The legal and ethical implications of drone warfare

Abstract

This article examines whether American drone-based targeted killing program represents a fundamentally new challenge to the traditional legal and ethical standards of armed conflict. It argues that the novelty of drones flows less from the technology itself than from the Obama administration's articulation of a presumptive right of anticipatory self-defense, which allows it to strike anywhere in the world where al Qaeda and its allies are present. It highlights five new legal and ethical dimensions to the Obama administration's drones policy, all of which may lower the traditional barriers to the use of force if other actors begin to follow contemporary American practice.

Keywords:

• drones
• law
• ethics
• international conflict, USA

In a speech at the National Defense University in May 2013, President Barack Obama described the US drone campaign in Pakistan, Yemen and elsewhere as part of ‘a just war – a war waged proportionally, in last resort, and in self-defense’.¹ Emphasising the dilemmas associated with confronting the threat of modern terrorism, President Obama cast the American drones programme as a legally and morally superior alternative to using Special Forces or ground troops to capture or kill terrorist suspects in places where the United States (US) was not formally at war. This address, along with an earlier speech by Obama's chief counterterrorism advisor John Brennan in April 2012, showed the Obama administration on the defensive against a rising tide of criticism about the civilian casualties associated with American drone strikes.² In describing its response to what it called a new, de-territorialised form of armed conflict against terrorist groups, the Obama administration fell back upon long-established concepts of the law of armed conflict and just war theory to argue that its drone strike campaign was legally appropriate and ethically sound. Yet beyond the rhetoric, its actual position on drones has long been marked by ambiguities and internal contradictions. At some points, the administration has been insistent that it is fully complying with all applicable law, including the laws of war, in its drone strike policies.³ At others, it has denied that some bodies of law – for example, international human rights law – should be applied to these conflicts.⁴ At still others, it has implied that the law of armed conflict, and traditional notions of self-defence, need to be updated to reflect the realities of twenty-first century warfare.⁵ Throughout its term in office, the Obama administration has tried to use the traditional legal and moral standards of armed conflict as a shield to protect itself from criticism, while simultaneously suggesting that the shield itself is fundamentally inadequate for the threats that the US and its allies face in the twenty-first century.

The fact that the Obama administration has continued to rely on the same legal and ethical frameworks used to restrain the behaviour of states during wars in the nineteenth and twentieth centuries to defend its drone campaign against a range of critics is not surprising.⁶ These standards have proven invaluable in checking the cruelty and moral callousness of armies and government states for over a hundred years; indeed, they have proven resilient in light of the vast technological changes in warfare that have occurred since the first Geneva Convention in 1864. Although they do pose some new legal and ethical dilemmas, drones do not fundamentally undermine the applicability of traditional legal and ethical standards of armed conflict.⁷ Drones are merely the latest iteration of a process of rapid technological change in warfare that has continued for much of the last one hundred years, and they are not the biggest technological change during this period. The challenges posed by drones are not larger than those posed by nuclear weapons or other twentieth-century innovations such as inter-continental ballistic missile systems (ICBM). As Stephanie Carvin argues in her contribution here, there is no reason to assume that the legal and ethical problems posed by drones are wholly unprecedented or unique to this form of technology.⁸ Some of the literature on drones makes this mistake by assuming that the new aspects of drone technology mean that unprecedented legal and ethical questions are now being raised. In fact, the opposite is true. Even more than nuclear weapons and or ICBMs, drones can be discussed within the traditional legal and ethical frameworks governing armed conflict because they are so analogous to conventional or manned aircraft. For many of the relevant ethical questions, such as how many were killed and why, drone strikes can be measured in a straightforward way against bombing done by conventional manned aircraft. In most respects, there is not much of a difference between a bomb dropped by a piloted aircraft and one dropped by a drone provided that both are done by declared combatants in a recognised armed conflict.⁹

Further, some of the charges that are levelled against drones on the basis of the technology itself – for example, that drones automate killing, or that they produce a moral distance between the operator and the target that facilitates gratuitous killing – are either overstated or unproven. For example, while it is true that fully autonomous weapons would produce a new set of legal and ethical challenges for warfare, this is not the situation that prevails today.¹⁰ In all major or well-established drones programmes, the technology remains under the firm control of an individual from the military or intelligence services of an established government.¹¹ No modern drones programme is fully automated, and in all cases it remains the decision of a single individual – often located in a military chain of command – to strike a target and be held accountable for the consequences. The allegation that there is a ‘Playstation mentality’ to targeted killings by drones remains to be proven.¹² There is some evidence that the moral distance between the drone operator and target desensitises the operator from the act of killing and may make targeted killings easier.¹³ Some drone operators’ use of derogatory terms for their targets – ‘bug splats’ and ‘squirters’ – points to this dynamic at play.¹⁴ But equally there is evidence that drone operators feel a surprising degree of intimacy with their targets because they monitor them for such long periods of time. Drone operators report relatively high rates of post-traumatic stress disorder (PTSD) in part because they are so acquainted with their target.¹⁵ Further, many of these charges of moral distance are not, strictly speaking, new as they were levelled against automated weapons systems such as cruise missiles and nuclear weapons during the Cold War. The possibility of an individual callously ending a life by pressing a button is as real with cruise missiles as it is with a drone, and even more real with the prospect of a nuclear strike.¹⁶ None of these forms of lethal violence preserve the direct confrontation between perpetrator and target that characterises most depictions of honour in warfare, yet all have been a reality of modern warfare for over half a century.¹⁷

Still, there may be something new about drone warfare that requires rethinking some of the rules of armed conflict in the twentieth century. While the technology itself may not present wholly new moral and legal dilemmas, it may exaggerate some of them and accelerate other trends – towards speed, precision, and targeting those who blur the distinction between combatants and civilians – present in other forms of armed conflict. If so, while it is certainly not necessary to rewrite the rulebook on armed conflict, it may be necessary to adapt it in ways that reflect the reality of the legal and ethical dilemmas that drones pose, either due to the technology itself or the way that the technology is used. This is the challenge that has yet to be fully met despite a decade of drone-based targeted killing. Although the Obama administration has implied that the rules of armed conflict need to change to meet the new threat of terrorism, and has stated on multiple occasions that the domestic legal authority for drone strikes (the Authorization to Use Military Force, passed in 2001) needs to be revised, it has shied away from explicitly stating what is new about the drone technology and why it requires changes in the rules of war for the twenty-first century. It has left this debate open for scholars and outside experts to identify what is new about drones and to spell out the legal and ethical implications of this technology.

The articles in this special edition identify and analyse the new legal and ethical implications of drones and illustrate how they relate to the existing legal and ethical frameworks that govern armed conflict. This introduction and the article by Stephanie Carvin wrestle with the key conceptual question about what is new with drones: the technology, or the policies surrounding the use of them? The second set of articles, by Craig Martin and by Daniel Brunstetter and Arturo Jimenez Bacardi explain why the legal and ethical debate over drone-based targeted killing remain so contested after nearly a decade of operation. The third set of articles, by David Whetham and by Caroline Kennedy and James Rodgers, look at some of the legal and ethical issues surrounding the next generation uses of drones for humanitarian relief and United Nations (UN) peacekeeping. Taken together, these articles show how American policies surrounding drone usage have themselves produced the dilemmas that the Obama administration now confronts while charting a way in which drone usage might be reconciled with the traditional legal and ethical frameworks in international politics.

This introductory article will address the central conceptual question about what is new with drone warfare, with a focus more on the new dimensions of the policy of targeted killing than the technology itself. Echoing some of Craig Martin's arguments here, it argues that the Obama administration is correct that drones have begun to challenge some of the assumptions that underlie the traditional legal and ethical frameworks for the use of force, but that it is American policy, rather than the technology itself, which is producing this challenge. The US drone-based targeted killing programme has five different characteristics which produce new challenges to the traditional interpretations of the legal and ethical standards for armed conflict. They are: (1) the legal authority for drone strikes under both domestic and international law; (2) the nature of the targets that the US engages with drone strikes – specifically non-state actors operating outside recognised conflict zones, including those with an indirect link to the original combatants; (3) the regular deployment of drones for near-perpetual surveillance and targeted killing, as opposed to periodic or time-limited kinetic action; (4) the institutional framework governing drone strikes, which can give rise to transparency and accountability problems; and (5) the precedent set by lowering the bar against assassinations and by allowing a state a presumptive right to violate the sovereignty of another state on the basis of a generally articulated threat. Each of these issues flow more from the American use of drones – and the precedent of presumptive self-defence that it sets – than from the technology itself.

Legal authority

The Obama administration has deployed a number of different arguments to provide a legal foundation for its drone programme under domestic and international law. The primary claim of legal authority is based on the Authorization to Use Military Force (AUMF), passed on 14 September 2001. The Congress provided this authorisation following the 9/11 attacks to permit the president to ‘to use all necessary and appropriate force against those nations, organizations, or persons he determines planned, authorized, committed, or aided the terrorist attacks that occurred on September 11, 2001, or harbored such organizations or persons, in order to prevent any future acts of international terrorism against the United States by such nations, organizations or persons’.¹⁸ The Bush administration interpreted the AUMF broadly as a legal rationale for worldwide operations against terrorist operatives. It also unsuccessfully cited it as legal authority for employing military tribunals at Guantanamo Bay in Hamdi v. Rumsfeld.¹⁹

The position of the Obama administration on the AUMF has been marked by ambivalence. Like his predecessor, President Obama has interpreted the AUMF broadly to allow for maximum discretion in pursuing terrorists abroad, including through the use of drone strikes. Even more than the Bush administration's interpretation of the AUMF, the Obama administration's interpretation has allowed for the US to pursue ‘associates’ of al-Qaeda wherever they are located, even if that is far from the original theater of war in Afghanistan. The Obama administration has recently extended this argument to suggest that it may pursue successor groups to al-Qaeda on the basis of the AUMF even if they did not exist at the point of the 11 September attacks. For example, under pressure from Congress, the Obama administration argued that it had the right to target the al-Nusra front, an al-Qaeda linked insurgent group located in Syria but operating only since 2012, on the basis of the 2001 AUMF.²⁰ Similarly, the Obama administration used the AUMF as the legal basis for aiding the Libyan rebels in the overthrow of Muammar Gaddafi and for engaging in air strikes against the Islamic State of Iraq and Syria (ISIS).²¹ Both of these objectives are not directly related to al-Qaeda or to the 11 September attacks, yet drone strikes for these purposes have been conducted on the basis of the 2001 AUMF.

The Obama administration has regularly cited the AUMF as the basis for expanding the use of drone strikes in Pakistan, Yemen, Somalia and elsewhere, and argued that its authority has no specified temporal or geographic limit. The Obama administration has defended these strikes as legitimate attacks against al-Qaeda or the Taliban, declared enemies of the US, who as a transnational enemy must be attacked even outside the original theatre of war. In a March 2012 speech at Northwestern Law School, Attorney General Eric Holder remarked that:

Indeed, neither Congress nor our federal courts has limited the geographic scope of our ability to use force to the current conflict in Afghanistan. We are at war with a stateless enemy, prone to shifting operations from country to country. Over the last three years alone, al Qaeda and its associates have directed several attacks – fortunately, unsuccessful – against us from countries other than Afghanistan. Our government has both a responsibility and a right to protect this nation and its people from such threats.²²

Similarly, John Brennan has argued that ‘there is nothing in the AUMF that restricts the use of military force against al-Qaida to Afghanistan’.²³ This interpretation of the AUMF has come under criticism for effectively providing a greenlight for US action anywhere in the world until some unspecified point in the future when al-Qaeda and its co-belligerents are defeated.²⁴

The administration has further implied that it has a series of additional criteria that determine whether a drone strike should take place against a prospective target outside a declared combat zone, such as Afghanistan. Former Legal Advisor for the State Department, Harold Hongju Koh, has said that the US drone policy takes into account specific conditions of that case, including ‘those related to the imminence of the threat to the sovereignty of the states involved, and the willingness and ability of those states to suppress the threat that the target poses’.²⁵ The measurements for each of these criteria – in other words, how the US government assesses whether a government is willing and able to respond to a terrorist threat – have never been released. Similar additional criteria also apply when determining if a drone strike may be used against an American citizen. According to an internal document with the legal justification for killing Anwar al-Awlaki, an American citizen, the US considers itself ‘in a non-international armed conflict against al-Qaida and its associated forces’, and believes that the determination of whether a proposed strike would require ‘consideration of the particular facts and circumstances of each case’ rather than a prior determination of the geographic scope of the AUMF.²⁶ For strikes against both foreign and American citizens, the criteria used to decide when a state is not willing or able to address the terrorist threat on their territory have never been made public. This information would be particularly important in cases like Pakistan, where it remains unclear which of the operative criteria – lack of ability or political will – is the deciding factor in enabling US drone strikes.

There are three major concerns with the interpretation of the AUMF adopted by the Obama administration. First, the lack of a temporal scope for the AUMF, and the fact that its termination is premised on the defeat of al-Qaeda, raises the question of whether any future American administration will ever be constrained from launching drone strikes against al-Qaeda or its associated forces. At this point, the AUMF has operated as an effective green light for worldwide terrorist operations for over 13 years. There is no sign that this war, or the AUMF's applicability, will end soon. In 2013, Michael Sheehan, the assistant secretary of defense for special operations, suggested that the defeat of al-Qaeda could take ‘at least 10 to 20 years’.²⁷ Similarly, former Defense Secretary Leon Panetta has predicted that the US faces a ‘30 year war’, with a campaign extending to Libya, Somalia, Yemen and other locations.²⁸ Despite a thoughtful effort by former counsel for the DoD Jeh C. Johnson to wrestle with the question of what defeating al-Qaeda would look like, the Obama administration has never articulated what strategic defeat of al-Qaeda and its forces would look like, or what will happen to American counterterrorism operations when that point is reached.²⁹ In other words, no one knows what the defeat of al-Qaeda would look like or what happens to American counterterrorism policy the day after. There is some evidence that the US would be reluctant to give up the broad remit of authority provided by the AUMF no matter what happens to al-Qaeda. Even when Panetta announced in 2011 that al-Qaeda was on the verge of strategic defeat, the Obama administration never relented on its use of the AUMF as the legal basis for drone strikes, and in fact increased the tempo of these strikes.³⁰

Second, the lack of geographic limits to the administration's interpretation is a broad remit for military action, including full-scale combat operations and preventive attacks, in any country where al-Qaeda may be present and planning an attack. Given that al-Qaeda and its affiliates are officially in as many as 30 or more countries, the US has effectively pre-authorised itself for drone strikes in any place it sees a growing threat and little evidence of a local government with the ability to deal with it.³¹ The administration has already shown that it will extend drone strikes to countries where there were no plotters of the 11 September attack and no evidence of prior knowledge of that attack, such as Somalia. While the US is less likely to use drones against terrorist targets in Europe, it has used drones across the Middle East (for example, in Iraq and Libya) for reasons unrelated, or only indirectly related, to the 11 September attacks. It remains unclear whether drone strikes in regions far outside South Asia, such as Indonesia and the Philippines, could be conducted on the basis of the legal authority of the AUMF, but the Obama administration's argument implies that it would be permitted to do so if the target was linked in some way to al-Qaeda or the Taliban.

Third, the administration's implication that it has private criteria for determining both the ability and willingness of a government to deal with terrorist threats on its own soil is troubling. On one level, its arguments echo the principal argument of the Responsibility to Protect (R2P) doctrine, which suggests that external states can decide when a sovereign state forfeits its right to non-interference on its territory. For many humanitarian activists, a state's systemic abuse of the human rights of its population can give rise to a duty to intervene for well-intentioned states to rescue a victimised group of civilians. For many R2P advocates, if a government cannot or will not protect its citizens, other actors are entitled to step in and do so as long as the UN or another legitimate body approves. In an ironic twist, the Obama administration has retrofitted this argument – designed to instill fear in dictators and enable states to intervene for legitimate humanitarian purposes – into a rationale for the American drone programme in ungoverned spaces around the world. According to what is known about the Obama administration's criteria, the forfeiture of the sovereign right of non-interference is based on a state's inability, or unwillingness, to address a growing terrorist threat within its borders. The right to make this determination is accorded solely to the US government without any explicit standards or outside authority to adjudicate the decision to violate another state's sovereignty. This raises the prospect that this decision to violate the sovereignty of another state will be used in an inconsistent or capricious manner, with deleterious consequences for the sovereign right of non-interference in the international system.

The Obama administration's ambivalence on the AUMF has also been reflected by different statements from key administration officials on whether a new, revised AUMF is desirable or necessary. In his May 2013 speech, President Obama said:

The AUMF is now nearly 12 years old. The Afghan war is coming to an end. Core al Qaeda is a shell of its former self. Groups like AQAP must be dealt with, but in the years to come, not every collection of thugs that labels themselves al Qaeda will pose a credible threat to the United States. Unless we discipline our thinking, our definitions, our actions, we may be drawn into more wars we don't need to fight, or continue to grant Presidents unbound powers more suited for traditional armed conflicts between nation states. So I look forward to engaging Congress and the American people in efforts to refine, and ultimately repeal, the AUMF's mandate. And I will not sign laws designed to expand this mandate further. Our systematic effort to dismantle terrorist organizations must continue. But this war, like all wars, must end. That's what history advises. That's what our democracy demands.³²

Yet the administration has done nothing to revise or amend the AUMF; moreover, it has since invoked the AUMF to justify military operations in Iraq and Syria against ISIS, a group that did not exist in 2001 and that is not formally allied with al-Qaeda. As Jack Goldsmith has argued, ‘if this remarkably loose affiliation with al Qaeda brings a terrorist organization under the 2001 law, then Congress has authorized the President to use force endlessly against practically any ambitious jihadist terrorist group that fights against the United States. The President's gambit is, at bottom, presidential unilateralism masquerading as implausible statutory interpretation.’³³ Other Obama administration officials have expressed scepticism about the need for a new AUMF, noting that even a new legal authorisation for drones under a new AUMF would be neither necessary nor wise.³⁴ Despite growing pressure from some of his Democratic allies in Congress, the president has made no moves towards repealing or revising the AUMF, thus allowing this framework to stand as the predominant legal authority for drone strikes.³⁵

The Obama administration has also invoked international law to provide an additional legal foundation for its drone strike programme, arguing that it has the right to strike at al-Qaeda under the inherent right for self-defence provided by the UN Charter.³⁶ The interpretation of self-defence offered by the administration argues that individuals who are part of a group like al-Qaeda and the Taliban can be considered belligerents, and become targetable, even if they are not located in the country where the original self-defence justification was invoked. By this logic, an al-Qaeda member in Yemen could be killed by a US drone strike authorised in part on a claim of self-defence for an attack prepared and organised from Afghanistan. In the administration's view, this strike would not qualify as an assassination because, in the words of Koh, ‘under domestic law, the use of lawful weapons systems – consistent with the applicable laws of war – for precision targeting of specific high-level belligerent leaders when acting in self-defense or during an armed conflict is not unlawful, and hence does not constitute “assassination”’.³⁷ This argument has been backed by some scholars who argue that when a country is facing a potential attack, a self-defence rationale may be invoked for drone strikes against transnational actors, no matter where they are located.³⁸

This assertion of a right of anticipatory self-defence – which allows for an unrestricted geographic scope of military action, even in the absence of evidence of an immediate threat – has not been widely accepted by the international community. Even if one concedes that states are entitled to self-defence against non-state actors not controlled by another state – and the International Court of Justice has suggested that they are not – this argument has at least two major limitations.³⁹ First, as Philip Alston has pointed out, the self-defence argument implies higher standards for necessity and proportionality than the US government typically applies with its drones programme.⁴⁰ Specifically it would require that the government would only use force for defensive objectives, and for that it would only do so once other non-lethal options had been exhausted. There is no evidence in many drone strikes that the US is using them defensively or that they have exhausted other options before electing to strike. In many cases the Obama administration's arguments have emphasised preventive, rather than defensive, actions in that drone strikes are authorised before an attack can be launched or even fully planned. Moreover, the Department of Justice (DoJ) White Paper on drone strikes showed how the Obama administration adopted tendentious readings of ‘imminence’ and ‘necessity’ to superficially meet these requirements. According to this memorandum, ‘the condition than an operational leader presents an “imminent” threat of violent attack against the United States does not require the United States to have clear evidence of a specific attack in the future’.⁴¹ Similarly, the definition of necessity, according to the Obama administration's interpretation in the DoJ White Paper, requires only that it make the determination that a capture mission or some other non-lethal option be considered infeasible or too risky.⁴² The evidentiary standard for a capture operation to be deemed infeasible has never been publicly released.

Second, the case most often cited as providing precedent for using drone strikes for anticipatory self-defence – the Caroline case, from 1837 – requires that a state must show that there was an ‘instant, overwhelming, and leaving no choice of means, and no moment for deliberation’ situation before invoking a right of preventive self-defence.⁴³ This depiction is at odds with the narrative depiction of the Obama administration's drones programme, which shows a deliberate, non-hurried process to identify potential targets with the president's direct input before the drone strike occurs.⁴⁴ Although this explanation is an additional one to the AUMF, the anticipatory self-defence rationale adopted by the Obama administration leaves the US as an outlier in the international system, according for itself a right of anticipatory self-defence that it opposes for other states, with some exceptions.⁴⁵ Senior administration officials have recognised this fact. Former CIA Director Michael Hayden remarked in 2012 that ‘right now, there isn't a government on the planet that agrees with our legal rationale for these operations, except for Afghanistan and maybe Israel’.⁴⁶ The Obama administration is not an outlier due to its possession of drones and use of them for targeted killing, but rather due to its legal interpretation that enables it to continually strike against targets only indirectly related to the original combatants on the grounds of self-defence.

Nature of targets

The approach that the Obama administration has taken to identify the targets of drone strikes is also new and stands in stark contrast with the American approach during previous armed conflicts. According to the law of armed conflict, the identification of a target in an armed conflict involves a positive determination of their conflict status – i.e., whether that individual serves as a direct combatant, such as a soldier, in a recognised armed conflict. This determination, supported by both state practice and customary international law, involves designating as a combatant all those involved in combatant armies, even those not directly fighting, with the exception of religious and medical personnel.⁴⁷ In classical interstate wars, this is a relatively straightforward determination as the combatant is typically clothed as a soldier, carries some clear national identification, and is participating in conflict in a direct way. In these cases, states are not generally required to pinpoint the precise identity of an individual before targeting them, but they are required to determine if their actions are materially related to the armed conflict before attacking. They are also not required to provide combatants with access to legal process if they are in hors d'combat, but they must treat them as a privileged combatant and accord them prisoner of war status if captured.

For irregular wars, the designation of combatant status has always been more complex. The chief problem with these conflicts is that insurgents and other irregular fighters have often not been clothed or self-identified as combatants. For many insurgent groups, blurring their combatant status by hiding amidst the civilian population has allowed them to fight a stronger power and exploit the fact that governments are obliged to use force with the principle of distinction, avoiding civilian casualties if at all possible. Nevertheless, Article 4 of the Third Geneva Convention requires governments to accord to these irregular fighters prisoner of war status if they fulfil the following conditions: (1) that of being commanded by a person responsible for his subordinates; (2) that of having a fixed distinctive sign recognisable at a distance; (3) that of carrying arms openly; (4) that of conducting their operations in accordance with the laws and customs of war.⁴⁸ During the Vietnam War, the US generally applied the Geneva Conventions to the Viet Cong forces and accorded captured insurgents humane treatment, though not always full prisoner of war status. During the Bush administration, the US adopted a different approach and unsuccessfully advanced the argument that al-Qaeda and Taliban members were unlawful combatants because they failed to fulfil these conditions and should not enjoy trial by jury in civilian courts. Their argument – that unlawful combatants should be tried by military tribunals at Guantanamo Bay and elsewhere – was ultimately rejected by the Supreme Court.⁴⁹

The Obama administration has confronted similar problems with its drones programme. Many of the targets of drones do not wear uniforms and are part-time combatants, fighting at one moment but engaging in peaceful civilian activities at the next. The fact that many insurgent and terrorist groups do not have a clear command structure or distinction between political and military leaders produces a series of important moral dilemmas.⁵⁰ Many facilities and vehicles used by insurgents are dual use: they are used for military activity at one moment, and for combat at another. Even more so than guerrillas during the Cold War, today's insurgents are embedded in the civilian population, intermixing with them in ways that make analytically separating combatants and civilians difficult, if not impossible, in some cases.⁵¹ How governments can justly and carefully fight non-state actors, while maintaining the principle of distinction and not slipping into the barbarism that its abandonment would imply, is a key question that any government facing insurgent warfare in the twenty-first century must answer.

The Obama administration has acknowledged these problems but has maintained that the difficulties associated with identifying combatants in today's messy intra-state wars should not be an insurmountable barrier against using drones for targeted killings. It insists that the monitoring technology associated with its drones allows operators to abide by the principle of distinction and carefully distinguish between combatants and civilians.⁵² Koh has argued that the American drone strike programme easily meets the international humanitarian law requirements of necessity, distinction and proportionality. On the question of necessity, the US has consistently argued that it is legally permitted to attack members of al-Qaeda and the Taliban as irregular combatants because they are engaged in a protracted armed conflict with the US. For distinction and proportionality, the Obama administration has insisted that it has carefully avoided civilian casualties and abided by the principle of proportionality in its drone strikes, killing relatively few civilians compared to other available military means.⁵³ President Obama has admitted that some civilian casualties are inevitable with drone strikes, but has insisted that reports of high levels of civilian casualties are overstated. From an ethical vantage point, the Obama administration argues that the drones programme is humane in that it kills relatively few civilians, all of whom would be covered under the principle of double effect given the military necessity of the strikes.

Yet the Obama administration's arguments both presume, and are dependent upon, a novel definition of the nature of the enemy not widely accepted by the international community. Recognising the problems associated with the Bush approach, the Obama administration has expanded the definition of the target in a way which avoids the difficulties associated with using the terms illegal or enemy combatants. In 2009, the Attorney General announced the Obama administration was abandoning the term ‘enemy combatant’ while reaffirming its broad powers to fight and detain terrorist suspects.⁵⁴ This allowed the Obama administration to avoid many of the legal headaches that the Bush administration experienced with that definition, although in practice it meant little for those held as prisoners in Guantanamo Bay. At the same time, the Obama administration advanced, without fanfare, a more expansive definition of the potential targets of American military action than President Bush articulated. First, the Obama administration rejected the notion that it was limited to striking only al-Qaeda and the Taliban and insisted that it could also target ‘associated forces’. Although the AUMF language requires some connection be drawn to the 11 September attacks or that the enemy be at the least a co-belligerent of al-Qaeda and the Taliban, the Obama administration has in practice attacked groups linked to al-Qaeda or the Taliban in a variety of indirect, often immaterial, ways.⁵⁵ These ‘associated forces’ are not precisely defined by the US government and in practice have ranged from Pakistani Taliban to even Somali militants. All are now considered co-belligerents of al-Qaeda under the Obama approach. In effect, it has drawn a functional equivalence between ‘associated forces’ and ‘co-belligerents’, thus providing legal cover to strike a growing array of targets. It has used this expansive definition of the enemy to target the Islamic State in Iraq and Syria (ISIS), which was expelled from al-Qaeda in February 2014 and had no contemporary organisational link to it.⁵⁶ In fact, its authorisation for the Iraq operation declared that ISIS was a successor to al-Qaeda, and thus it can be considered a co-belligerent, a determination now also supported by the UN.⁵⁷ Its expansive interpretation of ‘associated forces’ or ‘co-belligerents’ is also not public, as the US refuses to make the list of what it considers an associated force of al-Qaeda or the Taliban public.⁵⁸ While the Pentagon maintains that a target must be defined as a co-belligerent of al-Qaeda or the Taliban, it does not specify how an actor operating far outside a recognised theater of conflict (Afghanistan) can be considered a co-belligerent of an actor in that conflict. While there is a plausible case that transnational terrorist groups could be considered co-belligerents even if they are not physically co-located with a combatant, the legal standard for making that determination remains a mystery. By equating the term ‘associated force’ with ‘co-belligerent’, the Obama administration has enabled itself to attack any group operating in the same radical Islamist milieu as al-Qaeda by deeming it an ‘associated force’.

Second, the Obama administration has adopted vague language about a group being ‘linked’ to a recognised combatant group without explaining exactly what that link is.⁵⁹ If the linkage was material support, that connection could be considered relatively obvious as a group supplying weapons or key equipment to a fighting group could be considered a combatant under the normal interpretation of the laws of war. Equally, groups that are fighting alongside an original combatant – for example, the Haqqani network and the Islamic Movement of Uzbekistan (IMU) in Afghanistan – could plausibly be considered a combatant, and hence a target, under international law. Yet the administration has acted in ways that suggest that the linkage could consist of purely ideological or political support for al-Qaeda or the Taliban, rather than direct material support. For example, the Obama administration has reportedly used drone strikes to target elements of al-Shabaab in Somalia without specifying how it is materially linked to the original combatants in Afghanistan and Pakistan. Similarly, some Pakistani Islamist groups have been targeted on the basis of a tangential connection to al-Qaeda or the Taliban. This administration's expansive definition of linkage allows for many possible connections, even political or ideological links, to be sufficient to make a group ‘linked’ to al-Qaeda and hence targetable.

Third, the Obama administration has adopted a policy of engaging in periodic signature strikes, directing drone strikes against training camps and compounds which were suspected of harbouring militants.⁶⁰ These strikes are premised on the basis that the patterns of behaviour of the potential target are sufficient to indicate that they intended to engage in combat against American or local allied forces. Critics have argued that signature strikes carry with them a greater risk of civilian casualties than so-called personality strikes, as the risk of false positives – that is, a mistaken identification of a target – is greater in these cases.⁶¹ This approach may also fall below the standard for distinction that is required under international law, as it is generally accepted that the person launching the strikes must be certain, within reason, that the activity witnessed is materially aiding the enemy's military effort. Conducted at a substantial remove from the actual target, signature drone strikes do not always achieve that standard and can fall afoul of the principle of proportionality, depending on the nature of the target and what they appear to be doing. For example, striking people simply because they were moving towards a conflict zone, or stopping at a rest area known to be used by al-Qaeda, is insufficient as grounds for a signature strike.⁶² Although he did not mention signature strikes by name, President Obama promised to employ a higher standard for drone strikes, which would eliminate many of these strikes, after his May 2013 speech. It remains unclear whether this has taken place, although drone strikes in Pakistan and Yemen declined sharply after that speech.

Drone usage

A third new aspect to the American policy on drones has to do with the way in they are used for monitoring, and striking, populations in foreign countries. Since 2009, the US has used drones as a regular form of surveillance in places such as Pakistan, Afghanistan and Yemen. This usage takes advantage of one of the significant technological features of drones: that they can loiter for significant periods of time over territories and provide detailed readouts on the movements of people and equipment. This means that unlike manned surveillance flights – where the time in the air is limited by human endurance and difficulties loitering – drones are able to remain over a territory for a long time. For example, the Predator B drone (otherwise known as the Reaper) can stay airborne for up to 14 hours while fully loaded with cargo.⁶³ Others have an even greater endurance capacity. The RQ-1A Predator drone can remain airborne for up to 40 hours and cover a large area at a distance of 400 miles from its operating base.⁶⁴

The US has taken advantage of the endurance capacity of these drones to extend the length of its surveillance in ungoverned spaces, such as Pakistan and Yemen, and in Afghanistan and Iraq. It has also deployed some of its more capable lethal drones, such as Predators and Reapers, as a semi-regular presence in these territories as the drones scan for known targets. Unlike surveillance flights during the Cold War, surveillance drones can loiter over a territory and provide a constant feed of data on what is occurring below. In this respect, they operate more like hyper-capable, low-altitude modern satellites than the manned overflights that occurred over major nuclear and defence installations during the Cold War. The regular deployment of drones for target killings is crucially different than the use of cruise missiles or other temporary, time-limited kinetic (or lethal) options. A cruise missile attack, like the US launched against alleged al-Qaeda sites in Sudan and Afghanistan in 1998, is by definition a limited event, where the destruction is over in a matter of moments. Even systematic bombing campaigns, like the Rolling Thunder campaign in the Vietnam War, occurred over a period of hours but then ceased on a daily basis as the manned aircraft returned to bases for refueling and rest. In these cases, it was certainly the case that bombers were seen over the territory on a semi-regular basis, but the duration of the operations, and the attendant risks to civilians, were relatively short-lived.

By contrast, the deployment of drones over ungoverned spaces in places like Pakistan and Yemen has become a regular event, where a foreign population is routinely monitored, and sometimes killed, by the US on an ongoing basis. The effect of the near-constant presence of drones on the wider population has been noted by a number of critics. The journalist David Rohde, for example, has described the effect of hearing drones whirring above him for hours on end and called them a ‘potent, unnerving symbol of unchecked American power’.⁶⁵ Similarly, the Stanford/NYU report, Living under Drones, has described the waves of fear that individuals feel from drones hovering overhead and the extent to which their presence has begun to interrupt normal political, economic and social life in these countries.⁶⁶ Others have reported from Yemen that drones fuel a pervasive sense of helplessness for those under their flight paths.⁶⁷ This sense of fear is clearly not universal; some critics have found evidence that drones are not as disruptive or hated as these accounts imply, for many educated, urban-dwelling Pakistanis and Yemeni see drones as preferable to the greater evils confronting them from local militants and their own government.⁶⁸ The actual psychological effect of drones may vary by social class and location, with rural dwellers in the ungoverned spaces where drones operate feeling their effects most directly.⁶⁹ But even given this fact, it is hard to dispute that for those living under the flight paths drones have become a constant, enduring reminder that they are being watched, and may be killed, by a foreign government. According to Human Rights Watch, some Yemenis have even concluded that they fear the US as much as they fear al-Qaeda in the Arabian Peninsula.⁷⁰

The key question that arises from this particular use of drones is whether the American routine deployment of drones over ungoverned spaces – and the psychological effect that this deployment has on the subject populations – has produced novel ethical and legal problems. If drones have the effect of terrorising parts of the rural population in the areas where militants are active, there may be a prima facie case that drone deployments are violating the fundamental criterion of proportionality, even if many of the strikes do not. The administration has employed a narrow calculation of proportionality, focusing exclusively on the body count associated with the drones. The US has insisted that drones are not violating the principle of proportionality because the tallies of civilian deaths have been low and that the ratio of civilians killed per combatant is favourable.⁷¹ The Obama administration's method of calculating civilian casualties has run into a wide array of critics, but even many of its critics implicitly accept that body counts should determine whether drones are more proportionate than other weapons of war.

Yet there are three reasons why a narrow calculation of proportionality – which focuses exclusively on measuring body counts, both of ‘bad guys’ and civilians – provides a misleading account of the real effects of drones. First, as John Kaag and Sarah Kreps argue, the typical calculation of proportionality is sometimes skewed as the goals associated with drone strikes – for example, ending ‘evil’ and preventing catastrophic terrorist attacks – are inflated, thus making civilian casualties and other psychological costs associated with the strikes more palatable.⁷² Seen from this vantage point, the psychological costs associated with drones would be considered more acceptable if they were in the service of countering a greater ‘evil’. Second, as Daniel Brunstetter and Arturo Jimenez Bacardi point out in their article in this volume, the principle of proportionality is difficult to measure when the psychological consequences of drones are added into the equation.⁷³ Much of the debate is conducted with a narrow calculation of proportionality, which measures only civilian deaths against the military gains associated with a strike, and engages in a form of ‘proportionality relativism’ that uses impertinent comparisons to claim that drones are proportionate.⁷⁴ Such calculations tend to downplay many of the psychological effects of drones and the longer-term consequences on America's global image, treating them as ‘off the books’ costs when measuring the casualties associated with each strike. Third, the degree of fear inflicted on the wider population from drone strikes may be disproportionate to the military advantages accruing from the drone strike itself. According to Geneva Conventions, if the incidental harm to the civilian population exceeds the advantages of removing an operational leader from the battlefield, a military operation may be deemed disproportionate even if its intention was not to terrorise the civilian population.⁷⁵ The Obama administration's decision to keep drones as a near-constant presence in areas in Pakistan and Yemen could constitute a violation of proportionality if the generalised fear in the population were seen as producing real costs, exceeding the advantages of removing some terrorist operatives from the battlefield.

If the psychological effects of drones are real and widespread, the regular deployment of drones by the US may also constitute a violation of the principle of distinction, which requires governments to shield non-combatants from the direct effects of conflict as much as they can. By using drones as near-constant force over these ungoverned territories, as opposed to an occasional presence, the US may be subjecting the civilian population to an environment of constant fear and uncertainty that is at variance with the normal interpretation of non-combatant immunity. While this action would not nearly be as egregious a legal and ethical violation of the principle of distinction as the actual bombing of civilian targets, it would still stand in contrast to a traditional interpretation of non-combatant immunity which requires those using violence to direct their attacks only at open participants in the conflict. By making the American deployment of drones a regular, nearly constant feature of life in these countries, and by allowing the psychological consequences of that decision to impact upon the local population outside the circumstances of a supreme emergency, the US may be violating the spirit of the principle of distinction even if it is not violating the letter of it.⁷⁶

Transparency and accountability

A further element of novelty in the Obama administration's drones policy has to do with its relatively weak mechanisms for transparency and accountability. As described by the Geneva Convention, the use of military force should be conducted by recognised armies who have clear chains of command and standards for releasing information about their operations. These government-led military forces are considered ‘lawful combatants’ and they have clear rights and responsibilities on the battlefield. Information on battlefield activities should flow up the chain of command and should be released, albeit selectively, to the wider public. No government has ever been perfectly transparent about its use of military force or willing to show ugliness of war to its population; in wartime, censorship and the manipulation of information for propaganda purposes has always been a serious problem. Yet on balance democracies have gone further than other states towards providing clear mechanisms of transparency and accountability to ensure that their use of force is governed by applicable legal and ethical principles. Throughout the twentieth century advanced democracies in Europe and North America created sophisticated mechanisms of transparency and accountability to ensure that progress of a military campaign was regularly assessed through legislative scrutiny and through the investigation of an active media. For most of the Cold War, the US government conducted armed conflict under the watchful eye of Congress and of the media, which demanded a regular release of information on their activities. These mechanisms of transparency and accountability were flawed and often met with government-led evasions and outright lies, as most famously happened with the Pentagon Papers during the Vietnam War. But even the Nixon administration conceded the basic principle that American military activity should be subject to harsh scrutiny from external forces, including academia and non-profit organisations, as well as Congress and the courts. Recognising that the domestic need for transparency was crucial, the US government built its own set of public relations bureaucracies to relay information about its military campaigns and hopefully keep the balance of public opinion on its side.

With the targeted killing programme, the Obama administration has reversed much of this trend and kept most of its activities out of the gaze of Congress, the courts and the independent media. Former White House spokesperson Robert Gibbs has admitted that he was told to not even admit the existence of a drones programme.⁷⁷ Even after President Obama admitted in a town hall meeting that there was a drones programme in 2012, White House spokesperson Jay Carney rebuffed questions about its existence.⁷⁸ By May 2012, the Obama administration began to open up the drones programme and to engage in a selective public relations campaign to release some information about it. It emphasised that the administration had a careful, deliberative process for selecting targets, but shied away from laying out the details of that process in the public domain.⁷⁹ By May 2013, President Obama promised to be more forthcoming with details about the drones programme, yet concrete action on revealing details of it have been very limited. Although the government has released some information on personality strikes, and on strikes against US citizens, it has failed to acknowledge publicly that it even engages in signature strikes, despite how often that allegation is made.⁸⁰ Moreover, it has released few details on key issues surrounding the targeted killings of non-Americans, despite the fact this group constitutes the vast majority of those targeted by drone strikes. The administration has not even confirmed the number of countries where drones are present or where al-Qaeda or ‘associated forces’ have been attacked.

The Obama administration has not been much more forthcoming with Congress or the courts. The administration has regularly argued that Congress is fully briefed about the fundamental elements of the drones programme, but in practice only select members of Congress on the Intelligence Committees are briefed on the fundamentals of the drone operations. Relatively few Congressional briefings have been held on the mechanics or effects of the drones programme, although sparely attended hearings on the drones programme began to increase as the administration contended with a wave of public criticism.⁸¹ These hearings have largely been directed to the administration's assertion that it is entitled to target American citizens with drone strikes, despite the fact that strikes against American citizens are far less common than strikes against foreign citizens. With respect to the rest of the programme, Congress has ‘essentially abdicated oversight responsibility’ for much of the drones programme.⁸² There has never been a single vote for or against the drones programme, as most of it has been conducted with tacit Congressional approval.⁸³ Until 2013, American courts regularly refused to hear drone cases and met administration opposition that they lacked jurisdiction over the matter. The Obama administration has only recently allowed the courts to weigh in on drone strikes, though they have signalled some willingness to consider a drones court to weigh up the justness of a proposed strike.

The administration also has uneven mechanisms for Congressional accountability in the event of a mistake or careless use of a drone strike. The US drones programme is divided into two parallel programmes, one under the control of the Joint Special Operations Command (JSOC) and another under the control of the Central Intelligence Agency (CIA). The JSOC programme is governed by the military's rules and standards for targeting individuals and goes through a rigorous process of target selection and review. In other words, the JSOC programme has a chain of command that allows for proper accountability.⁸⁴ There is some evidence that this pays off in more careful discrimination about civilian casualties.⁸⁵ The CIA programme, on the other hand, is entirely classified, with almost no data in the public domain about its activities. It remains unclear who is involved in the process of selecting targets and how the chain of command for the CIA drone programme is designed.⁸⁶ The absence of clear accountability for the CIA's portion of the drones programme is important because it raises key questions about whether the CIA will be as careful and discriminating as JSOC. It also raises some key questions about the legal accountability of the CIA officials involved in the programme. As Mary Ellen O'Connell points out, the CIA would not be considered a lawful combatant according to the law of armed conflict and may not be as well trained or mindful of the requirements of distinction, necessity and proportionality as military officers.⁸⁷ This means that unlike soldiers they may be later held criminally accountable in foreign courts for their involvement in the drones programme. The Obama administration is attempting to prevent this by denying the existence of the programme, by refusing to reveal the mechanics of its targeting decisions, and by trying to limit the applicability of international human rights law in countries where drones are used frequently.

The result of the Obama administration's approach is that the drones programme has become an anomaly: an unaccountable and non-transparent form of democratic warfare. The administration has argued that the programme is accountable, but it has emphasised that the accountability for it should be relegated to the executive branch and select Intelligence Committees, as opposed to Congress, the courts or the general public. In this respect, it is treating the drones programme as analogous to spying, which receives scant Congressional and judicial oversight and remains a preserve of the Executive Branch. Yet it is hard to argue that the drones programme – which has killed as many as 4000 people, according to a leaked government estimate – is closer to spying than it is to armed conflict.⁸⁸ Moreover, this effort by the Obama administration to keep drones in the shadows stands in opposition to many of the global trends towards greater accountability seen with warfare around the globe, as other democratic governments are increasingly holding the agents of violence, including soldiers, criminally responsible for war crimes and other acts that violate the laws of war. Whether a democracy can engage in this kind of sustained violence in the absence of mechanisms for transparency and accountability, and remain democratic, is an open question. But there is a serious possibility that the drones programme will become corrosive of the democratic character of the US government as long as these two conditions remain unfulfilled.

Precedent

The final new dimension of the Obama administration's use of drones concerns the precedent that it sets for the use of assassination and targeted killings abroad. In general, the use of assassinations is not widely accepted by states, as relatively few governments – notably Israel – have argued publicly that they should be permitted outside of armed conflict. This is not to say that assassinations are rare; governments and intelligence agencies conduct assassinations from time to time, and in rare cases, such as Israel, make a policy of it.⁸⁹ Yet most governments are loathe to articulate publicly that this should be permissible by others for two reasons. First, it is widely accepted that assassinations and targeted killings are legally prohibited. Under international law, assassinations are forbidden outside the context of armed conflict because they violate international human rights law, as well as an array of domestic laws.⁹⁰ Assassinations are generally considered a violation of customary international law and of a state's obligation to protect the right to life. They are permitted only during the context of armed conflict and under exceptional circumstances, where overwhelming necessity to prevent further loss of life dictates that they should be conducted.⁹¹ The US government is officially barred from conducting assassinations by both domestic law (for example, the 1976 executive order signed by President Gerald Ford) and by international law. Although the CIA has committed extrajudicial killings, it operates under a legal authority (Title 50 of the United States code) that requires it be consistent with the US constitution and all other applicable laws.⁹² Second, there is a strong norm against assassinations that is widely respected by states, if only for reasons that they fear others will engage in assassinations against their officials.⁹³ This norm has proven to be durable despite the growth of informal violence in the system, as even states that conduct assassinations are reluctant to concede that this should become a generalisable right. Even during wartime, when some assassinations would arguably be permitted, states have been reluctant to kill wartime leaders, in part because it would deprive them of someone who could negotiate an end to the war.

The use of drones for targeted killings challenges this legal and normative prohibition on assassinations and targeted killings outside of armed conflict. The Obama administration has maintained that its drone programme does not constitute a form of assassination, and therefore it is not in violation of legal and ethical prohibitions on that act.⁹⁴ By declaring that it is in a geographically unrestricted state of armed conflict against al-Qaeda, the US has argued that its drone strikes should be considered as part of wartime conduct, rather than an assassination conducted in peacetime. This interpretation enables the US to strike at the leaders and operatives of non-state actors around the world under the broad rubric of armed conflict, while avoiding the legal restrictions on assassinations. The problem with this argument is that the US is not at war with the governments in the territories in which it is conducting targeted killings; its interpretation is essentially that it can be at war with non-state actors in foreign countries and strike them without being in a state of war against the government which is legally sovereign in the territory where the non-state actor is present. By this logic, a state at war with an ideological or nationalist terrorist group could kill their operatives on the territory of another state, without declaring war on that government or admitting that it is a form of assassination. In other words, the US policy creates a gray area between war and assassination that enables a variety of different forms of informal violence, especially government on non-state actor targeted killing.

From a legal and ethical vantage point, this articulation of a space for killing between war and peace poses a number of challenges. The Obama administration's position is not consistent with international law which requires states to either declare war on a government or leave it to that government's responsibility to deal with armed threats emanating from its borders. To articulate a generalised right to strike within another state's territory, without first declaring war on that state, could be interpreted as an act of aggression unless the state granted consent to do so. Even if they did, some have argued that states cannot grant other governments the right to strike on their territory in this way outside an official armed conflict.⁹⁵ Arguably, the US may be articulating a new form of state practice, as recognised by a number of legal scholars, which permits assassinations or targeted killings to regularly occur in the gray area between armed conflict and peacetime.⁹⁶ One danger of this approach is that recurring state practice may itself begin to produce followers and generate new norms of customary international law that undermines the strict prohibition on assassinations and enables a new range of prospective targets for targeted killings. In other words, the US may be gradually normalising the practice of extrajudicial murder by states, particularly if other states do not accept that the right of anticipatory self-defence belongs to the US alone.

There is already some evidence that this loosening of the barriers against assassinations or targeted killings is now in play. The US has already begun to expand its range of targets beyond traditional insurgents by drawing up a ‘hit list’ of Afghan drug lords.⁹⁷ Other states have begun to follow suit and adapt extrajudicial killings to their own purposes. For example, Israel has begun to use drones to strike at militants in the Sinai and in the Palestinian territories.⁹⁸ According to a New York Times report, China seriously considered using a drone strike to kill a notorious drug lord in Myanmar.⁹⁹ At present, the chief reason that more states have not joined the practice of using drones to target their enemies is that their technology has lagged behind industry leaders such as the US and Israel. But as more states catch up, the prospect of an increasing number of states engaging in targeted killings against their domestic enemies becomes a real one. A high-level panel at the Stimson Center recently sketched out a number of scenarios where drones could be used by others for targeted killings, such as Russia using drones to kill its enemies in Ukraine without admitting its legal reasoning for doing so. The danger of the American practice of drones is that it has produced a new pattern of state behaviour that may lead more states to join the assassination game. The Stimson Center report asks: ‘is the United States inadvertently handing abusive foreign regimes a playbook for murdering those it considers politically inconvenient under the guise of combating terrorism?’¹⁰⁰

A growing number of ethics scholars have also begun to recognise that this gray area between war and peace may be necessary, but should not be seen as a free-for-all zone where any form of violence is permitted. In the areas which Michael Walzer calls the ‘in between zones’, where terrorists are operating in ungoverned or less governed spaces, drone strikes must still be assessed on the basis of jus en bello standards or, as Braun and Brunstetter argue, a modified form of jus en bello with even stricter standards of proportionality and discrimination.¹⁰¹ In this realm, increasingly called jus ad vim, Braun and Brunstetter argue that the resort to force in these ethical gray zones should not be permissible on the basis of a generally articulated threat, as is often the case with the US targeted killings programme. In these areas, harm to civilians must be measured in a way that stretches beyond body counts and incorporates a wider notion of harm, including psychological harm. Even if one concedes that the US may be right that there is a space short of war where some form of force should be permitted, it does not follow that the standards currently applied by the CIA's drone programme, which presume a broadly articulated threat and a narrowly articulated standard of proportionality, must be accepted.

In this context, the technology of drones becomes relevant once more and interacts with American policy in dangerous ways. One of the chief ethical dangers arising from drones is moral hazard, where the fact that drone technology is low-cost and increasingly accessible to a number of states means that targeted killings becomes easier and therefore more frequent. As Kaag and Kreps argue, the expediency of drone-based targeted killing can be confused with the moral justification for it; just because something can be done quickly and with a modicum of personal risk does not mean that it is justified.¹⁰² Drone technology is seductive because it can create the illusion of riskless war and lead more states into taking risks to kill more enemies with drone strikes in other lands. Here the US policy on targeted killing is perilous if it creates a precedent and leads other states to follow suit with their own drones-based targeted killing programmes. The spread of drone technology and the American precedent could combine to create a world in which more and more states engage in targeted killings because they are easy and because so many of their rivals do it. Such a world would be closer to the rule of the jungle than to the rule-based international order that the US sought to create and sustain over the last 50 years.

Conclusion

One of the unfortunate consequences of the rise of drone-based targeted killings is that much of the debate over the technology and its uses focuses exclusively on that function and its legal and moral consequences. It is important to remember that targeted killings are not the only function of drones; they can be put to a wide range of other, often beneficial purposes. For example, as David Whetham points out in his contribution here, drones could play an important role in complex humanitarian emergencies as they can conduct surveillance in regions (like the Syrian civil war today) where outside monitors find it too difficult to operate.¹⁰³ Similarly, Caroline Kennedy and James Rodgers argue that drones can provide a solution for the UN in peacekeeping operations, both by reducing the risks that peacekeepers face and by monitoring the compliance of signatories to peace agreements.¹⁰⁴ Kennedy and Rodgers go even further by suggesting that armed drones might be able to play a useful role in peace enforcement. Both of these uses of drones have legal and ethical questions; there are important issues surrounding the legal authority for their use, the control of the drones, the legal accountability for their use, and the moral hazard that flows from their employment by the UN or other actors. But they point to an important fact: a world of drones is not inevitably one in which targeted killings are their only, or even primary, function.

Whether this world ever emerges has to do largely with what the US – as a leader in the research and development of drones and as the world's largest and most innovative user of drones – does next. In other words, the legal and ethical implications of drones are not set in stone but are in fact being created and reinforced by discrete American policies. At present, the use of drones by the Obama administration has been directed towards establishing a right of anticipatory self-defence, enabled and accelerated by drone technology, which is accorded exclusively to the US government. Such an articulation of the purpose of drones is dangerous because it may begin to crowd out other, more beneficial, uses of drones, and to produce a set of precedents that may be exploited by other states. The policies employed by the Obama administration for drone-based targeted killings have begun to slowly erode some of the traditional legal and ethical restraints on the use of force, creating a growing gap between current American practice and how armed force should be used according to the law of armed conflict. Such an approach is short-sighted for it surrenders the opportunity for the US to take the lead in articulating precisely what the legal and ethical implications of drones are and how the rules on the use of force must be adapted to accommodate and control them. By developing a targeted killing programme as an exceptional, US-only proposition, and by keeping that programme in the shadows, the Obama administration may be hastening a day in which others use drones in ways hostile to both American interests and to a just international order.

Acknowledgements

The author is grateful to Emma Leonard Boyle for feedback on this piece and wishes to thank all of the authors and reviewers who made this special edition possible.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Michael J. Boyle is an Associate Professor of Political Science at La Salle University in Philadelphia. His most recent book is Violence after War: Explaining Instability in Post-Conflict States (Baltimore: Johns Hopkins Press, 2014).

Notes

1. President Barack Obama, ‘Remarks by the President at the National Defense University’, The White House, 23 May 2013. http://www.whitehouse.gov/the-press-office/2013/05/23/remarks-president-national-defense-university (accessed 28 September 2014).

2. John Brennan, ‘The Ethics and Efficacy of US Counterterrorism Strategy’, Address at the Woodrow Wilson Center, Washington DC, 30 April 2012. http://www.wilsoncenter.org/event/the-efficacy-and-ethics-us-counterterrorism-strategy (accessed 28 September 2014).

3. See, for example, Harold Hongju Koh, ‘The Obama Administration and International Law’, 25 March 2010. http://www.state.gov/s/l/releases/remarks/139119.htm (accessed 28 September 2014).

4. See Daniel R. Brunstetter and Arturo Jimenez Bacardi, ‘Drones and Human Rights: A Clash of Discourses', in this volume.

5. Barack Obama, ‘Nobel Lecture’, 10 December 2009. http://www.nobelprize.org/nobel_prizes/peace/laureates/2009/obama-lecture_en.html (accessed 3 November 2014).

6. For example, former General Counsel of the Department of Defense, Jeh Charles Johnson, remarked that ‘this is a new kind of war. It is an unconventional war against an unconventional enemy. And, given its unconventional nature, President Obama – himself a lawyer and a good one – has insisted that our efforts in pursuit of this enemy stay firmly rooted in conventional legal principles.’ See Jeh Charles Johnson, ‘The Conflict Against Al Qaeda and Its Affiliates: How Will It End’, Speech to the Oxford Union, 30 November 2012.

7. This article refers to these standards in general terms, but they are comprised of the Geneva Conventions and the jus ad bellum and jus in bello requirements.

8. Stephanie Carvin, ‘Getting Drones Wrong’, in this volume.

9. This point is made well by Bradley Jay Strawser, ‘Moral Predators: The Duty to Employ Uninhabited Aerial Vehicles', Journal of Military Ethics 9, no. 4 (2010): 352–68. See also Strawser's debate with Asa Kasher in ‘Distinguishing Drones: An Exchange’, in Killing by Remote Control: The Ethics of an Unmanned Military, ed. Bradley Jay Strawser (Oxford: Oxford University Press, 2013), 47–65.

10. For good discussions of some of the philosophical and ethical questions arising if fully autonomous drones emerge, see Peter Finn, ‘A Future for Drones: Automated Killing’, The Washington Post, 19 September 2011; Patrick Lin, ‘Drone-Ethics Briefing: What a Leading Robot Expert Told the CIA’, The Atlantic, 15 December 2011.

11. For Robert Sparrow, this fact means that drones are permissible, but fully autonomous drones would not be due to a lack of clear human accountability. See Robert Sparrow, ‘Killer Robots', Journal of Applied Philosophy 24, no. 1 (2007): 62–77.

12. Human Rights Council, ‘Report of the Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions, Philip Alston’, Fourteenth Session, 28 May 2010, para. 84.

13. Mary Ellen O'Connell, ‘Unlawful Killing with Combat Drones: A Case Study of Pakistan 2004–2009′, Notre Dame Law School, Legal Studies Research Paper No. 09-43, July 2010, 8–10.

14. Glenn Greenwald, ‘Bravery and Drone Pilots', Salon, 10 July 2012.

15. See Matthew Power, ‘Confessions of a Drone Warrior’, GQ Magazine, 22 October 2013; Mark Bowden, ‘The Killing Machines', The Atlantic, 14 August 2013.

16. David Whetham, ‘Killer Drones', RUSI Journal 158, no. 3 (2013): 22–33.

17. This fact has led a number of scholars to argue that we are now living in a world of post-heroic warfare. See particularly Edward Luttwak, ‘Towards Post-Heroic Warfare’, Foreign Affairs (May/June 1995); and for an application to drones see Christian Enemark, Armed Drones and the Ethics of War: Military Virtue in a Post-Heroic Age (London: Routledge, 2013); and Robert Sparrow, ‘War Without Virtue’, in Killing by Remote Control: The Ethics of an Unmanned Military, ed. Bradley Jay Strawser (Oxford: Oxford University Press, 2013), 84–105.

18. Authorization to Use Military Force, Pub L. 107-40, 115 Stat. 224, 14 September 2001.

19. For an effective summary, see Hamdi v. Rumsfeld (2004), 542 U.S. (507), Lawfare, 12 November 2012. http://www.lawfareblog.com/wiki/the-lawfare-wiki-document-library/post-911-era-materials/post-911-era-materials-court-cases/hamdi-v-rumsfeld-542-u-s-507-2004/ (accessed 21 October 2014).

20. Jack Goldsmith, ‘Congress Must Figure Out What Our Government is Doing in the Name of the AUMF’, Lawfare, 17 May 2013. http://www.lawfareblog.com/2013/05/congress-must-figure-out-what-our-government-is-doing-in-the-name-of-the-aumf/ (accessed 8 October 2014).

21. Charlie Savage, ‘White House Invites Congress to Approve ISIS Strikes, but Says It's Not Necessary’, The New York Times, 10 September 2014.

22. Remarks by Attorney General Eric Holder, Northwestern Law School, 5 March 2012.

23. Brennan, ‘The Ethics and Efficacy of US Counterterrorism Strategy’.

24. Jennifer Daskal and Stephen I. Vladeck, ‘After the AUMF’, Harvard National Security Journal 5 (2014): 115–46.

25. Hongju Koh, ‘The Obama Administration and International Law’.

26. Department of Justice White Paper, ‘Lawfulness of a Lethal Operation Against a U.S. Citizen Who is A Senior Operational Leader of al Qai'da or an Associated Force’, obtained by NBC News, quotes at pp. 3, 4.

27. Quoted in Andrew Rosenthal, ‘The Forever War’, The New York Times, 17 May 2013.

28. Susan Page, ‘Panetta: “30 Year War” And a Leadership Test for Obama’, USA Today, 6 October 2014.

29. Jeh Charles Johnson, ‘The Conflict Against Al Qaeda and Its Affiliates: How Will It End’, Speech to the Oxford Union, 30 November 2012.

30. Craig Whitlock, ‘Panetta: U.S. “Within Reach” of Defeating al Qaeda’, The Washington Post, 9 July 2011.

31. The number of 30 countries where al-Qaeda operates comes from The New York Times, ‘Map of Countries Where al Qaeda and Its Affiliates Operate’, 12 May 2011.

32. Obama, ‘Remarks by the President at the National Defense University’.

33. Jack Goldsmith, ‘Obama's Breathtaking Expansion of a President's Power to Make War’, Time Magazine, 11 September 2014.

34. Harold Hongju Koh, ‘How to End the Forever War’, Speech at the Oxford Union, 7 May 2013. DoD backs up, see Goldsmith.

35. Ed O'Keefe, ‘Kaine: If Congress Doesn't Authorize Military Action, “We Will Have Created a Horrible Precedent”’, The Washington Post, 23 September 2014.

36. Hongju Koh, ‘The Obama Administration and International Law’.

37. Ibid.

38. For a good statement of this view, see Jordan J. Paust, ‘Self Defense Targetings of Non-State Actors and the Permissibility of US Use of Drones in Pakistan’, Journal of Transnational Law and Policy 19, no. 2 (2009): 237–80.

39. United Nations General Assembly, Human Rights Council, ‘Report of the Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions, Philip Alston’, 14.

40. Ibid.

41. Department of Justice White Paper, ‘Lawfulness of a Lethal Operation Against a U.S. Citizen Who is A Senior Operational Leader of al Qai'da or an Associated Force’, quote at p. 7.

42. Ibid., 8–9.

43. Thomas M. McDonnell, ‘Sow What You Reap? Using Predator or Reaper Drones To Carry Out Assassinations or Targeted Killings of Suspected Islamic Terrorists', George Washington International Law Review 44 (2012): 243. http://digitalcommons.pace.edu/cgi/viewcontent.cgi?article=1857&context=lawfaculty, 289.

44. Jo Becker and Scott Shane, ‘Secret “Kill List” Proves a Test of Obama's Principles and Will’, The New York Times, 29 May 2012.

45. The US is generally supportive of Israel's right to preventive self-defence to strike terrorist operatives in the West Bank and Gaza, as well as in Lebanon, Syria and elsewhere.

46. Quoted in John Mearsheimer, ‘America Unhinged’, The National Interest (January/February 2014): 29.

47. International Committee of the Red Cross, ‘Combatant Status'. https://www.icrc.org/customary-ihl/eng/docs/v1_rul_rule3 (accessed 11 October 2014).

48. Convention (III) Relative to the Treatment of Prisoners of War. Adopted on 12 August 1949 by the Diplomatic Conference for the Establishment of International Conventions for the Protection of Victims of War, held in Geneva from 21 April to 12 August 1949. Entered into force 21 October 1950.

49. Charles Lane, ‘High Court Rejects Detainee Tribunals', The Washington Post, 30 June 2006.

50. Michael Walzer, ‘Targeted Killing and Drone Warfare’, Dissent Magazine, 11 January 2013. http://www.dissentmagazine.org/online_articles/targeted-killing-and-drone-warfare (accessed 7 November 2014).

51. General Sir Rupert Smith calls this ‘war among the people’. See his The Utility of Force (London: Allan Lane, 2005).

52. Hongju Koh, ‘The Obama Administration and International Law’.

53. See Koh for the administration's defence. Ibid.

54. Del Quintan Wilber and Peter Finn, ‘US Retires “Enemy Combatant”, Keeps Broad Right to Detain’, The Washington Post, 14 March 2009.

55. Daskal and Vladeck, ‘After the AUMF’, 115–46.

56. Liz Sly, ‘Al Qaeda Disavows Any Ties with Radical Islamist ISIS Group in Iraq, Syria’, The Washington Post, 3 February 2014.

57. Marty Lederman, ‘The Legal Theory Behind the President's New Initiative Against ISIL’, Just Security, 10 September 2014.

58. Cora Currier, ‘Who Are We At War With? That's Classified’, Pro Publica, 26 July 2013.

59. Michael J. Boyle, ‘Is the US Drone War Effective?’, Current History 113, no. 762 (2014).

60. Becker and Shane, ‘Secret “Kill List” Proves a Test of Obama's Principles and Will’.

61. Danya Greenfield, ‘The Case Against Drone Strikes on People Who Only “Act” Like Terrorists', The Atlantic, 19 August 2013.

62. Kevin Heller, ‘One Hell of a Killing Machine: Signature Strikes and International Law’, Melbourne Legal Studies Research Paper 634, 2012.

63. MQ-9 Reaper Hunter/Killer UAV. http://www.defense-update.com/products/p/predatorB.htm (accessed 13 October 2014).

64. RQ-1A/MQ1 Predator UAV. http://defense-update.com/products/p/predator.htm (accessed 13 October 2014).

65. David Rohde, ‘The Obama Doctrine: How the President's Secret War is Backfiring’, Foreign Policy 192 (March/April 2012): 66.

66. Stanford Law School and NYU School of Law, Living Under Drones (2013): 73–99.

67. Adam Baron, ‘In Yemen, Drones Ill Effects Linger Long After Dust Settles', Christian Science Monitor, 17 July 2013.

68. C. Christine Fair, Karl Kaltenhalter, and William J. Miller, ‘Pakistani Opposition to American Drone Strikes', Political Science Quarterly 129, no. 1 (2014): 1–33; Christopher Swift, ‘The Blowback Fallacy’, Foreign Affairs, 1 July 2012.

69. C. Christine Fair, Karl Kaltenhalter, and William J. Miller, ‘You Say Pakistanis All Hate the Drone War? Prove It’, The Atlantic, 23 January 2013.

70. The Telegraph, ‘Yemen Drone Strikes Cause Civilians to “Fear the U.S. As Much As al Qaeda”’, 22 October 2013.

71. Becker and Shane, ‘Secret “Kill List” Proves a Test of Obama's Principles and Will’; Michael J. Boyle, ‘The Costs and Consequences of Drone Warfare’, International Affairs 89, no. 1 (2013): 1–29.

72. John Kaag and Sarah Kreps, Drone Warfare (Cambridge: Polity Press, 2014), 97.

73. See Daniel R. Brunstetter and Arturo Jimenez Bacardi, ‘Drones and Human Rights', article in this volume.

74. See Megan Braun and Daniel R. Brunstetter, ‘Rethinking the Criterion for Assessing CIA-Targeted Killings: Drones, Proportionality and Jus ad Vim’, Journal of Military Ethics 12, no. 4 (2013): 304–24.

75. Dieter Fleck, The Handbook of International Humanitarian Law, 3rd edition (Oxford: Oxford University Press, 2013), Section 509.

76. On supreme emergency, see Michael Walzer, Just and Unjust Wars: A Moral Argument with Historical Illustrations, 2nd edition (New York: Basic Books, 1992).

77. Michael Calderone, ‘Robert Gibbs Told Not to Acknowledge Drone Program Exists As White House Press Secretary’, The Huffington Post, 24 February 2013. http://www.huffingtonpost.com/2013/02/24/robert-gibbs-drones-white-house_n_2753183.html (accessed 20 October 2014).

78. Karen DeYoung, ‘After Obama's Remarks on Drones, White House Rebuffs Security Questions', The Washington Post, 31 January 2012.

79. Becker and Shane, ‘Secret “Kill List” Proves a Test of Obama's Principles and Will’.

80. Kaag and Kreps, Drone Warfare, 40.

81. Micah Zenko, Reforming U.S. Drone Strike Policies, New York: Council on Foreign Relations, Report No. 65, January 2013. See also Tim Starks, ‘Unmanned Oversight’, The New Republic, 7 February 2013.

82. Kaag and Kreps, Drone Warfare, 67.

83. Peter W. Singer, ‘Do Drones Undermine Democracy?’, The New York Times, 21 January 2012.

84. See the discussion in the Stimson Center, The Task Force on Drone Policy (Washington, DC: June 2014), 33.

85. Braun and Brunstetter, ‘Rethinking the Criterion for Assessing CIA-Targeted Killings', 310–11.

86. Stimson Center, The Task Force on Drone Policy, 33.

87. O'Connell, ‘Unlawful Killing with Combat Zones', 7.

88. Senator Lindsay Graham has estimated that drones have killed 4700 people. Spencer Ackerman, ‘Senator Lists the Death Toll from Drones at 4,700 People’, Wired, 21 February 2013.

89. See the discussion in Gabriella Blum and Philip B. Heymann, ‘Law and Policy of Targeted Killings', Harvard National Security Journal 145 (2010): 145–70.

90. UN Special Rapporteur on Extrajudicial Assassinations, Handbook (Center for Human Rights and Global Justice, New York University), ‘Chapter Two: Use of Force by Law Enforcement Officials'. http://www.extrajudicialexecutions.org/application/media/Handbook%20Chapter%202%20use%20of%20force%20in%20LE.pdf (accessed 21 October 2014).

91. Philip Alston, ‘The CIA and Targeted Killings Beyond Borders', Harvard National Security Journal 2 (2011): 283–446. See also Blum and Heymann ‘Law and Policy of Targeted Killings'.

92. Stanford Law School and NYU School of Law, Living under Drones, 121.

93. Ward Thomas, ‘Norms and Security: The Case of International Assassination’, International Security 25, no. 1 (2000): 105–33.

94. Hongju Koh, ‘The Obama Administration and International Law’.

95. O'Connell, ‘Unlawful Killing with Combat Zones', 16.

96. See David Kretzmer, ‘Targeted Killings of Suspected Terrorists: Extra-Judicial Executions or Legitimate Means of Self-Defence?’, The European Journal of International Law 16, no. 2 (2005): 171–212.

97. Blum and Heymann, ‘Law and Policy of Targeted Killings', 148.

98. Associated Press, ‘Israeli Drone Kills Suspected Islamic Militants in Egypt’, 9 August 2013.

99. Jane Perlez, ‘Chinese Plan to Kill Drug Lord with Drone Highlights Military Advances', The New York Times, 20 February 2013.

100. Stimson Center, The Recommendations and Report of the Task Force on Drone Policy (Washington, DC: 2014), 37.

101. Michael Walzer, ‘On Fighting Terrorism Justly’, International Relations 21 (2007): 480–4; Braun and Brunstetter, ‘Rethinking the Criterion for Assessing CIA-Targeted Killings', 306.

102. Kaag and Kreps, Drone Warfare, 112.

103. David Whetham, ‘Drones to Protect’, in this volume.

104. Caroline Kennedy and James Rodgers, ‘Virtuous Drones', in this volume.