The Nature of Drone-Based Violence

The drone strikes that the U.S. Central Intelligence Agency (CIA) reportedly carries out have occurred mostly inside Pakistan. Unofficial sources indicate that, using uninhabited Predator and Reaper aircraft, the CIA has prosecuted more than 350 strikes there since 2004.Footnote ¹ Drone strikes against individuals identified as terrorists are also reportedly underway in Yemen and Somalia, but on a much smaller scale. When a drone is flying over a particular location within the territory of one of these states, the aircraft and its onboard weapons are controlled via satellite by a U.S. agent seated in a control station thousands of miles away. Assisted by an analyst sitting immediately adjacent, the drone operator uses a keyboard and a joystick-like steering device while monitoring screens that display live video feed, a navigation map, and technical data on the drone.Footnote ² The agent's act of killing—pressing a button that causes a munition to detach from the drone's wing and fly to a target beneath it—thus takes place in the absence of any physical risk to that agent or any U.S. personnel.

For the purpose of informing ethical analysis of this set of actions and circumstances, an important question is: do drone strikes count as war? Every military organization and every individual military professional understandably prefers to fight in a way that involves a maximum of risk to the enemy and a minimum of risk to one's own side. To be able to kill from a great distance is, as Shane Riza describes it, “The Oldest Dream of the Second-Oldest Profession.”Footnote ³ Accordingly, numerous lethal technologies have been introduced that, by avoiding hand-to-hand combat, reduce the physical risk experienced by their user: the slingshot, the longbow, the rifle, the submarine, and so on. However, when a mode of killing is risk-free to the individual killer, it is worth asking whether “war” is going on at all. This is a critical question because of the relationship between violence and ethics. War is a state of affairs, but “war” is also a political term of art that potentially bestows legitimacy upon some forms of violence. Ethics is thus constitutive of the practice of war as a form of violence that is morally distinguishable from other forms—for example, from violence carried out for law enforcement or murderous purposes. Although the character of war—how, by whom, and for what purpose it is waged—can and does change, the nature of war is immutable: war, to be war, must be a contest. In the nineteenth century, the Prussian general Carl von Clausewitz wrote that it is “the element of the thing itself” that “war is nothing but a duel on an extensive scale,” and he likened it to a match between “two wrestlers,” each of whom “strives by physical force to compel the other to submit to his will.”Footnote ⁴ Perfect symmetry (as in a wrestling match or a game of chess) never occurs in war, and just war thinking has never required war to be a fair or evenly balanced fight, but it does require that a fight of some sort is going on.

Fundamentally, according to Clausewitz, “War is always the shock of two hostile bodies in collision, not the action of a living power upon an inanimate mass, because an absolute state of endurance would not be making war.”Footnote ⁵ Arguably, the U.S. government places drone operators in such a state by having them apply force in a radically asymmetric fashion. If so, drone technology is better conceived as not merely transforming the character of war (as so many other technologies have done in the past) but rather as enabling a form of violence so fundamentally different in nature that it is difficult to conceive of as war. Whereas the mutual experience of physical risk is elemental to any violent contest, a one-sided experience of risk resembles merely (to use Clausewitz's words) “the action of a living power upon an inanimate mass.”

It is important to acknowledge that, in the broadest sense, drone-based violence is not entirely uncontested and thus not absolutely risk-free for the United States. American civilians, for example, could be targeted by terrorist enemies who are enraged by drone strikes, and it is possible to imagine any number of ways in which diverse U.S. interests globally could be harmed by someone striving somehow to fight back. For this reason, and for present purposes, it is sufficient to test the limits of the concept of war by reference only to the condition of those who directly perpetrate or suffer violence. My focus, in other words, is on the experience of physical risk when and where killing and dying happens.

Following the 1999 NATO intervention in Kosovo, during which U.S. and allied casualties were avoided by high-altitude bombing, Michael Ignatieff contrasted “war” with what he regarded as “a turkey shoot”Footnote ⁶ in the Balkans. Paul Kahn, referring to the same set of circumstances, observed that “without the imposition of mutual risk, warfare is not war at all.”Footnote ⁷ In his view, combat establishes “a relationship of reciprocal risk” in which killing is “linked to a willingness to be killed.”Footnote ⁸ This would indeed be the case if, following Clausewitz's analogy, two individual “wrestlers” were willingly engaged in a fight to the death. But whereas Kahn appears to focus on the individual combatant (who kills or is killed), Clausewitz was referring to the metaphorical “wrestling” that is collectively engaged in by an entire army. Some modern-day scholars insist that when discussing the use of drones “the relevant threat” is that which exists “at the level of armed forces considered as a whole,” rather than between individual combatants.Footnote ⁹ However, the image of armed forces arrayed against each other is not generally apposite in the war on terror. Nor does it accommodate specifically the highly individualized character of drone-based violence whereby, strike by strike, one individual places another in the crosshairs, and so America's enemies are dispatched one by one. For the purposes of this article, therefore, a focus on individuals’ experience of risk at either end of a drone strike is at the very heart of the matter. Projecting strategic power in a way that avoids physical risk to individual users of force is the raison d'etre of this mode of remote-control killing.

When contemplating modern-day violence using advanced military technologies, it is perhaps too much to require, as Kahn does, that the only morally relevant source of physical risk is the reciprocated threat of an opposing combatant. But in order that violence of any kind may plausibly be conceptualized as war, it is surely a minimum requirement that a would-be killer experiences some risk that is intrinsic (and not merely incidental) to the process of killing. In certain circumstances, the sheer dangerousness of a situation would be enough. For example, a submariner deep underwater who fires a torpedo or a pilot high in the air who drops a bomb can expose others to violence without fear of reciprocation, but the very participation in such activities imposes significant risks. By contrast, ground-based drone operators deep inside U.S. territory are exposed neither to the violence of others nor to an inherently dangerous environment, and this is hard to reconcile with the traditional expectation that warfare is a contested activity. In a 2013 article, Michael Walzer argued that “the easiness” of killing by drone “should make us uneasy.” Drones are “a dangerously tempting technology,” as they “[make] our enemies more vulnerable than ever before, and we can get at them without any risk to our own soldiers.”Footnote ¹⁰

Related to Walzer's last point, Tony Coady has argued that “in the conceptual and moral analysis of war . . . the target's incapacity to fight back [should not] absolve the attacker of the need to justify the attack by the moral standards appropriate to war.”Footnote ¹¹ However, it seems both unreasonable and futile to insist upon the application of just war rules to that which, in Clausewitzian terms at least, is a nonwar. If, instead of war, drone strikes resemble “a godlike power to call down destruction from the skies,”Footnote ¹² the rules for restraining such strikes would need to be derived from a different concept of violence.

One alternative might be to conceptualize drone-based violence as law enforcement. In that paradigm, a radical imbalance of lethal power is sometimes legitimate. Circumstances of “insurmountable tactical asymmetry” present no problem, as Kahn argues, provided that the “ethos of policing” (rather than the ethos of warfare) applies: “The criminal has no right to use force against those seeking to enforce the law.”Footnote ¹³ Similarly, when judicially-authorized capital punishment is carried out, the moral inequality between the person killing and the person killed permits the former to use force without experiencing physical risk.

Are drone operators thus better likened to police officers or executioners than to warriors? Since the war on terror began, the language of the U.S. government has often contained mixed messages regarding the moral status of America's use of force, with the result that it has variously taken on the appearance of war, law enforcement, or some exotic mixture of the two. Although there may be political advantage to be derived from attempting to straddle the war and law enforcement paradigms, the ethical principles that apply within one are incompatible with those in the other, and the likelihood is that neither paradigm can maintain moral purchase on drone-based violence. For just as it is difficult to conceptualize the use of armed drones against individual terrorists as “war,” the portrayal of such violence as the stuff of “crime and punishment” is for at least three reasons even less plausible.

First and most obviously, there is a fundamental problem with the idea of enforcing (as against an individual) a criminal law in the international realm. Sovereign states with a monopoly on violence inside their territory may violently enforce law there because their authority to do so is established. This is less clearly the case when one state purports to engage in law enforcement on an extraterritorial basis, although international war-crimes tribunals are a rare example of how this can legitimately occur. Even if the domestic/international distinction is set aside, a second problem is that the United States, as the (would-be) victim of a terrorist attack as well as the judge and executioner, does not have the impartiality that is usually required of those who work in a system of criminal justice. Third, drone-based violence necessarily bypasses processes and responsibilities that are essential to the law enforcement paradigm. If, pursuant to that paradigm, state action against terrorists is based on the notion of criminal culpability (as distinct from dangerousness in war), such framing triggers the application of peacetime human rights norms, including the right to a fair trial and protection against the arbitrary taking of life.Footnote ¹⁴ The dangerousness of a violent criminal—on a rampage and resisting arrest—may justify suspending the requirement of arrest (preceding trial and possible punishment), but this is so only when the use of lethal force is necessary to protect law officials and/or innocent bystanders.Footnote ¹⁵ By contrast, in the way a U.S. drone strike is carried out in, say, Pakistan, the killing done is preventive rather than reactive, and without the immediate need for a drone operator to defend himself or others. More generally, whereas (lethal) punishment ought always to be an end-point in a process of law enforcement, killing is at once the foremost and ultimate rationale of a drone strike.

If it is difficult to conceptualize drone strikes as law enforcement, and if they do not constitute war (qua war-as-contest), we are left with the problem of lacking a conceptual basis for restraining the resort to this unfamiliar form of violence. There might be some advantage, therefore, in conceiving of such strikes as something “between” law enforcement and war, and accordingly to considering the suggestion by some scholars that the development of a jus ad vim (just resort to force) framework could serve to fill the ethical vacuum. Such a solution would involve equating drone-based violence with the concept of vim and, prior to that, establishing vim as possessing a distinct meaning that is sustainable in practice.

In a previous issue of this journal, Daniel Brunstetter and Megan Braun called for a jus ad vim framework to be developed that could capture modern forms of violent action that do not fit within the rubric of jus ad bellum.Footnote ¹⁶ In seeking to distinguish “war” and “force short of war” for ethical purposes, the authors address the genuine concern that violence in world affairs is sometimes ungoverned and that unjust harm can sometimes be inflicted with impunity. If the nature of a drone strike, for example, is such that it is something other than or short of “war,” does it escape judgment according to principles of justice applicable only to war? Brunstetter and Braun emphasize “the important distinction between force short of war and war,” and they describe the ethical contexts raised by the former as “unique.”Footnote ¹⁷ Indeed, only by drawing a clear distinction between these two forms of violence can jus ad vim have any meaning and purpose.

From the outset, however, this is made difficult at the conceptual level because Brunstetter and Braun define vim negatively—that is, as the absence of war. Logically, the knowledge of what something isn't does not nearly equate to understanding what it is. Thus, to define vim as force “short of war” is to be cryptic about the precise nature of such force. One is left wondering what the essence of the thing is, and the definitional void is only partially and contingently filled by dealing in examples. This is unsatisfactory if “force short of war” in some way involves actual violence that harms people and damages property; the prevention or mitigation thereof is not well served by an understanding only that that force is not war. Indeed, when bodies are bleeding and buildings are exploding, a negative definition would seem to be unhelpful and, in an immediate sense at least, implausible. One actor's vim could readily be conceived as another's bellum.

Because the vim/bellum distinction is in practice so often hard to discern, the tendency of some scholars is instead to fall back on a specious small/large distinction. It becomes, then, a distinction of degree rather than kind. The focus is on the quantum rather than the nature of violence, with large (or larger) amounts vaguely associated with “war.” Steven Lee, for example, defines war as “large-scale armed conflict between states or other large organized groups,”Footnote ¹⁸ and for Coady war is “the resort by an organized group to a relatively large-scale act of violence for political purposes to compel an enemy to do the group's will.”Footnote ¹⁹ Similarly focused on scale, Brian Orend argues that “the onset of war requires a conscious commitment and a significant mobilization on the part of the belligerents in question. There's no real war so to speak until the fighters intend to go to war and until they do so with a heavy quantum of force.”Footnote ²⁰ These authors' ideas provide ample precedent for Brunstetter and Braun to adopt the same approach in advocating the development of a jus ad vim framework, referring as they do to “the large quantum of force associated with war” and the “widespread destructive consequences of war.”Footnote ²¹ Throughout their article, the term “war” is routinely preceded by such qualifiers as “widespread,” “large-scale,” “full-scale,” and “full-blown.” However, by discussing war only in this way, the distinction being drawn is really between “large-scale war” and “war-short-of-large-scale-war” rather than between “war” and “force-short-of-war.” The trouble with the former distinction is that it is highly vulnerable to arbitrariness and contestation in the delineation between “small” and “large.” Most obviously, a war that seems small-scale to the stronger side may seem large-scale to the weaker side. A small/large distinction is thus too fragile a basis for jus ad vim to stand as a credible moral framework accessible to all would-be users of force. For vim to be meaningfully distinct from bellum, such that jus ad vim can have practical purchase as a moral framework when jus ad bellum cannot, there has to be more to the story than quantum of force alone.

If, by this process of reasoning, drone strikes cannot be rendered governable by conceptualizing them as vim, the ethical problem of impunity remains. Beyond the conceptual realms of war, law enforcement, and vim, a governance solution other than outright prohibition might one day be found. For now, however, I aim to introduce an alternative conceptualization of drone strikes, one that envisages yet more potential for injustice without remedy: the idea of such strikes as a kind of vis perpetua (perpetual force). I argue that the immorality of drone-based violence is attributable to its status as an exercise in risk management. The chosen means of violence involves the deliberate and systematic transfer of military risk away from U.S. combatants and toward foreign noncombatants, and the potential therein for injustice is compounded by the pursuit of a strategic end that is intentionally not time limited.

Justice and Perpetual Force

When the U.S. government uses drone strikes and other violent methods to engage its terrorist enemies on a global scale, its narrative is one of war: a war on terror. As discussed in the first part of this article, such framing purportedly enables the United States to avail itself of the special moral permissions that are traditionally afforded to war fighters. In addition, however, the notion of war necessarily implies the notion of peace as a state of affairs that follows victory (by one side over the other) or a political settlement that otherwise brings violence to an end. At the conceptual level, just as war is by nature a contest, so too is it something that exists in opposition to peace; the existence of one is verified by reference to the other. The ethical significance of this traditional dichotomy lies in the promise that war, and all the misery that attends it, is temporary. War is instrumental to achieving a just peace but is not an end in itself; there must be, one could say, a time for war and a time for peace.

Violence of a kind that is intended to be used unendingly is therefore difficult, for ethical purposes, to conceptualize as war. But that is not to say that all such violence is necessarily illegitimate. For example, if one accepts that the problem of criminality within a given society cannot be eradicated but only controlled, the occasional yet repeated resort to violence by police officers and executioners is legitimized by the perceived need to “manage” crime on an ongoing basis. Critically, and only in a peacetime context, this open-ended commitment is tempered by strict limitations on the use of violence for law enforcement purposes. The problem of terrorism, by contrast, is difficult politically to categorize as “crime,” and the idea of (merely) “managing” it has at times been controversial. Arguably, however, U.S. counterterrorism efforts (including CIA drone strikes) have been conducted according to a logic of risk management—one that deliberately ignores limits of space and time. And although the idea of war is evoked during such efforts, the war/peace dichotomy is denied in practice, and this inconsistency carries the potential for injustice.

In September 2001 the U.S. Congress authorized the use of “all necessary and appropriate force” in order to prevent future acts of terrorism against the United States. Two salient features of the 2001 Authorization for the Use of Military Force (AUMF) are, first, that it grants the U.S. president sweeping power to determine who counts as an enemy and, second, that it does not impose geographical limitations of any kind.Footnote ²² Accordingly, drone strikes have been conducted against insurgents in Iraq and Afghanistan, but also against individuals identified as terrorists in countries where the United States is not at war: in Pakistan, Somalia, and Yemen.

In a move also consistent with the AUMF, recent U.S. presidents have exercised their power to loosen certain limitations on the use of force. In 2008, as part of a dramatic expansion of the Pakistan drone campaign, the CIA was reportedly authorized to attack not only high-value individuals whose names are on an approved list (so-called “personality strikes”) but also suspected militants of lower value whose identities have not been confirmed.Footnote ²³ This expanded authority, granted by President George W. Bush and maintained by President Barack Obama, permits the CIA to rely on “pattern-of-life” analysis using information about individuals and locations collected by cameras mounted on drones and from other sources. In the words of one senior U.S. official (speaking anonymously): “We might not always have their names but . . . these are people whose actions over time have made it obvious that they are a threat.”Footnote ²⁴ As a further example of the increasing reliance on only circumstantial evidence in determining targets, it was reported in 2012 that President Obama had embraced a method of counting casualties that “in effect counts all military-age males in a strike zone as combatants . . . unless there is explicit intelligence posthumously proving them innocent.”Footnote ²⁵

The transcending of identity- and geography-based limitations on the use of force in combating terror has clearly been facilitated by drone technology. However, as the U.S. government carries out its plans to increase the scale and intensity of its drone use, the issue likely to come to the fore is that of temporal limitation: for how long is this form of violence intended to continue? From an ethical perspective, this is a vital question. As the death and destruction resulting from violence are prima facie wrongs, they can begin to be legitimized by trading them away only temporarily in the expectation that the promise of a better peace will thereby be fulfilled. By contrast, the intentional waging of an endless campaign of violence using drones would be an ethical impossibility: a kind of permanent or perpetual force (vis perpetua), indefinitely subordinating right to might. In the study of physics, “perpetual motion” describes motion that continues indefinitely without any external source of energy, and it is held to be impossible in practice because of friction and other sources of energy loss. Likewise, for the purposes of this article, vis perpetua is not a label to describe the permanent application of force as a matter of fact. Force originally described in this way could, due to internal and/or external factors (“friction”), later slow to a halt or be brought to an abrupt end. Rather, the morally significant factor to which the term vis perpetua refers is an actor's intent to sustain (perpetuate) the use of force in a way that is temporally unlimited.

Initially, under the presidency of George W. Bush, the potential for a war-on-terror-without-end was mostly bound up in the idea that America was engaged in a contest between good and evil. In the week following 9/11, for example, Bush declared: “I have faith in our military. And we have got a job to do . . . we will rid the world of evil doers.”Footnote ²⁶ Such powerful rhetoric served the immediate purpose of reassuring and motivating a demoralized citizenry, but in reality the “job” of defeating evil is in practice so boundless a task as to be an impossible one. However, Bush's successor, Barack Obama, took care to explain his country's participation in the war on terror in less ambitious terms. In May 2013, amid mounting political pressure to detail and justify his government's not-so-secret drone campaigns in Pakistan, Yemen, and Somalia, Obama delivered a major speech asserting that:

By so abandoning his predecessor's ambition to “rid the world of evil doers,” Obama purported to address domestic concerns about “a perpetual war on terror.”Footnote ²⁸ As the president continued, he outlined his intentions

Here the president was employing simultaneously the language of termination and continuation. A counterterrorism adviser to President Bush later told the New York Times: “This is both a promise to an end to the war on terror, while being a further declaration of war, constrained and proportional in its scope.”Footnote ³⁰ Thus, in the U.S. government's “systemic effort” to “continue” targeting terrorists—albeit not in the manner of Madison's “continual warfare”—the potential for vis perpetua remains.

In large measure, this potential is a function of the powerful and enduring temptation to violence offered by drone technology. Political friction—in the form of domestic concern about strategic objectives and friendly casualties—can hinder or halt a state's enterprise of political violence. By contrast, in the case of the U.S. drone program, such friction is largely avoided by pursuing an objective (management) that is sustainably modest and by using a method (remote-control killing) that exposes U.S. personnel to no physical risk. This is the “easiness” (the source of Walzer's moral uneaseFootnote ³¹ ) that provides a foundation for intentional open-endedness in the application of force.

In the context of the war on terror, conceptualizing drone-based violence in this way is plausible because it appears to be a deliberate and indefinite exercise in the management of terrorist risks. Christopher Coker, for example, has described “the concept of a ‘long’ or ‘never-ending war’” as “an astrategic, tactically driven risk management policy which locks the West into an endless process of risk management.”Footnote ³² In contrast to the Long Peace in Europe in the nineteenth century, which resulted from states’ collective maintenance of a stable balance of power, the (long) war on terror is about managing nonstate challenges to a preponderance of U.S. power. But if, as another author suggests, the “negative, if not dystopian” outlook of risk management is that “risks can only be managed, not completely eradicated,”Footnote ³³ this presents an ethical problem if the chosen method is a violent one. Peace (following the ending of that violence) would be neither promised nor in prospect, because “success” would be measured only by reference to the length of uninterrupted sequences of nonevents—that is, foiled or failed terrorist attacks.

Even if it is not possible to conceptualize drone strikes as war, law enforcement, or vim, this does not avoid the general proposition that violence of every kind should be subject to some form of limitation. And, at a minimum, based on the unassailable assumption that violence is not always going to be a good thing, the most straightforward limitation is a temporal one. At issue is not whether the violence being contemplated will ever end as a matter of fact (for that is unknowable). Rather, it is a matter of intention: the would-be user of force should not be permitted to intend their violent enterprise to be a permanent one. Commitment to a precise time limit is neither practical nor necessary; all that is required is that the eventual termination of violence be both conceivable and desired. If, however, the drone-based violence carried out by the U.S. government is or becomes bound up in the process-oriented logic of risk management, there is a danger that it will fail to satisfy even this most basic requirement of limitation. As a violent enterprise of intentionally indefinite duration, the drone program would constitute vis perpetua and thus be ethically untenable. It is unfortunate, therefore, that the political temptation to perpetuate a drone-based approach to risk-managing terrorism does not and is unlikely to encounter much resistance in a domestic context.

The reality that using force can itself be a directly and immediately risky endeavor has, in the past, served as a brake on using it too often and for too long. In ordinary circumstances, when a country's military personnel are putting their lives on the line in pursuit of a political cause, domestic concern about the duration of a war is largely attributable to the accumulation over time of friendly casualties. The lengthy, ground-based campaigns fought by U.S. troops in Afghanistan and Iraq were wound up for a multiplicity of reasons, but a reluctance to continue the shedding of American blood (in pursuit of aims that many Americans came to believe were not worth the cost) was prominent among them. As of June 2014, U.S. military casualties in Afghanistan (Operation Enduring Freedom) stood at 2,186 fatalities and 19,622 wounded. And in Iraq (Operation Iraqi Freedom), the United States suffered 4,411 military fatalities and 31,941 personnel wounded.Footnote ³⁴ Even in the case of the Kosovo intervention, during which there were no U.S. ground troops deployed and military pilots flew at altitudes beyond the range of Serbian anti-aircraft fire, President Bill Clinton warned: “This action is not risk-free . . . . And I ask for the prayers of all Americans for our men and women in uniform in the area.”Footnote ³⁵ By contrast, a president need not invoke protective “prayers” for U.S. drone operators.

In 2009, CIA Director Leon Panetta stated that U.S. drone strikes in Pakistan are “the only game in town in terms of confronting or trying to disrupt the al Qaeda leadership.”Footnote ³⁶ Setting aside the question of whether a game must be a violent one, there are in fact alternatives to using drones: for example, using inhabited aircraft and/or deploying ground troops. Indeed, the latter “game” achieved the killing of Osama bin Laden himself inside Pakistan in 2011. Nevertheless, non-drone operations are in general far less palatable politically to the U.S. government because they involve the exposure of U.S. personnel to physical risks. The perception that U.S. drone operators can go on killing America's enemies one by one, without experiencing any danger to themselves, means there is little or no reason for the operators’ families, friends, and fellow citizens to desire or demand the cessation of such activity. A humanistic concern for the fate of non-Americans at the receiving end of drone strikes has caused considerable concern within Congress and among the public, but the administration has sought to assuage this concern by routinely describing the use of drones as precise and discriminate. In a live Internet forum in 2012, for example, President Obama said that “drones have not caused a huge number of civilian casualties . . . . For the most part they have been very precise precision strikes against al Qaeda and their affiliates.”Footnote ³⁷ Of course, establishing the exact meaning of “huge” and “the most part” is impossible while a shroud of official secrecy hangs over these CIA activities.Footnote ³⁸

Later, in his May 2013 speech, Obama sought to compare the risk to noncombatants resulting from terrorism on the one hand and from U.S. drone strikes on the other: “Remember that the terrorists we are after target civilians, and the death toll from their acts of terrorism against Muslims dwarfs any estimate of civilian casualties from drone strikes.”Footnote ³⁹ Even if this is true, however, the relevant comparison when contemplating vis perpetua is between the risks experienced by those at opposite ends of a drone strike. To compare drone strikes to terrorism, as Obama did, is unhelpful even if it is done favorably. The salient problem from a justice perspective is not that the principle of noncombatant immunity is (or might be) willfully and repeatedly ignored by individual drone operators in a manner comparable to terrorism. Rather, it is that when drone-based violence is considered at the system level, the transfer of physical risk away from U.S. combatants and towards noncombatants in targeted territories results consistently (and, therefore, predictably) in indiscriminate harm. Driven by a concern to avoid the friction of domestic aversion to American casualties, drone-based violence is used in a way that privileges national identity over noncombatant status as regards the experiencing of physical risk. Contrary to the spirit of jus in bello discrimination, the lives of U.S. combatants at home are valued more highly than the lives of noncombatants in the vicinity of a drone strike.

It is a characteristic of what Martin Shaw calls “risk-transfer militarism” that some aerial bombing “is undertaken in the firm knowledge that it will increase the risk to civilians compared with other possible means, military as well as non-military.”Footnote ⁴⁰ An individual drone operator can be punctilious in upholding the discrimination principle, and a drone campaign certainly has the potential—from one air strike to the next—to be less harmful to noncombatants than a campaign involving other forms of violence. Nevertheless, it is a deliberate and systematic feature of drone use that a drone operator experiences no physical risk while, at the same time, noncombatants are endangered by imperfections of technology and human judgment. Noncombatants cannot be removed from these dangers like a pilot can be removed from both a cockpit and a country. Whereas a drone operator cannot be killed (qua drone operator), noncombatant deaths from drone strikes are or could be the result of bad targeting intelligence, a mechanical or communication malfunction, or of simply being too close to an air-to-surface munition. The latter danger, in particular, explodes the notion that drone strikes are precise to a “pinpoint.”Footnote ⁴¹ For example, Reaper aircraft carry 500-pound Paveway II bombs that have been shown to yield a “lethal [to 50 percent of exposed persons] blast range” of about 20 meters,Footnote ⁴² and in general the “recommended safe distance” (for unprotected troops) from the impact point is 500 meters.Footnote ⁴³

Even if, at the level of individual decision-making, intentional injustices are ethically distinguishable from those that are merely foreseeable, such a distinction is undermined by the systematic avoidance (by U.S. drone operators in general) and transfer (to foreign noncombatants in general) of physical risk. When a program of violence as a whole, working exactly as intended, consistently spares combatants on one side and endangers noncombatants on the other, it is difficult to describe and excuse the killing of innocents as having been merely foreseen. Rather, such a program has the appearance, across time, of dishonoring the principle of discrimination.