Living in The Drone Age?
The subtitle of Laurie Calhoun’s We Kill Because We Can: From Soldiering to Assassination in the Drone Age (2015) suggests that we have entered a new epoch: the Drone Age.1 Calhoun is not alone in expressing the belief that the recent introduction and reliance upon armed drones by the United States stands as a novel and troubling turn in regard both to armed conflict and to our moral views of armed conflict. However, when pressed as to what precisely is novel, most are unable to pinpoint what makes the deployment of drones so transformative. Is it that drones are operated by remote control from great distances and thus place their pilots at no direct risk? Is it that drones are used to find, target, and kill named individuals—as opposed to attacking anonymous members of a military? Is it that drones drastically reduce the economic and political costs of engaging in armed conflict and, thus, make it more likely that governments will resort to lethal force to address global security concerns? Though Calhoun—like many others—affirms all three of these factors as contributing to the novelty of the Drone Age, she also provocatively asserts both that “[drone] technology has guided policy and not vice versa” and that a logic of “lethal centrism” has unabashedly emerged and helped to define the Drone Age.2
The rhetoric of asserting that a new age has dawned is typically aimed at heightening either our alarm or enthusiasm for the novelties identified. Calhoun’s book is clearly directed at raising an alarm regarding the ongoing drone-based campaigns of targeted killing and, while alarm is warranted, her emphasis upon the purported novelty and epoch shifting character of drone technology may cloud our understanding of how we are witnessing an extension of other recent military practices. By placing this current practice into historical context, we needn’t mute our moral alarm. Instead, by gaining a fuller historical perspective, we may be able to better understand what we ought to be alarmed about and why.
I will now briefly consider the putative novelties of the Drone Age mentioned above to put them into some historical context. First, on the issue of fighting from great distances, drones have not ushered in something especially novel. As M. Shane Riza documents well, the goal of killing from greater distance and with greater force is “the oldest dream of the second-oldest profession.”3 Whether one considers aerial bombardment or long-range missiles, there has been a steady trend in recent military history to move soldiers farther and farther from danger. While the riskless position of drone pilots might appear novel, the phenomenon of riskless warfare was first acknowledge during the NATO intervention in the Balkans, where pilots operated at an elevation where ground forces effectively posed no threat to them and no NATO casualties were incurred. The riskless operations of drone pilots sitting half-way around the world from the attack site is an extension of what the U.S. military has done in the preceding decades. The policy to remove military personnel from risk was explicit during the Balkan humanitarian intervention and was politically motivated so as to remove domestic political resistance to that intervention. There was a similar political motivation for the Pentagon’s budgetary mandate in 2000 that one-third of all armed aircraft be unmanned by 2010. John Warner, then the chairperson of the Senate Committee on Armed Service, was the architect of this legislative mandate and he has been quoted that the policy was in response to the public’s growing intolerance to human casualties.4 If U.S. foreign policy was ever to use military force and not meet overwhelming, domestic political resistance, he reasoned that the risk of casualties would have to be nearly eliminated. In contrast to Calhoun’s claim that the technology has “guided policy,” it appears that the technology has caught up with policy.
Some will find the image of a person sitting at a computer terminal killing others at no personal risk unsavory and liken it to extermination, not warfare. Others have argued that if greater distance neither negatively affects military capabilities nor the capability to abide by the laws of war protecting non-combatants, then military planners have a moral obligation to move their personnel as far away as possible from the dangers of combat.5 Neither intuition is baseless. The extreme asymmetry of using drones for targeted killings against populations that lack such technology and who cannot reciprocate directly does resemble an act of extermination. Still, we may not feel as uncomfortable if the targeted individuals were truly being stopped from carrying out either ethnic cleansing (as in the Balkans) or a significant terrorist attack. On the flip-side, it is morally negligent to put military personnel at unnecessary risk. However, we might not find an obligation to remove soldiers from danger so compelling, if they were following orders to achieve an unjust military objective. If these contrasting, initial intuitions regarding distance and the absence of risk can be reversed depending upon the justice or injustice of the military aims, then perhaps there is nothing inherently moral or immoral regarding either the distance or riskless nature of drone deployments.
Is the morally salient novelty instead found in the way targets are named, tracked and killed by drone strikes? While drones have allowed targeted killing campaigns to expand, the practice has a longer history.6 During World War II, Allied Command named German Field Marshal Erwin Rommel as a target, tracked him, and bombed his location with the intention of killing him. Though this targeted killing mission was unsuccessful, attacking a named opponent during war is neither illegal nor thought to be morally problematic in itself. More recently, Saddam Hussein was targeted during the second Gulf War; Colonel Muammar Gaddafi was targeted during the 2011 NATO intervention in Libya; and Israel used helicopters for the targeted killings of Hamas leaders in Palestine long before using drones. In fact, the Clinton administration drew up a kill-list for terrorists, headlined by Osama bin Laden, who were deemed direct threats to the U.S. That many of these past attempts at targeted killing were unsuccessful and Clinton’s policy bore no fruit only shows that these policies and strategies were awaiting drone technology to be “successful.” Here, again, policy is not being driven by the technology, as Calhoun asserts, but technology is catching up to U.S. policy aims.
That targeted killings by means of drones is an extension of recent practices and policies does not somehow morally justify the current practice. In an effort to express the moral problems with this practice, Calhoun draws a number of analogies between targeted killings and other forms of homicide, viz. assassinations, mafia “hits,” a tyrant’s execution of citizens, and extrajudicial executions. Common to all of these proposed analogues is that the person being killed is named beforehand. However, the intended aims and moral problems with them vary greatly. In times of peace, assassination is typically conceived as being politically motivated in the narrow sense of aiming to eliminate a high ranking person and alter the political landscape. In times of war, assassination is reserved—under international law—for the treacherous killing of an opponent’s leadership, for example, under a truce or when negotiating peace. Both versions seem rather distant from the current practice of targeted killing as neither political advantage is sought nor is there a treacherous violation of trust. On the other hand, when tyrants execute citizens and when mobsters execute individuals on a hit-list, the aim tends to be to maintain or expand power and control over their territory or turf. Though drones could conceivably be used this way domestically by a tyrant, this does not reflect their current usage abroad.
Drawing an analogy to extrajudicial executions is more promising. However, the typical justifications offered for the targeted killing campaign are not punitive in nature.7 Instead of being retrospective as punishment is, these strikes anticipate harms the targets will supposedly contribute to and ‘defend’ against them. However, like extrajudicial executions that lack public trials and due process for the accused, no evidence is made public for why the named targets are deemed threats, nor why this threat should be judged to be imminent, nor why lethal force is a necessary and proportionate means for defending innocent people against this imminent threat. Given that individuals are named and targeted (and not the anonymous members of an army), it’s difficult to deny that case-by-case justifications are required. We can imagine a hypothetical case where all of these conditions are publicly demonstrated to be satisfied. If the individual truly poses an imminent threat to innocent people and lethal force is both necessary and proportionate, then the naming, tracking and killing of the target would conceivably be justified. In principle, then, it’s not the mere fact that individuals are named, tracked, and targeted to be killed that should have us alarmed; rather, it’s the secrecy surrounding both the general rules governing this institutional practice and the way these rules are applied in each concrete case.
Unlike CIA Black Ops of the past, where missions were never formally acknowledged—forget about justified—we are now witness to openly acknowledged secret missions where “credit” is taken for targeted killings but the decision-making process for whom to target and when to strike is completely opaque. That such Grey Ops could be so unquestioned by the U.S. public may be understood, in part, by the lack of U.S. casualties. Here, the lack of domestic political cost for engaging in a drone-based campaign shows itself not to be the primary moral concern but a novel condition that tends to mute the public’s alarm.
Though both the Bush and Obama Administrations have asserted the right to use military force against all members of al-Qaeda and its associated forces based upon the congressional authorization granted in the immediate aftermath of the 9-11 attacks (and now Obama seeks a similar authorization from Congress to use military force to pursue ISIL and its associated forces), there has been very little in the public record from either Administration explaining how individuals are identified as “legitimate” targets. Based upon reports from investigative journalists and what has been leaked to the press, we are gaining a confusing picture: there appears not to be a single targeted killing campaign but at least a few different versions, each using contrasting standards and sources of intelligence.
First of all, targeted killings have been carried out by both the CIA and the Pentagon and each has apparently different methods and criteria for determining the legitimacy of a target. Personality strikes is the name given to when information gathered from diverse sources (e.g. paid informants, local state intelligence agencies, surveillance of communications, and visual surveillance by drones) is used to build a case that a named individual is a member of al-Qaeda or its associated forces. While both agencies have followed such an approach, the CIA has also carried out an untold number of signature strikes that identify a target based upon observed “suspicious” behavior without knowing who exactly they are. The reliability of targeting individuals based upon behavioral profiling is dubious at best, but even the reliability of the information used to justify personality strikes is questionable. When an informant is motivated by profit, the truthfulness of the information they provide may be of secondary concern. Concerns have also been expressed that local state intelligence agencies have provided distorted information so the U.S. would target their political opponents, as opposed to dangerous terrorists. Worse still, recently leaked documents indicate that the Pentagon directed strikes in Yemen and Somalia rely solely upon communications surveillance and visual surveillance by means of drones without any human intelligence “on the ground” to help either: build the case against the individual, identify the target (e.g. by planting a SIM card on them), or verify that the intended target was the actual person killed.8 This is a third protocol for developing a target and is very different than the protocols in Iraq and Afghanistan, where U.S. ground forces are nearby enough to establish informants who can assist at all three of these levels. Given the lack of transparency regarding the decision making procedures in all three versions, the public has nothing to counter their moral alarm concerning the apparently unreliable intelligence that grounds the varieties of targeting killings carried out by the U.S. military and CIA.
‘Continuing, Imminent Threats,’ Defensive Force, and Risk-Management
In May 2013, the White House responded to criticisms and outlined a set of standards and procedures for carrying out drone based targeted killings “outside areas of active hostilities.”9 Though this last phrase, like much of the language in the document, is open to wide interpretations, I want to draw attention to one key element of the standards. The document asserts: “the United States will use lethal force only against a target that poses a continuing, imminent threat to U.S. persons.” First, it should be acknowledged that, without the transparent application of this standard to any case of targeted killing, the world must take it on faith that the standards are actually being satisfied. Second, the neologism of a “continuing, imminent threat” contained within this standard is an astonishing example of governmental doublespeak. No amount of verbal gymnastics can make the notion of imminence reflect an ongoing or continuous state of affairs. Even if one argued to relax Daniel Webster’s famous and widely accepted definition of an imminent threat as one that is “instant, overwhelming, and leaving no choice of means, and no moment for deliberation,” to make a sudden and impending threat continuous is oxymoronic.
Though it is tempting to simply dismiss this standard as an example of intentional obfuscation, it might prove useful to ask how a “continuing, imminent threat” is interpreted in practice. The recently leaked documents outlining the military’s Joint Special Operations Command (JSOC) procedures for placing individuals on the kill-list in Yemen and Somalia provide some clues.10 While these procedures were followed in these regions by the JSOC in 2011 – 2012, we should not extrapolate to other contexts or agencies. Still, given that these were the procedures being used “outside areas of active hostilities” just prior to the White House’s assertion that “the United States will use lethal force only against a target that poses a continuing, imminent threat to U.S. persons,” it may be a fair place to seek some understanding for how this standard has been practically implemented. The leaked documents show that after intelligence on a suspect had been gathered and summarized in a “baseball card,” a recommendation to put the individual on the kill-list was submitted to a five-step review and approval process prior to the target being established. The final stage involved President Obama’s review and approval and, once he signed off, the JSOC had a sixty-day window to kill the target before having to initiate the approval process again. Sixty days of continuing, imminent threat.
As previously mentioned, to justify an anticipatory resort to defensive force requires an imminent threat. However, to interpret imminence as lasting upwards of two months and to be renewable makes this limit effectively meaningless. Even if the intelligence regarding these targets is accurate and they do contribute to a real danger through affiliation with a terrorist group, the fact that the danger is deemed to last sixty days at a time suggests that it isn’t serious enough to ground a claim of self or other defense. A resort to defensive violence must somehow stop or eliminate a threat by interrupting the causal chain that will otherwise result in unjustified harms to an innocent person. Given the sixty-day window, the likelihood that the intelligence generated about a potential target would establish such a direct causal threat is slim. After all, it’s nearly laughable to suggest that one could know that killing someone fifty-nine days from now will actually save the lives of innocent people; it’s even more worthy of our derision to suggest that this act of killing in fifty-nine days will be necessary to defend innocent people.
If targeted killings within this sixty-day window do not appear to be direct acts of self or other defense, how can they best be characterized? The danger posed by individuals simply by their association with terrorist groups is neither an imminent threat nor even a direct causal threat. Instead, their affiliation with terrorist organizations contributes to a general security risk—and risks are not the same as threats. To help illustrate the difference, consider the analogue of a street gang that has a history of criminal activity and violent conflicts with rival gangs. Their mere presence in a neighborhood imposes a security risk to all residents but, without further knowledge of specific activities or plots, they don’t pose a direct threat to anyone. The distinction here is neither between a high versus low probability of harms coming to be nor between the greater versus lesser severity of the potential harms. (These factors would help distinguish major from minor threats.) Instead, a security risk is distinguished from a direct threat by the partial ignorance or significant uncertainty regarding the danger posed. In the case of the neighborhood gang, its presence imposes a security risk upon all residents but we can—to varying extents—be ignorant of or uncertain about: who is a gang member, what roles individuals play in the gang, what criminal activities they are planning, who might be harmed as a result of their activities, how severe those harms might be, and the likelihood of anyone actually being harmed. The partial ignorance or uncertainty concerning such details does not make the risk imposed less real; however, it does mean that any action taken against the gang by police or residents of the neighborhood is not a straightforward case of self or other defense wherein a direct threat is stopped or eliminated. Instead, in the absence of a determinate threat, the goal is to continuously manage the unjustly imposed security risk.
If we assume for a moment that the intelligence identifying someone as a member of a terrorist network is reliable but that the uncertainty or partial ignorance embodied in the rest of analogy still holds, then perhaps many, if not most, targeted killings can best be characterized as acts of attempted security risk management. Though military engagements aiming at risk-management or risk-reduction are not entirely novel, there are a few controversies I would like to highlight regarding a targeted killing campaign that primarily holds these aims.
First, reducing security risks ought not to be confused with defense. Whether one considers domestic law, international law, or a wide variety of moral frameworks used to justify violent means of defense, an attack must be either underway or imminent. Improving security, though generally valuable, is not a goal that can readily justify lethal violence from either a legal or moral point of view. The tendency to mislabel all attacks upon dangerous terrorists as acts of defense only confuses the matter and provides a false sense of justification. The complication that we face, however, is that security risks can escalate to direct threats as details are learned and we reduce our partial ignorance and uncertainty. So long as we conflate security risk management with self-defense, we avoid addressing the crucial questions of what knowledge is specifically needed to establish a direct threat, as opposed to a mere risk, and when that direct threat is sufficiently severe and imminent to justify lethal defensive force. The current response to these difficult questions is apparently to err on the side of caution and assume that all security risks are direct and imminent threats. If this is an accurate characterization, then the U.S. targeted killing campaigns are often engaged in ‘precautionary killings’ as opposed to defensive ones, and this ought to raise the volume of our moral alarm.
Second, for a campaign that aims to continuously manage security risks by eliminating members of a terrorist group, we are confronted with the difficulty of predicting outcomes when acting within such a complex social system. Risk-management is generally open to many variables and does not follow a simple linear logic. Addressing one risk within a complex system typically gives rise to novel and/or difficult to foresee risks. This stands in contrast to the more linear logic of self or other defense where one’s resort to force is a direct means of stopping an unjust attack by interrupting a causal chain producing an unjust harm. Actions like drone strikes often result in difficult to assess outcomes as they don’t just eliminate intended targets (when successful) but often impose unintended harms, terrorize those who live in the area, and generate resentment from the local population—all of which can inspire increased recruitment for the terrorist/militant group and blowback in general. As a result, we can fairly ask several critical questions. Have recent targeted killing campaigns generally increased security for all or have they strengthened the support base and recruitment efforts of the terrorist and militant groups we confront? Has the U.S. created a dangerous precedent by using lethal force to manage security risks that others will likely follow, especially as they gain drone technology? Have security risks been shifted away from U.S. citizens and onto innocent people living in close proximity to the drone strikes? The first two questions raise the possibility of a failure to manage security risks well by increasing security risks in both the short and long term. The third question raises the issue of a possible unjust distribution of risk.
Third, and last, while security risk-management is not widely acknowledged as a legitimate aim of military engagements, it is a legitimate aim of domestic law enforcement. Contemporary criminologists have abandoned the notion that law enforcement does much in the way of either deterring crime or reforming criminals, as the social scientific data doesn’t support such conclusions. Instead of eliminating crime, policing largely aims to continuously manage the risks of criminal activity by means of increased surveillance within society and the removal of potential opportunities for criminality. The parallels to counterterrorism efforts are abundant but so too are the differences. Though surveillance of suspected terrorists and making potential targets less accessible are central to managing our security risks, targeted killings depart from the standard restrictions placed upon law enforcement, which prioritizes capture and trial as well as prohibit the foreseeable killing of bystanders. Since within domestic law enforcement we reject a presumption to “kill, not capture” (or what Calhoun calls lethal centrism), why might it ever be found permissible for the management of terrorist risks? The typical responses emphasize either that: terrorists are more dangerous than typical criminals, or we lack effective international institutions of law enforcement, or the limits imposed upon law enforcement are not appropriate because counterterrorism is something between warfare and policing. Even if there is some accuracy to these three claims, they don’t convincingly excuse the presumption to kill and not capture suspected terrorists given the lack of transparency surrounding the targeted killing campaigns. The absence of public evidence that targets pose an imminent threat undermines the suggestion that the specific targeted individuals are more dangerous that typical criminals. The refusal to either bring evidence to a domestic court or to bolster international criminal courts to serve this function undercuts the second excuse. Lastly, writing new legal or moral rules that blend the risk-management aims of law enforcement with the liberties of warfare to kill without public review, and to permit proportionate “collateral damage,” is a transparent attempt to avoid public accountability. Until such excuses are rejected, we will likely remain mute regarding what is so alarming about the institutional homicide embodied within our targeted killing campaigns.
- Laurie Calhoun, We Kill Because We Can: From Soldiering to Assassination in the Drone Age (London: Zed Books, 2015).
- Calhoun, We Kill Because We Can, p. xii and p. 65.
- M. Shane Riza, Killing Without Heart: Limits on Robotic Warfare in an Age of Persistent Conflict (Washington, DC: Potomac Books, 2013), 38.
- For Warner’s comments, see P.W. Singer, Wired for War: The Robotics Revolution and Conflict in the 21st Century (New York: Penguin Books, 2009), 59 – 60.
- B.J. Strawser, “Moral Predators: The Duty to Employ Unihabited Aerial Vehicles,” Journal of Military Ethics 9 (2010): 342 – 368.
- A recent study traces the practice to ancient Rome. See, Kenneth Himes, Drones and the Ethics of Targeted Killing (London: Rowman and Littlefield, 2015).
- One notable exception was the targeted killing of Osama bin Laden by a special forces team where much of the justifying rhetoric embraced the notion of ‘justice being served’ and of retributive punishment.
- “The Drone Papers,” The Intercept, accessed January 10, 2015, https://theintercept.com/drone-papers/
- “U.S. Policy Standards and Procedures for the Use of Force in Counterterrorism Operations Outside the United States and Areas of Active Hostilities,” White House Whitepaper, accessed January 10, 2015, https://www.whitehouse.gov/sites/default/files/uploads/2013.05.23_fact_sheet_on_ppg.pdf/
- “The Drone Papers,” The Intercept, accessed January 10, 2015, https://theintercept.com/drone-papers/the-kill-chain/
SHAWN KAPLAN teaches philosophy at Adelphi University.