The war against vague threats: The redefinitions of imminent threat and anticipatory use of force

Abstract

This article explores how the United States (US) has redefined the concept of ‘imminent threat’ in order to relax the rules for anticipatory use of armed force against insurgents. The article focuses on how two new definitions of imminent threat have changed the conduct of specific combat activities, namely, drone strikes and ground combat operations.

The central part of the article is divided into four sections. The first section examines the redefinition of imminent threat in the context of drone warfare, while the second section provides an analysis of the redefinition of imminent threat in ground combat operations. Both sections show how the new definitions of imminent threat abandoned two key elements of the classic definition, that is, the immediacy and certainty of the threat. The third and fourth sections of the article explore how the new definitions of imminent threat prevented the application of two key principles governing the use of armed force: the principles of necessity and proportionality. Both sections show how successive US administrations enabled the US military to conduct operations without observing these two key principles regulating the use of force.

Keywords

Anticipation drone strikes war on terror US military use of force imminent threat

Introduction

Since the attacks against the United States (US) on 11 September 2001, we have witnessed how the US, as well as many other countries, have increasingly resorted to preventive actions in attempts to eliminate new terrorist threats (Aradau and van Munster, 2007). The emergence of preventive measures became visible in both law enforcement and military operations against alleged terrorists. In law enforcement, for example, we have seen a significant shift from the post-crime paradigm, which defines crime (e.g. terrorist attacks) as a harm or wrongdoing that has to be dealt with post hoc, to a pre-crime paradigm, which regards crime as a potential risk to which we need to respond with preemptive measures (Zedner, 2007: 261–262; Loader and Sparks, 2007: 84–87). Under the pre-crime paradigm, providing security by reacting to terrorist crimes that had already happened became less relevant than preventing crimes by relying on data trails and pattern-of-life analysis that indicate when and where crime may occur in the future (Zedner, 2007: 265; Amoore and Raley, 2017; Završnik, 2019).

The emergence of ‘anticipatory reason’ (Hong and Szpunar, 2019: 315) became evident also in military operations against terrorist groups and their supporters. As Massumi (2015: 5) argued, preemption became, under the Bush administration, a new operative logic of power that defined our political epoch and thus replaced the logic of ‘deterrence’ that defined the Cold War era. When using the term ‘preventive war,’ scholars usually refer to two things: first, a policy supporting the use of military force against imminent threats; and, second, a policy promoting the preventive use of force against future, non-imminent, broadly defined threats (e.g. an adverse shift in the balance of power that benefits an enemy state) (Trachtenberg, 2007: 29; Levy, 2008: 4).

This article will explore how ‘anticipatory reason’ transformed the conduct of military operations in the post-9/11 era. The objective is to shed new light on the reconceptualization of two under-researched aspects of preventive war, that is, the reconceptualization of preventive war that enabled the US military to respond with force against future, vaguely defined imminent threats, and the reconceptualization of anticipatory individual and unit self-defence that enabled US troops to preemptively eliminate perceived threats on the battlefields of the ‘war on terror’.

The classic doctrine of anticipatory use of force against an imminent threat, which is today still accepted by many legal scholars and commentators, emerged in the mid-19th century, in the aftermath of the Caroline incident, a preemptive attack of the British Empire on US soil against a small steamer, the Caroline, believed to be carrying rebels from the US into British-controlled Canada (Jennings, 1938: 82–84; Rogoff and Collins, 1990: 493–496). After the incident, in which British soldiers stormed the Caroline and set it on fire, the then US secretary of state, Daniel Webster, conceded – in a letter in April 1841 – that the British Crown had the right to preemptive self-defence if there was a ‘necessity of self-defence, instant, overwhelming, leaving no choice of means, and no moment of deliberation’ (Webster, 1841: 1137–1138). In addition, Webster stated that anticipatory use of force must not be ‘unreasonable or excessive’, and thus supported the idea that an armed response must not be disproportionate to the perceived threat (Webster, 1841: 1138). Based on Webster’s formula, legal scholars developed a doctrine of anticipatory use of force that consisted of four key components: the immediacy of the threat, the certainty of the threat, the necessity of the armed response and the proportionality of armed response (Rockefeller, 2004: 131; Lubell, 2015: 698–700). Under this doctrine, anticipatory armed response to an imminent threat was lawful only if the threat was immediate and certain, and the response to the threat was necessary and proportional.

Until the adoption of the UN Charter in 1945, Webster’s definition of imminent threat was accepted as the cornerstone of anticipatory self-defence (Schmitt, 2003: 530; Byers, 2003: 180). With the adoption of the UN Charter, however, the international community rejected the idea of anticipatory self-defence by stipulating that states have the right of individual or collective self-defence only ‘if an armed attack occurs’. As the majority of legal scholars pointed out, Article 51 allows a state to invoke the right to self-defence only after an armed attack has taken place (Kunz, 1947: 877–878; Higgins, 1963: 197–203; Kelsen, 2000: 797–798; Glennon, 2002: 546–547; Rostow, 2009: 550–551; Simma et al., 2012: 1041–1043). Despite the clear wording of Article 51, many scholars refused to abandon the idea of anticipatory self-defence by insisting that the inherent right to self-defence formulated in Article 51 includes the right to use military force even before an armed attack has taken place, that is, when conclusive evidence indicates a real and imminent threat of an armed attack (Bowett, 2008: 187–192; Schachter, 1984: 1633–1634; Franck, 2002: 98; Shah, 2007: 98–101). Bowett (2008: 187–192), for example, argued that Article 51 did not restrict but safeguarded the traditional right of self-defence, which included anticipatory self-defence.¹By adopting the doctrine of anticipatory use of force established after the Caroline incident, the US traditionally supported the position that it was lawful to launch an anticipatory armed attack against an imminent threat (Dinstein, 2005: 182). This position, however, changed significantly after the 9/11 terrorist attacks against the US, as successive US administrations advanced a new definition of anticipatory self-defence that included a redefined concept of imminent threat. The debate on the new version of anticipatory self-defence started to gain traction when the Bush administration introduced its doctrine of preventive war that culminated with the invasion and occupation of Iraq in 2003 (Arend, 2003; Williams et al., 2003; Taft and Buchwald, 2003; Schmitt, 2003; Rockefeller, 2004). A significant part of the debate focused on how the Bush administration redefined, in its 2002 National Security Strategy, the concept of imminent threat to justify the launch of preventive wars against vaguely defined threats. Although many scholars and commentators critically examined the new definition of imminent threat included in the ‘Bush doctrine’ of preventive war, other redefinitions of imminent threat, which were just as important in the process of creating a new doctrine of anticipatory self-defence, failed to gain much attention.

This article attempts to fill that gap by exploring two other redefinitions of imminent threat: first, the redefinition of imminent threat in the context of drone warfare, which was introduced by the Obama administration; and, second, the redefinition of imminent threat, introduced by the Bush administration and later embraced by the Obama administration, in the context of ground combat operations, in particular in kill-or-capture missions. The article aims to show how the US relied on the two redefinitions of imminent threat to create its new doctrine of anticipatory use of force.

The central part of the article is divided into four sections. The first section examines how the Obama administration redefined, in late 2011, the concept of imminent threat in jus ad bellum in order to establish the legal basis for targeted killings of individuals suspected of being senior leaders of Al-Qaeda and affiliated groups. The section shows how the Obama administration changed the definition of imminent threat by eliminating two of its key traditional components – that is, the immediacy and certainty of the threat. The new definition of imminent threat enabled the US military to start launching anticipatory attacks with unmanned aerial vehicles, or drones, which emerged as the ‘weapon of choice’ in military operations against alleged insurgents (Byman, 2013; Badalič, 2016: 158–159; Schwarz, 2016; Schmidt and Trenta, 2018: 202). In particular, the Obama administration used the new definition of imminent threat to portray drone strikes conducted in undeclared war zones (e.g. Pakistan, Yemen) as anticipatory self-defence operations.

The second section of the article provides an analysis of how the Bush administration redefined, in 2005, the concept of imminent threat in ground combat operations. The redefined concept was unveiled in the new version of the US Rules of Engagement, the internal directives delineating the circumstances and limitations under which US troops are allowed to use force against enemy troops (Chairman of the Joint Chief of Staff, 2005: 89; Joint Chiefs of Staff, 2016: 207). The section shows how both Bush and Obama administrations embraced a redefinition of imminent threat that abandoned one of its key traditional elements – that is, the immediacy of the threat. The new definition was used to relax the rules for anticipatory self-defence in ground combat operations, in particular in kill-or-capture missions or night raids. While night raids, usually carried out by US Special Operations Forces and their local partners, were used for years during the Bush administration, it was during Obama’s presidency that they became one of the cornerstones of the US military tactic in the fight against insurgent groups (Shanker et al., 2010; Niva, 2013: 186; Scahill, 2013: 114). In Afghanistan, for example, the Obama administration dramatically increased the number of night raids, which resulted in thousands of people, both combatants and civilians, being killed and injured (Porter, 2011; Porter and Shah, 2011; van Linschoten and Kuehn, 2011: 1). The section shows how the new definition of imminent threat enabled US troops to introduce very broad threat categorizations that blurred the line between civilians and legitimate military targets.

The third section of the article explores how both new definitions of imminent threat prevented the application of the principle of necessity during the preemptive use of force against perceived imminent threats. The section focuses on the principle of necessity incorporated in jus ad bellum, the norms regulating the use of armed force in international relations, in order to show how the US military failed to observe this principle in drone strikes. In addition, the section also examines the principle of necessity in jus in bello, the norms governing the conduct of hostilities in armed conflicts, in order to explain how the US military ignored this principle in ground combat operations.

The fourth section examines how the new definitions of imminent threat prevented the application of the principle of proportionality, another fundamental principle governing the use of armed force. The section focuses on the principle of proportionality in both jus ad bellum and jus in bello.

The final section shows how the redefinitions of imminent threat – first – established a legal framework for a war against vaguely defined potential threats, and – second – how these redefinitions blurred the line between civilians and legitimate military targets, thus increasing the risk of civilians getting killed or injured during combat operations.

The redefinition of imminent threat in the context of drone warfare

In order to legitimize the use of military force, including drone strikes, against what it perceived as imminent threats posed by insurgent groups, the Obama administration invoked – by relying on the expanded interpretation of Article 51 of the UN Charter – the inherent right of every state to individual and collective self-defence (US Department of Justice, 2011: 2).² In the process of exercising what they believed was their right to anticipatory self-defence, Obama administration officials decided to redefine the concept of imminent threat and thus relax the rules governing the use of military force.

What was the difference between the classic definition of imminent threat and the one introduced by the Obama administration? The classic definition of imminent threat consists of two key components. First, an imminent threat is defined as an immediate threat, that is, a threat that is just about to materialize (Lubell, 2015: 702). The term ‘imminent’, therefore, must be treated as nearly synonymous with the term ‘immediate’ (Rockefeller, 2004: 131; Fink, 2018: 116). Second, an imminent threat must be a concrete threat, that is, a specific, identifiable threat whose existence must be substantiated by verifiable facts (Lubell, 2015: 702). The dominant view among scholars is that the defending state must provide clear and convincing evidence that an armed attack is about to happen. For example, Franck (2001: 68) contended that a state had a right to launch an anticipatory self-defence operation if it possessed ‘clear evidence of an impending, planned, and decisive attack by a state’. Charney (2001: 836) argued that a state seeking to invoke the right to self-defence, including anticipatory self-defence, ‘should be required to furnish the international community with credible evidence . . . that the attack or threat of attack is continuing’. Shah stated that an imminent threat ‘should be real, verifiable’ (2007: 103) and must be ‘identified credibly, specifically, and with a high degree of certainty’ (2007: 118). The available literature provides several examples of immediate and concrete threats, such as the visible mobilization of the enemy’s armies, navies and air forces preparing to attack (The White House, 2002: 15; IIFFMCG, 2009: 254).

Seeking a way to legitimize its military operations against insurgent groups operating in areas outside declared war zones, the Obama administration contended that the traditional definition of imminent threat was inadequate to address contemporary terrorist threats (Trenta, 2018: 86–89). According to Obama government officials, we entered – after the 9/11 terrorist attacks against the US – a new era that needed a more ‘flexible’ (Brennan, 2011) and ‘broader concept of imminence’ (US Department of Justice, 2011: 7). The new concept was unveiled in a white paper prepared by the US Department of Justice, which stated that the condition that a senior operational leader of Al-Qaeda or an associated force ‘present an “imminent” threat of violent attack against the United States does not require the United States to have clear evidence that a specific attack on US persons and interests will take place in the immediate future [emphasis added]’ (US Department of Justice, 2011: 7). As we can see, the US administration significantly reformulated the definition of imminent threat by abandoning two key components of the classic definition. First, the Obama administration altered the temporal dimension of imminent threat by arguing that an imminent threat was not necessarily an immediate threat. By stating that it was not required to prove that the threat was to take place in the immediate future, the Obama administration enabled the US military to carry out military operations against threats that were expected to emerge in a more distant future.

Second, the Obama administration contended that an imminent threat did not have to be a specific, concrete threat whose existence had to be corroborated with clear evidence. This position became even more evident in one of the main arguments used by administration officials to justify the redefinition of imminent threat. The argument was that the new definition of imminent threat was needed because of a specific nature of contemporary terrorists – that is, individuals who continuously represent an imminent threat for the United States. The white paper (US Department of Justice, 2011: 6–8) referred to this new type of terrorists as individuals ‘continually planning attacks against US persons and interests’, individuals ‘personally and continually involved in planning terrorist attacks against the United States’, individuals ‘continually planning to kill Americans’ or individuals ‘continually plotting attacks against the US’. This new definition of a terrorist, which was based on the assumption that it was somehow possible to know in advance that a terrorist from a specific group was permanently on the verge of launching an armed attack against the US, was used by the Obama administration to advance the new definition of imminent threat. According to Obama government officials, an imminent threat was a continuous threat that emanated from the very nature of a terrorist, and, therefore, it was not necessary to gain evidence about where and when a specific imminent threat might emerge.

By defining rebels from specific groups as a constant imminent threat, the US administration enabled its military to shift from conduct-based targeting to status-based targeting. Under the traditional definition of imminent threat, the military had to identify a legitimate target by the conduct of the target, or, in other words, it had to identify a target by examining whether the activities of the target truly constituted an imminent threat. With the new definition of imminent threat, however, the military did not have to examine the conduct of the potential target, but only to verify its status. If an individual was believed to be a senior member of Al-Qaeda or affiliated groups, the military automatically assumed that he or she posed an imminent threat and was, consequently, ‘targetable’. With the new, overbroad definition of imminent threat, it became irrelevant in what kind of activities such an individual was involved, because the US military was authorized to presuppose that such an individual was ‘continually planning to kill Americans’.³

The new – ‘flexible’ and ‘broad’ – definition of imminent threat significantly broadened the temporal limits within which the US military was permitted to conduct military operations, including drone strikes. The classic definition of imminent threat, which described it as a concrete threat that emerged at a specific time and place, allowed the military to respond with armed force only when such a threat became evident. The new definition, however, described such a threat as a continuous threat that was present at any time because terrorists supposedly continuously planned attacks against the United States. The US military was therefore permitted to launch continuous armed attacks against presumed continuous imminent threats without having to collect evidence showing when and where such threats were about to emerge. To put it differently, the new definition of imminent threat established very broad temporal limits for conducting continuous ‘self-defence’ military operations against vaguely defined threats.

Moreover, the new definition of imminent threat also widened the geographical scope of anticipatory ‘self-defence’ operations. Given that members of Al-Qaeda and associated forces operated in dozens of countries around the world, the new concept of imminent threat enabled the US military to launch operations in all of those countries. In 2002, for example, the US Department of State estimated that Al-Qaeda was a ‘multinational enterprise with operations in more than 60 countries’, which meant that the US believed it was legitimate to conduct anticipatory attacks in those countries (US Department of State, 2003: 4). Under this doctrine, it became irrelevant whether the countries in which members of Al-Qaeda and their affiliated forces were hiding were at war with the US.

The redefinition of imminent threat in ground combat operations

In order to relax the rules governing anticipatory individual and unit self-defence in ground combat operations, including in kill-or-capture missions, the Bush administration redefined the term ‘imminent threat’ in the US Rules of Engagement. Under the US Rules of Engagement, US forces are allowed to use lethal force in self-defence against two categories of individuals: first, individuals who carry out hostile acts, that is, individuals who attack or use other force against US troops or other designated persons or property; and, second, individuals who display hostile intent, that is, individuals who make movements that suggest the threat of imminent use of force against US troops or other designated persons and property (Chairman of the Joint Chiefs of Staff, 2005: 88–89). This section focuses exclusively on the concept of imminent threat, the key element that triggers the right to anticipatory individual and unit self-defence, also called on-the-spot reaction (Dinstein, 2005: 219–221).

Although the debate over the legal source of the concept of individual and unit self-defence is still ongoing, it seems that the dominant view among scholars is that this kind of self-defence is a manifestation of national, or sovereign, self-defence (Dinstein, 2005: 219–221; Hosang, 2010: 420; Trumbull, 2012: 126–127). This is also the view adopted by the United States (Gaston, 2017a: 296). Like anticipatory national self-defence, anticipatory individual and unit self-defence are based on the principles established in the aftermath of the Caroline incident (Stephens, 1998: 135). The preemptive use of force in cases of individual and unit self-defence is, therefore, permissible only when soldiers face an imminent threat of an armed attack. Such a threat must consist of the following two elements: first, the threat must be immediate; and, second, the threat must be concrete, that is, its existence must be substantiated with evidence. With regard to the need to have evidence about an imminent threat, the North Atlantic Treaty Organization (NATO), which defines hostile intent as ‘a likely and identifiable threat’, insists that the threat must be recognizable on the basis of ‘evidence . . . which indicates an intention to . . . inflict damage’ (NATO Legal Deskbook, 2010: 255). Stephens (1998: 148, emphasis added) contended that the assessment of whether or not an enemy force represents an imminent threat must be based on ‘a combination of generally objective factors relating to well established tactical indicia which discloses an intention to attack’. Examples of objective factors relevant to identifying an imminent threat include maneuvering into weapons launch positions, deployment of remote targeting methods, the presence of a third party to coordinate targeting, the speed and bearing of an opposing unit, pointing a weapon in the direction of a soldier and placing an improvised explosive device on the road (Stephens, 1998: 148; NATO Legal Deskbook, 2010: 255; CALL, 2012: 6).

In 2005, the Bush administration advanced a new definition of imminent threat that significantly differed from the classic definition. The redefinition of imminent threat was based on the alteration of the common meaning of the term ‘imminent’. The common meaning of imminent – for example, ‘coming or likely to happen very soon’ (Cambridge Dictionary) – clearly refers to an action that is just about to happen, that is, an immediate and instantaneous action (IHRC, 2016: 20). The Bush administration abandoned that common meaning by contending that ‘[i]mminent does not necessarily mean immediate or instantaneous’ (Chairman of the Joint Chief of Staff, 2005: 89).⁴ The new description of the term ‘imminent’ indicated that the Bush administration, on the one hand, wanted to retain its common meaning, while, on the other hand, sought to add a new meaning that was the exact opposite of the traditional meaning. The Bush administration claimed that ‘imminent’ means both instantaneous and not necessarily instantaneous. Although it is contradictory to claim that ‘imminent’ means both instantaneous and non-instantaneous, the Bush administration, and later the Obama administration, wanted us to believe that both these terms can coexist in the description of imminence.

The new meaning of the term ‘imminent’ altered the concept of imminent threat by including in it non-immediate threats that were expected to emerge at some unspecified time in the future. Given that the 2005 US Rules of Engagement did not provide any clear instructions about how to interpret the non-instantaneous aspect of imminent threat, US soldiers started to interpret an imminent threat as a threat that has no temporal limitations – as if ‘it almost goes to infinity’ (IHRC, 2016: 20). Consequently, in ground combat operations, including in kill-or-capture missions, US troops stopped focusing on whether suspicious individuals posed an immediate threat before opening fire at them. The redefined concept of imminent threat enabled the troops to use force even when they believed that the threat could emerge in a more distant future (IHRC, 2016: 20–21; Badalič, 2019: 59–60).⁵

The broad temporal scope of the new definition of imminent threat negatively affected the second traditional key component of imminent threat – that is, the certainty of the threat. The notions of immediacy and certainty are closely interlinked. If, on the one hand, the threat is truly immediate (for example, we see someone pointing a gun at us), the level of certainty of the threat is high. If, on the other hand, the threat is not immediate, if there exists only a possibility that a threat could emerge at an unspecified time in the future (for example, we believe, based on movements of an individual, that the individual could point a gun at us at some moment in the future), the level of certainty is low. Less immediacy, therefore, usually means less certainty. This link between immediacy and certainty became apparent after the introduction of the expanded version of imminent threat in the US Rules of Engagement as the new, too-broad notion of imminence undermined the requirement that a threat must be certain. The expanded notion of imminence enabled US troops to categorize as an imminent threat a range of movements and actions that did not represent an immediate threat but merely indicated that a threat might emerge at some unspecified moment in the future. Mere speculations about which movements and actions of suspicious individuals could evolve into a threat in the future thus became the basis for decisions on when to preemptively use force in self-defence. The level of certainty of these non-immediate threats was much lower than the certainty of truly immediate threats.

The movements and actions of individuals that US troops believed to indicate that a threat might emerge in the future were as follows.⁶ First, during the war in Iraq, US troops many times deemed individuals using a mobile phone in the vicinity of passing troops to be an imminent threat (Gaston, 2017a: 293). The theory was that those individuals were probably unarmed ‘spotters’ working for an insurgent group, presumably sending information about US troop movements to facilitate an ambush or trying to remotely detonate an improvised explosive device placed on the road to harm the passing troops (Gaston, 2017b: 28). Although US troops also looked for other signs indicating that the individual was a ‘spotter’ – for example, they looked whether an individual had binoculars, a radio or any military-grade equipment – it was not necessary for them to obtain evidence that the threat posed by the ‘spotter’ was truly immediate (Gaston, 2017b: 28–29). To gain permission to open fire on a ‘spotter’, it was enough for US troops to speculate that the behavior of a ‘spotter’ might lead to a threat in the near or distant future.

Second, during the war in Iraq, US troops recognized an imminent threat in individuals digging a hole, particularly at night, in the vicinity of a road (Gaston, 2017a: 293). The theory was that if an individual was digging in the ground, he was probably preparing to plant an improvised explosive device to attack members of the US-led military coalition (Gaston, 2017b: 28). In such situations, US troops did not need to have evidence that the individual was truly planting a bomb and that he posed an immediate threat. In order to resort to anticipatory use of force, it was enough for the troops to speculate that the threat could materialize at some unspecified moment in the future.

Third, in kill-or-capture operations in Afghanistan, US assault troops interpreted as imminent threat movements of individuals who stepped out of their house when they were attacked (Starkey, 2010). In the aftermath of a botched night raid conducted in eastern Afghanistan in 2010, Greg Smith, director of communication at the International Security Assistance Force (ISAF), stated that the assault force fired on and killed civilians because they stepped out of their compound. ‘If you have got an individual stepping out of a compound, and if your assault force is there, that is often the trigger to neutralize the individual. You don’t have to be fired upon to fire back,’ said Smith (Starkey, 2010). The US military believed that individuals who merely stepped out of their compounds during kill-or-capture operations posed an imminent threat, and were, consequently, legitimate military targets.

Fourth, US troops also shot at, and sometimes killed, individuals who tried to escape when their homes came under attack in military operations conducted in Afghanistan and Iraq (Gaston, 2017b: 11; Gregory, 2019: 142). Members of US forces opened fire at those individuals because they believed that running away from the scene of a raid was a demonstration of imminent threat – the theory behind such interpretation was that those individuals were not running away to save their lives but trying to find a better position to counterattack (Gaston, 2017b: 11; Gregory, 2019: 142).

Fifth, in kill-or-capture missions conducted in Afghanistan, US troops many times interpreted as an imminent threat the fact that the attacked individuals possessed a weapon. If US troops positively identified a weapon, or they believed they saw a weapon, they fired on the individual based on the assumption that the individual could pose a threat in the future. The mere presence of a weapon, without being used, was reason enough to shoot at an individual (Mazzetti et al., 2015; Sahak and Rubin, 2011; Rubin, 2011). This practice was extremely dangerous for Afghan civilians, because many of them kept weapons at home to protect their families and property from criminals (Alston, 2009: 9; Porter, 2011). The fact that Afghans had weapons at home did not necessarily mean that they were members of an insurgent group.

Based on the above criteria for determining an imminent threat, it is clear that the expanded concept of the term paved the way for broad threat categorizations. The new concept enabled US troops to engage perceived threats even when the certainty of the threat was low. With such broad threat categorizations, the line dividing civilians and legitimate military targets became blurred, which increased the risk of civilians getting killed or injured (Montalvo, 2013: 26–27).

Ignoring the principle of necessity

Since the Caroline incident, the anticipatory use of force in self-defence operations had to be subjected to two fundamental limitations, that is, the principles of necessity and proportionality (Jennings, 1938: 89–90; Shah, 2007: 95). One of the problems that emerged with the introduction of the above-examined redefinitions of imminent threat was that the new concepts prevented the application of those two key principles governing the use of military force. This section aims to show how both redefinitions of imminent threat allowed the US military to conduct operations without observing the principle of necessity, while the next section focuses on the principle of proportionality.

The principle of necessity is incorporated in both jus ad bellum, the norms describing the conditions under which states may resort to the use of armed force, and jus in bello, the norms regulating the conduct of the belligerent parties involved in an armed conflict (Gardam, 2004: 4–8). In the context of anticipatory use of force in jus ad bellum, the principle of necessity consists of two elements (Roberts, 1987: 277). First, necessity refers to the need for instant and immediate anticipatory armed response to an imminent threat (1987: 277). Second, necessity refers to the absence of any other alternative to the anticipatory use of armed force. This means that all reasonable non-military solutions to effectively respond to an imminent threat must be exhausted so that the anticipatory use of force remains the only viable option to prevent an imminent attack (Schmitt, 2003: 530; Van den hole, 2003: 99–103; Švarc, 2006: 181).

By taking the position that an imminent threat was not necessarily an immediate threat, and that the defending state did not have to provide clear evidence about a threat, the Obama administration made it impossible to determine whether the response to such a threat was truly necessary. When the exact nature of the imminent threat is unclear, and when it is unclear where and at which time such threat is about to emerge, it also remains unclear whether the use of military force is necessary. First, if the time and location of the threat are unclear, it is not possible to ascertain that an instant and immediate anticipatory response to the threat is really necessary. There always exists the possibility that the threat could emerge in a more distant future, thus leaving enough time to respond with force later on. Second, if it is unclear when and where an imminent threat could take place, it is also not possible to determine with certainty that the anticipatory use of force is truly the only viable option for eliminating the threat. If the threat is to emerge in a more distant future, there is enough time to find an alternative solution to the anticipatory use of armed force.

Although the Obama administration continued to assert that the US military had to conduct its operations in a manner consistent with the applicable principles of the law of war, including the principle of necessity (US Department of Justice, 2011: 1), it was evident that the new definition of imminent threat made it impossible to ascertain that any anticipatory armed response to a vague, uncorroborated threat was truly necessary. It can be argued that the broad temporal scope of such vague threats enabled the US military to conduct operations without observing the principle of necessity. Even though it was not possible to determine whether anticipatory armed attacks against vague threats were truly necessary, the US military remained authorized to conduct such attacks, including drone strikes.

A similar development occurred in jus in bello. Although the meaning of the principle of necessity remains somewhat elusive (Gardam, 2004: 7–8; Forrest, 2007: 181), the US Rules of Engagement contain a clear definition of this principle. Drawing on the definition of necessity in jus ad bellum, the US Rules of Engagement define necessity in jus in bello as the absence of any alternative to the anticipatory use of force. Referring to situations of individual and unit self-defence, the US Rules of Engagement stipulate that when ‘time and circumstances permit’, individuals demonstrating an imminent threat ‘should be warned and given the opportunity to withdraw or cease threatening actions’ (Chairman of the Joint Chief of Staff, 2005: 89). This indicates that the anticipatory use of force in self-defence is permissible only when there is no alternative to the use of force, or – to put it differently – when an individual continues, despite being warned, to exhibit a threat of imminent use of force (2005: 89).

By contending that an imminent threat was not necessarily an immediate threat, the Bush administration made it impossible to determine that the armed responses by US troops to perceived threats in ground combat operations were truly necessary. If it is unclear when the threat is to take place, it is not possible to determine with certainty that the anticipatory use of force is the last resort for eliminating that threat. If the threat is to emerge in a more distant future, there is always time to seek an alternative solution to the anticipatory use of force. As demonstrated earlier, US troops many times used force against people who did not pose an immediate threat (e.g. people talking on a mobile phone in the vicinity of US troops, people digging a hole in the ground . . .), which indicated that they resorted to the use of force even though there was enough time to seek a non-military solution to the perceived vague threat.

Although the Bush administration kept the principle of necessity in the US Rules of Engagement, it became evident that the redefinition of imminent threat enabled US troops to conduct operations without observing this principle. With the new concept of imminent threat, US troops were authorized to resort to anticipatory use of force even when the perceived threats were not immediate, which meant that they were allowed to use force when the use of force was not yet truly necessary. In other words, the troops were authorized to use force in situations in which there was still time to find an alternative solution.

Ignoring the principle of proportionality

Like the principle of necessity, the principle of proportionality is incorporated in both jus ad bellum and jus in bello (Gardam, 2004: xv).

In jus ad bellum, the principle of proportionality requires that any use of military force in self-defence must be proportionate to the armed attack or the threat of an imminent armed attack (Van den hole, 2003: 103–104; Shah, 2007: 123). When evaluating whether a military response against an armed attack or a threat of an attack is proportionate, we have to take into consideration two criteria: first, we have to evaluate whether the degree of the armed force used in self-defence was proportional in terms of intensity and magnitude (Gardam, 2004: 156); and, second, we have to determine if the duration of the attack was limited to the removal of the threat (Shah, 2007: 123). Many scholars have argued that it is possible to determine whether those two criteria are met only if an armed attack, or a series of attacks, against the defending state have already taken place (Roberts, 1987: 282; Beck and Arend, 1993: 206–207). Only if we know the impact of a prior armed attack, can we compare it with the impact of the armed response to that attack, and thus determine whether the response was proportionate. If, on the other hand, the armed attack has not yet happened, it is not possible to compare its impact with the impact of the anticipatory armed response. As Roberts (1987: 282) put it, ‘preventive application of force . . . provides no ready reference point for the calculation of the proportional response’.

Some scholars, however, have contended that it is possible to gauge proportionality even in situations in which the defending state launches an anticipatory armed attack against potential future attacks (O’Brien, 1990: 472; Baker, 1987: 47). Baker (1987: 47), for example, suggested that the yardstick against which proportionality can be measured includes both specific incidents that have already occurred and ‘unknown but certain future attacks’. Similarly, O’Brien stated, when referring to how Israel measured the proportionality of its ‘counterterror’ operations, that an assessment of the proportionality of an armed response may take into consideration the ‘overall pattern of past and projected [hostile] acts’ (O’Brien, 1990: 472). The key problem with this argument is that the assessment of the proportionality of anticipatory use of force is not based on facts about a prior attack but on speculations about the potential impact of an imminent threat. Even if the defending state collects information about an imminent threat, it is still not possible to determine with certainty what the impact of the threat will be. It is not even possible, for example, to establish with certainty that the threat will materialize and cause harm to the defending state.

This problem became even more evident after the Obama administration introduced the new definition of imminent threat. The new concept of imminent threat, which was based on the assumption that evidence about an imminent threat was not needed because individuals from specific ‘terror groups’ represented a continuous imminent threat, in effect prevented the application of the principle of proportionality. Without information about the nature of the potential threat, without evidence showing where and when a threat may emerge, it was not possible for the US military to assess the potential impact of such a threat, and, therefore, it was not possible to determine what would be a proportionate response to such a threat. The US military was not able to meet the two above-mentioned requirements needed to assess proportionality. First, given that US troops did not know – or, to be more precise, did not have to know – what exactly was the nature of an imminent threat, they were not able to evaluate whether the degree of the armed force used in self-defence was proportional in terms of intensity and magnitude. Second, given that US troops did not know exactly what was the imminent threat, they were not able to determine whether the duration of the armed response was limited to the removal of the threat. Therefore, by green-lighting self-defence military operations, including drones strikes, against vague hypothetical threats the time and place of which were unknown, the Obama administration allowed the US military to conduct self-defence attacks the proportionality of which was not possible to measure.

The definition of proportionality in jus in bello differs from the definition used in jus ad bellum. Taking individual civilian lives as its main focus, the principle of proportionality in jus in bello prohibits all belligerent parties from carrying out military operations that ‘may be expected to cause incidental loss of civilian lives, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated’ (Protocol Additional I, 1977). Although the incidental loss of civilian lives and injury to civilians are inevitable in armed conflicts, the principle of proportionality aims at ensuring that such losses are not excessive (Gardam, 2004: 98). One of the key requirements that has to be met by belligerent parties trying to apply this principle is the distinction between combatants and civilians. Only if the belligerent parties do everything feasible to verify that the targets of military operations are legitimate military targets and not civilians nor civilian objects, can the belligerent parties comply with the principle of proportionality. Without a clear distinction between civilians and combatants, it is not possible to evaluate whether the harm inflicted on civilians is excessive in relation to the anticipated military advantage.

However, the US Rules of Engagement ignore this definition of proportionality and instead provide a definition that does not include the need to avoid excessive harm to civilians. Drawing on the definition of proportionality in jus ad bellum, the US Rules of Engagement broadly stipulate that troops are authorized to use force that ‘may exceed the means and intensity of the . . . hostile intent, but the nature, duration and scope of force used should not exceed what is required’ (Chairman of the Joint Chief of Staff, 2005: 89). Therefore, when assessing proportionality in situations of anticipatory individual and unit self-defence, US troops only have to evaluate whether the use of force exceeded what was needed to eliminate the threat, but they do not have to determine whether the use of force caused excessive harm to civilians. There is no instruction in the US Rules of Engagement that the harm inflicted on civilians must not be excessive in relation to the anticipated military advantage. On the contrary, the Rules of Engagement state that the ‘concept of proportionality in self-defence should not be confused with attempts to minimize collateral damage [loss of civilian lives, injury to civilians] during offensive operations’, thus suggesting that the principle of proportionality has nothing to do with the protection of civilians (2005: 89).

The problem with this position is that it replaces jus in bello proportionality with jus ad bellum proportionality. As we have seen above, the US treats anticipatory individual and unit self-defence as a subset of anticipatory national self-defence, and, consequently, it believes that the principle of proportionality incorporated in jus ad bellum can be used in jus in bello. This article rejects such a position and insists that the US military is bound to observe the principle of proportionality as defined in Protocol Additional I (1977), and, therefore, must prevent harm to civilians that is excessive in relation to the anticipated military advantage.

How did the redefinition of imminent threat in the US Rules of Engagement affect the application of the principle of proportionality and its demand to limit the harm inflicted on civilians? As demonstrated earlier, one of the key problems with the new definition was that it enabled US troops to use lax criteria for determining which movements and actions of targeted individuals represented an imminent threat. By recognizing imminent threats in movements and actions that in normal circumstances would have been perceived as unthreatening, US troops blurred the line between civilians and combatants, and, thus, in effect, failed to observe the principle of proportionality because they were unable to determine whether the harm to civilians was excessive in relation to the military advantage. When too-broad target selection criteria prevent the distinction between the two key elements – that is, civilians and combatants – needed to gauge proportionality, and, consequently, civilians start to be treated as legitimate military targets, then it is not possible to assess the impact of the military operation in terms of determining whether the use of force was proportionate or not.

Conclusion

Although Grotius supported the idea of anticipatory use of force against imminent threats, he dismissed as illegitimate armed responses against too-remote and vague threats. In his view, ‘to maintain that the bare probability of some remote, or future annoyance from a neighboring state affords a just ground of hostile aggression, is a doctrine repugnant to every principle of equity’ (Grotius, 2001: 70). As we have seen above, it is precisely such a doctrine of anticipatory use of force against remote threats that successive US administrations have created during the ‘war on terror’.

In order to redefine the concept of anticipatory armed attacks, the Bush and Obama administrations dismantled the strict criteria determining when, where and how such use of force was legitimate. In the pre-9/11 era, legal scholars and commentators relied on Webster’s formula to devise a narrow concept of imminent threat and put clear limits to the anticipatory use of force. After 9/11, however, the Bush and Obama administrations advanced a new, too-broad concept of anticipatory self-defence based on a definition of imminent threat that represented a radical departure from Webster’s formula. Both administrations ignored the traditional normative restraints to the anticipatory use of force in order to relax the rules governing the use of military force in drone warfare and ground combat operations.

This shift to a new version of anticipatory use of force created two major problems. The first problem is that the redefinition of preventive use of force in jus ad bellum provided cover for the expansion of US state power through the use of military force. By redefining the preventive use of force in jus ad bellum, the US granted itself the right to use military force against uncorroborated threats, and thus eased the restrictions for the use of force against other states. With the new definition, there is a danger that the US and other powerful states could use vague, unsubstantiated threats as a pretext to launch preventive attacks not to defend themselves but to pursue their narrow national interests (Švarc, 2006: 187–188).

The second major problem is the increased risk for civilians living in war zones. The too-broad definition of imminent threat in ground combat operations enabled US troops to be more flexible in deciding when to pull the trigger in ‘self-defence’ engagements, which led to civilians being more often put in danger of getting killed or injured (Gaston, 2017b: 7). While it is true that imminent threat determinations were always, to a certain extent, subjective, the new concept of imminent threat allowed excessive subjectivity that undermined civilian protection (IHRC, 2016: 18–21). According to Montalvo (2013: 26), the difficulties in assessing imminent threat, which was the result of the unclear meaning of the term, became a primary contributor to civilian casualties.

The redefinition of imminent threat in ground combat operations and the subsequent practices of targeting civilians provided new insights into the reasons behind violence against civilians in armed conflicts.⁷ Some authors – for example, Kahl (2007: 26) – observed that civilian casualties in anticipatory ‘self-defence’ incidents could be partly attributed to mistakes made by US troops due to the ‘fog of war’ (e.g. the difficulty in distinguishing civilians and combatants) and the constant stress under which the troops operated. While that was true in many cases, it is also important to acknowledge what was a more systemic source of civilian harm. As we have seen above, civilian casualties can be attributed, to a large extent, to the new concept of imminent threat that was part of a risk management strategy the goal of which was to reduce the risk for soldiers by increasing risk for civilians. Risk management strategies usually consist of force protection rules that lower the threshold for the use of military force, and, consequently, transfer the risk from soldiers to civilians (Smith, 2008: 144–145). That is what happened with the introduction of the new definition of imminent threat – killings of civilians were the result of a policy that aimed at minimizing casualties among US troops by using force protection rules that came at the expense of civilians.

Moreover, by lowering the threshold of proof to justify the preventive use of force, its redefinition in ground combat operations allowed US troops to operate with greater impunity, as, based on mere assumptions about future threats, they shot at and killed innocent civilians.

Footnotes

Acknowledgements

I would like to thank the reviewers for their valuable comments and suggestions on how to improve the article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was financially supported by the Slovenian Research Agency (research core funding No. P5-0221, and project Shifting boundaries in criminology and crime policy, J5-8242 B). The research also received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement no. 734855.

ORCID iD

Vasja Badalič

Notes

Vasja Badalič is a senior researcher at the Institute of Criminology at the Faculty of Law in Ljubljana, Slovenia. His primary fields of research are contemporary imperialism and migration. He has published many peer-reviewed articles in academic journals and monographs, including in Drones & Unmanned Aerial Systems (Springer, 2016). He is the author of the book The War Against Civilians: Victims of the ‘War on Terror’ in Afghanistan and Pakistan (Palgrave Macmillan, 2019).

References

Alston

(2009) Report of the Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions: Mission to Afghanistan. Geneva: UN Human Rights Council. Available at: www.refworld.org/docid/4a1cfb632.html (accessed 11 October 2017).

Amoore

Raley

(2017) Securing with algorithms: Knowledge, decision, sovereignty. Security Dialogue 48(1): 3–10.

Aradau

van Munster

(2007) Governing terrorism through risk: Taking precautions, (un)knowing the future. European Journal of International Relations 13(1): 89–115.

Arend

(2003) International law and the preemptive use of military force. The Washington Quarterly 26(2): 89–103.

Badalič

(2016) The predators’ rule of terror. In: Završnik

(ed.) Drones and Unmanned Aerial Systems: Legal and Social Implications for Security and Surveillance. Cham: Springer, 157–181.

Badalič

(2019) The War against Civilians: Victims of the ‘War on Terror’ in Afghanistan and Pakistan. London: Palgrave Macmillan.

Baker

(1987) Terrorism and the inherent right of self-defence (a call to amend article 51 of the united nations charter). Houston Journal of International Law 10(1): 25–50.

Beck

Arend

(1993) Don’t tread on us: International law and forcible state responses to terrorism. Wisconsin International Law Journal 12(2): 153–220.

Bowett

(2008) Self-defence in International Law. New Jersey: The Lawbook Exchange.

10.

Brennan

(2011) Strengthening our security by adhering to our values and laws. Available at: www.whitehouse.gov/the-press-office/2011/09/16/remarks-john-o-brennan-strengthening-our-security-adhering-our-values-an (accessed 10 October 2019).

11.

Byers

(2003) Preemptive self-defence: Hegemony, equality and strategies of legal change. Journal of Political Philosophy 11(2): 171–190.

12.

Byman

(2013) Why drones work: The case for Washington’s weapon of choice. Foreign Affairs 92(4): 32–43.

13.

CALL (Center for Army Lessons Learned) (2012) Afghanistan civilian casualty prevention: Observations, insights, and lessons. Available at: https://info.publicintelligence.net/CALL-AfghanCIVCAS.pdf (accessed 30 January 2018).

14.

Chairman of the Joint Chiefs of Staff (2005) Standing rules of engagement/Standing rules for the use of force for US forces. Available at: www.jag.navy.mil/distrib/instructions/CJCSI%203121.01B13Jun05.pdf (accessed 11 October 2019).

15.

Charney

(2001) The use of force against terrorism and international law. American Journal of International Law 95(4): 835–839.

16.

Dinstein

(2005) War, Aggression and Self-defence. Cambridge: Cambridge University Press.

17.

Fink

(2018) Maritime Interception and the Law of Naval Operations. Cham: Springer.

18.

Forrest

CJS

(2007) The doctrine of military necessity and the protection of cultural property during armed conflicts. California Western International Law Journal 37(2): 177–219.

19.

Franck

(2001) When, if ever, may states deploy military force without prior security council authorization. Washington University Journal of Law Policy 5: 51–68.

20.

Franck

(2002) Recourse to Force: State Action against Threats and Armed Attacks. Cambridge: Cambridge University Press.

21.

Gardam

(2004) Necessity, Proportionality and the Use of Force by States. Cambridge: Cambridge University Press.

22.

Gaston

(2017a) Reconceptualizing individual and unit self-defence as a combatant privilege. Harward National Security Journal 8(2): 283–332.

23.

Gaston

(2017b) When Looks Could Kill: Emerging State Practice on Self-defence and Hostile Intent. Berlin: Global Public Policy Institute.

24.

Glennon

(2002) The fog of law: Self-defence, inherence, and incoherence in Article 51 of the United Nations charter. Harvard Journal of Law and Public Policy 25(2): 539–558.

25.

Gregory

(2019) Dangerous feelings: Checkpoints and the perception of hostile intent. Security Dialogue 50(2): 131–147.

26.

Grotius

(2001) On the Law of War and Peace. Kitchener: Batoche.

27.

Hamid

(2007) The legality of anticipatory self-defence in the 21^st century world order: A re-appraisal. Netherlands International Law Review 54(3): 441–490.

28.

Higgins

(1963) The Development of International Law through the Political Organs of the United Nations. Oxford: Oxford University Press.

29.

Hong

Szpunar

(2019) The futures of anticipatory reason: Contingency and speculation in the sting operation. Security Dialogue 50(4): 314–330.

30.

Hosang

(2010) Force protection, unit self-defence, and extended self-defence. In: Gill

Fleck

(eds) The Handbook of the International Law of Military Operations. Oxford: Oxford University Press, 415–427.

31.

IHRC (International Human Rights Clinic – Harvard Law School) (2016) Tackling Tough Calls: Lessons from Recent Conflicts on Hostile Intent and Civilian Protection. Boston: Harvard Law School. Available at: www.justsecurity.org/wp-content/uploads/2016/03/Tackling-Tough-Choices-Hostile-Intent-HLSIHRC-2016.pdf (accessed 10 October 2019).

32.

IIFFMCG (Independent International Fact-Finding Mission on the Conflict in Georgia: Volume II) (2009) Available at: www.mpil.de/files/pdf4/IIFFMCG_Volume_II1.pdf (accessed 10 October 2019).

33.

Jennings

(1938) The Caroline and McLeod cases. The American Journal of International Law 32(1): 82–99.

34.

Joint Chiefs of Staff (2016) Department of Defence dictionary of military and associated terms (Joint Publication 1-02). Available at: https://fas.org/irp/doddir/dod/jp1_02.pdf (accessed 22 February 2019).

35.

Kahl

(2007) In the crossfire or the crosshairs? Norms, civilian casualties, and US conduct in Iraq. International Security 32(1): 7–46.

36.

Kelsen

(2000) The Law of the United Nations: A Critical Analysis of its Fundamental Problems. New Jersey: The Lawbook Exchange.

37.

Kunz

(1947) Individual and collective self-defence in Article 51 of the Charter of the United Nations. American Journal of International Law 41(4): 872–879.

38.

Levy

(2008) Preventive war and democratic politics. International Studies Quarterly 52(1): 1–24.

39.

Loader

Sparks

(2007) Contemporary landscapes of crime, order and control: Governance, risk and globalization. In: Maguire

Morgan

Reiner

(eds) The Oxford Handbook of Criminology, 78–101.

40.

Lubell

(2015) The problem of imminence in an uncertain world. In: Weller

(ed.) The Oxford Handbook of the Use of Force in International Law. Oxford: Oxford University Press, 697–719.

41.

Massumi

(2015) Ontopower: War, Powers, and the State of Perception. Durham, MD: Duke University Press.

42.

Mazzetti

Kulish

Drew

Kovaleski

Naylor

Ismay

(2015) SEAL Team 6: A secret history of quiet killings and blurred lines. The New York Times, 6 June. Available at: www.nytimes.com/2015/06/07/world/asia/the-secret-history-of-seal-team-6.html?mcubz=3 (accessed 11 October 2019).

43.

Montalvo

(2013) When did imminent stop meaning immediate? Jus in bello hostile intent, imminence, and self-defence in counterinsurgency. The Army Lawyer, 24–35.

44.

Murphy

(2002) Terrorism and the concept of armed attack in Article 51 of the UN Charter. Harvard International Law Journal 43(1): 41–51.

45.

NATO Legal Deskbook (Second Edition) (2010). Available at: https://info.publicintelligence.net/NATO-LegalDeskbook.pdf (accessed 18 January 2019).

46.

Niva

(2013) Disappearing violence: JSOC and the Pentagon’s new cartography of networked warfare. Security Dialogue 44(3): 185–202.

47.

O’Brien

(1990) Reprisals, deterrence and self-defence in counterterror operations. Virginia Journal of International Law 30(2): 421–478.

48.

Porter

(2011) ISAF data show night raids killed over 1,500 Afghan civilians. Inter Press Service, 2 November. Available at www.ipsnews.net/2011/11/isaf-data-show-night-raids-killed-over-1500-afghan-civilians/ (accessed 11 October 2019).

49.

Porter

Shah

(2011) U.N. Tally Excluded Most Afghan Civilian Deaths in Night Raids. IPS, 25 October. Available at: www.ipsnews.net/2011/10/un-tally-excluded-most-afghan-civilian-deaths-in-night-raids/ (accessed 11 October 2019).

50.

Roberts

(1987) Self-help in combatting state-sponsored terrorism: Self-defence and peacetime reprisals. Case Western Reserve Journal of International Law 19(2): 243–294.

51.

Rockefeller

(2004) The imminent threat requirement for the use of preemptive military force: Is it time for non-temporal standard. Denver Journal of International Law and Policy 33(1): 131–149.

52.

Rogoff

Collins

(1990) The Caroline Incident and the development of international law. Brooklyn Journal of International Law 16: 493–527.

53.

Rostow

(2009) International law and the use of force: A plea for realism. Yale Journal of International Law 34(2): 549–557.

54.

Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of International Armed Conflicts (Protocol I), 8 June 1977. Available at: https://ihl-databases.icrc.org/applic/ihl/ihl.nsf/Treaty.xsp?documentId=D9E6B6264D7723C3C12563CD002D6CE4&action=openDocument (accessed 11 October 2019).

55.

Rubin

(2011) Raid on wrong Afghan house kills Afghan girl, 12. The New York Times, 13 May, p. A12.

56.

Sahak

Rubin

(2011) For second time in 3 days, NATO raid kills Afghan child. The New York Times, 14 May. Available at: www.nytimes.com/2011/05/15/world/asia/15afghan.html?mcubz=3 (accessed 11 October 2019).

57.

Scahill

(2013) Dirty Wars: The World is a Battlefield. New York: Nation Books.

58.

Schachter

(1984) The right of states to use armed force. Michigan Law Review 82(5–6): 1620–1646.

59.

Schmidt

Trenta

(2018) Changes in the law of self-defence? Drones, imminence, and international drone dynamics. Journal on the Use of Force and International Law 5(2): 201–245.

60.

Schmitt

(2003) Preemptive strategies in international law. Michigan Journal of International Law 24(2): 513–548.

61.

Schulzke

(2017) The unintended consequences of war: Self-defence and violence against civilians in ground combat operations. International Studies Perspective 18(4): 391–408.

62.

Schwarz

(2016) Prescription drones: On the techno-biopolitical regimes of contemporary ‘ethical killing’. Security Dialogue 47(1): 59–75.

63.

Shah

(2007) Self-defence, anticipatory self-defence and pre-emption: International law’s response to terrorism. Journal of Conflict and Security Law 12(1): 95–126.

64.

Shanker

Bumiller

Nordland

(2010) Despite gains, night raids split US and Karzai. The New York Times, 15 November. Available at: www.nytimes.com/2010/11/16/world/asia/16night.html (accessed 28 August 2017).

65.

Simma

Khan

Nolte

Paulus

(eds) (2012) The Charter of the United Nations: A Commentary (Volume 2). Oxford: Oxford University Press.

66.

Smith

(2008) Protecting civilians . . . or soldiers? Humanitarian law and the economy of risk in Iraq. International Studies Perspective 9(2): 144–164.

67.

Starkey

(2010) Nato ‘covered up’ botched night raid in Afghanistan that killed five. The Times, 13 March. Available at: www.thetimes.co.uk/tto/news/world/asia/afghanistan/article2464783.ece (accessed 31 August 2017).

68.

Stephens

(1998) Rules of Engagement and the concept of unit self defence. Naval Law Review 45(126): 126–151.

69.

Švarc

(2006) Redefining imminence: The use of force against threats and armed attacks in the twenty-first century. ILSA Journal of International and Comparative Law 13(1): 171–191.

70.

Taft

Buchwald

(2003) Preemption, Iraq, and international law. The American Journal of International Law 97(3): 557–563.

71.

The White House (2002) The National Security Strategy of the United States of America. Available at: https://2009-2017.state.gov/documents/organization/63562.pdf (accessed 10 October 2019).

72.

Trachtenberg

(2007) Preventive war and US foreign policy. Security Studies 16(1): 1–31.

73.

Trenta

(2018) The Obama administration’s conceptual change: Imminence and the legitimation of targeted killings. European Journal of International Security 3(1): 69–93.

74.

Trumbull

(2012) The basis of unit self-defence and implications for the use of force. Duke Journal of Comparative and International Law 23(1): 121–148.

75.

US Department of Justice (2011) White paper: Lawfulness of a lethal operation directed against a US Citizen who is a senior operational leader of Al Qa’ida or an associated force. Available at: https://fas.org/irp/eprint/doj-lethal.pdf (accessed 11 October 2019).

76.

US Department of State (2003) Response to UNHCR SR on Yemen incident of 3 Nov 2002. Available at: www.aclu.org/files/dronefoia/dos/drone_dos_20110720DOS_DRONE001925.pdf (accessed 10 October 2019).

77.

Van den hole

(2003) Anticipatory self-defence under international law. American University International Law Review 19(1): 69–106.

78.

Van Linschoten

Kuehn

(2011) A Knock at the Door: 22 Months of ISAF Press Releases. Kabul: Afghanistan Analyst Network. Available at: www.afghanistan-analysts.org/wp-content/uploads/downloads/2012/10/AAN_2011_ISAFPressReleases.pdf (accessed 11 October 2019).

79.

Webster

(1841) Letter from Daniel Webster, US Secretary of State, to Henry S. Fox, Esq., Envoy Extraordinary and Minister Plenipotentiary of Her Britannic Majesty (Apr. 24, 1841), reprinted in British and Foreign State Papers, 1840–1841. London: British Foreign Office. Available at: https://babel.hathitrust.org/cgi/pt?id=mdp.39015019751299;view=1up;seq=7 (accessed 11 October 2019).

80.

Williams

Lyons

Neuwirth

(2003) Preemption in the 21st century: What are the legal parameters. ILSA Journal of International and Comparative Law 10: 353–360.

81.

Završnik

(2019) Algorithmic justice: Algorithms and big data in criminal justice settings. European Journal of Criminology. Epub ahead of print 18 September 2019. DOI: 10.1177/1477370819876762.

82.

Zedner

(2007) Pre-crime and post-criminology? Theoretical Criminology 11(2): 261–281.