may be an overstatement; it is reasonable to conclude that a more accurate statement is that NATO will not strike until proportionality balancing and a decision is made by the President or the Secretary of Defense).

(24) McNeal, supra n. 8, 19. (Notably this authority may be delegated. Depending on the ROE.)

(25) Ibid. (This acceptability threshold is established by the President of the United States or the Secretary of Defense.)

(26) Ibid.; Declaration of Jonathan Manes, supra n. 18, 38.

(27) Ibid.

(28) See McNeal, supra n. 8. See also, “The Dissident Voice”, supra n. 19.

(29) See McNeal, “Collateral Damage and Accountability,” supra n. 8. The Rules of Engagement in Iraq made complex and significant distinctions between types of targets and approval authority. These distinctions ranged from facilities with significant cultural or political value to individuals such as former regime members to members of specified terrorist groups.

(30) Usually an officer holding the rank of General (4-star).

(31) Usually an officer holding the rank of Major General (2-star). The Rules of Engagement may allow for delegation of this authority, depending on the weapons used and circumstances.

(32) See n. 23. For a lengthier discussion of these concepts, see Gregory S. McNeal, Collateral Damage and Accountability.

(33) See Declaration of Jonathan Manes, supra n. 18. Reports indicate that this authority has been delegated to the commander of US and ISAF forces (previously General Petraeus, currently General Allen), delegation confirmed in Interview 1 and Interview 3.

(34) Adam Entous, “Special Report: How the White House Learned to Love the Drone,” May 18, 2010, Thomson Reuters, available at 〈http://www.reuters.com/article/2010/05/18/us-pakistan-drones-idUSTRE64H5SL20100518〉 accessed November 4, 2011.

(35) See Brian Glyn Williams, “Accuracy of U.S. Drone Campaign: The Views of a Pakistani General,” CTC Sentinel, March 2011, Vol. 4, Issue 3, 9.

(36) See Matt J. Martin and Charles W. Sasser, Predator: The Remote-Control Air War Over Iraq and Afghanistan: A Pilot's Story (Zenith Press, 2010) 136. (Martin, a Predator pilot, noted that U.S. Army lawyers would arrive in post-strike areas “with suitcases full of cash to award people with damage claims.” He further states, “Most of the people who collected were righteous bystanders. Others who showed up to collect cash were undoubtedly the same people we were trying to kill.”) See also, “Drone Attacks: Altaf Demands Blood Money For Victims,” The Express Tribune, March 18, 2011, available at 〈http://tribune.com.pk/story/134294/drone-attacks-altaf-demands-blood-money-for-victims/〉 accessed November 4, 2011.

(37) Simon Bronitt (ed.), Shooting To Kill: The Law Governing Lethal Force In Context, forthcoming, available at 〈http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1501144〉; David Kilcullen and Andrew Exum, “Death from Above, Outrage Down Below,” New York Times, May 16, 2009 (O'Connell principally relies on this source to assert that “the U.S. was killing 50 unintended targets for each intended target”); Hearing of the House Armed Services Committee, Effective Counterinsurgency: the Future of the U.S. Pakistan Military Partnership, April 23, 2009; available at (Testimony of David Kilcullen) (claiming that “Since 2006, [the U.S.] has killed 14 senior Al Qaeda leaders using drone strikes. In the same time period [the U.S.] has killed 700 Pakistani civilians in the same area”); Peter Bergen and Katherine Tiedemann, “Revenge of the Drones,” New America Foundation, October 19, 2009, available at 〈http://newamerica.net/publications/policy/revenge_of_the_drones〉 (claiming that this article supports the ratios provided by Kilcullen and Exum, noting that Bergen and Tiedemann state, “Since 2006, our analysis indicates 82 U.S. drone attacks in Pakistan have killed between 750 and 1000 people. Among them were about 20 leaders of Al Qaeda, the Taliban, and allied groups, all of whom have been killed since January 2008.” O'Connell claims that this article supports the figures provided by Kilcullen and Exum; however, she downplays the fact that Bergen and Tiedemann only distinguish Al Qaeda and Taliban leaders from the remaining number of people that have been killed by drones. In fact, Bergen and Tiedemann even criticize Kilcullen and Exum's figures, stating, “in [Kilcullen and Exum's] analysis, 98 percent of those killed in drone attacks were civilians . . . Our analysis suggests quite different conclusions than those of . . . Kilcullen and Exum.”).

(38) O'Connell, supra n. 37, 10.

(39) O'Connell, supra n. 37, 6. (Curiously, she does not explain what “computer data” means.)

(41) Jane Mayer, “The Predator War, What are the Risks of the C.I.A.’s Covert Drone Program?,” The New Yorker, October 26, 2009, 36, 37, available at 〈http://www.newyorker.com/reporting/2009/10/26/091026fa_fact_mayer〉 accessed November 4, 2011.

(42) O'Connell, supra n. 37, 7.

(43) Mayer, supra n. 41.

(44) Ibid. (emphasis added)

(45) O'Connell, supra n. 37, 7.

(46) Martin and Sasser, supra n. 36, 104.

(47) “Drones and the man: Although It Raises Difficult Questions, The Use of Drones Does Not Contravene the Rules of War,” The Economist Online, July 30, 2011, available at 〈http://www.economist.com/node/21524876〉 accessed November 4, 2011.

(48) Mayer, supra n. 41.

(49) O'Connell, supra n. 37, 8.

(50) “Drones and the Man,” supra n. 47.

(51) Randall Hansen, Fire and Fury: The Allied Bombing of Germany 1942–1945 (Random House, 2008) 114.

(52) “The Ethics of Drones,” August 26, 2011, PBS, Religion & Ethics Newsweekly, available at 〈http://www.pbs.org/wnet/religionandethics/episodes/august-26–2011/ethics-of-drones/9350/〉 accessed November 4, 2011.

(53) Ibid.

(54) Ibid.

(55) O'Connell, supra n. 37, 6.

(56) Muhammad Idrees Ahmad, “Fighting Back Against the CIA Drone War,” The Nation, July 31, 2011, available at 〈http://www.nation.com.pk/pakistan-news-newspaper-daily-english-online/International/31-Jul-2011/Fighting-back-against-the-CIA-drone-war〉 accessed November 4, 2011.

(57) Walter Pincus, “Are Drones a Technological Tipping Point?” Washington Post, April 24, 2011, available at 〈http://www.washingtonpost.com/world/are-predator-drones-a-technological-tipping-point-in-warfare/2011/04/19/AFmC6PdE_story.html〉 accessed November 4, 2011.

(58) Steven A. Emerson, “In Defense of Drones,” The Invesitgative Project on Terrorism, October 17, 2010, available at 〈http://www.investigativeproject.org/2249/in-defense-of-drones〉 accessed November 4, 2011.

(59) Mayer, supra n. 41.

(60) Sarah Sewall and Larry Lewis, “Reducing and Mitigating Civilian Casualties: Afghanistan and Beyond: Joint Civilian Casualty Final Report,” 31 August 2010, 51 fn.110 (hereafter CIVCAS Report).

(61) Joint Publication 3-09.3, Close Air Support, June 2009.

(62) Ibid.

(63) See Declaration of Jonathan Manes, supra n. 18.

(64) O'Connell, supra n. 37, 7.

(65) U.S. Air Force, available at 〈http://www.af.mil/shared/media/epubs/AFI51-401.pdf〉 accessed November 4, 2011.

(66) Ibid.

(67) Specifically Air Force Doctrine Document 2-1.9 Targeting and Joint Publication 3-60 Joint Doctrine for Targeting.

(68) Excerpts from the 2-1.9 include passages such as “Targeting must adhere to the LOAC and all applicable ROE. It is the policy of the DOD that the Armed Forces of the United States will comply with the law of war during all armed conflicts, however such conflicts are characterized, and, unless otherwise directed by competent authorities, the US Armed Forces will comply with the principles and spirit of the law of war during all other operations. The ‘law of war’ is a term encompassing all international law for the conduct of hostilities binding on the United States including treaties and international agreements to which the United States is a party, and applicable customary international law. The ‘law of war’ is also commonly referred to as the LOAC.” (88) and “LOAC rests on four fundamental principles that are inherent to all targeting decisions: military necessity, unnecessary suffering, proportionality, and distinction (discrimination).” (88).

(69) Major Jerry Swift, “The Teaching of Morality in Warfighting in Today's Officer Corps,” 8, April 10, 2001, available at 〈http://www.au.af.mil/au/awc/awcgate/acsc/01-208.pdf〉 accessed November 4, 2011. (“Each service implements DoD 5100.77 through compliance with their individual service instructions on law of armed conflict. Air Force guidance is provided in Air Force Policy Directive (AFPD) 51–4, Compliance with the Law of Armed Conflict, and Instruction (AFI) 51–401, Training and Reporting to Ensure Compliance with the Law of Armed Conflict. Army guidance is through Field Manual (FM) 27–10, The Law of Land Warfare. Marine Corps implement MRCP 5-12.1A, The Law of Land Warfare. Navy guidance is provided through OPNAVINST 3300.52, Law of Armed Conflict (Law of War) Program to Ensure Compliance by the U.S. Navy and Naval Reserve and through Naval Warfare Publication (NWP) 1-14M, The Commander's Handbook on the Law of Naval Operations. Implementation is accomplished by assigning functional compliance, formalizing instruction and training, and institutionalizing reporting and investigations of violations of LOAC.” Focusing strictly on training, each Secretary of the Military Departments must ensure that “the principles and rules of the law of war will be known to members of their respective Departments, the extent of such knowledge to be commensurate with each individual's duties and responsibilities.”).

(70) Ibid. (“AFPD 51-4 ensures that ‘once each year, all commanders make sure their people are trained in the principles and rules of LOAC needed to carry out their duties and responsibilities.’ (AFPD 51-4, pg. 1.) It states that as a minimum, instruction is to include ‘training required by the 1949 Geneva Conventions for the Protection of War Victims and the Hague Convention IV of 1907, including annexes.’ (AFPD 51-4, pg. 1.) Furthermore, AFI 51–401 assigns Air University the additional responsibility to ‘include instruction on LOAC in Air War College, Air Command and Staff College, Squadron Officer School, Reserve Officer Training Corps, and Senior Noncommissioned Officer Academy curricula to ensure adequate knowledge of the subject commensurate with the nature of each enrollee's duties and responsibilities.’ (AFI 51–401, pg. 4.”)

(71) Martin and Sasser, supra n. 36, 92 (describing how “Ops officers, pilots, the MCC, and the sensor operators engaged in a spirited discussion over the best method for striking to destroy [a target]. Of course, the final decision over what action to take wasn't ours to make. The ground commander had the last word.”).

(73) See Maj Sean Condron (ed.), Operational Law Handbook (2011), International and Operational Law Department: The Judge Advocate General's Legal Center & School, U.S. Army. (The handbook summarizes the law of war “provisions for military personnel and commanders in the conduct of operations in both international and non-international armed conflicts”; citing DoDD 2311.01E, para. 3.1, the law of war is defined as “that part of international law that regulates the conduct of armed hostilities.” Section IV defines the purposes for the law of war as both “humanitarian and functional in nature.” Humanitarian purposes include: (1) protecting both combatants and noncombatants from unnecessary suffering; (2) safeguarding persons who fall into the hands of the enemy; and (3) facilitating the restoration of peace. Functional purposes include: (1) ensuring good order and discipline; (2) fighting in a disciplined manner consistent with national values; and (3) maintaining domestic and international public support. Section VI (A) states the Law of War “applies to all cases of declared war or any other armed conflicts that arise between the United States and other nations, even if the state of war is not recognized by one of them.”)

(74) All of these claims are substantied by the DoD Law Of War Program policy memo cited above. Moreover, The U.S. Institute of Peace in their study of Law of War Training (〈http://www.usip.org/files/resources/LawofWartext-final.pdf〉), accessed November 4, 2011, notes that such training is an annual requirement for all U.S. military personnel, and is required before deployment. See also the ICRC's report on U.S. compliance with Law of War training at 〈http://www.icrc.org/customary-ihl/eng/docs/v2_cou_us_rule142〉 accessed November 4, 2011.

(75) O'Connell, supra n. 37, 8.

(76) See, eg, specifically “Non International Armed Conflict,” International Law Conference, 2011, available at 〈http://www.usnwc.edu/Events/International-Law-Conference-2011.aspx〉 accessed November 4, 2011; “International Law and the Changing Character of Law” in 2010, available at 〈http://www.usnwc.edu/Events/International-Law-Conference-2010.aspx〉 accessed November 4, 2011.

(77) O'Connnell, supra n. 37, 8.

(78) Frontline, “Fighting for bin Laden” (2011), WGBH/Frontline.

(79) O'Connell, supra n. 37, 21, fn. 88.

(82) Entous, supra n. 34.

(83) Ibid.

(84) Ibid.

(85) Ibid.

(86) National Geographic Channel, “CIA Confidential: Inside the Drone War” (2011).

(87) Jeremy Page, “Google Earth Reveals Secret History of US Base in Pakistan,” The Times, February 19, 2009.

(88) Ibid.

(89) Jim Sciutto, “Wikileaks: Pakistan Asked for More, Not Fewer Drones,” ABC News, May 20, 2011, available at 〈http://abcnews.go.com/Blotter/wikileaks-cable-pakistan-asked-fewer-drones/story?id=13647893〉 accessed November 4, 2011.

(90) Human Security Report Project, “Accuracy of the U.S. Drone Campaign: The Views of a Pakistani General,” CTC Sentinel, March 2011, Vol. 4, Issue 3, 9.

(92) O'Connell, supra n. 37, 11.

(93) Boaz Ganor, The Counter-Terrorism Puzzle: A Guide for Decision Makers (New Brunswick, N.J.: Transaction, 2005) 128.

(94) See Gregory S. McNeal, “The bin Laden Aftermath: Why Obama Chose SEALS Not Drones,” Foreign Policy, May 5, 2011.

(95) Entous, “Special Report,” supra, n. 34.

(96) Ibid.