8.1 Overview
The purpose of this chapter is to examine the application of self-defence by military forces during armed conflict operations. While it may have been acceptable in the past to view soldiers as cannon fodder, it is now clear that “military lives are not simply expendable”.1 As has been pointed out in this book, there is not only a tendency to rely more heavily on the application of self-defence, the ability of military forces to do so during armed conflicts may seem exaggerated. Although members of military forces are individuals with individual rights, they are also State agents or tools for the use of force, and a balance must be found between these two roles. Military forces’ (over)reliance on self-defence as an exception to the requirement for roe authorisation directly challenges the need to control the use of force by State actors equipped and trained to kill. The question of when military forces may rely on self-defence is therefore central to this balance.
This chapter is divided into two parts. The first part focuses on the legal concept of personal self-defence as found in national criminal law, and applies the common principles of criminal law self-defence to the unique circumstances military forces may find themselves in during armed conflict. What is the realistic scope of application of self-defence during armed conflict? When and how may military forces use force in self-defence? Because reliance on self-defence by military forces appears to be more extensive than what the legal concept of self-defence would be expected to permit, the third part of this chapter will examine the existence of operational concepts of self-defence and the legal authorities for using forces under these concepts. This will include an assessment of whether an ‘operational self-defence’ concept has evolved, in addition to examining the better-known operational concepts of ‘unit self-defence’, ‘extended self-defence’, and ‘force protection’.
8.2 The Use of Force in Personal Self-Defence by Military Forces during Armed Conflict
8.2.1 Clarifications
Before examining the scope military forces have of using force in self-defence during armed conflict, some clarifications are in order.
First, this section is not intended as an argument in favour of applying self-defence in situations where loac applies. Admittedly, loac is likely to provide sufficient room to act in most cases;2 however, in those limited circumstances where loac does not provide proper room to act, self-defence becomes essential.3 If we accept the fact that soldiers also have a right to life and may use force in self-defence, it is important to define the parameters of this option.
Second, because the current focus is on the application of personal self-defence during armed conflict, the discussion will concentrate on the aspects that are most relevant to understanding the relationship between self-defence and loac. It is not intended to provide a complete overview of when and how self-defence may be relied upon by military forces.
Although most if not all States apply the same requirements for lawful self-defence, the interpretation and practical application of the principles vary. It should therefore be kept in mind that the following limitations on the use of force in self-defence may not apply to the same extent in all domestic systems. Finally, it should be emphasised that if the requirements of self-defence are not met, the use of force by military forces during an armed conflict will only be lawful when authorised by roe and in accordance with applicable law, primarily loac.4
8.2.2 Subjective Considerations
The use of force in self-defence is assessed on the basis of subjective and objective standards.5 The subjective aspect of self-defence is relevant both to how the threat is perceived and the requirement of a defensive intent. While the subjective intent may be sufficient to find that a person acted in self-defence even if the objective criteria are not met, the lack of defensive intent may prevent such a finding, even if, by coincidence, the objective criteria appear to be fulfilled.6
Both the necessity and proportionality requirements will be assessed in light of how an individual arguing self-defence perceived the situation. For instance, if a person truly believes that another is about to attack him and therefore defends himself, it may be accepted as self-defence even if the ‘attacker’ was only joking. Similarly, if a person being attacked thinks that her only way out is to shoot the attacker, the defensive act may be deemed proportionate even if the result is to kill the attacker who had no intention of killing, but only injuring the victim. However, in most jurisdictions there is an objective limit to what is acceptable; the perception must be reasonable.7
What is reasonable will depend on the circumstances, but it is clear that the unique training military forces are given, especially in the use of force, will affect the expectations of how they perceive and deal with threats.8 The relevance of military forces’ unique skill set is also emphasised by Boddens Hosang, who argues that “[t]he level of training and expertise that may be expected of military personnel increases the level of care and diligence expected from military personnel and may impose restrictions on the applicability of the criminal law concept of personal self-defence in an operational context”.9 Military forces are in other words likely to be assessed to a higher standard than civilians with little or no experience in being attacked. This higher standard may be applied in relation to the assessment of the initial threat, including whether it is imminent and whether it is possible to avoid the use of force. It may also affect how the choice of defensive measure is perceived and whether it meets the requirement to use no more force than necessary. Because of the training and experience military forces commonly have, it will be less likely that they become so stressed or panic-stricken that they misinterpret the situation or overreact and use disproportionate force in response to the threat.10
Military forces do not only have unique training and experience with dealing with threats in general, they are often provided with theatre-specific training to improve situational awareness and better enable them correctly analyse a situation.11 The classic example is the presence of weapons: the fact that someone is carrying a weapon should not be considered conclusive that the person is a threat if it is common for civilians to carry weapons for their own protection. This may affect the assessment of whether it was reasonable to mistake a harmless situation to be a threat.
Another aspect of the mind-set of the person using force in defence is his or her motive for doing so. As explained in Section 6.3.2.1, most jurisdictions require that the person acting in self-defence has a defensive intent or at least be aware of the circumstances that entitle him or her to act in self-defence.12 As Scaliotti explains, a person who raises a defence such as self-defence “usually claims that the actus reus was indeed committed, but for a good, or at least an acceptable, purpose”.13 The requirement of defensive intent is arguably central to self-defence as a justification for the use of force because it eliminates the possibility that self-defence may be relied upon to justify an offensive attack that happened to coincide with the opponent’s attack.14 Military forces will, however, commonly use force not only to stop an attack but also to ‘defeat the enemy’, and the use of force may be pre-planned rather than merely a response to a threatening situation.
The defensive intent must be honestly held, and if there is a mistake about the existence of the attack, that mistake must be genuine.15 Whether the mistake is also required to be reasonable, depends on the jurisdiction.16 However, human rights law requires the prosecution of state officials who cause death due to recklessness or gross negligence.17
8.2.3 The Impact of the Requirement of Unlawful Act
The requirement of unlawful act is relevant to the application of self-defence in two respects. First, the use of force causing the self-defence situation must be unlawful. Furthermore, the use of force in response to this initial threat must also be otherwise unlawful, that is, it would be unlawful if the criteria for lawful self-defence are not met.18 These requirements will be dealt with separately. If military forces have a legal right or authority to use force in domestic legislation, the use of force in response to an attack will normally not require any justification under criminal law. This issue will therefore be considered first.
8.2.3.1 ‘Lawful Acts of War’ and the Authority to Use Force
Due to the nature of personal self-defence as a defence in criminal law, it is a prerequisite that the use of force in self-defence would have been unlawful had the requirements for self-defence not been met. The application of personal self-defence during armed conflict therefore depends on how the respective States have integrated the concept of ‘lawful acts of war’ or ‘combatant immunity’ into their domestic legislation. As explained above in Section 5.1.2, there are different approaches to this. It may, inter alia, be included in an authority to use force on behalf of the state, or it may be included as a criminal law defence, justifying the use of force.
Where the authority of military forces to use force is provided independently, soldiers will not need to rely on personal self-defence to justify the use of force as personal self-defence. The use of force is per se lawful, provided it meets applicable criteria. As one commentator has explained: where the attacker is a lawful target according to loac, it is unnecessary to rely on self-defence, “as the case-in-chief would lack an essential element”.19 Soldiers from countries with this approach to combatant immunity will therefore primarily need to rely on personal self-defence in situations not covered by the authority, such as when the threat is unrelated to the conflict.20 If military forces use force beyond what is authorised by their Commander, for instance through roe, the use of force itself does not necessarily become unlawful, thereby making self-defence relevant. Instead it will be a case of failure to comply with lawful orders, which is a separate offence. The relationship between self-defence and military orders is examined in Section 8.2.10.
If ‘lawful acts of war’ is a criminal law exculpatory ground that justifies the use of force, whether as a separate concept or as part of a ‘public authority’ defence, it may apply in parallel to other exculpatory criminal law defences, including personal self-defence. Though the use of force in accordance with loac is perceived as lawful, in States where combatant immunity is incorporated as a criminal law defence, it is strictly speaking only lawful because it is justified, as is the case with self-defence. Combatant immunity and personal self-defence will be equally applicable in the same situation, even though the result will be different due to their different requirements and purposes. The existence or application of one exculpatory ground does not negate the application of another; they are not mutually exclusive. The question will then be which is most appropriate, an issue that will be examined further in this chapter.
The approach of including combatant immunity as a personal defence rather than a State’s positive right is unfortunate for several reasons. It departs from the general principle of State monopoly on the use of force, which implies that the State has the right to use force and that this use of force is in itself lawful. As will be further elaborated on in Section 8.3.2 when discussing the existence of an ‘operational self-defence’ concept, it shifts the responsibility for the use of force from the State onto the individual soldiers, who may have to prove before a court that force used during a military operation was justified and hence lawful. Finally, it also departs from the approach taken in the echr whereby lawful acts of war are treated as distinct from the list of exceptions to the right to life, permitted only when absolutely necessary.21
8.2.3.2 Illegality of Initial Threat or Attack
National rules on self-defence commonly require the initial threat or use of force that causes the self-defence situation to be unlawful.22 The use of force is unlawful when it lacks lawful authority; it does not need to be criminal.23 As Gill et al. explain, “[t]he right of self-defence is by definition limited to repelling illegal prior or impending use of force. It has no bearing upon responding to force which is legally sanctioned”.24 In a peacetime context, this means that it is not permissible to use force in self-defence against law enforcers when carrying out their official duties, provided the law enforcers are acting within the applicable law. Similarly, there is no self-defence against lawful self-defence.25 In the context of military personnel participating in an armed conflict, the requirement of the initial threat or act being unlawful has even wider practical implications. Attacks by the enemy’s military forces are lawful if directed at military objectives in accordance with loac, and the response to such attacks will be part of the conduct of hostilities.26 As a result, the requirement that the threat or attack is unlawful renders personal self-defence “mostly inapplicable or irrelevant”27 to military operations during armed conflict.
Since only combatants have a right to participate in hostilities and carry out lawful acts of war, and only State forces are combatants, the requirement that the threat or attack is unlawful will yield different results for iacs and niacs.28 Unlike civilians directly participating in hostilities or organised armed groups, combatants may lawfully use force against persons who are lawful targets, and because they themselves are lawful targets, an attack upon them by another combatant is also lawful. Although ‘combatant immunity’ is only included in the law governing iacs, State practice seem to be to afford the State’s own forces such immunity also during and after their participation in a niac. By contrast, although there is no prohibition under international law for non-combatants to use force, there is no requirement to provide immunity for such acts either. Attacks by persons taking direct part in hostilities without being a combatant, either because the requirements are not met or because it is a niac, will therefore rarely be considered lawful under domestic law.29 As a result, the potential for relying on self-defence is therefore greater in niacs and in relation to persons taking a direct part in hostilities.
The application of personal self-defence by a soldier during an armed conflict will primarily be limited to two instances: where the threat or attack is unrelated to the armed conflict (i.e. lacks belligerent nexus) or the connection is not sufficiently clear;30 and where the attack is unlawful under domestic law or international law. This will primarily apply in the case of a person using force without having the right to participate in hostilities, thereby violating domestic law. Alternatively, the use of force may violate loac, for instance because the attacker (combatant or non-combatant) uses unlawful means of warfare.31
In the first category, loac will not provide the legal authority to use force, and soldiers will in most cases have to rely on self-defence.32 Examples of situations that may fall within this category are attack by an angry farmer on soldiers marching through a field thereby destroying the farmer’s crops; violence by participants in a civilian demonstration, for instance, against the soldiers’ use of land or in response to a cultural or religious insult; and hostile acts by persons involved in other civilian violence, such as armed robbery or pirates interfering with a maritime mission.33 A self-defence situation may also arise when civilians violently try to breach a naval blockade.34 According to loac, if there is doubt as to whether a person is a lawful target or not, he or she should be dealt with as a protected person.35 This category will therefore also include situations where the nexus between the threat and the armed conflict may not be sufficiently clear to allow the soldiers to conclude that the person is a lawful target. This issue is particularly relevant in relation to civilians directly participating in hostilities without uniforms or other distinctive signs that enable them to be distinguished from protected civilians. As a result, self-defence is likely to play a greater role in niacs where the opposing forces fail to distinguish themselves from the civilian population.36
Second, self-defence may be relevant if the attack is related to the armed conflict, but the person carrying out the attack is not permitted to use force, and is for instance taking a direct part in hostilities. Self-defence may also be relevant if the force used is otherwise unlawful, as in the case of a combatant using unlawful means. In both cases, an authority for the use of force in response to the initial attack would be found in loac. However, the requirement for using force in self-defence may also be met, and where the facts leading to a loss of protection are unclear, self-defence may provide a sounder legal basis. This will be the case if the person attacking is entitled to special protection, such as medical or religious personnel, with the result that use of force against him will not be a lawful act of war unless that protection is lost.37 If the attack is to be carried out by a combatant using unlawful means or methods, such as using poison or poisoned weapons,38 feigning surrender,39 or misusing a protective emblem,40 the attacker will still be a lawful target and the use of force authorised by loac may be applied; however, the requirement for using force in self-defence may also be met and could be easier to fulfil. Despite this narrow potential for applying self-defence, it should be emphasised that, in most cases, loac will provide the troops with a more robust and suitable legal room of manoeuvre. As a result, unless the option of relying on loac is restricted by roe, self-defence will be superfluous.
A consequence of the requirement of unlawful attack is that the soldier must make an assessment of the legality of the act in order to know which law to apply: self-defence or loac. This is problematic because it is difficult to do under the circumstances and requires the soldiers to have extensive legal knowledge. There is therefore a need for clear instructions for how to deal with these issues, in roe or other places, especially during operations in which the soldiers are expected to face situations where loac will not provide sufficient authorities for dealing with such threats.
8.2.4 When May Force Be Used? The Limitations Imposed by the Principles of Necessity and Imminence
All legal systems require that the use of force in self-defence be necessary and that the threat be imminent. The principle of necessity requires the use of forceful means to defend against the threat or attack to be necessary, and that it be necessary to choose the particular means used to defend.41 As a means of ensuring that the use of force is ‘genuinely necessary’, some systems also emphasise that the threat must be imminent.42 Furthermore, if the threat or attack has passed, the need to defend is no longer present. Although some argue that the question of the degree or amount of force used is also an aspect of necessity in that it should be no more than necessary,43 this question will be dealt with as an aspect of proportionality.
8.2.4.1 Creating the Necessity: to Incite the Self-Defence Situation
If the person who is attacked caused the self-defence situation to arise, this can undermine an individual’s ability to rely on self-defence as a justification for the use of force. The role of the attacked person may be divided into two categories: active incitement (provocation) and passive incitement (to knowingly or deliberately place oneself in harm’s way).44 The issue is sometimes referred to as dolus and culpa in causa.45
Prior provocation of the defending party, to the extent that the use of force would not have been necessary had it not been for the defending party’s behaviour, will in many cases negate the ability to justify that the use of force was in self-defence.46 As Eser explains, self-defence may not be available if the defendant “provoked the attack for the purpose of getting a chance to counteract” since the defendant would be “predominantly motivated by other that [sic] defensive ends”.47 The challenge for military forces is that much of their activity is intended to provoke or must be expected to provoke. In fact, as explained in Chapter 2, nato roe regulate both the use of force and measures that may be construed as provocative. Whether the provocation is of a kind that should negate the defender’s ability to justify the use of force by reliance on personal self-defence will be context-dependent. This is emphasised by Boddens Hosang, who explains that the assessment will require “extensive interpretation and situational awareness on the part of the court in trials involving military personnel invoking the right of self-defence”.48 It has been argued, however, that courts will only deny an application of self-defence where “a person intentionally put himself in a state in which he could claim self-defence”.49 Where the need to use force to defend arises as a result of such provocation, either due to application of provocative roe or otherwise, soldiers may not be able to rely on self-defence to justify their use of force.50 The ability to use force in such situations will primarily be determined by loac. If loac does not permit soldiers to use force in the circumstances, they may rely on self-defence if the attack is disproportionate to the provocation.
The second form of incitement, passive incitement, is described by Boddens Hosang as “knowingly and willingly seeking out specific situations or deliberately placing oneself in harm’s way in order that a need for the use of force in self-defence will ultimately arise”.51 He concludes that because many ordinary tasks of military forces during armed conflict involve placing oneself in harm’s way or seeking out situations of danger, it will be difficult to rely on personal self-defence to use force in such situations.52 However, a distinction must be drawn between passive incitements and continuing a behaviour or continue to move in a certain direction despite knowing that it involves a risk of attack where the potential incitement is not intended. The freedom of movement of the person who is at risk of attack must also be taken into consideration and be protected. This was emphasised for instance in the American case of State v. Bristol (1938): “the question [is] whether or not the law can afford to encourage bullies to stalk about the land and terrorize citizens by their mere threats”.53
In the context of military operations, this is particularly important in relation to military forces’ interaction with the civilian population. Unless civilians directly participate in hostilities, it is prohibited to attack them. However, threats or attacks may occur that do not have the necessary belligerent nexus to render the civilian in question a lawful target. A merchant may for instance be angry because the military forces accidentally damaged his market stall when driving past. Where the risk of incitement is known, but not intended, it should not negate the ability to rely on self-defence should the need actually arise. As mentioned above, a case of culpa in causa will primarily be found where the incitement is intentional. Where the person attacking military forces is not a lawful target, soldiers may need to rely on self-defence to justify their use of force to defend themselves, and the fact that they somehow incited the attack does not necessary preclude that option. By way of conclusion, soldiers are unlikely to be able to rely on self-defence if they actively provoked the self-defence situation or intended to incite it. However, they may still be able to resort to self-defence where the risk was known but coincidental to their operations. This is likely to be particularly relevant where the source of the threat is a protected person.
As will be further discussed in Section 8.3.2, a trend appears to have developed whereby forces provoke or incite an attack that enables them to respond in self-defence because they lack the roe to carry out an attack. According to Husby, for example, “some patrol missions in Afghanistan have been designed to draw out adversaries, thereby triggering tic and self-defense authorities under the roe”.54 Corn refers to this “improper” practice as “baited self-defense”.55 The use of force outside the roe undermines the Commander’s decision not to authorise such use of force. The severity of such practice will depend on mission. For instance, a distinction may be drawn between operations designed to defeat as many members of the opposing forces as possible, and operations in which the mission is to enhance governance and foster a safe and secure environment. In the latter case, ‘trailing’ for tic without the roe to do so or using force in response to the threat it creates is likely to undermine the mission. Additionally, it is a misuse of self-defence that is unlikely to be accepted.
8.2.4.2 Duty to Retreat? The Use of Lethal Force as a Last Resort
The approach to whether retreat is required differs between legislation and has been subject to considerable change within the respective jurisdictions.56 For instance, the U.S. Supreme Court announced in 1895 that true men are not obliged to flee,57 while the Model Penal Code requires the victim to avoid the necessity of using force by retreating, when this may be done with complete safety.58 The question of retreat is one of the more controversial aspects of self-defence law.59 On the one hand, it may be argued that the culpable aggressor has forfeited his/her right to life, and that the victim is entitled to stand his/her ground.60 Furthermore, the emphasis on retreat has been criticised in the context of attacks in the home or work-place (the so-called ‘castle doctrine’). Even where there is such a duty, the castle-doctrine is generally accepted as an exception.61 On the other hand, the question is influenced by the development of ihrl. For parties to the echr, for instance, the use of lethal force in self-defence must be “absolutely necessary”.62 Ashworth has noted that the human rights approach to self-defence entails the maximum protection of all citizens, rather than “the law abiding citizen’s feelings of honour and self-respect at the expense of the criminal’s right to life or physical security”.63
It appears that the general trend is to consider retreat as an aspect of the necessity analysis.64 Force may only be used if necessary; if reasonably possible to retreat, this would negate the necessity of using force.65 The focus on the culpability of the aggressor and the view that the aggressor forfeits his or her right to life by causing the self-defence situation is less common. At the same time, the aggressor’s culpability may still be relevant to the question of retreat. Where it is known that the aggressor lacks culpability, for example when he or she is a child, an increased emphasis on the use of force only as a last resort is to be expected, if not for legal reasons then certainly for moral ones.
The application of this rule to military forces as well as civilians was emphasised by the Judge Advocate acting in the trial of Willi Tessmann and others before a British Military Court in Hamburg, 1947: “The law permits a man to save his own life by despatching that of another, but it must be in the last resort. He is expected to retreat to the uttermost before turning and killing his assailant; and, of course, such considerations as the nature of the weapon in the hands of the accused, the question whether the assailant had any weapon and so forth, have to be considered. In other words, was it a last resort? Had he retreated to the uttermost before ending the life of another human being?”.66 Similarly, in the U.S. sroe, it is stressed that when time and circumstances permit, the attacker should be warned and given the opportunity to withdraw or cease the threatening actions.67 These requirements clearly do not fit well with military operations, or as Boddens Hosang observes: “It is self-evident that this doctrine has no real use in making personal self-defence any more readily suitable as a basis for the use of force in the context of achieving military objectives”.68 There is no such duty to retreat or even deescalate the situation in loac, although those who do not participate in the hostilities, such as civilians and persons hors de combat, are entitled to protection against the dangers arising from military operations.69
8.2.4.3 Imminence
As explained above, imminence may be viewed as an aspect of necessity, serveing the function of ensuring that the use of force is truly necessary. The criterion that the threat of attack be imminent ensures that defensive force is not used too soon, before it is strictly necessary.70 As explained in the context of jus ad bellum self-defence,71 imminence should be distinguished from immediacy, which requires that the response not be too late, and thereby amount to revenge rather than defence.72
The imminence requirement raises particular challenges in relation to threats that are inevitable but not imminent, and where the victim will struggle to defend him/herself when the attack occurs. Imminence as an independent criterion, as opposed to an aspect of necessity, has therefore been criticised, especially in the context of domestic violence.73 In the context of military forces participating in an armed conflict, if military forces were to rely on self-defence, the requirement that they only use force when the threat is imminent would clearly hamper their ability to pre-plan operations and to carry out offensive operations. In many cases it is desirable and even necessary to deal with known threats before they become imminent. This is probably why, according to Bagwell and Kovite, “[t]he most constraining requirement, and perhaps the most systematically abused, is that U.S. soldiers may exercise self-defense only when a threat is imminent”.74
8.2.5 Limitations on the Defensive Act: Proportionality
8.2.5.1 Introduction
When the initial test of necessity has been met, the force used in defence must meet the further test of proportionality; it must not be disproportionate to its defensive purpose. Proportionality in the context of self-defence concerns in other words the relationship between the threat and the response, and must be distinguished from the separate concept, identically coined, in loac. As explained in Section 5.1.1, proportionality in loac deals with the relationship between the concrete and direct military advantage anticipated and expected (unavoidable) harm to civilians and civilian objects. The question relating to the use of force in self-defence and harm to innocent bystanders is, strictly speaking, not a question of the relationship between the threat and the response, and is therefore dealt with separately below (Section 8.2.7).75
8.2.5.2 Escalation of Force Procedures
One tool for ensuring that the force used is no more than necessary is the so-called escalation of force procedures (EoF).76 A graduated response enables the person attacked to a certain degree to tailor the response to the threat, thereby ensuring that it is not excessive. Examples of means that may be employed include oral warning, non-lethal force, warning shots,77 and disabling fire, before the final option, lethal use of force.78 The defender is only expected to take those measures that are reasonable, that is, the measure should be expected to effectively counter the attack and should not endanger the defender.79 The ability to adapt the response to the threat will depend on several factors, in particular the means available and, perhaps more importantly, training. This is an area where military personnel may be assessed differently than ordinary civilians, even when the force is used in personal self-defence. A person with no training in the use of force or in handling attacks is more likely to panic than those with such training. Furthermore, the training military personnel are given in handling such threats will provide them with knowledge of, and experience in, different ways of dealing with a threat. They should therefore have more options to exploit before having to resort to lethal force.80 Finally, they may be trained in de-escalatory techniques, thereby reducing the need to use defensive force, although this is more likely part of the training of law enforcement officials than military personnel.81
Although EoF procedures are clearly a useful tool for identifying the appropriate and proportionate response to a threat, it is important that they do not replace the good judgement of soldiers. The decision to use lethal force should not merely be based on the failure of person perceived as a threat to comply with certain procedures. As one commentator has explained, there is a risk that such procedures are “sanctioning violence based solely on a civilian’s (or insurgent’s) unlucky decision to cross an arbitrary line in the sand” such as “distance from the potential threat and the potential threat’s response to verbal, visual, and other warnings”.82 The application of EoF during armed conflict has also raised the concern that the soldiers may become too hesitant, thereby providing ineffective protection to their own forces or to civilians.
8.2.5.3 The Least Destructive or Damaging Choice of Defence
The proportionality requirement in self-defence is sometimes expressed as “minimum use of force”, which is further explained as requiring the response to be ‘no more than necessary’.83 The distinction between proportionality and necessity is complicated, something which is evidenced by how commentators characterise the choice of means of methods: as an issue of necessity; 84 both proportionality and necessity,85 or as an aspect of proportionality.86 Provided that all aspects are taken into consideration in the examination of legality of the use of force, what matters less is under which principle an issue is considered to fall. The current approach is to consider the choice of means and methods as related to the principle of proportionality. If a less destructive or damaging option of dealing with a threat is available, then the option chosen is excessive and hence unlawful. The same approach is taken by the icty Trial Chamber in the Gotovina case: “The Trial Chamber further considered the perpetrators’ conduct, even if an immediate illegitimate attack could be assumed, to be disproportionate, where other ways of thwarting any possible danger instead of firing lethal shots were available”.87 This does not mean that there is a requirement of equality of arms, so that firearms could only be used in response to firearms, for instance; it only means that it is not disproportionate to the threat.
The choice of means or methods will be assessed on the basis of the reasonable means and methods available at the time. In many States be prohibited to carry lethal weapons during peacetime. By contrast, military forces are likely to have at least one if not more lethal means at their disposal, and, depending on the mission, few if any non-lethal means. As a result, the level of force in response to a threat may be higher than during most peacetime situations.88 This is reasonable considering that military forces are primarily expected to use force in accordance with loac. However, if it is expected that they will need to rely on self-defence as a basis for the use of force, they should be provided the adequate equipment to apply graduated force. Failure to do so may result in State responsibility for human rights violations in relation to deaths resulting from the lack of adequate equipment.89 As emphasised by Hessbruegge: “[i]f state agents have to defend themselves with lethal force because the state failed to provide them with less lethal weapons or protective equipment, such force is not absolutely necessary and hence unlawful”. 90
The requirement of not using more force than necessary will also affect how the available equipment is used. In relation to escalation of force procedures as mentioned above, a firearm may be used to fire warning shots or disabling fire, in addition to lethal force. It may also be an option to shoot to wound rather than to aim to kill, depending on the circumstances and whether shoot to wound is permitted by national rules.91 Finally, proportionality will affect the amount of force used. As an example, although it may be proportionate to shoot back at someone having taken a shot at a soldier, even if he missed, it is likely to be disproportionate if the whole unit fires hundreds of rounds each in return.92 Similarly, it may be sufficient to set automatic weapons to single shot mode rather than constant fire mode.93
8.2.5.4 The Limits of Proportionality
There are differing approaches to what exactly should be taken into account as part of the proportionality analysis. For instance, while the approach taken here is that it concerns the degree of force used to avert the danger,94 others also include an assessment of the nature and duration of the force.95 Hessbruegge sets out four elements that have to be factored in: “the intensity and the extent of the harm inflicted versus that defended against; the probability of harm if the defensive measure is taken or not taken; the number of victims versus the number of aggressors harmed; and the culpability or non-culpability of the aggressor”.96 The question of probability of harm is arguably an aspect of necessity, and is dealt with above. Similarly, the question of whether the aggressor is culpable or not is arguably not a question of proportionality, but whether the person responsible may be held criminally responsible for the act.97 Although moral arguments may be made for limiting the degree of force used in situations where it is known that the person is for instance a child or mentally ill, the threat necessitating the use of a certain degree of force in self-defence remains the same. However, as mentioned above, the culpability of the aggressor, or at least known non-culpability, may be a factor to consider under necessity and whether all non-forceful or less forceful means were sufficiently exploited.
The requirement of proportionality for the use of force in self-defence indicates that excessive use of force will be excluded from legitimate self-defence. The consequences of disproportionate use of force are dealt with differently in different jurisdictions. If self-defence is rejected, the initial attack may function as a partial defence, that is, as a ground for reduction of sentence or reduction of the charge from for instance murder to manslaughter.98 The effect of excessive self-defence depends on the cause of the disproportionate use of force.
There are three likely causes of excessive self-defence: aggression; a mistaken belief; and fear and/or panic. If the use of force in self-defence is excessive because the purpose is not (or is no longer) to defend against the unlawful attack, it lacks the defensive or legitimate purpose,99 and is therefore unjustified. By contrast, if the force is excessive due to a mistake made as to the degree or amount of force necessary to repel the attack, the use of force will be justified provided the mistake was reasonable.100 Finally, a number of countries permit a person acting in self-defence to use excessive force if this is due to fear or panic. For instance, as mentioned in Chapter 6, the German Criminal Code states that “A person who exceeds the limits of self-defence out of confusion, fear or terror shall not be held criminally liable”.101 The inclusion of fear and panic as justification for excessive use of force has been criticised as being contrary to human rights obligations, especially in the context of State officials using force in self-defence.102
As discussed in Section 8.2.2, as a result of their training and experience, military forces are expected not to let fear or panic cause them to use excessive force. As Hessbruegge argues, they “must have the appropriate moral, psychological, and physical qualities for the effective exercise of their functions. They should therefore be particularly resistant to letting fear determine their actions”.103 He goes on to assert that in order for the fear to be justified, the situation must be so threatening that a reasonable officer would have responded in a similar manner. This seems a sound conclusion; military forces may only be able to rely on self-defence to justify excessive use of force caused by fear or panic in situations where other military personnel would also be expected to react disproportionately. However, the extensive training military forces are given in handling extreme situations of danger, beyond the training generally given to law enforcement officers, for instance, will in practice leave a very limited scope for excessive self-defence being justified.
8.2.6 The Use of Otherwise Unlawful Means and Methods in Self-Defence
The extent to which otherwise unlawful means and methods may be used in self-defence is particularly pertinent to situations of armed conflict because loac contains both prohibitions of and limitations on means and methods. The violations of some of these rules are considered grave breaches of loac and may therefore amount to a war crime.104 As explained above (Section 6.3.1.2), it is possible to plead self-defence as a defence to a war crimes indictment, provided it meets the requirements set out in Article 31(c). States party to the Rome Statute of the icc will therefore be expected to make their self-defence rules applicable to war crimes cases.105
Certain means, such as torture and inhuman treatment,106 are absolutely prohibited and will therefore not be available even in self-defence because they are per se disproportionate. As Hessbruegge points out, certain means are “categorically outlawed under international law and thus per se disproportionate”.107 Other means are either completely prohibited during armed conflict, or their use in a certain manner is prohibited. This distinction appears relevant also with regard to the question of whether they may be used in self-defence. If a particular weapon or ammunition is not lawful to use in any circumstances during the operation to which the military forces are deployed, it should not be present and available to the troops. For example, for States that have ratified the Ottawa Convention, anti-personnel mines should not be an available option, even in self-defence.108 Other examples of prohibited means include chemical weapons,109 biological weapons,110 poison or poisoned weapons,111 and for certain States, certain cluster munitions.112
If prohibited means are deployed in order to be available in cases of self-defence, the fact that they were only used in self-defence is unlikely to be accepted as a justification for using unlawful means if the person using it knew it was unlawful.113 The availability of such unlawful means would not be reasonable, and the deliberate attempt to circumvent the prohibition would likely undermine the requirement of defensive intent. The only potential exception is in relation to means that are not absolutely prohibited, but prohibited for certain States or certain circumstances. If a partner State has not ratified the same prohibition and have therefore deployed them, the means may become available even for military forces from States subject to the prohibition. In such cases, the question will be whether its use was proportionate and would be perceived as such by a reasonable person in a similar position.
The same test will have to be applied in the context of means that may only be used for certain purposes. Examples of such means include: white phosphor intended for illumination;114 exploding ammunition under 400 grams such as certain multipurpose ammunitions intended for anti-vehicle use;115 and tear gas deployed for riot control purposes.116 Would a reasonable person in a similar situation consider it proportionate to use such means to defend if the use is prohibited by loac? While the assessment will depend on the facts and how the defender perceived the situation, certain weapons are so damaging that they are unlikely to be accepted even in self-defence.
The use of incendiary weapons against combatants could have been expected to fall into this category. However, according to the icrc Customary International Law Study, the anti-personnel use of incendiary weapons is only prohibited if it is not feasible to use a less harmful weapon to render a person hors de combat.117 Presumably this also means that there is a potential for the use of incendiary weapons to be permitted in self-defence. The use of exploding ammunition against persons may also be accepted in self-defence, especially if there is insufficient time to reload reasonably available weapons with alternative ammunition. The use of tear gas is trickier to assess because, when it is available, it is likely to be the least harmful alternative and therefore proportionate. At the same time, the prohibition against its use as a method of warfare has become so entrenched after the First World War that it may be argued that it is absolute. Furthermore, while the prohibition is only on using it as a method of warfare and not in self-defence, the prohibition is an aspect of the proportionality assessment where the attacker is a lawful target in accordance with loac, such as a person taking direct part in hostilities during a niac.
The legality of use of prohibited methods in self-defence raises different questions. As illustrated by the examples in Section 6.3.2.3, self-defence rules may potentially be applied to all prohibited actions, not just use of force. 118 This means that self-defence may theoretically justify the use of prohibited methods, provided the requirements of necessity and proportionality are met. For instance, loac prohibits the improper use of a flag of truce, or the flag or of the military insignia and uniform of the enemy or of the United Nations, as well as of the distinctive emblems of the Geneva Conventions.119 It is also unlawful to feign hors de combat status or civilian status with the intent to abuse the status in order to kill or injure the adversary.120 If a soldier is under an imminent threat of attack, and the only way to survive is to give the attacker the false impression that the soldier is a person entitled to protection, such as by pretending to be injured or by putting on a white medical coat, doing so is clearly necessary and would probably also be considered proportionate. If the soldier decides to use the impression that he is protected to enable him to use force against the attackers, this would amount to a war crime.121 Whether the use of force was necessary would depend on the availability of other means, such as escape and, even if necessary, the force must of course be proportionate.
Most prohibited methods of warfare will, however, never meet the requirements of self-defence. For instance, prohibited methods such as starvation as a method of warfare,122 pillage,123 or orders that no quarters will be given124 are not individual acts suitable to deal with concrete and imminent threats; they will fail the necessity requirement. Furthermore, it is difficult to see how they would be carried out with a defensive intent.
In conclusion, there appears to be some scope for employing means and methods that are prohibited by loac in self-defence. However, this scope is narrow.
8.2.7 Self-Defence and Innocent Bystanders: Is ‘Collateral Damage’ Acceptable?
While there are clear requirements in loac to avoid harm to protected civilians when using force,125 the self-defence rules on the issue are less clear. As explained above, the threat to innocent bystanders is not part of the self-defence proportionality analysis as it is not related to the assessment of the relationship between the threat and its response. Instead, it requires a separate proportionality assessment. Strictly speaking, it is more a question of necessity than self-defence, as the innocent bystanders represent no unlawful threat.126 As explained in Section 6.3.2.1, during armed conflict, the only form of necessity that applies is loac and the principle of military necessity therein. The general principle of necessity may therefore not be applied as a justification for violating loac. In practice, however, there appears to be a tendency to assess the two effects of the defensive act together, as part of the self-defence assessment.127 Whether this is due to the limitations imposed on necessity by loac during armed conflicts, or perhaps done merely to simplify the analysis, is not clear, and will not be speculated on here.128
Even if the effects on the target and on innocent bystanders are considered as part of self-defence, the proportionality assessment remains different. According to Hessbruegge, “[w]hile defensive force against attackers need only be not disproportionate, a stricter standard applies vis-à-vis innocent bystanders. The risk of harm for the innocent bystanders must be outweighed by the harm the defensive force is expected to prevent”.129 If the use of force in self-defence results in harm to innocent bystanders, the lawfulness of that use of force must be assessed as part of the proportionality and reasonableness standards. What is clear, however, is that less incidental harm to peaceful bystanders is tolerated under self-defence than under loac.130
loac requires that those who plan or decide upon an attack take all feasible precautions to minimise the risk to civilians.131 The requirement of minimising the risk to civilians before using force reflects the unique character of military force as planned, decided and in many cases intentionally lethal. Self-defence, on the other hand, should be a measure of last resort in response to an imminent threat, and there is arguably less time and opportunity to assess and minimise harm to third persons prior to using force than what is required by loac.132 However, certain precautionary measures are still required, in particular when the defensive measure is carried out by State actors. According to Hessbruegge, the State is required “to factor in and take appropriate precautions covering all foreseeable risks of its defensive actions, including the possibility that innocent bystanders will be harmed by the aggressors”.133 The precautionary requirements for the use of force under self-defence are therefore more extensive. It includes all who are affected by the use of force and not merely bystanders. However, due to the imminence of the threat, it is likely that fewer precautionary measures are expected in a self-defence situation than under loac.134
If innocent bystanders have been harmed as a result of self-defence, a distinction should be made between unintended harm and harmed caused knowingly or recklessly. If the person using force in self-defence does not know of the risk of harm to bystanders and unintentionally causes them injury or death, the use of force is likely to be considered reasonable.135 For instance, in the Bakan case, a police officer accidentally killed an innocent bystander with a warning shot that ricocheted. The ECtHR concluded that this was not a violation of Article 2 as it was mere bad luck.136 The use of force may also be reasonable if there is potential for harm, but the alternative if no defensive measure is taken is foreseeable harm.137 This was established by the ECtHR in the Finogenov case, which dealt with the use of gas in the Moscow theatre hostage crisis, resulting in the incidental killing of 125 hostages. The ECtHR concluded that the use of gas “was not in the circumstances a disproportionate measure, and, as such, did not breach Article 2 [the right to life]”.138 Importantly, the Court made a distinction between means or methods that are supposed to kill, such as bombs and missile, and the gas used, which was dangerous but “left the hostages a high chance of survival”.139
By contrast, if the defender knowingly or recklessly injures or kills a bystander, the use of force is more likely to be considered excessive.140 For instance, it may be reckless to use automatic weapons in a constant fire mode where there are innocent bystanders nearby who are likely to be harmed.141 In the Tagayeva case, the ECtHR held that the massive use of indiscriminate weapons to fight 30 terrorists holding over a thousand civilians hostage in a Beslan school violated Article 2.142 In certain circumstances, however, it may be reasonable to knowingly use lethal force against innocent bystanders, such as when the killing of one innocent bystander is expected to save the lives of others. For instance, it may be reasonable and proportionate to shoot down a civilian airliner that has been hijacked and is expected to be crashed into a building killing both the passengers and crew and those on the ground. Because these are complex scenarios challenging the right to life of the person or persons sacrificed for some greater good, national jurisdictions can be expected to take different approaches. A distinction may be made, for example, between sacrificing the life of someone who has a chance of survival and someone who is expected to be killed if no defensive measure is taken.143
In conclusion, there may be scope for allowing injury to innocent bystanders as a result of use of force in self-defence, even when the risk is known. However, such harm will only in very rare and extreme circumstances be considered proportionate and reasonable. As Henderson and Cavanagh emphasise, “it is highly likely that reasonableness under the law of self-defence imposes a higher standard of care on a military member to avoid causing any injury or death to civilians”.144
8.2.8 Protection of Others
Many national self-defence rules permit the use of force in defence of others. Common law countries have a tendency to limit this to persons with a close connection to the defending party, while civil law countries are more likely to permit the use of force in self-defence of any person.145 When military forces operate in an armed conflict environment, both civilians and military personnel may find themselves in a situation of imminent threat and in need of help. At the same time, the State or nato mission has a need to control when and how military forces use force. For this reason there are roe regulating the use of force on behalf of others, and a distinction is made between one’s own forces, partner forces and organisations, and others. If military forces defend others in situations that are not authorised by roe, either because the roe is not approved or it is retained and thereby requiring authorisation from a higher authority, this would run the risk of undermining the mission. As will be explained below, this is therefore an area where it is more likely that military orders will be accepted as a limitation on the use of force in self-defence.
Acknowledging this challenge, Boddens Hosang proposes that the necessity consideration should include an assessment of proximity between the person attacked and the person carrying out the defence. In his view, “the geographical relationship between those to be defended and those coming to their rescue may be taken into consideration to determine whether the force used was part of the right of personal self-defence or part of a military engagement in the context of operational roe”.146 This entails that the further away the forces in need of help are, the less likely it is possible for other forces to rely on personal self-defence as a justification to help them. The deployment to defend other forces must instead be planned and executed on the basis of applicable roe and other operational orders. He emphasises that as a result, stand-by units such as close air support (cas) units will not be able to rely on personal self-defence but are bound by applicable roe.
This approach makes legal sense, even if it leaves a gap in the protection of one’s own forces where the roe are restrictive and the forces able to come to their rescue are not nearby. Because such situations are not uncommon, the ability to apply defensive force should not depend on the limited scope for relying on personal self-defence. The more appropriate authority for the use of defensive force on behalf of other forces is the operational concept of ‘unit self-defence’ (see below, Section 8.3.3). What Boddens Hosang’s solution fails to deal with, however, is the use of force in defence of third persons, for instance the local population or representatives of ngos. There may be several reasons for military forces not to get involved even if they are in close geographical proximity, especially with regards to the local population. The most important reason is that the use of force against a civilian may lead to increased tensions and undesired escalation of the conflict. This may be the case where the use of force to defend a person from one civilian group may cause nato forces to be perceived as impartial.147 Similarly, if the violence is culturally acceptable in the area, but not where the military forces come from, as could be the case with domestic violence, it may be perceived as inappropriate interference. At the same time, military forces should not stand by and watch atrocities or serious harm being inflicted, and roe will commonly include authorisations to use force to prevent the commission of serious crimes such as murder, rape, or aggravated result, or to defend civilians from acts that endanger life or are likely to cause serious bodily harm.
If the roe is retained and there is insufficient time to release it, the question becomes whether the military forces may still be permitted to intervene, and whether they are obliged to do so. As was explained above in Section 6.3.2.1, the legal question of self-defence as a duty is a question of the scope of the obligation to protect the human right to life. It is not a question of whether there is a duty to defend others, but rather when and whom.148 The State has a duty to protect the lives of those under its human rights jurisdiction. This clearly includes its own military forces and citizens. The question is less clear with regards to citizens of other countries. This is a question of extraterritorial application of human rights, and will depend on the extent to which the persons in question are under an effective control of the State agents who could act in their defence.149
The duty to protect others is, however, not absolute. According to Hessbruegge, a balance must be sought between the right to life of others and the risk to the lives of those intervening: “In deciding how best to defend others, the state and its agents may factor in whether a particular defensive measure would cause unreasonable risks for the intervening police officers”.150 Although the State agents should not be required to expose themselves to unreasonable risks, it is clear as Hessbruegge emphasises that “police officers in the line of duty can be expected to incur a certain level of risk to save that life”.151 This will apply to military forces as well, and the level of risk military forces should be expected to accept is arguably higher than for police officers, especially for operations involving armed conflict.
In military operations, it may also be that the military forces are ordered to protect certain groups or persons, beyond those subject to the human rights obligation of the soldiers’ State; in that case they have an obligation to comply with that order. For instance, and as will be explained in Section 8.3.4, nato States have agreed that under the concept of ‘extended self-defence’ their forces will protect forces from other nato States as well.
8.2.9 Protection of Objects
States have taken different approaches in their domestic legislation on the issue of using force in self-defence to protect objects. This topic is therefore usually regulated by roe. Where the use of lethal force in defence of objects is possible under self-defence, it is likely to only be permitted where the attack on the object is expected to result in an imminent risk of death or serious injury.152 As a result, the discussion above and below with regards to application and possible limitation of the use of force in defence of persons will apply equally to the protection of objects.
8.2.10 Self-Defence and Limitations Imposed by Military Orders
8.2.10.1 Introduction
The final issue to be considered in the context of the application of personal self-defence by military forces participating in armed conflict operations is the question of whether, and if so, when personal self-defence may be limited by military orders such as roe. While the common reference to self-defence as inherent and not limited by roe gives the impression that self-defence may never be limited, as the discussion above has indicated, this does not hold true for all aspects of self-defence. The possibility of roe limiting the right of self-defence is reflected both in roe doctrine153 and state practice. For instance, Admiral Woodward, Commander of the UK forces during the Falklands War, has stated in his account of that conflict that “I had, in effect, taken away some of my Commanders’ right of self-defence, further restricting the rules from home which allowed them to fire back. But I did not want this war to go off at half-cock, because that would likely cause disastrous confusion and loss of control”.154 The following discussion will examine the extent to which self-defence may lawfully be imposed limitations, and subsequently, the relationship between orders imposing such limitations and the application of self-defence.
8.2.10.2 When May Self-Defence Lawfully Be Limited?
When examining the question of when self-defence may be limited, a distinction must be drawn between the protection of others or objects, and the protection of self. The use of force, especially lethal force, to protect objects is normally dependant on whether it involves a risk of injury or death to persons, and limitations on the ability to defend objects beyond this are likely to be uncontroversial. The focus here will therefore be on the protection of persons.
The question of protection of others depends on whether the soldiers have a right or duty to protect them, while the ability to protect oneself relates to the soldiers’ right to life and the State’s duty to protect this right. A further distinction must therefore be made between situations where the military forces have a duty to protect because they are State agents, and where they would be legally justified to act but are not obliged to do so. Where there is no duty to defend, a decision based on operational or force-protection considerations to limit the ability of military forces to defensively use force would presumably be legally unproblematic.155 Such a limitation may, however, still be ethically or politically problematic, especially when those involved are likely to be innocent civilians in a war-torn area. Where there is a duty to protect others, this may only be restricted if the defensive act is expected to cause unreasonable risk to the military forces, bearing in mind that they must accept a certain level of risk due to their occupation.156 The same test should be applied to the defence of other members of the military forces, both their own forces and those covered by the nato concept of ‘extended self-defence’.157 The State may only limit their ability to use force to defend other nato forces where it would result in unreasonable risk to the military forces carrying out the defence.
This leaves the question of whether military forces retain the ability to use force to save their own lives, regardless of the risk to the mission or to uninvolved forces, or whether this also may be limited to where the State or Commander considers it reasonable. As Hessbruegge explains, “a balance has to be struck between legitimate interests regarding the command and control of military operations and soldiers’ right to life and physical security that underpin their right to personal self-defence”.158 Limitations may be imposed, but they may not be unreasonable. At the core of the issue is the question of how much risk military forces reasonably must accept due to their occupation. Because military forces also have a right to life,159 it is clear that they should not be ordered to commit suicide; however, it is also important that they do not use force that is not authorised by the roe and therefore subject to command and control unless it is absolutely necessary. They may be placed at risk, and even be ordered to show “courageous restraint”,160 but they cannot be ordered to not defend their lives in situations of risk, as this would violate their right to life.161
Commanders are therefore to some extent entitled to prohibit soldiers from using force in self-defence, such as to gain a tactical advantage or to enhance the protection of the civilian population.162 Operationally, limitations on the use of force in self-defence are more likely to be imposed for naval and air forces than ground forces. This is because the former forces are more likely to have direct communications with the Commander, and they have access to larger weapon systems which individual soldiers or sailors are less likely to be permitted to use without higher authorisations. Finally, it should be emphasised that the ability of Commanders to restrict the use of self-defence will, to some extent, shift the responsibility for ensuring that self-defence measures are appropriately exercised from the individual soldiers onto the Commanders.163
Two distinctions may be made to support the determination of when personal self-defence, in the form of defence of self, may be limited. The first distinction is between imminent threats and actual attacks. In the Canadian Use of Force Manual, for instance, it is stated that a Commander may order forces not to respond to an imminent attack, but no such authority to impose limitations is mentioned with regards to actual attacks.164 According to Cathcart, the authority to restrict the response to threats of attack is limited to exceptional circumstances “when there is a clear need to de-escalate a situation with no other alternative”.165 Although self-defence allows the use of force in response to imminent threats and not just actual attacks, military forces must accept a high level of risk. It is therefore logical that the threshold of necessity for the use of force in self-defence is higher for military forces participating in an armed conflict than for persons not having accepted that risk. It may also be that the Commander limiting the right of self-defence is not aware of the severity of the threat his or her forces are faced with. If the forces, unbeknownst to be Commander, are faced with a lethal threat, they are not obliged to comply with the order not to use force.
The second distinction that should be made is between a risk of light injuries and a serious risk of death or serious injury. This is also related to the assessment of the necessity to use force in self-defence. A certain degree of injury must be accepted as part of a military occupation. However, denying military forces the right to defend themselves when there is a risk of death or serious injury, and that risk is serious would, as Hessbruegge points out, be “tantamount to expecting them to commit suicide or self-mutilation”.166 This would clearly violate the military forces’ right to life.167
Finally, it may be that a State decides that the interests of the State outweigh the individual soldier’s human rights, even where the use of force in self-defence is lawful, and an order restricting that use of force would be unlawful. As Boddens Hosang explains, “an individual serviceman’s reactions in a self-defence situation may trigger military reactions or political reactions in or around the theatre of operations which change the entire context of the operation”.168 By way of illustration, he suggests a scenario where a soldier, who is patrolling the demarcation line between North Korea and South Korea, is fired upon from North Korea, and the response in self-defence has the potential for triggering a full-scale armed conflict.169 A similar situation could have arised in Syria in 2017, where forces from nato States and Russian forces were supporting either local groups or the Syrian government, and were therefore at risk of becoming an imminent threat to each other. A State may in other words determine that avoiding a full-scale armed conflict is more important than the life of individual soldiers, and therefore restrict the ability to use force in self-defence in violation of their human right to life. Although the denial of the right of self-defence would, as discussed above, generally be unlawful, it is clear that “the national interest requires that the law should accord the widest measure of appreciation to commanders on the ground who have the responsibility of planning for and conducting operations there”.170 Furthermore, the State could argue that any wrongfulness arising from limitations on military forces’ right to life is precluded by the demands of national security. The legality of such limitations would therefore be a question of the scope and application of constitutional necessity. This is, however, beyond the scope of this book.
8.2.10.3 Restrictive Orders and Self-Defence
It is important to distinguish between the legality of the use of force and the legality of the failure to comply with orders. As mentioned above in Section 8.2.3.1, the use of force may itself be justified in self-defence and at the same time involve a dereliction of duty.171 This is because military forces have a duty to comply with orders, unless they are manifestly unlawful.172 The use of force to defend third persons may be lawful under criminal law, but if it is done in disregard of an order to not intervene, there is a dereliction of duty. Furthermore, although many if not most unreasonable limitations on the ability to use force in self-defence will be manifestly unlawful, borderline cases will most likely not be considered unlawul. For instance, while so-called ‘hold-fire’ orders are mentioned by some as an example of a limitation on self-defence that may be permitted in some circumstances,173 others contend that such orders are not to be considered a limitation on self-defence.174 Furthermore, the person using force may perceive the situation as more threatening than his or her on-scene Commander, with the result that both the order and the use of force in self-defence may be lawful.
The relationship between military orders and self-defence is therefore not straight forward, and it is influenced by the requirement of an initial illegality for self-defence to apply. If a soldier uses force in lawful self-defence in violation of a lawful order restricting the use of force, the defence against the dereliction of duty cannot be self-defence because the order is not an unlawful act. The relevant defence is instead likely to be a form of necessity, with self-defence acting indirectly to explain why the use of force was necessary and reasonable.175 Necessity to defend oneself may also be relevant if a soldier is ordered not to use force in accordance with loac, but the consequence of that order is to place the person at imminent risk of death or serious injury, and the soldier therefore disregards the order. If the order not to use force is unlawful, but not manifestly unlawful, and the use of force is also lawful, self-defence may be relied upon to justify why the soldier did not comply with the order not to use force. Finally, if the order is manifestly unlawful because it clearly violates the soldier’s right to life, there is no duty to comply with the order, and no defence is required for disregarding it by acting in lawful self-defence.
As a result, it is arguably an oversimplification of the issue when Henderson and Cavanagh state that the question, “can a commander issue an enforceable order to a soldier not to shoot in situations where the soldier would otherwise be able to shoot in self-defence?” may be rephrased as “can a soldier rely on the defence of self-defence when charged with the offence of failing to comply with an order?”.176 Still, their conclusion is still sound, that it should be permissible to disobey the order when it is necessary and reasonable, even if the applicable defence is not always self-defence.177
8.3 Operational Concepts of Self-Defence
8.3.1 Introduction
This Section deals with operational ‘use of force’ concepts that are referred to as a form of self-defence, but do not necessarily meet the requirements of personal self-defence set out above in Section 8.2. During nato operations, the use of force or other provocative means must be authorised by roe. The only exception to this is ‘self-defence’. As a result of the relatively restrictive roe and other applicable procedures and orders applied in recent operations, in particularly isaf, the role of this self-defence exception has become more significant, and appears to have evolved into a term for all defensive force, regardless of its legal basis.178
Restrictive roe entail that the military forces involved in the operation are unable to apply all of loac. As a result, the use of force not authorised by roe may still be lawful under loac. loac regulates both offensive and defensive operations as forms of attack,179 which means that loac also applies to defensive operations. Calling something self-defence does not necessarily make it self-defence in a legal sense, and the law applicable to defensive force during an armed conflict is more likely to be loac than the legal concept of self-defence.180 In cases where loac does not apply, the relevant legal basis for the use of force in self-defence is most likely to be that of personal self-defence, as explained in Chapter 6. This is because the reliance on State self-defence during an ongoing armed conflict would in most cases entail an undesirable and unwarranted conflation of the jus ad bellum and jus in bello.181 As will be further discussed in Section 8.3.3, there appears to be little support for the existence of a separate legal concept of ‘unit self-defence’.
Ideally, there should be a legal authority for using defensive force not authorised by loac, distinct from self-defence.182 It may for instance be that the degree of belligerent nexus required to conclude that the person is taking direct part in hostilities is insufficient, but the threat is nonetheless imminent and requiring a response. Where force is used in relation to mission accomplishment, the military forces are acting as State actors rather than private individuals. They should therefore be able to rely on a legal authority authorising the use of force necessary to carry out their duties in such cases, rather than it being otherwise unlawful but justified due to the requirements of personal self-defence being met. However, this does not appear to be the general approach, at least not among the States studied in this book, nor is the existence of such authorities reflected in academic writing. Unless there is a clear national provision authorising the use of force in situations where there is a need to use defensive force beyond what is authorised by loac, courts are likely to consider those acts in the context of personal self-defence. It will therefore be misleading and unhelpful to military forces to construe a new customary international legal basis for the use of force in these situations, unless it is followed up by formalisation in either national or international law that enables its application by a court. As a result, the use of defensive force by military forces during an armed conflict must either be authorised by loac or be accepted as self-defence.
8.3.2 ‘Operational Self-Defence’183
8.3.2.1 The Existence of an Operational Concept of Self-Defence
As mentioned above, the combination of the nato policy to require all use of force to be authorised either by roe or self-defence and the existence of relatively restrictive roe and other ‘use of force’ procedures, has resulted in ‘self-defence’ as a roe exception becoming particularly important. Like the legal concept, this operational concept focuses on the defensive use of force. However, the legal basis will in many cases be loac, such as the notion of direct participation in hostilities,184 rather than self-defence. For example, the operational concept covers defensive force both in response to civilians not participating in the armed conflict, but nonetheless posing an imminent threat, and opposing forces who are lawful targets under loac. In other words, the operational concept of self-defence is wider than the legal concept, and is consequently a great cause for confusion. This section will further explain what this operational concept of self-defence is and how it relates to the legal concept of self-defence.
In complex conflicts such as counter-insurgency (coin) operations or otherwise politically sensitive operations, the use of force has been subject to extensive and detailed control. For example, in conflicts where the opposing forces fail to distinguish themselves from the civilian population, one way of reducing the risk of harm to innocent civilians may be to restrict the use of force to situations where there is a clear threat. In addition to the roe being restrictive, further limitations on the use of force are imposed, inter alia, in directives and standard operating procedures. For example, due to a very low tolerance for civilian casualties, the use of force in situations that involve a high risk to civilians, such as indirect fire in civilian populated areas and the search of civilian compounds, has been subjected to strict regulations.185 Limitations may also be imposed on the geographical area in which military forces are permitted to operate, especially in multinational operations where such limitations are important to de-conflict simultaneously ongoing operations.186
Although retained roe or target engagement authorities may be released upon request if the use of force is justified, obtaining such a release may be a time-consuming process. This has created a need to assure military force that they are still permitted to save their own lives or the lives of other members of their force in extreme situations. The result is arguably an overemphasis on the importance of the right to use defensive force, even for operations involving participation in an armed conflict.187 Whether the use of force needed to save their lives is based on a legal concept of self-defence or loac is probably not something military forces will take the time to contemplate.
As a result of this practice, a military concept of self-defence appears to have developed, a concept focused on the act of defending lives rather than on the legal basis for this use of force.188 In their report from an expert meeting on “The use of force in armed conflicts”, the icrc made the following observation: “The concept of self-defence as embodied in RoE was not understood as domestic self-defence or human rights-based self-defence, but rather as an autonomous military concept, which is more permissive. The extent of this notion of military self-defence varies however from one State to another (in the same way as the concept of domestic self-defence)”.189 This confusion of self-defence and warfighting is also evidenced in how States review defensive force incidents. For instance, in a U.S. investigation into alleged civilian casualties caused in Boz Village, Kunduz, on November 2–3, 2016, it was concluded that “U.S. forces acted in self-defense, in accordance with the Law of Armed Conflict”.190
In the context of nato operations, the application of self-defence as an exception to roe is emphasised in the introductory general text (gentext) of the roe for a nato operation: ‘nothing in these roe should be read as limiting the inherent right of self-defence’. As explained previously,191 it seems that the reference to an inherent right of self-defence not limited by roe (even in roe for operations involving armed conflict) is included for historical reasons rather than on the basis of a carefully considered legal analysis.192 Although the reference to ‘inherent’ and the heavy reliance on self-defence during military operations may have been legally sound when initially included in roe for a peacetime context, i.e., as a reference to the jus ad bellum, the continued reliance on the phrase for all nato operations and seemingly to all forms of self-defence is erroneous.193 As explained by Boddens Hosang: it “suggests that the concept has a significance and meaning beyond its legal limits, which can lead to the erroneous assumption that it confers upon nato military personnel a right to use force that applies at any time and under any circumstance. It does not”.194 A similar concern is voiced by Gary Corn, who in the context of the U.S. sroe explains that “the continued inclusion of ‘inherent right’ language (…), language lifted directly from Article 51 of the UN Charter – has generated an entrenched misunderstanding among many that individual servicemembers and unit commanders possess an inviolate ‘natural law’ right of self-preservation independent of their status as members of the military, a right that ultimately prevails over any command-imposed restraints on the use of force”.195
The challenge is, as explained in Section 8.2, that the application of the legal concept of self-defence by military forces during armed conflict is relatively narrow; it is considerably more limited than the impression given by the inclusion of and especially the emphasis on the declaration on inherent self-defence in the roe gentext. Even if the right of personal self-defence is inherent, its characteristics make it unsuitable as a legal basis for the use of force during warfare, and it would not permit the use of force required in the circumstances it is relied upon. Rather than being an indication of the legal basis for the use of force, the reference to self-defence as an exception to the requirement of roe authorisation for the use of force should therefore be understood as referring to a general concept of defending oneself and other members of the force, albeit in self-defence-like situations. The roe in general do not specify the legal bases for the various authorisations for using force, and there may be more than one relevant legal basis for the respective roe, depending on the circumstances. This has also become true for ‘self-defence’ as a roe exception.196
The importance of the declaration that roe will not limit the right of self-defence is therefore not the reference to the legal basis for the use of such force, but rather the permission it provides to defend oneself. It is being used to activate some emergency procedures that enable a use of force that would otherwise not be permitted by roe and other operational orders or regulations. As Corn explains, “self-defense is often cited as an exception to restrictive (…) roe and a basis to conduct hasty, tactical targeting whenever troops are engaged by insurgents (i.e. situations of troops in contact [tic])”.197 As a result of the term self-defence being used, the authority to use force beyond roe is understood as being limited to exceptional circumstances, and subject to the limitations of necessity and proportionality. During armed conflict, where the use of force by military forces will in most cases be authorised by loac, the limitations on the use of force to situations that are similar to self-defence are imposed for political and military reasons. The result is, as Henderson and Cavanagh also conclude, that self-defence is used “as a label (albeit a potentially confusing one) by command to impose an operational constraint on the use of force. In such cases ‘self-defence’ does not operate as a criminal defence but rather as an roe trigger”.198
Perhaps the biggest challenge with this approach is that no distinction is made between the operational self-defence concept and the criminal law concept. Although defensive force during armed conflict mostly will be based on loac, situations may also arise where the criminal law concept of self-defence must be relied upon. The ability to rely on the legal concept of self-defence to use force not authorised by roe is also included in the gentext reference to roe not limiting the right of self-defence. It seems, in other words, that the roe concept of self-defence refers to two forms of self-defence: a (limited) legal right; and a wider operational or military concept. The distinction between these concepts and the relationship between them will therefore be further examined.
8.3.2.2 Distinguishing the Operational Concept of Self-Defence from the Legal Concept
The following example may be used to explain how the operational concept of self-defence can be distinguished from the legal concept of self-defence.
During an armed conflict, a military unit becomes surrounded by opposing forces, and is fighting to survive. The unit requests close air support (cas) to defeat opposing forces and enable it to pull out of the situation. However, because there is a risk of collateral damage involved in using indirect fire, the use of cas is strictly regulated. The Commander has decided that, due to the importance of protecting civilians and maintaining their support, cas or other forms of indirect fire may not be used if there is a risk of civilian casualties – even if this risk is proportionate to the military advantage anticipated to be achieved and therefore lawful in accordance with loac. The only exception would be in extreme situations where cas would be the only way to save soldiers’ lives. In the current example, there are no units on stand-by for cas for the operation, and the necessary approvals for cas had not been sought in advance. Furthermore, there is no Forward Air Controller (fac) on the ground, that is, a person specialised in giving the necessary information to ensure the munition is dropped in the appropriate place and assessing potential collateral damage. The use of cas usually requires a fac.
The forces requesting support are emphasising that they are under overwhelming attack and it is therefore a case of ‘self-defence’. As a result, emergency cas (also known as cas in extremis, casiE) is authorised. The opposing forces are hit with a 500 lb. bomb, the smallest ammunition available on the plane that was sufficiently near to arrive in the area in time to rescue the unit. The attack results in a small number of civilians being killed and causes extensive damage to neighbouring civilian buildings and infrastructure.
The forces behind the attack have been identified as the opposing forces and are therefore lawful targets. Assuming that the pilot took the required precautions to minimise the risk to civilians and that the collateral damage expected was proportionate to the anticipated military advantage, the attack would be in accordance with loac. Although the use of cas was generally not permitted if it was expected to injure or kill innocent civilians, it could nonetheless be authorised because the operation may be perceived as a ‘self-defence’ operation.
It would not, however, be a case of self-defence as a matter of law. First of all, because loac is specifically designed to regulate the use of force during armed conflict, it would be natural to seek legal authority in this legal regime if possible. Furthermore, if it is an iac, there is no initial unlawful attack because the military forces are lawful targets and the opposing forces are entitled to attack them. If it is a niac, where the opposing forces are not combatants and are therefore considered not to have the right to participate in the hostilities, it may be argued that their attack is unlawful, thereby meeting this initial threshold for self-defence. However, even if the requirement of unlawful attack was met, the use of force in response to that attack would also be regulated by loac, making self-defence superfluous.199 Furthermore, self-defence will usually not permit killing innocent bystanders where the likelihood of that result is foreseen. Self-defence will, in other words, usually not be an appropriate legal basis for an emergency cas (or other forms of war fighting). Because the use of force is permitted by loac, the reference to ‘self-defence’ functions a trigger for authorising the use of force beyond roe or procedures imposing limitations on the use of indirect fire, rather than suggesting that the use of force is legally founded on the criminal law concept of self-defence.
From a legal perspective, it would be preferable if the reference to self-defence in roe, and the instruction that the roe do not limit or restrict the right of self-defence, only referred to the legal concept of self-defence.200 However, this does not reflect the practice of military forces which, as explained above, is to rely on the declaration regarding ‘inherent right of self-defence’ in a wider range of defensive operations involving risk to soldiers’ lives.201 Even if this practice is based on a misinterpretation of the application of the legal principle of self-defence, it has now become so entrenched in military practice that it may be difficult to explain to troops that the actions, which to the soldier appear similar and which they know by the same term, should be differentiated because the legal basis for the use of force differs. As Gary Solis points out, “It is unrealistic to expect that the term ‘self-defense’ not be used colloquially in instances where one combatant returns the fire of another”.202 As a result, rather than attempting to limit the content of the well-established military concept of self-defence referred to in roe, the better approach may be to accept the unfortunate reality of a legal term being used in a different and non-legal sense. It is basically necessary to accept, as Henderson and Cavanagh argues, that “[s]elf-defence is not a unitary concept, but rather has different legal and operational meaning”.203 Not insisting that the reference to self-defence in roe be read only as self-defence in a legal sense removes any need to change the terminology from ‘self-defence’ to, for instance, emergency defensive operations or words to that extent.
Maintaining an operational self-defence concept may also reinforce the perception of the use of force contrary to general orders as only being permitted in extreme circumstances akin to those giving rise to a right of self-defence. Because the application of this concept entails that force is being used in situations not subject to the Commander’s command and control, it is appropriate to limit such use of force to that which is necessary and proportionate to defend the forces under threat. A Commander may also choose to impose limitations on the weapons permitted in the use of ‘operational self-defence’. Provided the force used is in accordance with loac, the application of the self-defence requirements amount to self-imposed limitations. Such additional limitations on the use of force are important because they prevent the application of ‘operational self-defence’ from becoming too extensive. As Boddens Hosang points out, it could potentially include “the use of all measures which are considered necessary to protect the (international) military force”.204 Limiting the use of force beyond roe to that which is necessary and proportionate to deal with an imminent threat, should prevent the use of force which has undesirable effects on the operation, or worse, not justified by either loac or self-defence.
Finally, because ‘operational self-defence’ permits the use of force outside the general command and control framework, a broad interpretation and application of the concept would undermine the Commander’s ability to exercise effective command and control over the forces. This is undesirable both with regard to the ability to ensure mission success and because it reduces the Commander’s ability to influence how and when force is used in those circumstances that are perceived as self-defence, either in the operational or legal sense of the term.205 Furthermore, concerns have been raised that “incidents justified under self-defense are more difficult to scrutinize, and may frequently not be held to account. This is in part due to the nature of these incidents, in part due to the inherent or inalienable character of this right, and in part due to the ambiguity over what the standards are”.206 Narrowing the scope for relying on a form of ‘self-defence’ during armed conflict in situations that are legally premised on loac, preferably by incorporating and authorising the force necessary within the roe set itself, could therefore reduce the risk of and even enhance accountability for potential loac violations.207
8.3.2.3 The Relationship between the Operational and Legal Concepts of Self-Defence
This operational, or non-legal, concept of ‘self-defence’ would of course not affect the application or relevance of the legal concept of self-defence. This may still be applicable if the alternative legal bases are not sufficient or not applicable.208 However, personal self-defence should be perceived as “a safety net, a measure of last resort in the event that the mission-specific authorizations to use force, as set forth in the roe, and the ‘military’ types of self-defence (…) are somehow insufficient to save one’s life”.209 Although this safety net is important and comforting, its role in military operations should be downplayed. As explained in Section 8.2, when applying the requirements of unlawful attack, necessity and proportionality to the unique situation that military forces find themselves in during an armed conflict, the right of personal self-defence becomes very different from the one military personnel are familiar with in a civilian or peacetime context.210 As a result, although self-defence may be perceived as a broader legal authority than loac because there is no requirement of a nexus to an armed conflict or that the attacker is a lawful target, its actual application is generally more restrictive than loac. This broad yet restrictive scope of the application of self-defence is emphasised by Bagwell and Kovite: “[Self-defence] is broad because, when acting in self-defence, a soldier may use force against anyone, including a civilian, who presents an imminent threat. It is restrictive because once force is authorised, only as much force as is necessary to neutralize the threat may be used. Additionally, force may only be used where it is not possible to mitigate the threat by other means”.211
While self-defence will generally permit less use of force than loac, the situations in which force may be used differ. It is therefore important to clarify the difference between the legal concept of self-defence and the use of force under the operational self-defence concept, based on loac. It is particularly important that Commanders and their legal advisers are aware of the applicable law and its implications, and that they make certain that the regulations and guidance provided to troops ensures the use of force which complies with the applicable law.212 The distinction between self-defence and use of force based on loac relates both to when force may be used and how.
First, the legal concept of self-defence does not require the identification of the opposing force as lawful targets, but depends instead on the necessity of using force. As a result, the use of force will no longer be lawful under self-defence if the person no longer poses a threat. Under loac, a person who a lawful target continues to be so either at all times (combatants and members of organised armed groups) or at least for a period after taking a direct part in hostilities (civilians).213 Second, although the use of force permitted in self-defence generally is more restrictive than loac, it may also permit the use of force that would not be lawful under loac. For instance, there is no equivalent to the loac requirements of precautions in attack for the use of force in self-defence. As a result, where loac would require the forces to do everything feasible in their planning of operations to minimise the risk to civilians and civilian objects, the use of force in self-defence is assumed not to be pre-planned and the expectations of such precautions being taken are considerably more limited. Furthermore, certain means which are prohibited in warfare, such as expanding bullets and tear gas, may be permitted in self-defence.
Imposing the self-defence principles of necessity and proportionality as limitations on the use of force during armed conflict is therefore not sufficient in and of itself to ensure compliance both with loac and roe limitations. This is particularly the case for forces from States with a relatively wide right of self-defence.214 When applicable, loac principles that are more restrictive than self-defence rules must be complied with as well. Furthermore, a lack of awareness about the existence of two self-defence concepts may lead to the development of a ‘self-defence plus’ concept applicable in both instances. If soldiers are permitted to use more force than the legal concept of self-defence would permit in some ‘self-defence’ situations, because the use of force is actually regulated by loac, they may be misled to believe that they can use the same amount of force in ‘proper’ self-defence situations as well.215 The result could be the unlawful use of force.
Another reason the difference between the legal and operational concepts of self-defence needs to be clarified, is the question of responsibility. As explained in Section 4.3 in the context of the relationship between the jus in bello and jus ad bellum, one of the underlying principles of loac is the idea that the soldiers are acting on behalf of the State, and do not incur individual liability for those acts, provided they comply with loac (so-called ‘lawful acts of war’), other applicable legislation, and military orders. This is why there is a requirement for armed forces to be “under a command responsible to that Party for the conduct of its subordinates”.216 This concept ensures that military forces are not held responsible for the political decisions of their State, and encourages compliance with orders and hence control of the use of force. The use of force in personal self-defence, however, is primarily the responsibility of the individual.217 As Scaliotti explains, it is “a personal defence, which applies to the individual, who is charged with an act committed in order to defend himself or another person”.218 If operations are carried out on the basis of self-defence, and it is either considered to be personal self-defence or it is not clear whether this is a legal concept of self-defence or a military or operational concept, it is not clear how this underlying principle is adhered to. Self-defence is a plea that can enable the soldier to avoid criminal responsibility.219 However, if the act is a lawful act of war, there should be no individual responsibility in the first place.
If defensive military operations were to be conducted on the basis of personal self-defence instead of roe and the underlying loac authorisations, the result would be to shift the responsibility from the State to the individual soldier. States and Commanders would, in other words, be effectively placing an enormous burden on the shoulders of the individual members of their armed forces if they insist that their troops are operating on the basis of personal self-defence when carrying out defensive military operations. The question of responsibility would be particularly relevant in a situation in which a soldier is using force not permitted by personal self-defence but lawful in accordance with loac, and the forces are ordered to only use force in ‘self-defence’. If the forces are operating on the basis of the legal concept of self-defence, the use of force would be unlawful. If, however, the use of force is permitted by loac, the use of force itself would be lawful, even if the decision to use force would be subject to disciplinary or military penal action for violating the order to only use force similar to that permitted in self-defence.
The effect of combining restrictive roe with a requirement that use of force not in self-defence must be authorised by roe, even during armed conflict, was arguably pushed to the extreme during the involvement of nato States in the Afghanistan and Iraq conflicts from 2001 to 2014. The commentaries on the misapplication of self-defence by military forces during armed conflict all relate to these conflicts. While it may be argued that these trends are context-dependent and hence limited to those operations, the lack of awareness of this fact means that the heavy reliance on both self-defence as a legal and operational concept and the requirements of necessity and proportionality is likely to continue.
It is therefore necessary to recognise that current practice appears to include more than a criminal law defence in the roe concept of inherent self-defence, here referred to as “operational self-defence”. This may make sense especially at the tactical level, in that it is easier to train soldiers in one cautious standard without setting out the various legal complexities, and they should not be expected to make complex legal assessments when in situations akin to self-defence. However, it is important that those in charge of assessing the force used, including Commanders and legal advisers, realise that there is a difference and what this entails.
8.3.3 ‘Unit Self-Defence’
The concept of ‘unit self-defence’ presents further challenges. ‘Unit self-defence’ is a uniquely military concept which allows military units to use force to defend themselves regardless of applicable roe. A unit may be understood as an army platoon, a ship, an aircraft, or in some circumstances even an international task force; according to Boddens Hosang, “[t]he size or composition of the unit is not relevant, the operation as a single unit is”.220 Although there appears to be general agreement that the use of force in ‘unit self-defence’ is lawful, there is no consensus on its legal basis.221 For instance, the concept has not found its way into domestic criminal legislation.222 There are also differing approaches to what ‘unit self-defence’ actually entails: for example, does it require a threat to the unit as such, or is it sufficient that a member of the unit is threatened?
In 1983, Ashley Roach pointed out that “little or no operational guidance is given on how and where to exercise that right”,223 and although the guidance may be better today, it is still far from clear. Because there are differing views on what is included in the concept, there are significant divergences in the legal authorities proposed. Dale Stephens, for example, explains that “the right of unit self defense allows a commander, or an individual soldier, sailor or airman the automatic authority to defend his or her unit, or him or herself, in certain well defined circumstances.”224 Boddens Hosang appears to have a narrower definition of the concept, explaining that it consists of “the right of a commander to take all necessary measures to defend his unit against an (imminent) attack”.225 It is therefore not clear who will be authorised to make the decision to use force in ‘unit self-defence’, and whether ‘unit self-defence’ also include individual self-defence. As will be shown below, these two commentators have also reached very different legal conclusions.
Boddens Hosang suggests that there are four potential legal models that can be used to explain the concept of ‘unit self-defence’: the collective application of personal self-defence; the human right of the military unit to defend the lives of its members; the application of State self-defence by the military unit as State representative, or a right sui generis.226 As explained above in 8.2, the scope for relying on personal self-defence as a legal authority for military operations is limited, especially for armed conflict operations. Furthermore, domestic rules on self-defence differ, making it an unpredictable legal basis in the context of multinational operations.227 It is therefore unlikely to be a sufficient legal basis for the acts considered authorised in ‘unit self-defence’.228 ihrl is also inappropriate as a legal authority because the military unit is a State entity and the State cannot use force beyond the limited exceptions to arbitrary deprivation of life set out in ihrl. Furthermore, as Boddens Hosang points out, a right to use force to further State interests by State agents does not fit neatly with the purpose and nature of human rights.229 The individual members of the unit will of course retain their right to life, but the use of lethal force would only be permissible when absolutely necessary, and the challenges arising from relying on personal self-defence during an ongoing armed conflict would apply.230
The reliance by the State agents on State self-defence as expressed in Article 51 of the UN Charter may therefore seem the best solution; however, this is not unproblematic either. First, the use of force in State self-defence is usually not a tactical level decision because it requires authorisations and access to information that a unit commander may not have. The scale of the response accepted in ‘unit self-defence’ is also likely to be smaller than when defending the State. Furthermore, the jus ad bellum self-defence criteria, in particular the threshold of armed attack as applied in the Nicaragua case, appear inappropriate in the context of military forces facing threats in an ongoing operation.231 In practice, the threshold of State self-defence may be too high to be relevant for military forces involved in tactical-level operations.
Recognising these challenges, Boddens Hosang suggests that ‘unit self-defence’ could be viewed as a downscaled version of State self-defence, adapted to the applicable scale and command level. As a tactical level representation of State self-defence, the use of force in ‘unit self-defence’ will be lawful where the requirements set out in the Caroline incident are met (see Section 6.2.2).232 The authority is delegated from strategic level to tactical level, and due to the downscaled interests involved, the nature and scope of the force that may be employed will have to be adapted to meet the applicable criteria.233 This means, inter alia, that a small scale attack which is considered too limited to amount to an attack on the State may still be an attack on the unit and permit a response proportionate to the attack. A similar approach is taken by Yoram Dinstein, who views ‘unit self-defence’ as an ‘on-the-spot reaction’ taken by military forces in response to small-scale attacks on the basis of States’ right to self-defence. He explains that “[t]here is quantitative but no qualitative differences between a single unit responding to an armed attack and the entire military doing so. Once counter-force of whatever scale is employed by military units whatever size – in response to an armed attack by another State – this is a manifestation of national self-defence, and the legality of the action is determined by Article 51 as well as by customary international law”.234 As discussed in Section 6.2.3, the right of States to use force in self-defence is increasingly considered applicable also to non-State actors.
The challenge, however, in relying on the jus ad bellum right of State self-defence as the basis for ‘unit self-defence’, is that its scope of application during an ongoing armed conflict, in response to threats arising in connection with that armed conflict, is limited. While Dinstein only proposes that State self-defence is relevant outside the context of armed conflicts, Boddens Hosang suggests that it may be applied as a general legal basis for ‘unit self-defence’, albeit in a limited or tactical-level form.235 As explained in 4.3, the jus ad bellum defines the parameter for the armed conflict, but will not authorise violations of the jus in bello. The use of force during an armed conflict that is not authorised by loac may therefore not be based on the jus ad bellum right of State self-defence.236 To allow units to rely on State self-defence during an armed conflict would be to conflate the two areas of law. Because self-defence requires the initial force to be unlawful, the argument that jus ad bellum self-defence could apply during an armed conflict would imply that the use of force by an adversary considered to be acting in contravention of the jus ad bellum will be unlawful regardless of whether it is in accordance with loac. This position was clearly rejected in post-Second World War trials.237 As a result, State self-defence only provides a legal basis for the use of force in ‘unit self-defence’ against attacks not related to an ongoing armed conflict.
The fact nonetheless remains that States allow military units to defend themselves, beyond the limited scope of application of the criminal law concept of personal self-defence, even during armed conflicts. If personal self-defence and State self-defence are insufficient legal bases for the reliance on the concept during armed conflicts, two solutions remain: either it has become a separate customary norm, or, as is more likely, it is a non-legal operational concept referring to an act that has different legal bases depending on the circumstances. In peacetime or operations not involving armed conflict, it is based on the jus ad bellum right of State self-defence, while during armed conflict, the force will be authorised as personal self-defence or on the basis of loac.
The argument that ‘unit self-defence’ is a sui generis right under customary international law has been presented in particularly by Dale Stephens and Charles P Trumbull IV.238 A clear benefit of this approach is that the use of force in ‘unit self-defence’ would be authorised rather than merely justified on the basis of being an exculpatory ground for the purposes of criminal liability.239 In Stephens’ view, it is a non-derogable human right, and may therefore not be subject to limitations.240 Because State self-defence is a matter of national discretion and loac depends on the existence of an armed conflict, ‘unit self-defence’ as a right sui generis must in his view be independent from both.241 When it comes to identifying the applicable criteria for the customary rule on ‘unit self-defence’, Stephens argues that it “draws its jurisprudential authority directly from the ‘Caroline’ prescription”.242
The proposed customary rule, however, appears to be based on insufficient evidence of opinio juris. Trumbull, for instance, primarily bases his argument on references to a right of self-defence which is separate from State self-defence, but fails to explain why this should be interpreted as referring to a separate concept of ‘unit self-defence’ rather than the generally recognised personal right of self-defence.243 It may also be criticised on its legal analysis. Although it is clear that the individual members of military forces retain human rights, this is reflected in their right to defend their right to life as established under their domestic criminal law rules on self-defence.244 As mentioned above, it is not a right that may be conferred on a State entity, such as a military unit. This does not mean that Commanders do not have a duty to protect the lives of subordinates. In order to protect the individual members of military forces from arbitrary deprivation of life, the State is obliged to permit the use of force in self-defence, and Commanders will be obliged to do what is reasonable to effectuate the defence.245 However, this only applies to the same (relatively limited) extent as their personal right of self-defence, and military forces are also required to accept a certain degree of risk.246 The argument that ‘unit self-defence’ is founded on human rights is therefore stretching the personal right of self-defence too far.247 Furthermore, although the Caroline principles may reflect the general consensus on the use of force in self-defence,248 the result of applying jus ad bellum criteria to a human right is somewhat confusing.249
In order to achieve much needed clarity on this complex topic, it would be better to separate the widely recognised legal concepts of personal self-defence and State self-defence from the operational reality that forces may defend themselves. The fact is that none of the legal concepts completely address every aspect of ‘unit self-defence’.250 In such cases, the best way to achieve more clarity is not to stretch the application of the legal concepts, but rather to recognise the existence of a non-legal concept. As an operational concept, the authority for using force will differ for those who participate in an armed conflict and those who do not.251 A similar approach is taken by Henderson and Cavanagh who explain that from a legal perspective, ‘unit self-defence’ is either a delegated authority to use State self-defence “in limited circumstances and in a constrained fashion” or “a reminder of the criminal law authority to act in self-defence to protect oneself and others”.252
As was argued in relation to ‘operational self-defence’, since ‘unit self-defence’ is an exception to the restrictions imposed through roe and other military orders, it is important that it is narrowly defined. The criteria for the operational concept of ‘unit self-defence’ will primarily be imposed by the applicable legal authority. Under the jus ad bellum, the use of force must be necessary, proportionate and immediate, and must be in response to an unlawful imminent or actual attack (the Caroline criteria). States appear to have included the first three requirements for the use of force in ‘unit self-defence’ also during armed conflict. These requirements would operate as self-imposed restraints in addition to the legal requirements arising from the applicable legal authority for the use of force, namely personal self-defence or loac, depending on the circumstances.253 As Henderson and Cavanagh conclude, ‘unit self-defence’ should be thought of as “an order or command to use military force when certain ‘triggers’ are present”.254 Considering that force may be used in ‘unit self-defence’ in situations that are not authorised by roe, such limitations would be a way to limit and in some way control its application.
In addition to the restraints imposed on the force by the requirements of necessity, proportionality and immediacy, the application of ‘unit self-defence’ hinges on the definition of a “unit”. Who may be defended in ‘unit self-defence’? First, if State self-defence is the legal basis, ‘unit self-defence’ must be limited to the nationals of the State in question.255 It may potentially also be expanded to include other nato forces through an application of the concept underlying nato and its Article 5 of the North Atlantic Treaty on collective defence.256 Protection of units from other countries, on the basis of jus ad bellum and collective State self-defence, requires a request from the threatened State, and is likely to be a strategic-level decision.257 Agreement on collective ‘unit self-defence’ on behalf of all military forces where applicable could be reached between troop-contributing nations during the planning of a nato operation. Second, ‘unit self-defence’ is generally limited to forces close to those threatened, either by being in the vicinity or operating together with those forces as a single unit.258 This excludes forces arriving after the attack in order to support the forces, such as quick reaction forces (qrf) or close air support (cas), from relying on ‘unit self-defence’ as the authority for their use of force. Instead, they must rely on other authorities for their use of force, such as loac or personal self-defence or defence of others, where applicable.259
This narrow interpretation of ‘unit self-defence’ is important to avoid giving military forces the impression that they have an authority to use force that is unlikely to be accepted by the Courts which would hear cases of alleged unlawful use of force. The legal basis proposed for the use of force in ‘unit self-defence’ ought to be recognised in the jurisdiction in which the issue would be litigated.260 Kalshoven and Fontein have, for instance, expressed concern that “Units of (…) lower echelons may be strongly inclined to assist each other; and if this result in a violation of their RoE, the excuse that it was a matter of collective unit defence may lie readily at hand”.261 However, “due to the lack of a legal basis, [it] would be unlikely to succeed”.262 Furthermore, the misapplication of ‘unit self-defence’, in particularly against forces that are not opposing forces in an armed conflict, may have disastrous consequences. In the worst case, it may initiate an armed conflict. Unclear rules may also have the opposite effect, that is, Commanders may become unsure about when to act and wait too long before using defensive force, with potentially deadly consequences for those attacked. Clear rules are therefore of crucial importance, and the current lack of consensus is troubling. Considering that the legal basis and its requirements will define the scope for applying the force, this lack of consensus regarding the legal basis for the use of force in ‘unit self-defence’ is likely to cause operational challenges and friction.
8.3.4 ‘nato Self-Defence’? Self-Defence and ‘Extended Self-Defence’ as Defined in the MC 362/1
The nato roe doctrine, the MC 362/1263, includes a definition of self-defence. However, as explained previously, the use of force in self-defence by military forces participating in a nato operation will be defined by the respective domestic legislations.264 As emphasised in the nato training publication on roe, the definition provided in the MC 362/1 must therefore be viewed as a general description, useful as a reference point for the various national versions of self-defence, rather than as a legal basis or description of a common rule.265
The reference in nato doctrine to “extended self-defence” is more complicated because it appears to introduce an additional self-defence form. Under the heading “extended self-defence”, the MC 362/1 explain that “[w]ithin the general concept of self-defence, nato forces and personnel also have the right to take appropriate measures, including the use of necessary and proportional force to defend other nato forces and personnel from attack or imminent attack”.266 However, rather than being a separate self-defence concept, this reference to self-defence in the MC 362/1 should, according to stanag 2597, also be understood as a baseline for planning purposes for nato operations.267 The purpose of the concept is to extend the objects of protection to include other nato forces. It does not increase the amount of force that may be used or extend the circumstances where self-defence may be applied.268
The importance of the reference to ‘extended self-defence’ therefore appears to be a reminder of the purpose of nato as a defence organisation. As explained above in the context of protection of others,269 military forces from certain States have a limited ability to defend third persons with self-defence.270 For instance, it may be limited to persons in a close relationship with those able to assist, which may be interpreted to mean other members of the same unit or nationality. The inclusion in the MC 362/1 of the concept of ‘extended self-defence’ is therefore one way to ensure that nato forces come to the assistance of other members of the nato forces, of all nationalities.
Although stanag 2597 refers to national concepts of personal self-defence and the different approaches taken to the defence of others, ‘extended self-defence’ may also be an expression of State self-defence. Boddens Hosang considers it to be “an operational level reflection of the right to collective self-defense, which is the basis for the Alliance”,271 meaning State self-defence as reflected in Article 5 of the North Atlantic Treaty.272 It is therefore limited to nato forces, and because of the political sensitivity of visiting forces using force in another nato State, it should in his view only be applied during nato operations.273 He acknowledges that Article 5 will only be an appropriate legal basis for tactical level force during Article 5 operations;274 however, this position fails to take into account that such operations in many cases will involve participation in an armed conflict, which excludes the continued application of the jus ad bellum to authorise otherwise unlawful force.275 The same concern will apply to the contention that in non-Article 5 operations, the authority must be found in the mandate or legal basis for the operation as such.276
As a result, the better approach is to consider ‘extended self-defence’ as an application of the relevant self-defence provision. During peacetime and operations not involving participation in an armed conflict, this will either be State self-defence (to the extent the authority is delegated) or personal self-defence. During armed conflict, ‘extended self-defence’ must be based on personal self-defence as set out in the respective domestic legislation.
8.3.5 ‘Force Protection’ Distinguished from Self-Defence
‘Force protection’ is an operational concept closely related to self-defence, and is another area where the definitions, and therefore also the suggested legal bases, differ. In nato, ‘force protection’ is defined as a “set of measures and means to minimise the vulnerability of personnel, facilities, materiel, operations and activities from threats and hazards in order to preserve freedom of action and operational effectiveness thereby contributing to mission success”.277 According to stanag 2597, nato roe are supposed to provide Commanders with a sufficient range of force protection measures,278 implying that unlike self-defence, force protection is not a justification for acting beyond the authorised roe.
‘Force protection’ is, however, also used as a reference to a different concept, namely the ability or permission to protect military forces under attack from unnecessary suffering or death, even by scholars from nato States.279 Boddens Hosang, for instance, argues that “[i]n its simplest form, force protection is the collective exercise of personal self-defense or unit self-defense”, although aspects of force protection go beyond what is permitted in self-defence and therefore require a different legal authority.280 He suggests that this will be the mandate or other legal basis for the operation in question, in other words the jus ad bellum. For instance, if the military force is authorised in UN Security Council resolution to use “all necessary means”, this should be interpreted to mean that the forces are authorised to defend themselves.281 If the jus ad bellum authority to act is unclear, for instance during national evacuation operations, the authority to use force beyond self-defence will be equally unclear.282 Although Boddens Hosang appears to suggest that the UN mandate for the operation will provide the legal basis for the use of force even during armed conflicts, he also points out that loac will permit the measures necessary for force protection.283 As repeatedly emphasised in this book, the reliance on the jus ad bellum to authorise use of force during an armed conflict, presumably not permitted by loac, would amount to a conflation of the jus ad bellum and jus in bello which must be avoided.
The main challenge with the above proposition is not, however, conflation of two separate areas of law, but rather that ‘force protection’ in the nato context should not be perceived as an aspect of self-defence. According to the nato definition set forth above, ‘force protection’ include measures and means used to minimise vulnerability, not the use of force in response to an attack. Examples may include building fences around a military installation, the use of guards to protect personnel and equipment, the use of armoured vehicles, the use of warning zones, and the requirement to use vehicles with weapons mounted on the top (so-called ‘top cover’). ‘Force protection’ must therefore be distinguished from the authority to use of force in self-defence. To the extent that it involves the use of force or other provocative measures, the authority to employ measures or means for the purpose of force protection must be authorised by roe. There are no specific force protection roe in nato doctrine, but there are several roe that will enable the forces to ensure force protection, such as roe permitting the use of force to prevent unauthorised access to a military installation or to prevent someone from taking possession of military equipment.
Means and measures employed in force protection that do not involve the use of force will in many cases not require a legal basis because they do not infringe other persons’ rights. For operations not involving participation in an armed conflict, the scope for employing force protection will either be defined by the State itself for its own military forces or in the case of military forces operating in support of another State, in the agreement between the host nation and the military forces present on how to effectuate the mandate for the operation.284 During an armed conflict, force protection that results in the destruction of enemy property will be lawful provided it is “imperatively demanded by the necessities of war”.285 The use of force related to force protection, as in the case of someone who tries to destroy the gate to a camp or who fails to respect a warning zone, will have to comply with loac or personal self-defence as applicable.
8.4 Conclusions on the Relevance of Self-Defence as a Legal Basis for Use of Force during Armed Conflict Operations
“War is by definition a risky and hazardous business”.286 This was stressed in the so-called Krupp case before the Nuremberg Military Tribunals and still holds true today. Beyond armed conflicts, the acceptable risk to military forces is low, and an attack on military forces is likely to be considered a threat to the State, permitting the use of force in State self-defence to the extent that is necessary and proportionate under the circumstances. During armed conflicts, however, military forces must accept a certain degree of risk as part of their work, and they cannot set aside the governing legal regime as a result of any risk to their lives. As the judgement in the Krupp case made clear, the rules and customs of land warfare “are designed specifically for all phases of war. They comprise the law for such emergency. To claim that they can be wantonly – and at the sole discretion of anyone belligerent – disregarded when he considers his own situation to be critical, means nothing more or less than to abrogate the laws and customs of war entirely”.287 loac is designed specifically for the situations that military forces may face when participating in an armed conflict, and the scope for setting this legal regime aside in order to defend oneself, is accordingly limited. The need for setting certain rules aside due to the necessities caused by an armed conflict is the justification for the existence of loac. As a result, the justification of necessity only applies to the extent permitted by this lex specialis legal regime, namely in the form of military necessity, and the scope for relying on necessity to set further rules aside in self-defence must be narrowly construed.
Because the primary legal basis for the use of force during an armed conflict will be loac, self-defence should only be relied upon in the limited circumstances where loac is insufficient and the more restrictive requirements of personal self-defence are met. This was emphasised by the icty Trial Chamber in the Gotovina case. When examining the legality of the use of force against another person in an armed conflict, the Court should first establish the status of the victims. If they were not taking a direct part in hostilities, the next question would be whether the victims had carried out an immediate illegitimate attack on the perpetrator, and whether the response was proportionate.288
As explained in detail in Section 8.2, the generally recognised criteria for personal self-defence entail that self-defence by its nature is unsuitable as a legal basis for the use of force during military operations. The same conclusion is reached by Hans Boddens Hosang who points out that self-defence “may be a controversial basis on which to plan and carry out premeditated military operations to achieve political or strategic objectives”.289 During armed conflict, military forces may mainly rely on personal self-defence in the context of civilians threatening them in a way which is not perceived as amounting to direct participation in hostilities.
The extensive focus and reliance on self-defence appears therefore unfounded, unless the self-defence concept referred to is something other than the legal concepts. The use of operational concepts to permit the use of force beyond roe in exceptional cases makes operational sense; the main rule is that the use of force is subject to command and control, but military forces must nonetheless be able to defend themselves. This does not, however, have any impact on when or how legal regimes become applicable. Regardless of what the operational concepts are referred to as, the use of force during armed conflict will primarily be regulated by loac, and only exceptionally by self-defence, primarily personal self-defence. The use of force not permitted by applicable roe may still be authorised by loac, especially when the roe are restrictive and permitting only a fraction of the acts loac allows. The result is that use of force beyond roe, on the basis of the roe exception referred to as ‘self-defence’, will in most cases be legally premised on loac.
Although States are free to impose policy restrictions akin to respective national rules on personal self-defence and to refer to such defensive force based on loac as ‘self-defence’,290 they must also ensure that the use of force complies with loac. Furthermore, States must be aware of and take measures to mitigate the risks resulting from the terminological conflation. There is a danger that both the criminal law concept of self-defence and the operational concept of self-defence are viewed as the same concept, with the result that too much force is used in ‘true’ self-defence situations. The operational concept of self-defence, which may be based on loac, is “quasi-offensive in nature”,291 and should not be applied in response to unlawful threats posed by civilians, neither in peacetime nor during armed conflicts (provided the civilians are not taking direct part in hostilities).292 According to Bagwell and Kovite, the tendency to require military forces to fight offensive operations on the basis of self-defense means that they “are left with little recourse but to stretch some aspects of self-defense, while modifying or ignoring others”.293 The requirement that the attack must be imminent is particularly problematic to apply to armed conflict operations, with the result that it has been stretched to no longer mean immediate.294
Rather than restraining the use of force, the concept of self-defense has expanded beyond legally permissible limits, and the traditional dividing line between defensive and offensive uses of force has eroded. While the vast majority of these combat engagements are otherwise justifiable under loac, the co-opting of self-defense authorities to justify offensive targeting risks misapplication of both regimes in combat and overbroad application of self-defense rules in future, less hostile environments.295
From a legal perspective, the better approach would be to make a clearer distinction between the use of force on behalf of the State, and the use of force in personal self-defence. In the nato context, a solution would be to enhance the roe and guidance on the use of force in response to attacks or threats of attacks carried out by persons who may be considered lawful targets, either because they are combatant or they directly participate in hostilities.296 Only the use of force in true self-defence should be excluded from the regulation of command and control tools such as roe. This would reduce the need for an operational self-defence concept permitting the use of force beyond roe, and it would make the distinction between ‘lawful acts of war’ and the exceptional role of personal self-defence clearer. The introduction of additional roe in nato roe doctrine authorising the use of force in response to attacks or imminent attacks by opposing forces, as suggested in Chapter 9, would be one way to enhance the distinction between warfighting and self-defence.
Dale Stephens, ‘Rules of Engagement and the Concept of Unit Self-Defense’, 45 Naval Law Review 126 (1998), p. 22.
For instance, it has been argued that the application of personal self-defence by soldiers during armed conflict if merely theoretical because the roe should provide for the necessary authorisations in most situations requiring the use of force. J.F.R. Boddens Hosang, ‘Self-Defence in Military Operations: The Interaction between the Legal Bases for Military Self-Defence and Rules of Engagement’, 47 Military Law and Law of War Review 25 (2008), p. 84. However, the same author emphasises elsewhere that “this does not mean that the right itself is never applicable and that there is no right to defend one’s life”. Hans F. R. Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence: Their Relationship to Rules of Engagement’, Chapter 24 in Terry D. Gill and Dieter Fleck (eds.), The Handbook of the International Law of Military Operations (Oxford University Press, Oxford, 2015) p. 499, §24.25 with commentaries. See also Jens Ohlin, ‘Self-defence’, in Antonio Cassese (ed.), The Oxford Companion to International Criminal Justice (Oxford University Press, Oxford, 2009) p. 507.
The following example, provided by Bagwell and Kovite, is useful in illustrating the need for self-defence, even during armed conflict: “A U.S. armor unit is involved in combat operations overseas. While returning from a mounted patrol, a tracked vehicle swerves to avoid debris on the road and accidentally crushes a young boy who was waiting on the shoulder for the column to pass. When the convoy stops to render aid, a small group of local men gathers to see what happened. One man pushes through the crowd to see the boy, who he recognizes as his son. The soldiers recognize the man as a local farmer who has always been friendly toward U.S. and coalition forces. Inconsolable, the man runs back to his house, and moments later reappears running toward the soldiers with what appears to be an Ak-47.” Randall Bagwell and Molly Kovite, ‘It is Not Self-Defence: Direct Participation in Hostilities Authority at the Tactical Level’, 224(1) Military Law Review 1 (2016), p. 12.
In exceptional circumstances, military forces may also be authorised to use force for law enforcement purposes.
See further Section 6.3.2.1.
Jan Arnold Hessbruegge, Human Rights and Personal Self-Defense in International Law (Oxford University Press, New York, 2017) p. 198.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 495, §24.21 with commentaries.
On the limits of adapting the reasonable man standard in accordance with the person’s characteristics, see e.g. Victoria Nourse, ‘Self-Defence’, in Markus D. Dubber and Tatjana Hörnle (eds.), The Oxford Handbook of Criminal law (Oxford University Press, Oxford, 2014) pp. 617–619.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 495, §24.21.
The same consideration applies to law enforcement officers, although perhaps to a lesser extent, seen as they are less likely to have to use lethal means in their work than military forces. On the reasonable law enforcement officer, see Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 143–146.
See also Erica L. Gaston, ‘Reconceptualizing Individual or Unit Self-Defense as a Combatant Privilege’, 8(2) Harvard National Security Journal 283 (2017), p. 309, where she explains that in her experience, military forces faced with case studies on self-defence would often add information regarding contextual factors that would influence their response on the case studies.
Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 66–67. See also George P. Fletcher, Basic Concepts of Criminal Law (Oxford University Press, Oxford, 1998) p. 137.
Massimo Scaliotti, ‘Defences before the international criminal court: Substantive grounds for excluding criminal responsibility – Part 1’, 1 International Criminal Law Review 111 (2001), p. 171; Kai Ambos, ‘Other grounds for excluding criminal responsibility’, in Antonio Cassese, Paola Gaeta, John R.W.D. Jones (eds.), The Rome Statute of the International Criminal Court – A commentary, Volume I (Oxford University Press, Oxford, 2002) p. 112.
Ohlin, ‘Self-Defence’ (n 2) p. 507.
Fiona Leverick, Killing in Self-Defence (Oxford Monographs on Criminal Law and Justice, Oxford University Press, Oxford, 2006) p. 160–161. See also Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 137 and 210, and Ohlin, ‘Self-Defence’ (n 2) p. 507.
The UK Criminal Justice and Immigration Act of 2008, for instance, only requires a genuine belief, even if this is not reasonable (Section 76(4), UK Criminal Justice and Immigration Act of 2008, available at
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 211, citing Öneryildiz v. Turkey [GC] (Judgement), App. No. 48939/99, icty, 30 November 2004, para. 93.
See also Ian Henderson and Bryan Cavanagh, ‘Military members claiming self-defence during armed conflict – Often misguided and unhelpful’ [hereinafter: ‘Claiming self-defence on the battlefield’], in Jadranka Petrovic (ed.), Accountability for Violations of International Humanitarian Law – Essays in honour of Tom McCormack (Routledge, London/New York, 2016) p. 77.
Hannah Tonkin, ‘Defensive Force Under the Rome Statute’ 6(1) Melbourne Journal of International Law 86 (2005), p. 93. See also Gary P. Corn, ‘Should the Best Offense Ever be a Good Defense? The Public Authority to Use Force in Military Operations: Recalibrating the Use of Force rules in the Standing Rules of Engagement’, 49 Vanderbilt Journal of Transnational Law 1 (2016), p. 28.
See e.g. the Norwegian Armed Forces, Manual i krigens folkerett (Trans: Manual of the Law of Armed Conflict), issued under the authority of the Chief of the Defence, [2013] [hereinafter referred to as Norwegian loac Manual], pp. 24–26.
European Convention for the Protection of Human Rights and Fundamental Freedoms (echr)[1950], as amended by Protocols Nos. 11 and 14, ETS 5, Articles 2(2) and 15(2).
See Section 6.3.2.1. See also Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 63–64; Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 497, §24.23 with commentaries.
The requirement that the initial attack is unlawful should not be interpreted to mean that the act must be criminal. It will remain a self-defence situation even if the perpetrator may not be held criminally responsible, for instance due to insanity or being below the age of criminal responsibility. See Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 63 and pp. 129–131, and Robert Cryer, ‘Defences/Grounds for Excluding Criminal Responsibility’, in Robert Cryer et al. (eds.), An Introduction to International Criminal Law and Procedure (Cambridge University Press, 2010) p. 405.
Terry Gill, Carl Marchand, Hans Boddens Hosang, and Paul Ducheine: ‘General Report’, in Stanislav Horvat and Marco Benatar (eds.), Legal Interoperability and Ensuring Observance of the Law Applicable in Multinational Deployments (Proceedings of the 19th International Congress, Quebec, xix Recueil of the International Society of Military Law and Law of War, Brussels, 2013) p. 123, fn. 2.
This was first expressed in the context of jus ad bellum in the ‘Ministries case’ and has become a much cited maxim. United States v. Von Weizsaecker et al, Trials of War Criminals before the International Military Tribunal Under Control Council Law No. 10, Nuremberg, October 1946-April 1947 (Nuremberg, 1947), vol. xiv (‘Ministries Case’),
See also Hessbruegge, Human Rights and Personal Self-defense, ibid, p. 101, especially fn. 41. However, rather than concluding that the use of defensive force is allowed by loac, he argues that the right to use force is derived from the soldiers’ right to life and physical integrity.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 497, §24.23 with commentaries. See also Timothy L.H. McCormack, ‘Self-Defence in International Criminal Law’, in Hirad Abtahi and Gideon Boas (eds.), The Dynamics of International Criminal Justice: Essays in the honour of Sir Richard May (brill, 2005) pp. 237–238, emphasising that “individual combatants do not need to be excused for killing enemy combatants by lawful means”.
Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of victims of International Armed Conflicts (Protocol I) [1977], printed in Dietrich Schindler and Jiri Toman, The Laws of Armed Conflicts (Martinus Nijhoff Publisher, 1988) pp. 711ff, Article 43(2). See further Section 5.1.2.
Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of Non-International Armed Conflicts (Protocol II) [1977], printed in Schindler and Toman, The Laws of Armed Conflicts, ibid, pp. 775ff., encourages States, at the end of hostilities, to grant amnesty to persons who have participated in the armed conflict (Article 6.6), thereby acknowledging the State practice to sanction those persons. See also Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 81.
See also, Michael N. Schmitt, ‘Targeting and International Humanitarian Law in Afghanistan’, 85 International Law Studies 307 (2009), p. 327.
Kalshoven and Fontein propose an alternative scope of application for self-defence during armed conflict: “when a soldier carries out an act which, whether or not part of the tack he was ordered to perform, appears to have resulted in a violation of the applicable law of armed conflict”. Although self-defence may be applicable in a case of loac violation, they choose an unfortunate example. They refer to the case of a soldier who “kills an unidentified person who turns out to have been an unarmed civilian”. Frits Kalshoven and Thyla Fontein, ‘Some Reflections on Self-Defence as an Element in Rules of Engagement’ in Mariëlle Matthee etc. (eds.), Armed Conflict and International Law: In Search of the Human Face (tmc Asser Press, The Hague, 2013) p. 106. This argument appears erroneous: if the unidentified person was believed to have been a lawful target, and the necessary precautions had been taken to support that determination, the fact that it later was revealed to be wrong will not make it a violation of loac (see Section 5.4).
Except for the exceptional circumstances where soldiers are given law enforcement authority and may use force under this authority. See also icrc, Interpretive Guidance on the Notion of Direct Participation in Hostilities under International Humanitarian Law [hereinafter icrc Interpretive Guidance] (May 2009, prepared by Nils Melzer, available at
See also Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 221–223.
ibid, pp. 224–225.
The requirements for becoming a lawful target and the issue of doubt are dealt with in more detail in Sections 5.2 and 5.4.
See also Gaston, ‘Reconceptualizing Individual or Unit Self-Defense’ (n 11) p. 293.
The protection may be lost if medical or religious personnel carry out acts harmful to the enemy, cf. Geneva Convention (I) for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field (GC I) [1949] printed in Schindler and Toman, The Laws of Armed Conflicts (n 28) pp. 459ff., Article 21, and AP II (n 29) Article 11(2). See also Chapter 5, footnote 79.
U.S., ‘Instructions for the Government of Armies of the United States in the Field’ (Lieber Code) [1863], printed in ibid pp. 3ff, Article 70, and Hague Convention (IV) Respecting the Laws and Customs of War on Land of 1907 and its Annex: Regulations Respecting the Laws and Customs of War on Land (‘1907 Hague Regulations’), printed in ibid, 1988, pp.66ff, Article 23(a). See also Rome Statute of the International Criminal Court [hereinafter: Rome Statute], opened for signature 17 July 1998, 2187 unts 3 (entered into force 1 July 2002), Article 8(2)(b)(xvii).
According to AP I (n 28) Article 37, this would be perfidy if the intent is to kill, injure or capture an adversary.
AP I, ibid, Article 28. See also McCormack, ‘Self-Defence in International Criminal Law’ (n 27) p. 238.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 491, §24.16 with commentaries.
Leverick, Killing in Self-Defence (n 15) pp. 87–89. See also Nils Melzer, Human Rights Implications of the Usage of Drones and Unmanned Robots in Warfare (2013, European Parliament, EXPO/B/DROI/2012/12) p. 31. For an example of a different approach, see e.g. Nourse, ‘Self-Defence’ (n 8) pp. 611 – 614, arguing that necessity should not be viewed as a separate element of self-defence, but rather as an aspect of imminence and proportionality.
See e.g. Amnesty International, Guidelines for implementation of the UN basic principles on the use of force and firearms by law enforcement officials (August 2015, available at
See e.g. American Law Institute, Model Penal Code: Official Draft and Explanatory Notes (Complete Text of Model Penal Code as Adopted at the 1962 Annual Meeting of the American Law Institute at Washington, D.C., May 24, 1962, The Institute, Philadelphia, 1985) Section 3.04(2)(b)(i), and United States Manual for Courts-Martial (mcm), 2016 Edition, available at
See e.g. Elies Van Sliedregt, Individual Criminal Responsibility in International Law (Oxford University Press, Oxford, 2012) pp. 238–239, and Gill et al., ‘General Report’ (n 24) p. 164.
Nourse, ‘Self-Defence’ (n 8) pp. 611–612 and Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 492, §24.17 with commentaries.
Albin Eser, ‘Article 31. Grounds for excluding criminal responsibility’, in Otto Triffterer, Comments on the Rome Statute of the International Criminal Court – Observer’s Notes, Article by Article (2nd ed, Beck/Hart, München, 2008) p. 883, fn. 128. The requirement of defensive intent is dealt with further above, in Sections 6.3.2.1.1 and 8.2.2.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 492, §24.17 with commentaries.
Sliedregt, Individual Criminal Responsibility in International Law (n 45) p. 240, citing Gur-Arye, Miriam, Actio libera in causa in criminal law (Hebrew University, Jerusalem, 1984) pp. 82–91.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 492, §24.17 with commentaries.
ibid, pp. 492–493, §24.18 with commentaries.
ibid.
(1938) 53 Wyo. 304, cited in A. J. Ashworth, ‘Self-defence and the right to life’, 34(2) Cambridge Law Journal 282 (November 1975), p. 295.
Eric C. Husby, ‘A Balancing act: In Pursuit of Proportionality in Self-defense for On-Scene Commanders’, Army Law 11 (May 2012), p. 11.
Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) p. 11. See also Jonathan Owen, ‘British Soldiers Resort to ‘Baiting’ Taliban to Beat Rules of Engagement’, The Independent (27 August 2012,
See e.g. Leverick, Killing in Self-Defence (n 15) pp. 69–76. She sets out four approaches to retreat: an absolute retreat rule, a strong retreat rule, a weak retreat rule and no retreat rule. See also Nourse, ‘Self-Defence’ (n 8) pp. 614–615.
Beard v. the U.S., 158 U.S. 550 (1895), p. 561.
U.S. Model Penal Code (n 44) Section 3.04(2)(b)(ii), with the exceptions of retreat not being required from the victim’s dwelling or work, or by police officers (or equivalent) performing their duty (Subsections 1 and 2).
Nourse, ‘Self-Defence’ (n 8) p. 614 and pp. 619–621.
Leverick, Killing in Self-Defence (n 15) pp. 76 and 79, citing the German Criminal Code Section 32 as an example.
See e.g. Steven P. Aggergaard, ‘Criminal Law–Retreat from Reason: How Minnesota’s New No-Retreat Rule Confuses the Law and Cries for Alteration–State v. Glowacki’, 29(2) William Mitchell Law Review 657 (2002), Article 5, especially pp. 664–666; Criminal Justice and Immigration Act of 2008 (n 16) Section 76 as amended by the Crime and Courts Act 2013 Section 43 (UK); David Ormerod et al., Smith and Hogan’s Criminal Law (University Press, Oxford, 2015) pp. 435–437; Leverick, Killing in Self-Defence (n 15) pp. 83–85; and Ashworth, ‘Self-defence and the right to life’ (n 53) p. 294.
echr (n 21) Article 2(2).
Ashworth, ‘Self-defence and the right to life’ (n 53) p. 290. See also Leverick, Killing in Self-Defence (n 15) p. 76.
See e.g. Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 92; Tonkin, ‘Defensive Force Under the Rome Statute’ (n 19) p. 103; and Ashworth, ‘Self-defence and the right to life’ (n 53) pp. 289–290 and 293–294. For a detailed discussion of the different approaches taken by U.S. States to retreat in the context of self-defence, see Aggergaard, ‘Criminal Law–Retreat from Reason’ (n 61) pp. 659–673. By contrast, Geert-Jan Alexander Knoops, Defenses in Contemporary International Criminal Law (Martinus Nijhoff Publishers, Leiden, 2008) pp. 69–71, claims that American law endorses the no retreat rule.
Note, however, that the Criminal Justice and Immigration Act of 2008 (n 16) Section 76(6A) deals with retreat as an aspect of proportionality, stating that rather than retreat being a duty, it is a factor to be taken into account when deciding whether the degree of force was reasonable in the circumstances.
Willi Tessmann and others, British Military Court in Hamburg, 1–24 Sept. 1947, cited Law Reports of Trials of War Criminals, Vol. XV (published for the United Nations War Crimes Commission by His Majesty’s Stationary Office, Digest of Laws and Cases, London, 1949) p. 177.
U.S. Chairman of the Joint Chiefs of Staff (cjcs), Standing Rules of Engagement (sroe)/Standing rules for the use of force (sruf) for U.S. Forces [hereinafter: sroe/sruf] (cjcs Instruction 3121.01B, 13 June 2005) p. A-3. In the U.S. Code of Military Justice, there is no requirement of retreat from a place that the accused has a right to be, but it may according to the Military Judges Benchbook be a factor in deciding whether or not the accused acted in self-defence. See U.S. Department of Army Pamphlet 27–9: Military Judges’ Benchbook, Headquarters, Department of the Army, Washington, DC, 10 September 2014, p. 977. Note, however, that the sroe also stress that self-defence “includes the authority to pursue and engage forces that have committed a hostile act or demonstrated hostile intent, if those forces continue to commit hostile acts or demonstrate hostile intent” (U.S. cjcs, sroe/sruf, ibid, p. A-4). To some extent, these are conflicting guidelines, especially in the context of threats that have not yet amounted to an attack (U.S. hostile intent).
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 493, §24.19, fn. 46. See also Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) p. 41, and Gaston, ‘Reconceptualizing Individual or Unit Self-Defense’ (n 11) pp. 314–315.
See e.g. AP I (n 28) Articles 48 and 51(1) (civilians), and Articles 41 and 42 (hors de combat). By contrast, the duties of law enforcement officers is to protect civilians from other civilians, and they will therefore only have a duty to retreat in exceptional circumstances. Hessbruegge, Human Rights and Personal Self-defense (n 6) pp.148–150.
See also Fletcher, Basic Concepts of Criminal Law (n 12) p. 134.
See Section 6.2.3.
Not all adhere to this distinction. See e.g. Nourse, ‘Self-Defence’ (n 8) pp. 613–614. Note that in the U.S. Model Penal Code, the imminence requirement is substituted for a test of immediately necessary. See Fletcher, Basic Concepts of Criminal Law (n 12) p. 134. The U.S. sroe (cjcs, SROE/SRUF (n 67) p. A-3), on the other hand, state that imminent “does not necessarily mean immediate or instantaneous”, but it fails to provide further clarification or explanation for its departure from the U.S. rcm requirement of the threat being “about to” occur, which in the Military Judges’ Benchbook is interpreted as meaning immediate (see e.g. U.S. Manual for Courts-Martial (n 44) rcm 916(e)(1)(A), p. II-111, and U.S. Military Judges’ Benchbook (n 67) p. 965). For further discussion on the sroe requirement of imminent, see references in Chapter 6, footnote 369.
See e.g. Leverick, Killing in Self-Defence (n 15) pp. 89–108. For a defence of the imminence criteria, see Kimberly Kessler Ferzan, ‘Defending imminence: from battered women to Iraq’, 46 Arizona Law Review 213 (2004).
Bagwell and Kovite, ‘It is not self-defence’ (n 3) p. 13. Note that the statement in the U.S. sroe that immediacy does not need to mean imminent has been met with much criticism. See e.g. John J. Merriam, ‘Natural Law and Self-Defense’, 206 Military Law Review 43 (2010), pp. 76–87, and Eric D. Montalvo, ‘When Did Imminent Stop Meaning Immediate? Jus in bello Hostile Intent, Imminence and Self-Defense in Counterinsurgency’, Army Lawyer, August 2013, p. 24, pp. 24–34.
Hessbruegge, however, deals with it under the heading of proportionality, though emphasizes that it is a separate proportionality analysis. Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 180.
See also Section 3.5.3.3 on the use of escalation of force procedures to identify hostile act and hostile intent.
The use of warning shots by law enforcement agents is limited in many States due to the risk to innocent bystanders and because it may be misinterpreted as an attack. See e.g. Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 159.
An example of an EoF is “shout, show, shove, shoot”. Bagwell and Kovite, ‘It is not self-defence’ (n 3) p. 17.
See e.g. Douglas Guilfoyle, International Criminal Law (Oxford University Press, Oxford, 2016) pp. 371–372, and Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 154–155 (in the context of law enforcement where such graduated response is formally required, but still only if it is expected to counter the attack).
See also Section 8.2.2 concerning how military training may affect what is considered reasonable self-defence.
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 167.
Aaron Pennekamp, ‘Standards of Engagement: Rethinking Rules of Engagement to More Effectively Fight Counterinsurgency Campaigns’, 101 The Georgetown Law Journal 1619 (2013), p. 1635. The argument is made in the context of military operations involving participation in armed conflict where EoF procedures have been included to identify threats and persons directly participating in hostilities, however, the concern is equally applicable to self-defence situations.
See e.g. Bagwell and Kovite, ‘It is not self-defence’ (n 3) p. 16.
The question of how self-defence is carried out when there is a need to use force is also argued to be an aspect of necessity. See for instance Fletcher, who argues that “necessity speaks to the question whether some less costly means of defense, such as merely showing the gun or firing a warning shot into the air, might be sufficient to ward off the attack. The requirement of proportionality addresses the ratio of interest threatened both on the side of the aggressor and of the defender.” Fletcher, Basic Concepts of Criminal Law (n 12) p. 135. Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 65–66, views the requirement of least harmful means as an aspect of necessity.
See e.g. Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) pp. 490–491, §§24.15–16 with commentaries.
See e.g. McCormack, ‘Self-Defence in International Criminal Law’ (n 27) p. 238.
Prosecutor v. Gotovina and others (Judgement Vol II), Case no. IT-06-90-T, icty, Trial Chamber, 15 April 2011, §1730.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 490, §24.15 with commentaries.
See e.g. Gülec v. Turkey (Judgement), App. No. 21593/93 (ECtHR, 27 July 1998), para. 71.
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 169.
For a summary of the debate on ‘shoot to wound’ versus ‘shoot to kill’, see e.g., Jonathan P. Edwards, ‘The Law and Rules of Engagement against Suicide Attack’, in Center of Excellence Defence Against Terrorism (ed), Suicide as a Weapon (Amsterdam, ios Press, 2007) pp. 144–145.
Gaston provides the following example of disproportionate use of force in self-defence, taken from an interview with a U.S. military adviser: “We’re on patrol and get a pop shot at us. No one’s hit, but 30 people are suddenly on line and they fire like 1,000 rounds each”. She also suggests that the so-called ‘Haditha case’ may be another example of disproportionate force. Gaston, ‘Reconceptualizing Individual or Unit Self-Defense’ (n 11) p. 323.
The use of automatic constant fire in the context of law enforcement was hotly debated at the Sanremo Round Table ‘Weapons and international rule of law’ after the presentation by Laurent Gisel (icrc) on the panel “Case study: Law enforcement by military personnel”, on Friday 9th September 2016. See also Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 158.
See also Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 89.
See e.g. U.S. cjcs, sroe/sruf (n 67) p. A-5.
Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 170–171, bullets and reference omitted.
This issue is briefly covered in 8.2.2 on the requirement of an unlawful attack.
Scaliotti, ‘Defences before the international criminal court’ (n 13) p. 161, and Leverick, Killing in Self-Defence (n 15) p. 169.
On the requirement of defensive purpose, see Sections 6.3.2.1.1 and 8.2.2.
Leverick, Killing in Self-Defence (n 15) p. 169.
Section 33 of the German Criminal Code (translated by Prof. Dr. Michael Bohlander, 2016 juris GmbH, Saarbrücken, available at
See e.g. Aydan v. Turkey (Judgement), App. No. 16281/10 (ECtHR, 12 March 2013), para. 101. See also Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 211–212.
Hessbruegge, Human Rights and Personal Self-defense, ibid, p. 212, making reference to Aydan v. Turkey, ibid, para. 99.
See e.g. Rome Statute (n 38) Article 8(2)(b) xvii-xx.
Curiously, Albin Eser only consider the application of jus ad bellum self-defence in his chapter on defences in war crime trials. Although state actors may to some extent rely on Article 51 of the UN Charter, the more relevant form of self-defence in an ongoing operation involving participation in an armed conflict is personal self-defence. Albin Eser, ‘“Defences” in War Crime Trials’, in Yoram Dinstein and Mala Tabory (eds.), War Crimes in International Law (Martinus Nijhoff Publishers, The Hague, 1996) p. 263. On the importance of upholding the separation between jus ad bellum and jus in bello, see Section 4.3.
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 189.
See also ibid, p. 314.
Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-Personnel Mines and on their Destruction [18 September 1997] in Schindler and Toman, The Laws of Armed Conflicts (n 28) pp. 285ff.
Convention on the prohibition of the development, production, stockpiling and use of chemical weapons and on their destruction [13 January 1993] in ibid, pp. 239ff.
Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on their Destruction [10 April 1972] in ibid, pp. 135ff.
Lieber Code (n 38) Article 70, and 1907 Hague Regulations (n 38) Article 23(a). See also Rome Statute (n 38) Article 8(2)(b)(xvii).
Convention on Cluster Munitions [30 May 2008] (available at
In any case, the State may be held responsible for deaths resulting from the use of an inappropriate weapon. See Gülec v. Turkey (n 89) para. 71.
ccw Protocol iii on Prohibitions or Restrictions on the Use of Incendiary Weapons, Geneva [10 October 1980] U.N. Doc. A/CONF. 95/15, 27.10.1980, Article 1(1)(i).
St. Petersburg Declaration (n 20) Although generally considered to reflect customary law, the U.S. have objected this and do not consider herself bound by the prohibition. See U.S. Department of Defense (DoD), Law of War Manual (December 2016 update, Office of the General Counsel of the Department of Defense, Washington, 2016, available at
Article I(5) of the Convention on the prohibition of the development, production, stockpiling and use of chemical weapons and on their destruction [13 January 1993] in Schindler and Toman, The Laws of Armed Conflicts (n 28) pp. 239ff, states: “Each State Party undertakes not to use riot control agents as a method of warfare”. The list of purposes not prohibited under this Convention includes law enforcement including domestic riot control purposes (Article II(9)(d)).
Jean-Marie Henckaerts and Louise Doswald-Beck, Customary International Humanitarian Law [hereinafter: icrc cil Study] (Cambridge University Press, Cambridge, 2005 available at
See also Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 91.
AP I (n 28) Articles 38 and 39.
AP I, ibid, Article 37.
Rome Statute (n 38) Article 8(2)(b)(vii) (iacs) (misuse of emblem etc.) and Article 8(2)(b)(xi) (iacs) and Article 8(2)(e)(ix) (niacs) (treacherous killing).
AP I (n 28) Article 54(1) and AP II (n 29) article 14.
1907 Hague Regulations (n 38) Article 28 and Article 47; Geneva Convention (IV) Relative to the Protection of Civilian Persons in Time of War (GC IV) [1949], printed in Schindler and Toman, The Laws of Armed Conflicts (n 28) pp. 575ff., Article 33, second para; and AP II (n 29) Article 4(2)(g).
1907 Hague Regulations (n 38) Article 23(d); AP I (n 28) Article 40; and II (n 29) Article 4(1).
See e.g. AP I (n 28) Article 57(2)(a)(ii) (“take all feasible precautions in the choice of means and methods of attack with a view to avoiding, and in any event to minimizing, incidental loss of civilian life, injury to civilians and damage to civilian objects”), Article 57(2)(a)(iii) (“refrain from deciding to launch any attack which may be expected to cause incidental loss of civilian life, 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”), and Article 57(2)(b) (“an attack shall be cancelled or suspended if it becomes apparent that the objective is not a military one or is subject to special protection or that the attack may be expected to cause incidental loss of civilian life, 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”). See further Sections 5.4.1 and 5.5.
Note that not all jurisdictions accept necessity as defence against the killing of another. Shlomit Wallerstein, ‘Justifying the right of self-defense: A theory of force consequences’, 97 Virginia Law Review 999 (2005), p. 1010.
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 180.
Although it is relevant, this issue is not sufficiently central to the topic of the present research to warrant the space required to go beyond mere speculation.
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 180.
Melzer, Human Rights Implications of the Usage of Drones and Unmanned Robots in Warfare (n 42) p. 32.
AP I (n 28) Article 57(2)(a)(ii). See also Sections 5.4.1 and 5.5.
See further Section 4.2 on the human right to life and the prohibition on arbitrary deprivation.
Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 182–183. See also Gloria Gaggioli, Expert Meeting on the Use of Force in Armed Conflict: Interplay Between the Conduct of Hostilities and Law Enforcement Paradigms (icrc, Geneva, 2013,
See also Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 90.
By contrast, others argue that the killing of innocent bystanders is never permitted in self-defence. See e.g. Wallerstein, ‘Justifying the right of self-defense’ (n 126) p. 1002.
Bakan v Turkey (Judgement), App. No. 50939/99 (ECtHR, 12 June 2007), para. 55.
In the Tagayeva case, the Court stated that “after the first explosions in the gymnasium and the terrorists had opened fire upon the escaping hostages, the risk of massive human loss became a reality, and the authorities had no choice but to intervene by force. Accordingly, the Court accepts that the decision to resort to the use of force by the State agents was justified in the circumstances, under Article 2 § 2 (a) of the Convention.” Tagayeva and Others v. Russia (n 68) para. 591.
According to Hessbruegge, three factors must be considered in making this assessment:
- –The intensity and extent of harm innocent bystanders may suffer as the result of the envisaged defensive action versus the expected harm for the defender or innocent bystander resulting from the non-action or other defensive action;
- –The number of innocent bystanders endangered by the envisaged defensive action versus the number endangered by non-action or other defensive action;
- –The probability of the harm risked by the defensive action becoming reality versus the probability of such harm in case no action or other defensive action is taken.
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 181.
Finogenov and Others v. Russia (Judgement), App. Nos. 18299/03 and 27311/03, (ECtHR, 20 December 2011), para. 236.
ibid, para. 232. See also Andronicou and Constantinou v. Cyprus (Judgement), App. No. 86/1996/705/897 (ECtHR, 9 October 1997), para. 184–188, and Melzer, Human Rights Implications of the Usage of Drones and Unmanned Robots in Warfare (n 42) p. 32.
See e.g. Isayeva v. Russia (Judgement), App. No. 57950/00, Former First Section (ECtHR, 24 February 2005), paras. 181–191. See also Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) pp. 90–91, citing U.S. domestic case law: People v Adams, 9 Ill App 3d 61 (1972); Henwood v. People, 54 Colo 188 (1913) [8]; Annot, 18 alr 917 (1922) 928; Ringer v. State (1905) 74 Ark. 262, 85 S. W. 410; and Scott v. State (1905) 75 Ark. 142, 86 S.W. 1004.
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 158.
Tagayeva and Others v. Russia, App. No(s) 26562/07, 14755/08, 49339/08, 49380/08, 51313/08, 21294/11, 37096/11, Judgment (Merits and Just Satisfaction) (ECtHR, 13 April 2017), para. 609, confirming the position in Isayeva v. Russia that “the primary aim of the operation should be to protect lives from unlawful violence. The massive use of indiscriminate weapons stands in flagrant contrast with this aim and cannot be considered compatible with the standard of care prerequisite to an operation of this kind involving the use of lethal force by State agents.” Isayeva v. Russia (n 140) para. 191.
For an interesting and detailed discussion of these issues, see Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 183–189.
Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 90.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) pp. 497–498, §24.24 with commentaries, especially fn. 53. For instance, in the Canadian publication ‘Use of Force for CF Operations’, it is made clear that Canadian forces are permitted to use force in self-defence to protect oneself, other members of the Canadian forces, and “non-Canadian military personnel who are attached or seconded to a Canadian force”. Canadian Forces, Use of Force for CF Operations (Joint Publication 5.1, Document B-GJ-005-501/FP-001, issued under the authority of the Chief of the Defence Staff, 2008) p. 2–2.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) pp. 497–498, §24.24 with commentaries.
Todd C. Huntley, ‘Balancing Self-Defense and Mission Accomplishment in International Intervention: Challenges in Drafting and Implementing Rules of Engagement’, 29 Maryland Journal of International Law 83 (2014), p. 115.
See also Hessbruegge, Human Rights and Personal Self-defense (n 6) pp. 102–103.
The question of extraterritorial human rights application is examined in Section 4.2.3.
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 103.
ibid.
The use of force to protect objects, the destruction or damage to which will not result in threat to life or limb, is likely violate human rights. See above, Sections 4.2.2 and 6.3.1.3.
See e.g. U.S. cjcs, sroe/sruf (n 67) p. A-3: “When individuals are assigned and acting as part of a unit, individual self-defense should be considered a sub-set of unit self-defence. As such, unit commanders may limit individual self-defense by members of their unit.” See also p. I-1, quoted in Christopher D. Amore, ‘Rules of Engagement: Balancing the (Inherent) Right and Obligation of Self-Defense with the Prevention of Civilian Casualties’, 1 National Security Law Journal 39 (spring 2013), p. 50: “[U]nit commanders may issue supplemental measures to limit self-defense by members of their units. The use of force for mission accomplishment may sometimes be restricted by specific political and military goals that are often unique to the situation”. See also the Norwegian Manual of the Law of Armed Conflict, which makes it clear that the ability to defend others or objects may be subject to limitations due to political or operational considerations, while the ability to defend oneself in self-defence may not limited. Norwegian loac Manual (n 20) pp. 24–25 and 281.
Sandy Woodward, One hundred days: The memoirs of the Falklands Battle Group Commander. (HarperCollins Publishers, 1992) pp. 107–108, cited in Guy R. Phillips, ‘Rules of Engagement: A Primer’, The Army Lawyer, Department of Army Pamphlet 27-50-248 (July 1993), p. 16.
See also Peter Dreist, ‘Rules of engagement in nato operations – application in Germany’s legal system’, in Barbara Janusz-Pawletta (ed.), Zasady użycia siły (ang. Rules of Engagement) – wybrane problem prawne [Rules of Engagement – legal problems] (Towarzystwo Wiedzy Obronnej, Warsaw, 2011) p. 127, arguing that “the right to defense in aid of a third person can be restricted in scope up to complete prohibition”.
See above, Section 8.2.8.
‘Extended self-defence’ is further examined in Section 8.3.4.
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 102. See also Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) p. 55, emphasising that the moral duty to protect has to be considered in light of the legal duty to control the use of force, and Gill et al., ‘General Report’ (n 24) p. 163.
This is examined in Section 4.2.
“Courageous restraint” was an isaf doctrine introduced by Gen Stanley McChrystal, the former American commander, to reduce the number of civilian casualties. The concept has been much criticised, see e.g. Amore, ‘Rules of Engagement’ (n 153) especially pp. 59–74. He concludes that “[w]hile the strict roe and the concept of “courageous restraint” were initially successful at reducing civilian casualties caused by U.S. and nato forces, the total number of civilian casualties actually increased due to greater insurgent activity”. ibid, p. 75. The policy was also much debated and criticised in the media for putting the lives of military forces at risk. See e.g. Thomas Harding, ‘‘Courageous restraint’ putting troops lives at risk’, The Telegraph (6 July 2010,
Stephens, ‘ROE and the Concept of Unit Self-Defense’ (n 1) pp. 147–148. The right to life is examined in Section 4.2.
See also Danish Ministry of Defence/Defence Command Denmark, Military Manual on international law relevant to Danish armed forces in international operations [hereinafter: Danish Military Manual], Rosendahls, København, 2016, available at
Amore, ‘Rules of Engagement’ (n 153) p. 50.
Canadian Forces, Use of Force for CF Operations (n 145) p. 2–2: “commanders may legitimately order individuals or units under their command not to respond to hostile intent. Such an order would be based on that commander’s responsibility, in certain circumstances, to control the escalation of force.” Note that hostile intent in this context refers to the requirements for self-defence and is therefore different to the nato concept of ‘hostile intent (not constituting imminent attack)’. See also Blaise Cathcart, ‘Enforcement and Peace Enforcement Operations’, Chapter 11 in Terry D. Gill and Dieter Fleck (eds.), The Handbook of the International Law of Military Operations (Oxford University Press, Oxford, 2015) pp. 144–145, para. 26.
Cathcart, ‘Enforcement and Peace Enforcement Operations’, ibid, p. 144, para. 20.
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 102. See also Stephens, ‘ROE and the Concept of Unit Self-Defense’ (n 1) p. 147, although he argues that for this reason, self-defence may never be limited.
On the right to life, see further Section 4.2.
Boddens Hosang, ‘Self-Defence in Military Operations’ (n 2) p. 32.
ibid, fn. 10.
Smith and others (FC) (Appellants) v The Ministry of Defence (Respondent) [2013] uksc 41, para. 71. The case concerned the application of echr Article 2 (n 21) to military forces participating in an armed conflict.
Johs. Andenæs, ‘Okkupasjonstidens «likvidasjoner» i rettslig belysning’, Tidsskrift for rettsvitenskap (TFR) 1948, 1–31, p. 15: “Det kan tenkes at en likvidasjon er utført av underordnede i organisasjonen på egen hånd i strid med de regler som ble fulgt. En slik ordreoverskridelse kan ikke i og for seg gjøre likvidasjonen rettsstridig overfor den drepte.” See also the Danish Military Manual (n 162) pp. 150 and 152 where it is made clear that if an roe limits the ability to use force in self-defence, the roe must be complied with even if the force would be lawful, and failure to do so would be a dereliction of duty.
See e.g. the Norwegian Military Penal Code (Militær Straffelov), LOV-2016-04-22-3, §24 and Norwegian LOAC manual (n 20) p. 271, where it is explained that military forces have a duty to follow orders unless they know the order to be unlawful or the order is manifestly unlawful. Similarly, the Uniform Code of Military Justice (ucmj), 64 Stat. 109, 10 U.S.C. Chapter 47 (
Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 102; and Danish Military Manual (n 162) p. 151.
Cathcart, ‘Enforcement and Peace Enforcement Operations’ (n 164) p. 144, para. 21.
An examination of the defence of necessity is unfortunately beyond the scope of this book.
Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 92.
ibid.
See e.g. Brian Bengs (Lt Col, then of nato School), ‘NATO Rules of Engagement and Use of Force’, presentation available at
AP I (n 28) Article 49(1). See also Section 5.1.3.
During operations below the threshold of armed conflict, the legal base for such defensive actions cannot be loac, and is more likely to be a legal concept of self-defence, either in the form of personal self-defence or State self-defence. Due to the limitations of the book, however, this will not be examined further here. See for instance Boddens Hosang, ‘Self-Defence in Military Operations’ (n 2) pp. 46–47.
On the relationship between the jus ad bellum and jus in bello, see also 4.3. See also Gaston, ‘Reconceptualizing Individual or Unit Self-Defense’ (n 11) pp. 289 and 330.
The existence of such a concept is argued by Corn. In his view, “while universally accepted principles of self-defense are useful to understanding the law’s tolerance of self-help uses of force generally, it is the public authority justification more broadly, interpreted through the lens of domestic and human rights law governing the use of force by state actors, which should form the basis of formulating defensive use-of-force rules for the military”. Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) p. 48. He goes on to state that the “public authority self defense” is a general principle of international law and that it overlaps with or replaces the individual right of self-defence of the service members. See ibid, pp. 48–49.
The term “operational self-defence” as used in the current context must be distinguished from the use of the same term by Albin Eser to describe defensive operations under the jus ad bellum. Eser, ‘Article 31’ (n 47) pp. 879–880.
On direct participation in hostilities, see further Section 5.2.3.
See, for instance, the 2007–2011 isaf Tactical Directives on the avoidance of civilian casualties (so-called civcas directives), commented on in Sections 2.4.1 and 5.5.
In isaf for example, the military forces assigned to a particular Regional Command would not have the authority to enter another Regional Command area.
See e.g. Bagwell and Kovite, ‘It is not self-defence’ (n 3) especially pp. 4–5, 8 and 11, explaining how failure to adapt the sroe application in Afghanistan to provide authority to deal with persons directly participating in hostilities resulted in confusing ‘use of force’ guidance and self-defence becoming the default ‘use of force’ authority in many tactical situations. See also Gaston, When Looks Could Kill (n178) p. 23, arguing that the tactical directives introduced to reduce civilian casualties “may have had the additional side effect of increasing reliance on self-defense, where other types of force were more limited”.
See also Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 73, where they explain that there is “an apparent trend towards relying on self-defence under criminal law as a justification for the use of force during armed conflicts. One possible cause for this trend is a combination of restrictive rules relating to the use of offensive force in counter-insurgency operations, combined with an emphasis on the paramount nature of the ‘inherent right of self-defence’ in the military training and doctrine of some States.” (footnote omitted).
Gaggioli, The use of force in armed conflict (n 133) p. 27.
nato Resolute Support Press Release of 12.01.2017, Civilian casualties confirmed in Boz village, Kunduz, available at
See further Sliedregt, Individual Criminal Responsibility in International Law (n 45) p. 234, where she refers to defensive loac force as self-defence.
See Section 6.3.1.
See also Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) pp. 6–7 and p. 19.
See also Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) pp. 485–486, §24.11.
ibid, p. 477, §24.02. Although the comment concerns nato ‘extended self-defence’ (see Section 8.3.4), it is equally applicable to the reference to and reliance on self-defence in the nato roe context.
Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) p. 7.
Gaston also voices the concern that military manuals, guidance, and other legal documents tend to treat self-defence as “an assumed and inalienable right” but that its “scope and outer bounds is not discussed in guidance or legal literature”. See Gaston, ‘Reconceptualizing Individual or Unit Self-Defense’ (n 11) pp. 290–291.
Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) p. 11. The reference to the U.S. concept of mission-accomplishment roe is removed to avoid confusion with the slightly different application of the same term in the context of nato roe. See further Section 3.3.
Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 77.
See also Charles Garraway, ‘The ‘War on Terror’: Do the Rules Need Changing?’, Chatham House International Law Briefing Paper 06/02, September 2006, p. 7.
Boddens Hosang, for instance, argues that the statement can only apply to unit self-defence and personal self-defence, not other forms such as extended self-defence or force protection. Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 485, §24.11.
See also Gary D. Solis, The Law of Armed Conflict: International humanitarian law in war (Cambridge University Press, Cambridge) 2016, pp. 487–488.
ibid, p. 488.
Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 74.
Boddens Hosang, ‘Self-Defence in Military Operations’ (n 2) p. 84, in the context of ‘extended self-defence’.
The use of force in self-defence is for instance reported to have been the main cause of civilian casualties caused by U.S. forces in Afghanistan, begging the question whether the rate of civilian casualties could have been reduced if the use of force had been subject to further control and regulation. See U.S. Joint and Coalition Operational Analysis, Reducing and Mitigating Civilian Casualties: Enduring Lessons (2013, Suffolk, Virginia, available at
Gaston, ‘Reconceptualizing Individual or Unit Self-Defense’ (n 11) p. 325. She suggests that “It may be that many, perhaps even the majority, of incidents involving unnecessary or excessive responses are simply violations of the U.S. self-defense standards and roes” (p. 325). However, because they were reported as self-defence incidents, they are less likely to be investigated, and soldiers “tend to be given the benefit of the doubt” (p. 326).
See also ibid, pp. 325–327. See further Chapter 9 below, where it is argued that ideally, the use of force in combat operations should be dealt with as a roe ‘use of force’ category rather than self-defence.
As a result, although it is less complicated and hence clearer, the solution proposed by Gaston, whereby all force a soldier uses to defend himself during an armed conflict should be viewed as legally premised on loac and hence a form of “combatant self-defence”, would fail to take into account all the situations military forces may be faced with, giving rise to a need for defensive force, that are not regulated by loac. ibid, pp. 329–332.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 499, §24.25. See also Scaliotti, ‘Defences before the international criminal court’ (n 13) p. 161, fn. 330.
See also Gaston, ‘Reconceptualizing Individual or Unit Self-Defense’ (n 11) p. 321, where she emphasises that “the very narrow conception of self-defense held by European soldiers may not be sufficient to deal with the full panoply of ambiguous or indirect threats that present themselves” for instance in conflicts like Afghanistan.
Bagwell and Kovite, ‘It is not self-defence’ (n 3) p. 3 (footnotes omitted). See also U.S. DoD, Law of War Manual (n 115) pp. 2233–234.
For an example of confusion of the two areas of law, see e.g. Husby, ‘A Balancing act: In Pursuit of Proportionality in Self-defense for On-Scene Commanders’ (n 54) p. 6. For instance, at p. 10, he states that “[t]he jus ad bellum focus in the sroe regarding ihl principles related to self-defense does not obviate the need to apply jus in bello ihl [sic] principles, including military necessity, distinction, and humanity, once a self-defense engagement is initiated”, and further that “proportionality balancing is one of the core jus in bello principles that determines whether a self-defense action is ‘appropriate’”.
On the loss of protection due to dph, see further Section 5.2.3.
See further Section 6.3.2 for an introduction to national approaches to self-defence.
See also, Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) pp. 11 and 46, voicing concerns about “overbroad application of self-defense rules in future, less hostile environments”. The issue was for instance raised in connection with the 2010 disaster relief operation in Haiti.
AP I (n 28) Article 43(1).
The issue of legal responsibility was raised by Boddens Hosang, ‘Self-Defence in Military Operations’ (n 2) pp. 30–31, with regards to the distinction between national self-defence and personal self-defence, but is equally applicable to the distinction between roe and self-defence.
Scaliotti, ‘Defences before the international criminal court’ (n 13) p. 158.
Antonio Cassese et al., International Criminal law (Oxford University Press, Oxford, 2013) p. 211.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 481, footnote 18 and commentaries to §24.07.
Stephens, ‘ROE and the Concept of Unit Self-Defense’ (n 1) p. 126; Charles P. Trumbull IV, ‘The basis of unit self-defense and implications for the use of force’, 23 Duke Journal of Comparative and International 121 (2012–2013), p. 122; Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) p. 21; and Alan Cole et al., Sanremo Handbook on Rules of Engagement (International Institute of Humanitarian Law, 2009) p. 3.
Kalshoven and Fontein, ‘Some Reflections on Self-Defence’ (n 31) p. 106.
Ashley Roach, ‘Rules of engagement’, 36(1) US Naval War College Review 46 (Jan/Feb 1983), p. 49.
Stephens, ‘ROE and the Concept of Unit Self-Defense’ (n 1) p. 126.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 481, §24.07.
Hans Boddens Hosang, Rules of Engagement (PhD thesis, University of Amsterdam 2017, available at
See also Boddens Hosang, Rules of Engagement (n 226) p. 113.
See also Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 482, §24.08, paragraph 2.
Boddens Hosang, Rules of Engagement (n 226) p. 115.
ibid. See also Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) pp. 482–483, §24.08, paragraph 3.
Trumbull, ‘The basis of unit self-defense and implications for the use of force’ (n 221) pp. 143–145; Stephens, ‘ROE and the Concept of Unit Self-Defense’ (n 1) pp. 138–140; and Boddens Hosang, Rules of Engagement (n 226) p. 112.
Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) pp. 483–484, §24.08, paragraphs 4–7 and §24.09 with commentaries.
The difference in the level of authorisation and the nature and scope of the defensive force has caused others to reject the relevance of State self-defence as the legal basis for the use of force in unit self-defence. See e.g. Trumbull, ‘The basis of unit self-defense and implications for the use of force’ (n 221) pp. 127–132.
Yoram Dinstein, War, Aggression and Self-Defence (Cambridge University Press, Cambridge, 2011) p. 243, footnote omitted. See, however, Trumbull, ‘The basis of unit self-defense and implications for the use of force’ (n 221) p. 131, claiming that there is in fact a qualitative difference.
Boddens Hosang, Rules of Engagement (n 226) pp. 112–113. Note, however, that he considers unit self-defence to be “somewhat irrelevant” in the context of international operations because the use of force will be founded on the right of force protection or nato ‘extended self-defence’ instead.
See also Hessbruegge, Human Rights and Personal Self-defense (n 6) p. 232.
See further Section 4.3.
Stephens, ‘ROE and the Concept of Unit Self-Defense’ (n 1) pp. 128–140, and Trumbull, ‘The basis of unit self-defense and implications for the use of force’ (n 221) especially pp. 133–139.
Gill et al., ‘General Report’ (n 24) p. 144.
He views unit self-defence as “an example per excellance” of the State’s human rights obligations owed to military members. Stephens, ‘ROE and the Concept of Unit Self-Defense’ (n 1) 144.
ibid, pp. 128 and 137–138. He explains that “[t]his right is one which has evolved from bitter experience”, such as the 1983 Beirut bombing of U.S. Marine Headquarters, the 1987 attack on uss Stark, and the 1988 Vincennes shootdown (see pp. 128–129). See also Stephens, ‘Human Rights and Armed Conflict’ (n 162) pp. 21–23.
Stephens, ‘ROE and the Concept of Unit Self-Defense’ (n 1) p. 140.
Trumbull, ‘The basis of unit self-defense and implications for the use of force’ (n 221) pp. 133–139. See also Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) p. 21, criticising Dale Stephens for failing to provide sufficient evidence for this claim. See also Kalshoven and Fontein, ‘Some Reflections on Self-Defence’ (n 31) p. 106, and the differing national approaches reported in Horvat and Benatar (eds.), Legal Interoperability (n 24) ‘National Reports’, pp. 180–181 (Germany: State self-defence); pp. 197–198 (Norway: collective application of personal self-defence); and pp. 212–213 (Czech Republic: collective application of personal self-defence).
The existence of a human right to self-defence was examined, and rejected, above in Section 6.3.1.3.
On the right to life of military forces, see Section 4.2.
See also discussion in Section 6.3.2.1 concerning self-defence as a right or a duty.
See also Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) p. 22, arguing that “[t]hose who argue that the obligation stems from human rights law fundamentally misconstrue the nature and purpose of that body of law as a check on the use of force by state actors.”
Stephens, ‘ROE and the Concept of Unit Self-Defense’ (n 1) pp. 131–135.
See also Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) p. 21, fn. 87.
Boddens Hosang also concludes that “no legal basis by itself fully satisfies every aspect of the right of unit self-defense without requiring at least some adaption or interpretation of the legal principles on which it is founded.” Boddens Hosang, Rules of Engagement (n 226) p. 117. As mentioned, his answer is to apply a limited form of the jus ad bellum rule of self-defence as the sole authority for unit self-defence, however, he considers it of limited importance during international operations. See ibid, p. 113, fn. 68.
See also Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 76, arguing that unit self-defence is an operational concept.
ibid, p. 76.
It should be noted, however, that the U.S. sroe defines its unit self-defence concept wider, permitting force in response to “force used directly to preclude or impede the mission and/ or duties of US forces”. U.S. cjcs, SROE/SRUF (n 67) p. A-4. See also Stephens, ‘ROE and the Concept of Unit Self-Defense’ (n 1) pp. 142–143.
Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 76.
The U.S. sroe, for instance, limits unit self-defence to the defence of other U.S. forces. U.S. cjcs, SROE/SRUF (n 67) p. A-3. By contrast, according to the German contribution to Horvat and Benatar (eds.), Legal Interoperability (n 24) ‘National Reports: Germany’, p. 181, unit includes all personnel and equipment of a mission. The reliance on unit self-defence by UN forces, a topic beyond the scope of this book, would also be permitted on the basis of the jus ad bellum, but in the form of the applicable UN Security Council resolution rather than State self-defence. For a contrary view, see Trumbull, ‘The basis of unit self-defense and implications for the use of force’ (n 221) pp. 129–130.
North Atlantic Treaty, Washington, D.C [4 April 1949] (available at
On collective self-defence, see Section 6.2.3.
U.S. cjcs, SROE/SRUF (n 67) p. A-3 and Boddens Hosang, ‘Force Protection, Unit Self-Defence, and Personal Self-Defence’ (n 2) p. 484, §24.10 with commentaries.
Boddens Hosang, ‘Force Protection, Unit Self-defence, and Personal Self-defence’, ibid, p. 485, §24.10, paragraphs 3–4.
Henderson and Cavanagh, ‘Claiming self-defence on the battlefield’ (n 18) p. 76.
Kalshoven and Fontein, ‘Some Reflections on Self-Defence’ (n 31) p. 105.
ibid, p. 111.
nato, Military Decision on MC 362/1 – nato Rules of Engagement [hereinafter MC 362/1], 30 June 2003. The MC 362/1 is nato unclassified, however, permission to use parts of the document for the purposes of this research is granted by the nato Military Commission in document imstam(O&P)-0006-2018 (copy on file with the author).
Note that the self-defence concept included in the MC 362/1 may also be relevant to State or jus ad bellum self-defence, however, as the focus of the book is on the use of force during an armed conflict, this perspective is not examined further here.
According to stanag 2597, the MC 362/1 definition is not a legal definition but a planning tool to know where roe begins/what roe are needed. See nato, stanag 2597: Training in Rules of Engagement, ATrainP-4 [hereinafter: stanag 2597], 4 May 2015 (available at
nato, MC 362/1 (n 263) p. 4. See also nato, STANAG 2597 (n 265) p. B-20.
nato, STANAG 2597, ibid, p. B-19.
See however Dreist, ‘Rules of engagement in NATO operations’ (n 155) p. 127, who appears to consider the nato concepts of ‘hostile act’ and ‘hostile intent’ to be part of ‘extended self-defence’, thereby extending the circumstances where self-defence may be relied upon beyond situations of imminent or actual attack. As explained in 3.5, these nato roe concepts are not self-defence concepts but rather mission accomplishment roe.
See Section 8.2.8.
See also U.S. Army, Operational Law Handbook (U.S. Army Judge Advocate General’s Legal Center and School, International and Operational Law Department, Charlottesville, Virginia, 2017) p. 79, for an example of such an approach.
Boddens Hosang, Rules of Engagement (n 226) p. 102. See also Gill et al., ‘General Report’ (n 24) p. 170.
Article 5 is set out in note 256.
Boddens Hosang, Rules of Engagement (n 226) pp. 102–103.
ibid, pp. 103–104.
See the discussions in Sections 4.3 and 8.3.3.
Boddens Hosang, Rules of Engagement (n 226) p. 104.
Allied Joint Doctrine for Force Protection AJP-3.14 (November 2007) cited in nato, STANAG 2597 (n 265) p. B-81. The definition found in the Sanremo roe Handbook is similarly focused on preventive measures: “force protection: actions taken to prevent or mitigate hostile actions against personnel (to include family members), resources, facilities, and critical information. Force protection does not include actions to defeat the enemy or protect against accidents, weather, or disease.” Cole et al., Sanremo Handbook on Rules of Engagement (n 221) p. 82.
nato, STANAG 2597 (n 265) p. B-81.
See e.g. Jens David Ohlin and Larry May, Necessity in international law (Oxford University Press, Oxford, 2016) Chapter 11: ‘Force Protection’, pp. 259–272, and Gill et al., ‘General Report’ (n 24) p. 154.
Boddens Hosang, Rules of Engagement (n 226) p. 104.
ibid, pp. 104–105.
ibid, p. 106.
ibid, p. 105.
A relevant topic in this regard is the use of force by UN forces, however, this is not within the topic of this book. For a selection of the extensive literature on the topic, see e.g. Simon Chesterman, The Use of Force in UN Peace Operations (External Study, Best Practices Unit, UN Department of Peacekeeping Operations, New York, 2004); Trevor Findlay, The Use of Force in UN Peace Operations (sipri, Oxford University Press, Oxford, 2002); Katherine E. Cox, ‘Beyond Self-Defense: United Nations Peacekeeping Operations and the Use of Force’, 27 Denver Journal of International Law and Policy 239 (1999); Boddens Hosang, Rules of Engagement (n 226) pp. 106–110; and Trumbull, ‘The basis of unit self-defense and implications for the use of force’ (n 221) pp. 138–139.
1907 Hague Regulations (n 38) Article 23(g) reads: “To destroy or seize the enemy’s property, unless such destruction or seizure be imperatively demanded by the necessities of war”. This rule is reflected in customary law and international criminal law, see the ICRC CIL Study (n 117) Rule 50, and the Rome Statute (n 38) Article 8(2)(b)(xiii). See also Rule 97 of Yoram Dinstein et al., Oslo Manual on Select Problems of the Law of Armed Conflict: Rules and Commentaries (per August 27 2018, forthcoming, copy on hand with the author): “It is specifically prohibited to destroy, damage or seize enemy private or public property unless such destruction is justified by military necessity under the principles and rules of loac”.
Krupp et al., Judgment of 31 July 1948, U.S. Military Tribunal Nuremberg, in Trials of War Criminals Before the Nuremberg Military Tribunals, Vol. IX (available at
ibid.
Prosecutor v. Gotovina and others (n 87) §1730.
Gill et al., ‘General Report’ (n 24) p. 157.
See also Gary P. Corn, ‘Developing Rules of Engagement: Operationalizing law, policy and military imperatives at the strategiunic level’, in Geoffrey S. Corn, Rachel E. VanLandingham and Shane R. Reeves (eds.), U.S. Military Operations – Law, Policy and Practice (Oxford University Press, Oxford, 2016) p. 224.
A term used by Husby to describe self-defence as applied by U.S. forces in recent conflicts. Husby, ‘A Balancing act: In Pursuit of Proportionality in Self-defense for On-Scene Commanders’ (n 54) p. 11.
See also Gaston, ‘Reconceptualizing Individual or Unit Self-Defense’ (n 11) pp. 327–329, citing a U.S. airstrike on a al-Shabaab training camp in Somalia, killing an estimated 150 persons, as an example of such expansive self-defence being applied outside of an armed conflict.
Bagwell and Kovite, ‘It is not self-defence’ (n 3) p. 13.
ibid, pp. 13–16. For further discussion on the sroe requirement of imminence, see references in Chapter 6, fn. 369.
Corn, ‘Public Authority to Use Force in Military Operations’ (n 19) p. 11.
See also Bagwell and Kovite, ‘It is not self-defence’ (n 3) especially pp. 40–42; and Corn, ‘Public Authority to Use Force in Military Operations’, ibid, p. 56.