On April 15–16, 2011, some three dozen philosophers, legal scholars, and military officers gathered for a conference in Philadelphia under the auspices of the University of Pennsylvania to examine the moral and legal issues surrounding the targeted killing of suspected or known terrorists. The topic had been discussed and debated in the media, mainly in connection with the increased use of unmanned armed drones to target terrorists by the Obama administration, but also in relation to Russian policies in Chechnya and Israeli policies in the Occupied Territories and elsewhere. Critics and supporters of the targeted killings had traded arguments and counterarguments that raised complex and contested questions of fact, law, and morality. The premise of the Philadelphia conference was that the differing perspectives of moral theory, legal doctrine, and military experience could jointly shed light on those difficult questions.
Two weeks after the conference took place, U.S. Navy Seals killed the infamous leader of Al Qaeda, Osama bin Laden, in his redoubt in Pakistan, and targeted killing became a front-page topic around the globe. Although media attention to the topic has naturally diminished since the first few weeks after bin Laden was killed, targeted killing is virtually certain to remain a controversial instrument in the arsenal that states use to counter threats stemming from those non-state organizations and movements that engage in terrorism.
The questions raised by targeted killing are not going away any time soon: they are at once timely and enduring. This volume is the first appearance in print of a collection that brings together scholars from across disciplines for a sustained and reasoned discussion of these questions. In this introduction, I provide material intended to orient readers, coming as they will from a broad range of academic and non-academic backgrounds. Section I explains what is meant by “asymmetric” armed conflict and how terrorism is connected to such conflict. Section II examines the term, “terrorism,” sketching and defending a concept of terrorism that informs the various contributions to this volume. Section III describes the two main approaches to assessing the legality and morality of targeted killing: the law-enforcement and the armed-conflict models. Section IV summarizes each subsequent chapter, drawing contrasts and remarking on similarities among them, and Section V offers some brief concluding thoughts.
The armed confrontation between states and their enemies who employ terrorist tactics is a case of the broader phenomenon of asymmetric conflict. Such conflict is characterized by opposing sides that bring to their armed contest radically different means and methods of violence and radically divergent levels of brute firepower. Asymmetric conflict is nothing new in the history of organized human violence, but in the modern world it has typically taken the distinctive form of a conflict between the regular armed forces of states, on the one side, and the irregular forces of non-state groups and movements, on the other. The “regular” forces of a state are characterized by a strict chain of command, precisely defined roles in the command hierarchy, and conformity to the legal requirement that members wear uniforms and insignia and carry their arms openly in order to distinguish themselves clearly from civilians. The “irregular” forces of non-state groups and movements are more loosely organized, and, although they typically have something akin to a chain of command, they do not abide by the legal requirements concerning uniforms and the open carrying of arms: they seek to blend into the civilian population, rather than to distinguish themselves from it.
The armed forces of states have much greater firepower and organizational strength and more advanced technological resources than do the irregular forces that they confront, and it is this fact that partly constitutes the asymmetry between states and their irregular opponents. Additionally, the firepower and technology of a state’s armed forces are supplemented by the resources of the state’s intelligence agencies, which can not only provide information about their adversaries to the state’s leaders, but also, in the case of the CIA at least, operate their own lethal technology.
It is in order to offset their disadvantages in firepower, technology and organizational strength, that irregular forces choose to blend into the civilian population and thus adopt a tactic explicitly prohibited by international law and constituting one of the defining features of terrorism: the deliberate killing of civilians. Thus, the leadership of Al Qaeda evidently believed that to drive the United States out of the Middle-East and to force the country to change its pro-Israel foreign policy, it needed to attack not just the Pentagon, but also the World Trade Center with its thousands of civilians. Non-state groups involved in attacks on states are not likely to take the deliberate killing of civilians off the table for another reason: the members of such groups tend to regard civilians of the adversary state as complicit in grave wrongs perpetrated by their state and as morally liable to attack for that reason. In addition, once violence is adopted as the method for accomplishing a goal, then the inclination is to ramp up the violence when it turns out that the goal has not been achieved, and civilians become tempting targets when the violence needs to be ramped up. (This tendency to ramp up violence in the absence of success is shared by state officials as well, with the U.S. war in Vietnam serving as a prime example.) In short, the asymmetric threats confronted by states in the decades ahead will almost invariably involve terrorist threats.
Much about terrorism is contested, including the very meaning of the term “terrorism” itself. And many people and state agencies have succumbed to the temptation to stretch the term for rhetorical and political purposes. For example, the FBI regards the property crimes that animal liberation and environmental groups commit to further their causes as terrorism,1 and the Animal Enterprise Terrorism Act makes it a crime to interfere with the operation of an animal enterprise by intentionally damaging real or personal property.2 Such bloated uses of the term “terrorism” have little to recommend them. Property crimes, even when they involve an undue risk to human life—and the so-called “terrorist” crimes of animal liberation and environment groups often involve no such risk—simply do not belong in the same moral or legal category as Timothy McVeigh’s Oklahoma City bombing, much less the 9/11 attacks. The justification for the law’s crude conception of terrorism is obscure, at best. It might be said that the property crimes in question and the crimes of Timothy McVeigh and the 9/11 terrorists share the feature that they are politically motivated. True enough, but one should expect terrorist crimes to involve the intention to commit violence against persons, and not just to be political in motivation.
A defensible approach would take the conceptually central cases of terrorism to be those acts undertaken with the intent (and realistic possibility) of killing in a single episode of violence a number of civilians for the purpose of advancing certain political goals by intimidating (a certain segment of) the population. Acts morally different from those undertaken with such an intent and purpose should not be counted as terrorist at all, or should be seen as peripheral instances of the concept.
The term “terrorism” and its cognates are almost always used to make a moral judgment about the acts and agents to whom the terms are applied, and it might seem to follow from that fact that, as a conceptual matter, a terrorist act is morally wrong. But this conclusion does not follow, and it is not necessary to include a moral element as part of the very concept of terrorism. Instead, one can treat the question of whether some act is an act of terrorism as a factual question about the intent and purpose behind the act, and then proceed to judge the act as an egregious moral wrong, if the act does meet the factual conditions for a terrorist act.
It should be noted that the acts counting as central cases of terrorism, on the account I have sketched, are not limited to acts of non-state groups; states can and do engage in terrorism too. This implication helps to rebut the suspicion that “terrorism” is a term that is rhetorically loaded in favor of states and against insurgent movements, a suspicion fed by such abuses of the term as we see in the FBI’s categorization of crimes. However, it might be argued that the concept of terrorism I have sketched is still tilted in favor of states over their non-state adversaries. The argument is this. States often pursue their goals with armed force that, although not deliberately directed at civilians, is used with indifference toward their lives, resulting in a substantial number of collateral deaths. Such a use of armed force can certainly be terrifying to populations, but, on my account, it would not count as terrorist. This implication seems questionable, because it might not be immediately clear what the moral difference is between a state foreseeably killing many civilians as “collateral damage” and a non-state organization killing the same number of civilians, albeit deliberately. Dead is dead. Thus, one might argue that a more even-handed account of the concept of terrorism would categorize both the state and the non-state killings as central cases of terrorism.
This argument is not entirely persuasive. A widely (though not universally) held moral view is that wrongful harms brought about intentionally are more serious wrongs than those that are brought about foreseeably but not intentionally. On that premise, the state’s killings, even if wrong, would be less so than the killings of the non-state group, and the state killings would arguably count as peripheral cases of terrorism, if they counted as terrorism at all. Moreover, even if the premise is mistaken, it is still the case that persons who bring about wrongful harms intentionally bear a greater burden of blame than those who bring about such harm foreseeably but not intentionally. Treating the state and the non-state killings as central instances of terrorism would elide this difference in appropriate blame.
It should also be noted that, even if the intentional killing of civilians can be morally justified in certain extreme situations, the general moral prohibition against the intentional killing of the innocent would seem to be stricter than the prohibition against foreseeably but non-intentionally killing the innocent. The relatively greater strictness of the former prohibition is connected to the premise mentioned in the previous paragraph that intentional wrongs are more serious than non-intentional ones, and the consequence of this greater strictness is that the intentional killings of civilians by the non-state groups are more likely to be wrong in the first place than the non-intentional killings by the state. Accordingly, this consequence is not chargeable to any inherent bias in my account of the concept of terrorism, but, rather, simply falls out of the different relative strength of different moral prohibitions.
Targeted killing is the intentional killing by a state of an individual identified in advance and not in the state’s custody.3 One can only speculate as to whether such killing would have been a topic of intense and widespread global attention had the attacks of September 11 been foiled and their perpetrators arrested. Certainly, the thousands of deaths resulting from the attacks, and the indelible images of the collapse of the Twin Towers, have shadowed any subsequent discussion of the use of targeted killing in response to the threats posed by terrorism. In the eyes of some, the events of September 11 were a wake-up call to governments around the world, demonstrating just how dangerous terrorism had become and just how inadequate to the danger the time-honored procedures and principles of investigating and prosecuting criminals before civilian courts now were. But in the eyes of others, those events did not demonstrate any such inadequacy: terrorism was a serious danger; but it called not for any substantial departure from the way that liberal states dealt with crime, but rather for better coordinated and more effective law-enforcement efforts.
These opposing ways of understanding September 11 have shaped the discussion of targeted killing by leading to the articulation of two competing models for assessing the response of governments to terrorism: the armed-conflict (or war) and the law-enforcement models. The models are not meant to be simply descriptive accounts of the current state of the law, but, at a deeper level, to provide moral frameworks for judging the actions of governments and determining what the law should be. And they lead to sharply conflicting assessments of the targeted killing by governments of suspected or known terrorists and to incompatible judgments about whether the law should ever permit such killing. The two models do not exhaust the approaches that might be taken to the issue, and several of the contributors to this volume suggest that both models are inadequate and some third way is needed. Nonetheless, the two models highlight the main questions to be addressed and provide answers that, if not beyond criticism, are plausible and widely held.
The proponents of the law-enforcement model argue that terrorism should be dealt with largely by the personnel, procedures, and standards used in responding to any kind of serious crime: police, prosecutors, timely public trials in civilian courts, warrants issued by such courts upon probable cause, and the rights to adequate legal counsel, to habeas corpus, to confront one’s accusers, and to the other elements of due process of law. Some marginal modifications from ordinary criminal prosecutions might be justifiable, for example, if state secrets are involved in the prosecution, but the aim is to stick very closely to the idea that alleged terrorists should be dealt with in the same way that all others accused of serious crimes are treated by a well-functioning liberal state. This model pointedly rejects the idea that the targeted killing of suspected or known terrorists is morally or legally permissible, apart from situations in which the targeted individual poses an imminent (or otherwise unavoidable) threat to the lives of civilians and killing him is the only way to stop the threat from being realized. Excluding such emergency situations, the authorities are morally and legally obligated to capture the suspect and forbidden from killing him. So argue the defenders of the law-enforcement model.
The proponents of the armed-conflict model argue that the law-enforcement model is inadequate to deal with the threat of terrorism and that, instead, suspected and known terrorists should be treated as enemy combatants whose very tactics—targeting civilians—violate the laws of war and whose threat should be met, in large measure, by military means on the basis of principles appropriately applied during a time of war. On this model, a military response to terrorism is justified as a matter of national and/or personal self-defense (and defense of others). Terrorists are not simply enemies of the states against which they conduct their attacks. They are “unprivileged belligerents,” who lack the legal and moral permission to kill anyone because they do not distinguish themselves from civilians, and they are war criminals because they deliberately kill civilians. If they are captured, it is morally and legally permissible to try them in military courts and accord them a less rigorous form of due process than is found in civilian criminal courts. Additionally, on the armed-conflict model, there is no moral or legal requirement to try to capture them, and it is morally and legally permissible to target them with lethal force, even when they are not posing an imminent threat to the lives of others. There is no “last-resort” requirement on operations aimed at killing the enemy in war time: a legitimate target can be permissibly killed, even if capture would be costless. Only if and when the enemy surrenders is it then forbidden to intentionally kill him. In contrast, the law-enforcement model imposes a last-resort requirement: intentional killing is permitted only if capture is not possible and the killing is “strictly indispensable to save human life from unlawful attack.”4
One of the notable features of the debate over targeted killing is that each side regards the other as proposing an approach that is not merely sub-optimal but unacceptable. The proponents of the law-enforcement model do not simply say that targeted killing is less than the best way to respond to terrorism; rather, they reject it as morally and legally impermissible. On the other side, defenders of the armed-conflict model insist that for a state threatened by terrorists to forego the practice, when the state has the requisite means, is an unacceptable abdication of its responsibility to its citizens.
The reason why the law-enforcement advocates regard the practice as impermissible is relatively straightforward: they see targeted killings as nothing more than unjustifiable homicide by the state, in other words, as murder by the officials who plan and execute the killings. Just as it is murder for a guard to intentionally kill a suspected or known terrorist who is held in detention and posing no immediate threat to any of the captors or other detainees, it is murder for CIA personnel to intentionally kill by drone a suspected terrorist in Pakistan who is posing no immediate threat to citizens of the United States or of any other country. In both cases, the suspected or known terrorist is not, literally speaking, on a battlefield confronting U.S. or other military forces with his own lethal force. Rather, he is just like any other criminal suspect who continues to hold his human right to life unless and until he is convicted of a capital crime in accordance with due process of law. Targeted killing is impermissible because it is a violation of that right.
Moreover, advocates of the law-enforcement model argue that targeted killing cannot be justified as a matter of the right to self-defense because the relevant instances of such killing do not conform to an essential condition of rightfully exercising lethal force in self-defense, viz., that the targeted individual pose an imminent threat to human life. Bin Laden was certainly a threat to the United States and other countries, but, at the time he was killed, his threat was not on the verge of being realized. According to the law-enforcement model, then, the targeted killing of bin Laden and other known or suspected terrorists cannot be justified as a matter of self-defense.
The armed-conflict advocates contend that the absence of a demarcated battlefield only shows that traditional ways of thinking about the use of armed force are woefully obsolete when it comes to addressing contemporary terrorism. We are in “a world of new wars,”5 and the traditional idea of war as a more or less symmetrical relation between two states (or two coalitions) whose armed forces clearly identify themselves as such and confront one another in delimited geographic regions cannot be sensibly applied to the asymmetrical relation between a state and the terrorist organizations that seek to attack it. There might be no specific battlefield on which opposing armies fight in the case of terrorism, but, the proponents of targeted killing argue, a war exists between the parties to the conflict, albeit an asymmetrical war.
Contrary to the law-enforcement model, the defenders of the armed-conflict approach argue that suspected and known terrorists are not just like other criminal suspects and not even like members of the Mafia and other organized crime enterprises. Unlike traditional organized crime, the purpose of terrorist organizations is to kill large numbers of civilians in the states that they regard as the enemy, in an effort to defeat the political will of those states. Thus, Al Qaeda attacks the United States for its support of Israel and the Saudi regime, seeking to drive America out of the Middle East. Moreover, terrorist organizations are often supported by friendly governments that clandestinely provide resources such as money, forged documents, weaponry, training camps and safe haven. If a state were to forego entirely the use of targeted killing in dealing with a terrorist threat, then it would be failing to meet its responsibility to protect the lives and the political independence of its people. So argue the proponents of the armed-conflict model.
As for self-defense, advocates of the armed-conflict model reject the idea that the imminence requirement, understood as a demand to wait until an attack is just about to happen, applies to actions taken against the enemy in a time of war. When a country is at war, it is not required to refrain from lethal action until the next attack from the enemy is on the verge of occurring before launching a strike. Accordingly, on the armed-conflict model, the countries threatened by bin Laden did not need to wait until his next attack was just days away to strike at him.
Few would argue against the claim that the targeted killing of individuals who are involved in terrorist activity is morally permissible under certain conditions, and none of the contributors to this volume argues that targeted killing is absolutely and universally prohibited by morality. Indeed, it is common ground among the contributors that such killing is morally permissible where the authorities know that a particular individual is about to deliberately trigger an explosion intended to kill many civilians and the only way to prevent the explosion is to kill the individual. But the actual practice of those governments that use targeted killing is by no means limited to such a case. The debate in this volume, and in public discussion more generally, concerns the permissibility of the practice insofar as it extends beyond the imminent-explosion type of case. Proponents of the armed-conflict model argue that the extended practice is morally permissible and that the law should also regard the practice as permissible. Defenders of the law-enforcement model can concede that, in some cases at least, the extended practice might be morally permissible, but they will argue that the legal rules should flatly prohibit any targeted killing deviating from the imminent-explosion scenario.
Mark Maxwell leads off Part I, which is devoted to exploring the question of who is permissibly targeted in asymmetric armed conflicts. He argues that the law-enforcement model unduly hampers the efforts of a state to protect its citizens against terrorist threats. Unlike most scenarios in which law enforcement can be effective, terrorists often plan their operations in safe havens outside of the jurisdiction of the states they attack and the master planners might never roam beyond those safe havens. Maxwell thinks that the United States rightly abandoned the law-enforcement model in the wake of 9/11 in favor of the armed-conflict model.
However, in Maxwell’s view, the pursuit of terrorists under the law of armed conflict is only the beginning of the story. He argues that there should be revisions in that sphere of the law so that states can deal even more effectively with the threats posed by terrorist groups, because the current law gives terrorist organizations a critical advantage over the states that are trying to stop them. The advantage derives, on Maxwell’s account, from the way in which the law allows terrorists to exploit a basic principle of the law of armed conflict, the principle of distinction. Under this principle, combatants are legally permitted to intentionally kill one another, but they are absolutely prohibited from the intentional killing of civilians. The legal permission of combatants to kill carries with it the legal obligation to distinguish themselves in a clear and public way from civilians: combatants must wear uniforms, carry their arms openly, and so forth. Terrorists deliberately disregard such requirements of legal armed conflict and so fail to count as combatants who have a permission to kill the enemy. So what do terrorists count as? Maxwell tells us that, under current international law, they count as civilians who directly participate in hostilities. Such civilians do not have full civilian immunity: it is permissible for them to be intentionally killed, but only “for such time as they directly participate in hostilities.”6 When they cease their direct involvement, even if only temporarily, they regain civilian immunity.
Because of their “now immune, now not immune” legal status, Maxwell analogizes terrorists to the moles in the children’s game “Whack a Mole”: the moles can be hit with a mallet but only when they pop out of their underground homes; they regain their safety as soon as they duck back down into their burrows. A clever mole is difficult to whack, and, under current international law, so is a clever terrorist. So Maxwell proposes a revision in the law that would substantially reduce the immunity currently enjoyed by terrorists. In particular, he argues that terrorists who are members of an armed group should forfeit their civilian immunity when their pattern of conduct shows that they are contributing to the military function of the group. Direct involvement in hostilities would be only one form of conduct that contributes to the military function.
Jens Ohlin constructs a normative framework that seeks to integrate elements of the law-enforcement and armed-conflict models in order to answer the question, “Who is permissibly targeted?” Discussions of the question typically proceed on the assumption that relying on law-enforcement principles to deal with terrorism would better serve the rule of law than would following the norms belonging to the armed-conflict model. However, Ohlin argues that this assumption is wrong: the rule of law is better served by norms taken from the armed-conflict model and modifying them to fit the asymmetric armed conflict between the state and terrorist organizations.
Ohlin points out that terrorist threats sufficiently grave to call for targeted killings stem largely from organizations rather than from individuals acting alone. Additionally, a targeted killing is a summary killing, and, in justifying such an action, the threatened state might invoke its right to self-defense or, alternatively, its rights under international humanitarian law to attack the enemy with lethal force. But, under either justification, some suitable link must be established between the individual who is targeted and the organization that poses the threat. Clearly, it is not a sufficient link merely for the individual to have a favorable attitude toward the terrorist organization, but what, then, is sufficient? Ohlin canvasses a range of linking principles.
One intuitively attractive principle is that the targeted individual must be a member of the terrorist organization. International humanitarian law relies on a membership principle to determine who is liable to lethal attack in the context of an armed conflict: the members of the military forces of the enemy. And, as Ohlin notes, in the case of a symmetrical conflict between regular forces who wear uniforms and carry their arms openly, the membership principle has the virtue that its linking criterion is public and easily applied by both sides. However, he also points out that, in the asymmetric conflict between a state and a terrorist organization, the membership principle encounters a potentially serious problem, because terrorists make a point of not wearing uniforms or carrying their arms openly.
Moreover, civil libertarians would reject membership as a linking principle because membership is ordinarily understood as a matter of status rather than conduct. In their eyes, state coercion against an individual should be predicated on what the individual has done—that is, on his conduct, not on who he is or what groups he belongs to—that is, not on his status. A linking principle more friendly to the civil-libertarian view would require the individual to be making a contribution to the criminal acts of the terrorist organization (complicity), or, alternatively, to have made an agreement to commit such acts (conspiracy). But Ohlin is concerned that such a principle would be quite elastic in application, potentially covering a far wider range of cases than seems justifiable, and would require an intensive factual inquiry to determine whether the individual was guilty of the offending conduct.
Yet, Ohlin is prepared to go part way toward accommodating the civil-libertarian viewpoint. He modifies the membership principle so that it includes a more robust conduct element, while also making the principle suitable for dealing with terrorist organizations. In Ohlin’s view, the link that makes an individual liable to a targeted killing is this: the individual has a self-declared, functional membership in an organization engaged in an armed conflict with the threatened state. Functional membership means that the individual plays a role within the organizational structure by carrying out the orders and missions given to him by his superiors and by, in turn, giving orders to those, if there are any, below him in the hierarchy. This kind of membership is a hybrid concept, incorporating elements of both status and conduct. Moreover, Ohlin argues that a principle employing the idea of functional membership would better serve the rule of law than would a purely conduct-based principle, such as the complicity or conspiracy standards. Those latter standards are more difficult to apply to the individual and to administer by state agencies than the more clear-cut principle of functional membership. Ohlin concludes that, surprisingly, the principle that contains a status element better serves the rule of law than a pure conduct-based principle. The demands of national security can thus be served in a way that is consistent with the nation’s commitment to legality.
Ohlin’s account of who is legitimately targeted for killing is in general agreement with Maxwell’s suggested revision to the rules of war, adding rule-of-law arguments to arguments about the activities that terrorists undertake as members of an organization. Daniel Statman seeks to broaden the issue beyond terrorism, arguing that targeted killing is not only a matter of how states should address terrorist threats, but, more generally, it is a question of how states should deal with threats posed by irregular forces in asymmetric armed conflicts. He presents three interpretations of just war theory and explores the implications of each for the moral permissibility of targeted killing, whether the targets are terrorists or individuals belonging to irregular forces that do not employ terrorist tactics. On the individualist interpretation, the moral principles governing war are continuous with those governing all other conditions of human life, and the moral permissibility of killing in war derives from the right that individuals have to defend themselves against others who are personally responsible for unjust and deadly threats against them. On the collectivist interpretation, combatants have such a right of individual self-defense, but they also have a moral permission to intentionally kill enemy combatants that derives from the role that combatants play in defending their society from the collective violence of the enemy. On the contractualist interpretation, the moral permission of combatants to intentionally kill derives from a tacit agreement among states (and, potentially, armed non-state groups) to diminish the death and destruction of war by drawing a firm line that permits lethal attacks directed against enemy combatants but absolutely prohibits such attacks against civilians.
Statman argues that all three interpretations of just war theory permit targeted killing and that they even provide reasons for preferring targeted killing to other forms of armed combat. He is persuaded that targeted killing is very effective in avoiding collateral deaths and in focusing lethal force against the individuals most responsible for unjust and deadly threats. Under all three interpretations of just war theory, such effective focusing of deadly force is a substantial virtue. Moreover, none of the three interpretations provides sufficient grounds for judging persons involved in terrorism as civilians immune to permissible targeting. Statman concludes by suggesting that in this new era of asymmetric warfare, there is a strong moral case in favor of legal rules that license the targeted killing of terrorists and other irregular fighters.
Jeremy Waldron examines a proposed legal norm that would license targeted killing: “N1: Named civilians may be targeted with deadly force if they are presently involved in planning terrorist atrocities or are likely to be involved in carrying them out in the future.”7 He expects that some readers will share Mark Maxwell’s view of the terrorist threat and judge that, in light of the need of society to protect itself against terrorism, N1 ought to replace the strict legal prohibition in both domestic law and the law of armed conflict on deliberately killing civilians. However, Waldron argues that such readers ought to be wary.
When a person moves from normative judgments about particular actions, say, the judgment that the killing of bin Laden was justifiable, to the endorsement of norms covering killing, the logical implications of those norms might prove troubling in ways the person did not anticipate. Exploiting this logical generality of norms, philosophers and legal thinkers sometimes use a “neutral-principles test” to evaluate the validity of a person’s endorsement of a given norm: can the person consistently maintain her endorsement of the norm in light of the troublesome implication? In an analogous way, making a norm into a law can have troublesome practical implications that are not anticipated. Waldron points out that such implications flow, not just from the logical generality of norms as such, but also from the fact that legal norms are part of the public world and so are likely to be invoked and applied by persons who have beliefs and agendas that diverge sharply from the viewpoints of many individuals who find the norm attractive as an abstract proposition. In the case of N1, Waldron argues, we should keep in mind that, if it were a legal norm, it could be invoked publicly by governments that are hostile to us and our liberal democratic values in order to justify their actions. He also points to the fact that governments in general have a terrible track record of abuse and misjudgment when they take their gloves off in dealing with those they judge to be terrorists. Liberal democracies have been no exception, Waldron reminds the reader.
All legal rules are subject to abuse, but Waldron argues that N1 is especially liable to egregious abuse because it deals with deliberate killing in a context in which there is no strong and independent judiciary, or its institutional equivalent, which could impartially examine each killing post hoc for the purpose of judging its consistency with the norm and sanctioning those responsible in cases of a violation. Accordingly, Waldron is unpersuaded by efforts to defend N1 by appealing to the idea of self-defense as it functions in domestic criminal law, a context in which there is a strong judicial back-stop.
Moreover, Waldron argues that the legal norm absolutely prohibiting the deliberate use of armed force against civilians (while permitting its deliberate use against enemy combatants) has proved to be a crucially important part of the law of armed conflict, helping to mitigate the human devastation of war. He agrees that the norm is easy to criticize as under- and overinclusive, but philosophical nuance is not what is needed in a norm that is to do the job of restraining those who wield weapons in the heat of battle and the encompassing homicide of war. Rather, what is needed is a norm that is widely acceptable, simple to apply, and capable of being inculcated in the people who pull the trigger. The strict prohibition on the intentional killing of civilians has become anchored in both law and the ethos of military service. In Waldron’s judgment, it would be reckless to experiment with loosening the prohibition in order to free state officials to kill more terrorists, and it would be philosophically misguided to think that the appropriate attitude toward human life and its deliberate destruction would be better reflected by abandoning the prohibition in favor of a norm to govern war that is more consistent with norms that we already accept for governing peace.
Waldron’s contribution can be understood as meeting the arguments of Maxwell and Statman on their own turf: the ground of the law of armed conflict. Whereas Maxwell seeks to loosen the legal rules of war that restrict the targeted killing of terrorists and Statman aims to show that just war theory, under any of its main interpretations, licenses targeted killing, Waldron argues that we should hold fast to restrictive rules of war that place such killing beyond the limits of permissible conduct. Ohlin’s analysis suggests that, pace Waldron, some terrorists are the functional equivalent of combatants, and so we can hold fast to the categorical prohibition on targeting civilians while still engaging in the targeted killing of terrorists, but Waldron might judge that the conceptual tidiness of Ohlin’s approach would break down in practice, weakening a crucial legal restraint on the violence of war.
Part II picks up the issue of whether the norms of war, or the norms of domestic law enforcement, are the appropriate ones to use in deciding what the legal rules should be for regulating and restricting targeting killing. Jeff McMahan begins the section by arguing that the targeted killing of terrorists can be morally justified only in terms of self-defense or defense of others, and not on retributive grounds. The most plausible version of the defensive justification, in his eyes, holds that a terrorist is liable to the use of force when he is responsible for a threat to innocent lives and the force reasonably aims to avert that threat. On this justification, the permissibility of using lethal defensive force is not based on the terrorist’s involvement with past episodes of terrorism, as would be the case in retributive justification, but, rather, on his current involvement by way of intending, planning, or assisting some envisioned act of terrorism. This justification might seem to fit well with the law-enforcement model, because domestic criminal law authorizes the use of deadly force for purposes of self-defense and defense of others. However, McMahan points out that the justification in question must depart from the law-enforcement model in a crucial respect: the latter model authorizes defensive force only when the threat is imminent, while targeted killing is preventive, aimed at persons whose threat is not imminent.
Turning to the question of how the law should handle the targeted killing of terrorists, McMahan finds that, under current legal norms, terrorists must be either criminals or combatants. Although he thinks that there might be no determinate legal answer to the question of which category applies, McMahan argues that there are strong reasons to regard terrorists as criminals and to treat them under a law-enforcement model, rather than as combatants, treating them under an armed-conflict model. Terrorists deliberately violate the legal requirements for having the status of combatant, a status that carries with it the legal privilege to use lethal force. McMahan thinks that such a status is reasonably restricted to the members of forces that generally observe the legal requirements in return for getting the privilege. Moreover, McMahan contends that the law-enforcement model minimizes the chances of incorrectly identifying someone as a terrorist and consequently killing him, a fatal mistake documented in several instances of targeted killing. The model also contributes to the effective deployment of defensive force by avoiding the waste of resources involved in tracking and trying to take out the wrong people. If the law were to treat terrorists as combatants, then it would be legally permissible for them to be hunted down and killed rather than arrested and tried before a court. Aside from the greater chances for mistaken identification, legally permitting such killing would create risks for innocent bystanders and be subject to abuse by regimes that target political opponents and then defend their actions by claiming that the victims are terrorists. McMahan agrees with Waldron that this risk of abuse argues strongly in favor of a categorical prohibition on targeted killing. However, in the end, McMahan is not entirely persuaded, because terrorists, though not combatants, are not ordinary criminals either. Rather, they occupy an “intermediate” position between combatants and criminals, thus creating a situation calling for new legal norms that are intermediate between the law-enforcement and armed-conflict models. Accordingly, McMahan is likely to be more sympathetic than Waldron to Ohlin’s idea that functional membership in a terrorist organization should make a person into a legally permissible target of lethal force.
Claire Finkelstein notes that the administration of President Obama has dramatically increased the number of targeted killings over that of the Bush administration, while radically reducing the number of detentions of terrorist suspects. Behind these changes in presidential policy, she detects a desire to circumvent the political, legal, and moral problems that afflicted President Bush’s policy of indefinite detention. However, Finkelstein argues that the shift to targeted killing fails to deal with the underlying source of those problems, viz., the lack of fit between the laws of war, on the one side, and the nature of the armed conflict between terrorists and the states they attack, on the other. Moreover, traditional arguments from personal self-defense also fail to justify targeted killing, due to the distinctive nature of the terrorist threat. However, Finkelstein does find in the law-enforcement model certain principles of pre-emptive killing that can provide solid ground for the justification of lethal attacks on terrorists who pose a danger to a state’s national security.
Finkelstein argues that efforts to justify targeted killing in terms of the laws of war confront an insuperable set of obstacles, stemming from the fact that those laws were designed for conventional war, fought by regular forces on a clearly demarcated battlefield. In the asymmetric conflict between states and terrorist organizations, however, the distinctions central to conventional war have unraveled. The laws of war license combatants to kill enemy belligerents who have not surrendered, and both the license to kill and the liability to be killed depend on the status of the soldier as a member of a regular, belligerent armed force. Civilians are not permissibly targeted. But, Finkelstein argues, the targeted killing policy deviates from the law-of-war model in several ways. The targeted terrorists are named in advance, in the manner of an assassination and in contrast to a status-based attack. Individuals on the kill list are sometimes not engaged in active combat. In the case of drone-missile killings, civilians in the CIA often operate the drones, and the use of drones means that there is no opportunity for targeted individuals to surrender when facing an imminent attack. Additionally, the clearly demarcated battlefield of conventional war has disappeared in wars between states and terrorist organizations. If Finkelstein’s analysis is sound, then serious doubt is cast on approaches such as those of Maxwell and Statman, which rely, respectively on the laws of war and the closely related principles of just war theory.
Finkelstein is also skeptical of the way in which McMahan tries to show that targeted killing is a morally permissible use of defensive force. She agrees with him that principles of ordinary morality, reflected in the rules of the criminal law, can ground the permissibility of targeted killing. But Finkelstein thinks that McMahan’s reliance on a principle of personal self-defense and other-defense is mistaken. She argues that the appropriate principles are to be found in a law-enforcement situation in which the police are pursuing a person suspected of a felony who is resisting arrest by fleeing from them. The police are legally permitted to use deadly force against such a suspect, as long as they believe that he poses a risk of committing a future felony and have warned him of their intent to use force should he not submit. This scenario illustrates the pre-emptive use of force, and Finkelstein emphasizes that the police are permitted to kill the fleeing individual even if his suspected felony is not a capital offense and his anticipated future felony is not imminent. She argues that an appropriately fashioned targeted killing policy would fit the requirements of justified pre-emptive killing.
Richard Meyer thinks that targeted killings now fall into a legal no-man’s land between the laws of war and domestic criminal law. This legal indeterminacy means that the rights and duties of the relevant parties, including the persons who carry out the killings, but also the members of the local police forces where the killings take place, are left in a fog of legal uncertainty. In order to provide for greater certainty and to conform more closely to the demands of the rule of law, Meyer argues that it is necessary for international law to draw a sharper line than it does at present between combat during war and conduct outside of war.
For Meyer, the current legal uncertainty surrounding targeted killing derives largely from a failure to appreciate the chasm that separates the morality of ordinary life from the morality of combat. The ordinary life of the members of society is governed by moral principles that are reflected in the criminal law: individuals are prohibited from using violence against others, except in a very narrow range of circumstances in which it is explicitly authorized, such as those involving legitimate self-defense. In contrast, Meyer argues, combat is governed by moral principles discontinuous with ordinary morality and reflected, not in criminal law, but in international humanitarian law. The heart of this alternative morality is the principle that members of a regular armed force who otherwise observe the laws of war are privileged belligerents permitted to kill enemy belligerents who are members of a regular armed force, except in a very narrow set of circumstances, such as when the enemy soldier has surrendered.
On Meyer’s account, the targeted killing of suspected terrorists is insufficiently regulated by law. States are using military force, but it is uncertain whether those who wield the force are privileged belligerents, because international law does not clearly provide for the situation of war between a state and a non-state organization, much less between a state and an individual. Individuals who might well be terrorists are killed. However, no evidence of their involvement with terrorism need be provided to anyone by the attacking state, the attackers are de facto immune from prosecution under the laws of the state where the killing takes place, and no viable legal forum exists before which the family of the victim can fairly challenge his designation as a terrorist. At the same time, Meyer agrees with Mark Maxwell that the states victimized by terrorism can legitimately complain that certain other states harbor terrorists, letting suspects melt into the population and refusing to capture them, even when their whereabouts are known to the local authorities.
In order to bring some legal order to a situation that he regards as anarchic, Meyer proposes three major changes in international law. First, privileged belligerency should be restricted to the uniformed military and militias of a state that has formally declared war. Second, the law should provide for declarations of war against non-state organizations and against individuals. Third, the International Court of Justice should be given jurisdiction to decide cases challenging any state’s designation of any individual as an enemy terrorist.
Meyer’s invocation of a morality of war, discontinuous with the morality of ordinary life, runs contrary to the views of Finkelstein and McMahan. Although the latter two disagree with one another on what the morality of ordinary life has to say about intentional killing, both of them reject the notion of a special morality of war that displaces ordinary morality in wartime, and both affirm the idea that intentional killing in war can be justified in terms of the moral principles that govern ordinary life. In Meyer’s judgment, though, any attempt to justify killing in war on the basis of the morality of ordinary life is akin to trying to square the circle.
The theme of Part III is self-defense, and it begins with Craig Martin’s account of the concept of self-defense that is part of the post-World War Two system of international law. He focuses on the jus ad bellum rules regulating the resort to armed force by states. Under the U.N. Charter, states are prohibited from using such force unless it is undertaken in self-defense or authorized by the Security Council. Targeted killing has been justified by U.S. officials and many commentators as legally permissible on the basis of the Charter’s self-defense provision, but Martin finds such a justification wanting.
Tracing major historical developments in just war theory and in the legal rules governing the use of armed force, Martin argues that the legal permission for states to use armed force in self-defense is much narrower than the proponents of targeted killing suppose it to be. The architects of the U.N. framework consciously and explicitly sought to impose stricter limits on the use of force than the failed, pre-World War Two system had incorporated. In Martin’s view, essential to the framework is the idea that self-defense does not involve the geographically unlimited right of a state to attack any and every group that threatens it, wherever the group is operating. If the group is a non-state entity, such as a terrorist organization, then attacking it will, in the modern global order of sovereign states, amount to an attack on the political independence of the state in which the organization’s members are found, unless the state consents to the attack. Martin contends that, without such consent, the attack is a legally permissible exercise of self-defense only if there is a more substantial connection between the terrorist group and the state than the simple geographical presence of the group within the state’s borders. Accordingly, he holds that there is no legal right of self-defense against non-state entities as such, but only against those entities insofar as there is a sufficiently strong connection between them and the state in which they are located.
Although a legal right of self-defense so understood might seem unduly constraining on states, Martin argues that the narrow scope of the right is important for limiting the scourge of war among states. And, for all its horror, terrorism has not rivaled international war in the scale of death and destruction unleashed. Additionally, Martin argues that relaxing the rules on the use of armed force would risk undoing the carefully calibrated interplay currently obtaining between jus ad bellum rules and the rules of international humanitarian law, determining when an armed conflict exists and regulating how it is to be conducted. Under the existing system, permissible acts of self-defense automatically bring into play the rules of humanitarian law, because what counts as permissible self-defense is aligned with what counts as an international armed conflict. However, this alignment would be thrown out of kilter were the rules of permissible self-defense relaxed in the way that is recommended by defenders of targeted killing. Martin warns that the result could be a situation in which military force is used but not legally constrained by important humanitarian provisions designed to mitigate the suffering caused by war.
Like Waldron, Martin finds that the existing rules of international law do an important job of restraining the homicidal impulses that lead to war and are manifested in war and thinks that there are grave risks in loosening those restraints. As Martin sees it, the proponents of targeted killing, especially U.S. officials, have often been too quick to justify such killing on grounds that imply sweeping changes in the law and are insufficiently attentive to those risks. He might be open to the more calibrated argument and proposals offered by Ohlin and Maxwell, but Martin insists that any acceptable change stop well short of re-establishing the lax legal rules regarding the resort to armed force found in earlier eras.
One of the elements of permissible self-defense under the criminal law is the requirement that the defender refrain from exercising defensive force until the attack against her is imminent. This requirement seems to stand in the way of justifying targeted killing on the ground of self-defense, a point noted by Jeff McMahan and Claire Finkelstein. Russell Christopher agrees with McMahan and Finkelstein on that score and develops an extended set of arguments for the abandonment of the imminence requirement.
In Christopher’s view, the requirement suffers from a series of moral and conceptual problems. It has the unacceptable consequence that persons who are physically unable to respond quickly enough to defend themselves against imminent aggression are deprived of a right to effective self-defense. Many proponents of the imminence requirement respond that, based on objective evidence, it serves as a good way to sort out cases of the justifiable use of defensive force from cases of the unjustifiable use. Accordingly, the argument is often made that imminence serves as a good proxy for when it is truly necessary for a person to use defensive force to protect herself against aggression. But Christopher replies that imminence is a poor proxy for the necessity of resorting to force. There are not only situations in which defensive force is necessary prior to the point of imminence, as in the case of persons too slow to mount an effective defense if they wait until an attack is imminent; there are also situations in which defensive force is unnecessary even at the point of imminence, for example, when the aggressor has a last second change of heart and decides not to pull the trigger of the gun pointed at my head. Christopher argues that the proxy here is an inadequate substitute for the underlying moral principle.
Some proponents of the imminence requirement contend that it is not simply a proxy for the real normative criteria that distinguish justifiable from unjustifiable uses of defensive force; they argue that imminence is itself one such criterion. But Christopher responds that such arguments end up, contrary to the original intention of their advocates, relapsing into a proxy-based view of imminence.
One of the key arguments for the requirement is that it distinguishes the aggressor from the defender, with the imminence of an attack serving as the objective manifestation of aggression. But Christopher argues that imminence cannot adequately serve that function. As a strictly temporal notion that specifies a certain time period just prior to a certain use of force, imminence cannot capture what is essentially a moral distinction between aggressive and defensive force. Indeed, Christopher argues that if imminence is understood strictly in terms of time, then the imminence requirement will, paradoxically, end up treating some defenders as aggressors and some aggressors as defenders.
In Christopher’s view, it is to no avail to add an action component to the temporal understanding of imminence, so that the aggressor would be the party who, within the specified “imminent” time period, first physically manifests the use of force against the other party. An action component brings with it its own problems. For example, Christopher argues that it unduly restricts the right of self-defense and unjustifiably favors aggressors by requiring defenders to wait, not only until an attack is temporally imminent, but also until the aggressors physically manifest their aggression during that period. In light of these and other problems with the imminence requirement, Christopher concludes that the requirement provides no valid ground on which to object to the permissibility of targeted killings.
Phillip Montague argues that ordinary moral principles regulating self-defense provide a justification for the targeted killing of terrorists, including the killing of those persons in terrorist groups who aid and assist the individuals who wear and trigger suicide belts or who otherwise directly kill their civilian victims. He points out that the rationale for an individual’s right to kill an aggressor in self-defense rests on a principle that also grounds the right of a third-party to kill the aggressor in other-defense. Rejecting the imminence requirement along with Christopher, Montague contends that the aggressor’s right to life has been forfeited if it is inevitable that the aggressor will culpably kill the victim unless someone kills the aggressor first. It is permissible for anyone, not only the intended victim, to kill the aggressor in this situation. However, in some circumstances, there are more characters in the story than simply the aggressor, the intended victim, and, possibly, a third-party other-defender. The aggressor might be deliberately enabled in his designs by an accessory. Indeed, there might be a whole raft of such culpable enablers acting jointly. The culpable enablers are not themselves aggressors, and so the paradigms of individual self- and other-defense do not apply. But killing one or more of the enablers might be necessary to prevent the aggressor from killing his victims. On the other side, the intended victims might need the assistance of a whole raft of individuals acting jointly in order that lethal aggression be prevented. It is precisely this kind of scenario, with joint action on the sides of both the aggressor and defender, that many targeted killings involve, Montague argues, and so the concept of joint action becomes crucial in showing how principles of self-defense ground the conclusion that such killings can be justifiable defensive homicides.
If culpable individuals act jointly for the purpose of bringing about the death of an individual who has a right to live, and a “closed choice” situation exists in which either the intended victim will be killed or at least one of the culpable joint actors will be killed, then, Montague argues, it is permissible for someone to kill the culpable actors in whatever number is necessary and sufficient to prevent the killing of the intended victim. Targeted killings can be justified on the basis of such a principle, which does not involve any assumption about the existence of a state of war or a special morality of war.
Part IV examines various aspects of the operational implementation of targeted killings and begins with Amos Guiora’s account of how targeted killing does, and should, work on the ground. Reflecting on his experience as a legal advisor in the Israeli Defense Forces, Guiora argues that the decision-making process culminating in the official authorization of the commander on the ground to commence lethal action must be regulated and constrained by certain procedural norms, else the process will fail to reliably respect the requirements of law and morality. Guiora calls these norms “criteria,” and he contrasts his criteria-based approach with “intuitionism,” which he understands as the view that the final decision to proceed with lethal action should be left to the discretion of the commander on the ground. Intuitionism fails to place adequate constraints on the commander’s discretion and places an excessive decision-making burden on the commander’s shoulders. The commander is in good position to decide when to request a go-ahead, but the request should be reviewed by a legal advisor for approval or disapproval.
The legal advisor’s job is often carried out under severe time and information constraints. Guiora argues that under such conditions, an advisor should have a checklist of questions to ask the commander. The checklist would not provide an algorithmic decision procedure for the legal advisor, but it would serve to concentrate his or her attention on the legally relevant variables so that a reasoned and defensible decision could be reliably made. In the absence of a procedure, such as a checklist, to regiment decision-making in the targeted-killing scenario, unjustifiable deviations from important rule-of-law principles—such as treating like cases alike—are bound to occur at an unacceptable rate.
Greg McNeal’s contribution tackles key empirical claims made by those who criticize drone attacks against suspected or known terrorists. Among those claims is the charge that U.S. drones, particularly those operating in Pakistan, have killed many times more civilians than terrorists. The drone program in Pakistan is operated by the CIA, and it remains shrouded in secrecy. However, a lawsuit has brought to light the nature of the program in Afghanistan, operated by the U.S. military, and McNeal points to that program in order to cast doubt on the claims of drone critics.
Prior to a drone strike, McNeal tells us, the U.S. military goes through a process that involves a scientific analysis of the likely blast damage that a strike would cause and an estimation of collateral causalities from such damage. Steps are then taken to mitigate the likely damage by altering various parameters of the strike, such as time of day. On McNeal’s analysis, this process is highly effective: when followed, less than one per cent of the attacks have resulted in collateral casualities. Moreover, if, notwithstanding the process of estimation and mitigation, one or more civilian casualties are expected, then the strike must be approved by the President or the Secretary of Defense, due to the potentially troubling foreign-policy ramifications of such a strike. McNeal finds it difficult to believe that the CIA program would permit attacks that are as indiscriminate as charged, even as the military’s program is so discriminating, especially because potentially troublesome foreign-policy implications attend civilian death from drone strikes in both Afghanistan and Pakistan. Moreover, a high-ranking military official, the head of CENTCOM (Central Command), is responsible for the drone programs both in Afghanistan and in Pakistan, and McNeal argues that it is illogical to think that the CENTCOM commander would have radically different standards for allowable collateral casualties for the CIA program than for the military’s program.
McNeal proceeds to take to task a prominent drone critic, Ellen O’Connell, for inconsistency in her use of sources, relying on a source when it supports her view of the indiscriminate nature of drone strikes but rejecting the same source when it contradicts her view. He also argues that O’Connell demonstrates a lack of familiarity with the way that the targeting process works. Contra O’Connell’s assumptions, potential targets are tracked by the drone’s camera for at least 24 hours and, in order to firmly identify the target, the information from the drone tracking is put together with intelligence from a range of human and technical sources. McNeal also charges O’Connell with having produced no evidence to substantiate her assertion that Pakistan has not consented to U.S. drone attacks on terrorists in its territory.
Kevin Govern provides a close look at the U.S. Navy Seal operation that resulted in the killing of Osama bin Laden. Code-named “Neptune Spear,” the operation was chosen from options that included an air strike on the compound where bin Laden resided in Pakistan. President Obama decided that the operation would proceed without the approval or even the notification of the Pakistani government, due to a lack of confidence that Pakistani officials would maintain the secrecy of the plan. In the course of the Navy Seal’s raid on the compound, three people were killed, in addition to bin Laden himself. According to a report quoted by Govern, the Seal team was intent from the start on killing bin Laden: capturing him alive was not an option. If that report is accurate, then the moral permissibility of the killing would be undermined in the eyes of some observers. However, in his contribution found in Part V of this volume, Michael Moore contends that bin Laden’s killing can be justified on retributive principles: The past conduct of the Al Qaeda leader merited death as punishment. As we have seen, Jeff McMahan rejects retributivist justifications of targeted killing, but he thinks that it would not have been unreasonable for President Obama to have concluded that any considerations in favor of capturing bin Laden were outweighed by the prospect that his followers would have kidnapped and killed, one by one, American hostages, in a futile effort to win their leader’s release.
Kenneth Anderson’s contribution replies to a commonly voiced objection to the use of drone-fired missiles for targeted killing. The objection is that, because of its precision in killing the intended target and avoiding collateral deaths, drone technology makes the resort to armed violence too easy. Put in the economic language of efficiency: although the use of drone missiles is efficient in killing the enemy without causing the deaths of many innocent bystanders, precisely the “in bello” efficiency of armed drones leads to the “ad bellum” inefficiency of excessive resort to the use of armed force in the first place. Among the reasons offered for this supposed inefficient outcome is that the drone operators work far from the battlefield in complete safety, thus removing one of the disincentives that political leaders ordinarily have for deciding to resort to armed force.
It is clear that not all targeted killing involves drone attacks, as the bin Laden case illustrates. But Anderson points out that it is also mistaken to think that all drone strikes are targeted killings. Even though drones can and do target individuals identified in advance (so-called “high-value targets”), the technology has also been used by the U.S. military in conventional attacks against columns of Taliban, where the only relevant distinguishing feature of the victims is that they are members of the enemy force. Anderson finds it reasonable to assume that drone missiles—whether used for targeted killings or conventional attacks—are in fact more precise than alternative weapons in discriminating the enemy from civilian bystanders. He also emphasizes that this greater precision would not be possible unless ground-level intelligence were effectively integrated into the use of weapon systems: technology alone would be insufficient.
Anderson points out that it would seem that a military technology that was highly discriminating in distinguishing enemy fighters from civilians and, at the same time, provided protection for one’s own forces, would be a non-controversial moral gain in the campaign to reduce the horror of warfare. Harm to civilians is more likely to occur when forces are under attack or otherwise insecure. But the drone critics argue that what appears to be a non-controversial moral gain is in fact a moral loss, because the availability of drones will make a country too willing to resort to armed force in the first place. The price of precision and force protection is, the critics argue, too much to pay due to the increased resort to armed violence.
Anderson is skeptical of various aspects of this argument. He contends that the degree of risk to the personnel of their own forces is only one of several considerations that political decision-makers take into account in determining whether to resort to armed force and that the most important considerations pertain to national security rather than force protection. The one major exception, Anderson argues, is armed humanitarian intervention, which is likely to occur more frequently with the increased force protection that drones allow. And he concedes that it is likely that, as armed drones become cheaper and more widely available around the world, there will be an increased propensity of states to use armed force, especially in discrete and limited ways, for dealing with international disagreements. But he doubts that there is any neutral standard by which one can judge that the resort to armed force has become too easy or that the total amount of armed force used in the world is inefficient, or efficient. The opposing sides in an armed conflict have competing substantive standards for judging when the resort to force is desirable, and one side’s value-promoting use of drones will be the other side’s value-destroying resort to armed force. Anderson points out that the idea of efficiency makes sense in a context where each side to a transaction aims to better its position in terms of some shared scale of value, such as money. Thus, bargaining produces an efficient outcome when one side buys out the other—for example, a railroad buys a farmer’s right not to have his crops exposed to sparks flying off the wheels of the railroad’s trains—and each side gets what it values the most. But Anderson argues that the idea of efficiency makes no sense in the context of armed conflict, where the notion that one side could buy out the other is preposterous.
Part V addresses the interplay between the desirable outcomes that targeted killings seek to achieve, such as the destruction of terrorist groups and networks and the minimization of the total killings of innocents, and the limits that moral principles place on the ways in which desirable outcomes are permissibly pursued. Fernando Tesón’s contribution opens the section, arguing that terrorism is a threat that is sui generis and cannot be adequately addressed by either the law-enforcement or the armed-conflict (that is, what he calls the “just war”) models. For Tesón, terrorism is an especially noxious and dangerous form of evil, different from ordinary crime in the extent of wrongful harm that it intentionally threatens to cause and in the non-self-interested motives that drive it. However, Tesón resists the idea that liberal states are in global war with terrorists and insists that liberal states remain true to the moral principles that underlie their institutions, even as those states fight the terrorist scourge.
Tesón regards some terrorists as being in a state of war with liberal states, and he believes that it is morally permissible to target those terrorists. Such terrorists are to be found in a combat context, such as exists in parts of Pakistan and Afghanistan. But terrorists in Paris or New York are outside of any combat context and cannot, consistent with sound liberal principles, be killed on sight. Rather, it is morally permissible to target them only if they have been given an opportunity to surrender and killing them is necessary to save many innocent lives. Nonetheless, because even liberal states are prone to mistake in identifying when a targeted killing is necessary and because the liberal principle of the rule of law suggests that, regardless of its possible good consequences, the practice of targeted killing would constitute a form of liberal vice, Tesón argues that there should be an absolute legal ban on targeted killing in a peacetime setting. In a final twist, he allows that the President should have the authority to suspend the ban in an emergency, as long as the suspension is publicly declared and publicly justified.
Michael Moore approaches the question of the morality of targeted killing by developing a general theoretical framework for answering questions about what morality permits, forbids, and requires. The framework consists of three levels. At the first level is the consequentialist principle that one ought to choose that action from among the alternatives that will bring about the greatest net good in the world. This principle is decisive in determining the moral status of an action across a wide range of cases, but it can be preempted by principles that belong to the second level. At that level are the deontological principles that specify “strong permissions,” which allow an agent to choose an action that does not have the optimal consequences, and “strict obligations,” which require the agent to perform an action that does not have the optimal consequences. The right of self-defense is a strong permission, and the prohibition on deliberately killing one innocent person in order to save, say, three other innocents is a strict obligation. Normally, such permissions and obligations are decisive when they apply to a case, but at the third level of Moore’s framework is a principle that overrides those permissions and obligations in cases in which catastrophic consequences would otherwise ensue.
Using his framework, Moore constructs a decision tree that shows at what point in the deliberation process, and how, consequentialist and deontological considerations are to figure in a decision about the moral permissibility of an action. The tree is not meant to constitute an algorithm that can be mechanically instantiated, but rather as a guide for coherent and cogent moral judgment. And deploying his decision tree, Moore finds that the targeted killings of suicide bombers and those who plan and otherwise aid such bombers are easy cases of justifiable killing. Additionally, he finds that the targeted killing of non-culpable individuals can be permissible, for example, in a situation in which a terrorist organization places a bomb in the brief case of an unsuspecting person who is (unknowingly) about to set off an explosion in a crowded railway station by opening the brief case. Although the person is not in the least blameworthy, killing him in order to prevent the explosion is justifiable. Even more controversially, Moore claims that morality permits targeted killings as punishment for those terrorists whose past conduct makes them deserving of the death penalty.
Leo Katz exposes and explores some of the puzzles and paradoxes that arise from the norms that govern killing. Unlike Michael Moore, Katz does not develop a systematic moral view, but rather focuses on our normative thinking when it is in the trenches, grappling with particular scenarios and cases. He begins by imagining a scenario in which an elite anti-terrorist squad deliberately provokes a known terrorist, so that they can then beat him to the punch and kill him in self-defense, without transgressing the imminence requirement for self-defense. The criminal law and most commentators would disallow the appeal to self-defense in this kind of scenario, but Katz does not buy into their view. He believes that in some cases, agents should not be disqualified from invoking a certain defense (self-defense, necessity, duress, etc.) even if they contrived to create, for the very purpose of performing an action for which they would be otherwise liable, conditions under which the defense applies. The agents of the elite squad have deliberately created just such conditions for themselves, and Katz thinks that, paradoxically, they have a valid self-defense claim.
One might think that this kind of contrived self-defense is dubious because it seems to be on all fours with a case involving another kind of contrived defense that is clearly invalid. A man hurls himself out of a window in order to land on, and kill, a certain pedestrian on the street below and then claims that his landing on the victim was an involuntary act because he had no control of his body’s fall at the moment of impact. But Katz counters that the two sorts of cases can be distinguished on the ground that the death in the case of his anti-terrorist squad was brought about by a causal chain whose links were broken by the intervening agency of the person who was killed, while the causal chain in the leaping-man case was unbroken. One might reply that the decisive factor in self-defense is the necessity of using force for self-protection and that the anti-terrorist squad did not need to use force because they did not need to provoke the terrorist in the first place. But Katz thinks that this appeal to necessity runs into the problem that the necessity of using force is always relative to the time frame in which the act of defensive force is placed. For any act of defensive force there is always some previous moment such that the act would have been unnecessary had the agent—or someone else—done something differently at that previous time. A woman has a right of self-defense against her partner’s violence, even if it would have been unnecessary for her to use defensive force against him had she decided at the start of the relationship that it was too risky to pursue due to the partner’s known history of violence.
Additionally, Katz argues that there are deeper reasons to think that contrived defenses cannot be easily banished: they rely on certain cyclical orderings that are entrenched in systems of normative rules. If the only way for me to avoid losing a dollar is to kill the thief who demands it, then the law says that I must hand over the dollar. So the thief’s life gets preference over my money. But if I would rather get killed as a result of refusing the thief’s demand than hand over the dollar, then the law says that I am permitted to refuse, in which case my money is preferred to my life. And if the thief attacks me with lethal force, the law says that I am permitted to kill him rather than let him kill me. Thus, we have a cyclical ordering: 1) the thief’s life is preferred to my money; 2) my money is preferred to my life; and 3) my life is preferred to the thief’s life. And this ordering can be exploited, Katz argues, so that, paradoxically, I am permitted to kill an aggressor in preference to giving up my money. I do this by appealing to steps 2 and 3 in order to defeat 1. Accordingly, I permissibly act so as to prefer my money over my life by telling the thief who has a gun at my head to get lost, and then I permissibly save my life by killing him when he starts to pull the trigger. Risky, to be sure, but permissible if I can carry it off.
Katz proceeds to consider a scenario involving further twists in the permissible use of lethal defensive force. The aggressive, albeit misguided, followers of a peaceable holy man are about to launch an attack against us which we are entitled to preempt with force, even though many innocent bystanders will be killed. The standard view is that it is impermissible to deliberately kill the holy man, even if doing so would so demoralize the aggressive followers that they would call off their attack. But Katz is doubtful. He argues that the case is no different in its moral essentials from one in which we permissibly redirect a trolley onto a side track, thereby killing one innocent person, in order to save the lives of a number of innocents who would otherwise be run over and killed on the main track in the course of our permissibly using the trolley to strike at aggressors who are about kill us. Now just imagine that it is the holy man on the side track and that it is his followers, surrounded by innocent bystanders, on the main track, and the conclusion is difficult to avoid that deliberately killing the holy man in the original, trolley-free scenario is permissible.
Katz opines that these and other counterintuitive conclusions about defensive killing have their source in the fact that there are multiple criteria that decisions about killing seek to satisfy. This fact makes those decisions analogous to collective decisions that rest on an aggregation of individual preferences, and Katz points out that, due to Arrow’s theorem and other results in social-choice theory, it has long been known that collective decisions cannot entirely escape paradox and irrationality. So while Michael Moore systematically combines consequentialist and deontological criteria so that we might rationally address questions of targeted killing and other contested moral issues, Katz suggests that, when we apply any such system in the trenches, it will at some point lead us into paradox and irrationality.
The taking of human life is the most serious of matters addressed by legal and moral norms. Prohibitions on homicide are part of every legal and moral system, but, aside from pacifist moral codes, the taking of human life is permitted, under certain highly limited conditions, by every moral and legal system. It is in this zone where the general prohibition starts to give way to the highly limited exceptions that the issue of the targeted killing of terrorists arises. Does such killing fall within the prohibition? Does it fall within one of the exceptions? What is the scope of the exception that permits the use of defensive force? Are there special exceptions that apply only in the context of war? These questions are only the beginning, however. In addition, there are questions arising from the complex interplay of moral and legal principles. Should the legal principles governing targeted killing restate the applicable moral principles? If not, what should the legal principles be and why?
The foregoing questions are among those at the center of public discussion and debate about targeted killing. This volume contributes to that discussion the carefully developed thoughts of scholars from a range of backgrounds. Those thoughts are not always in harmony with one another and will not end the discussion. But it is hoped that they will help to raise it to a higher level.