What’s in an Act? Conduct, Context and Time: A Framework for Analyzing Direct Participation in Hostilities (Sword and Scale, Fall 2014)

What’s in an Act? Conduct, Context and Time: A Framework for Analyzing Direct Participation in Hostilities Written by Jennifer Dumoulin, J.D. 2015 Candidate, Queen’s University INTRODUCTION Why does the definition of “direct participation in hostilities” matter? The purpose of international humanitarian law (IHL) is to regulate the conduct of military operations during armed conflicts and, in particular, to protect civilians and the victims of war. Under IHL, civilians are afforded “general protection against [the] dangers arising from military operations” in situations of international and non-international armed conflicts.1 This protection is extended to civilians, “unless and for such time that they take a direct part in hostilities”.2 The concept of direct participation in hostilities is, therefore, a determinative factor in armed conflict situations as it delineates those civilians who should be protected from those who have engaged in unlawful conduct and, as a result, have forfeited their protected status. Despite its importance, identifying the precise conduct that causes a civilian to lose his or her protection remains elusive. There has been much debate surrounding the concept of direct participation in hostilities. In an attempt to clarify this debate, the International Committee of the Red Cross (ICRC) published its Interpretive Guidance on the Notion of Direct Participation in Hostilities under International Law [hereafter Interpretive Guidance]. 3 To do this, the Interpretive Guidance considered three questions: who is a civilian, what constitutes direct participation and what causes the loss of protection. 1 Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of International Armed Conflicts, June 8 1977, 1125 UNTS 3, Article 51.1 [Additional Protocol I]; Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of NonInternational Armed Conflicts, June 8 1977, 1125 UNTS 609, Article 13.1 [Additional Protocol II]. 2 Additional Protocol I, supra note 1, Article 51.3; Additional Protocol II, supra note 1, Article 13.3. 3 See Nils Melzer, ICRC-0990-002, 2009 [Interpretive Guidance]. 1 Despite the issuance of the Interpretative Guidance, many aspects of direct participation in hostilities remain unsettled.4 This debate can be divided into five main categories: (1) when does direct participation apply (status of the conflict); (2) to whom does it apply (status of the actor); (3) what is meant by direct participation (”conduct-ual” nature of the act); (4) for what length of time does an individual directly participate in hostilities (temporal nature of the act); and, (5) what are hostilities (contextual nature of the act). While the focus of this paper is the “act” component of direct participation in hostilities, a brief overview of the debate surrounding the status of the conflict and the status of the actor is nevertheless informative. PART I: What Does “Direct Participation in Hostilities” Mean? What is an armed conflict? The notion of direct participation in hostilities is relevant in both international armed conflicts and non-international armed conflicts. Simply put, an international armed conflict is any difference occurring between two States that leads to an intervention by their armed forces.5 It also includes situations of rebellion against alien occupation and colonial domination.6 A non- 4 See e.g. William H Boothby, “Direct Participation in Hostilities – A Discussion of the ICRC Interpretive Guidance” (2010) 1 Journal of International Humanitarian Legal Studies 143 [Boothby, “Discussion of the Interpretive Guidance”]; Michael N Schmitt, “The Interpretive Guidance on the Notion of Direct Participation in Hostilities: A Critical Analysis” (2010) 1 Harvard National Security Journal 5; Bill Boothby, “‘And For Such Time As’: The Time Dimension to Direct Participation in Hostilities”, (2010) 42:3 NYUJ Int’l & Pol 741 [Boothby, “Time Dimension in DPH”]; Kenneth Watkin, “Opportunity Lost: Organized Armed Groups and the ICRC ‘Direct Participation in Hostilities Interpretive Guidance’”, (2010) 42:3 NYUJ Int’l & Pol 641; Nils Melzer, “Keeping the Balance Between Military Necessity and Humanity: A Response to Four Critiques of the ICRC’s Interpretive Guidance on the Notion of Direct Participation in Hostilities”, (2010) 42:3 NYUJ Int’l & Pol 831. 5 See Article 3 of the Geneva Convention for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field, Aug 4 1949, 75 UNTS 31; Article 3 of the Geneva Convention for the Amelioration of the Condition of Wounded, Sick and Shipwrecked Members of Armed Forces at Sea, Aug 4 1949, 75 UNTS 85; Article 3 of the Geneva Convention relative to the Treatment of Prisoners of War, Aug 4 1949, 75 UNTS 135; and, Article 3 of the Geneva Convention relation to the Protection of Civilian Persons in Time of War, Aug 4 1949, 75 UNTS 287 [Common Article 3]. 6 See Article 2 of the Geneva Convention for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field, Aug 4 1949, 75 UNTS 31; Article 2 of the Geneva Convention for the Amelioration of the Condition of Wounded, Sick and Shipwrecked Members of Armed Forces at Sea, Aug 4 1949, 75 UNTS 85; Article 2 international armed conflict, on the other hand, occurs between organized, non-governmental armed groups and the armed forces of a State.7 It also includes conflicts between two non-State armed groups.8 Each type of conflict has specific criteria that must be met in order for a situation to qualify as an armed conflict.9 In the case of a non-international armed conflict, for example, both the intensity of the violence between non-State armed groups10 and the organizational structural of non-State parties are factors.11 Ultimately, because the concept of direct participation in hostilities is only relevant in armed conflict situations, either an international or a noninternational armed conflict must exist to trigger its applicability.12 For the purpose of this paper, it is assumed that any situations referred to in Part II meet the definitional requirements of an armed conflict. Who is a civilian? Under IHL, civilian status is defined by who a person is not rather than who are they are. Article 50(1) of Additional Protocol I, which governs international armed conflicts, provides that a civilian is any person who is not a member of the armed forces of a Party to a conflict13 or who is not participating in a levee en masse. Unlike international armed conflicts, there is no set 2 of the Geneva Convention relative to the Treatment of Prisoners of War, Aug 4 1949, 75 UNTS 135; and, Article 2 of the Geneva Convention relation to the Protection of Civilian Persons in Time of War, Aug 4 1949, 75 UNTS 287 [Common Article 2]; Additional Protocol I, supra note 1, Article 1.4. 7 International Legal Protection of Human Rights in Armed Conflict, OHCHR, HR/PUB/11/01, (2011) at 36. 8 Ibid. 9 See e.g. Robert Chesney, “Who May Be Killed? Anwar al-Awlaki as a Case Study in the International Legal Regulation of Lethal Force” (2010) 13 YB Int’l Human L 3 at 30. 10 Prosecutor v Tadić, IT-94-1-T, Decision on Defence Motion for Interlocutory Appeal on Jurisdiction (2 October 1995) at para 70 (International Criminal Tribunal the Former Yugoslavia, Trial Chamber II). 11 See e.g. APV Rogers’ discussion in Law on the Battlefield, 3d ed (Manchester, UK: Manchester University Press, 2012) 298-301. 12 OHCHR, supra note 7 at 33. 13 For a definition of the term ‘armed forces’, see Additional Protocol I, supra note 1, Article 43.1 – 43.2. See also Interpretive Guidance, supra note 3 at 21-25. 3 definition for the term civilian in a non-international armed conflict setting. However, it is generally understood that a civilian is a person who is not associated with either the military or an organized armed group of a party to the conflict.14 Under international law, members of the armed forces of a state or of an organized armed group of a party to the conflict are combatants and are therefore “authorized” to directly participate in hostilities, as long as they do so in a manner that is consistent with applicable laws.15 While at first glance distinguishing between civilians and members of the armed forces seems intuitive and straightforward, difficulties arise with regards to the classification of noncombatant members of armed forces, such as medical and religious personnel, as well as mercenaries, spies and other persons who accompany the armed forces.16 A similar problem emerges with regards to identifying the membership of organized armed groups.17 It is important to note that while civilians are afforded special protections under international law, civilian status in and of itself does not mean that a person cannot be targeted during armed conflict. Civilians could nevertheless be casualties of an attack if, for example, the object of that attack is a military objective18 and that attack complied with proportionality and other requirements of the law of targeting.19 14 Gary D Solis, The Law of Armed Conflict: International Humanitarian Law in War (Cambridge: Cambridge University Press, 2010) at 202. 15 Kunt Ipsen, “Combatants and Non-Combatants”, in Dieter Fleck, ed, The Handbook of Humanitarian Law in Armed Conflicts (Oxford: Oxford University Press, 1999) 301; Interpretive Guidance, supra note 3 at 27. 16 See e.g. ibid at 11, 37-40. For the definition of a mercenary, see Additional Protocol I, supra note 1, Article 47. For a definition of a spy, see ibid, Article 46. 17 See generally Watkin, supra note 4. 18 See e.g. William J Fenrick, “ICRC Guidance on Direct Participation in Hostilities” (2009) 12 YB Int’l Human L 287 at 292. See also Interpretive Guidance, supra note 3 at 35. 19 See e.g. Anicée van Engeland’s discussion of the 1999 NATO bombardment of Radio Televizija Serbia in Civilian or Combatant? A Challenge for the 21st Century (Oxford: Oxford University Press, 2011) at 104-05 4 For the purpose of this paper, it is assumed that all acts referred to under Part II are committed by civilians.20 Whether the civilians at issue are located or working in an area or building that is, itself, a legitimate military objective and therefore targetable under IHL is also beyond the scope of this paper. The Threshold of Harm and Other Requirements of Direct Participation The Interpretive Guidance establishes three requirements that an act must meet in order to qualify as direct participation in hostilities: threshold of harm, direct causation and belligerent nexus. To satisfy the threshold of harm requirement, an act must “be likely to adversely affect the military operations or military capacity of a party to an armed conflict or, alternatively, to inflict death, injury or destruction on persons or objects protected against attack”.21 For direct causation, “there must be a direct causal link between the act and the harm likely to result either from that act, or from a coordinated military operation of which that act constitutes an integral part”.22 Lastly, a belligerent nexus exists when an act is “specifically designed to directly cause the required threshold of harm in support of a party to the conflict and to the detriment of another”.23 Each of these constitutive elements of direct participation will be discussed in further detail below. 20 For an in-depth discussion about the concept of civilian status, see Interpretive Guidance, supra note 3 at 20-40; Boothby, “Discussion of the Interpretive Guidance”, supra note 4 at 148-50; Fenrick, supra note 18 at 289-92; van Engeland, supra note 19 at 27-133. 21 Interpretive Guidance, supra note 3 at 50. 22 Ibid. 23 Ibid. 5 Threshold of Harm The threshold of harm required for direct participation in hostilities can be attained in two ways: (1) if the act at issue causes harm “of a specifically military nature” or (2) if it causes the death, injury or destruction of protected persons or objects.24 It is important to note that the threshold of harm is relatively low in that it only requires an objective likelihood of the occurrence of harm rather than a materialization of that harm.25 Under the assumption that the two additional constitutive elements of direct participation in hostilities are satisfied, the ICRC’s Interpretive Guidance provides a list of acts that would meet the threshold of harm. These activities include sabotage and other activities that adversely affect military deployment, logistics and communication, guarding captured military personnel to prevent their liberation and the electronic interference with military computer networks.26 The Interpretive Guidance also provides a list of activities that, “in the absence of adverse military effects”, would not satisfy the threshold of harm requirement, including building fences or road blocks, disrupting electricity, water or food supplies and appropriating cars and fuel.27 For the purpose of this paper, it is assumed that all acts referred to in Part II satisfy the threshold of harm requirement. Direct Causation The Interpretive Guidance provides that the direct causation requirement is met when there is “a direct causal link” between an act and its resulting harm.28 Acts that are integral to a 24 Ibid at 47. Ibid. 26 For a more detailed list of activities that could meet the threshold of harm, see ibid at 47-49. 27 Ibid at 50. 28 Ibid at 51 (emphasis added). 25 6 larger military operation resulting in harm also satisfy this requirement.29 These specifications attempt to balance an interpretation of direct causation that would be too narrow if it was restricted to combat and active military operations, with one that would be too broad if it included the entire war effort.30 There nevertheless remains much debate and criticism about the Interpretive Guidance’s test for direct causation.31 Whether direct causation is satisfied will depend upon the relationship between the activity at issue and the harm imposed on the enemy “at the time and the place where the activity takes place”.32 The Interpretive Guidance suggests that this means not only that there must be a “sufficiently close causal relationship between the act and the resulting harm” but also that “the harm in question must be brought about in one causal step”.33 It has been argued that this latter recommendation is both incorrect and too restrictive. Michael Schmitt, an expert who participated in the ICRC’s Interpretive Guidance consultation project, states that there are some activities that would not satisfy the single-step causal link but would nevertheless constitute direct participation. He argues that the reference to a single-step causal link requires further clarification and suggests that it may be redundant because the act in question must already be integral to the resulting harm.34 In contrast to the causal link requirement, some consensus has emerged from the literature with regards to the meaning of ‘integral’. Generally, it is agreed that integral is neither 29 Ibid. International Committee of the Red Cross, Commentary to Additional Protocol I at para 1679, online: ICRC Treaties and Customary Law <http://www.icrc.org> [Commentary to AP I]. 31 See e.g. Schmitt, supra note 4. Among other things, Schmitt argues that the standard for direct causation should be based on ‘but for’ causation, foreseeability of consequences and the mens rea of intent. See ibid at 29. The relationship between conduct and causation is discussed in further detail in Part II of this paper. 32 Commentary to AP I, supra note 30 at para 1679. 33 Interpretive Guidance, supra note 3 at 52. 34 Schmitt, supra note 4 at 29-30. See also Michael N Schmitt, “Deconstructing Direct Participation in Hostilities: The Constitutive Elements” (2010) 42:3 NYUJ Int’l & Pol 697 at 729-32. 30 7 “necessary” nor “indispensable” to the operation, as this would impose too high a threshold.35 Ultimately, the purpose of this specification is to allow for the inclusion of conduct that may cause harm when it is performed with other acts, even though it may not be able to do so on its own. Regardless of which specification is applied, the classification of acts as direct participation remains a case-specific and fact-specific analysis, based on “the person’s behaviour, location, attire and other information [available] at the time”.36 The purpose of this paper is, ultimately, to provide a framework for this analysis by examining the nature of an act. Part II therefore considers in further detail the ‘conduct-ual’, temporal and contextual components of an act. Belligerent Nexus The presence of a belligerent nexus speaks to the purpose – but not necessarily the intent – of an act. It requires that an act be “specifically designed to support one party to the conflict by causing harm to another”.37 This means that hostile acts that occur in an armed conflict which do not support a party to the conflict by causing harm to another party would be excluded. From this perspective, stealing military equipment for private use or committing a bank robbery or other violent crimes for reasons unrelated to the conflict would not have a belligerent nexus to the conflict and would therefore not constitute direct participation in hostilities.38 35 See e.g Interpretive Guidance, supra note 3 at 30; ibid at 54. US, Department of the Navy, The Commander’s Handbook on the Law of Naval Operations (NWP 1-14M) (Washington, DC: Navy Warfare Publications Library, 2007) at 8.2.2. See e.g. Schmitt’s discussion on intelligence activities, the production of weapons and the use of unmanned aerial vehicles, supra note 4 at 29-31. 37 Interpretive Guidance, supra note 3 at 61. 38 Ibid at 60. 36 8 There has been some criticism surrounding the Interpretive Guidance’s notion of harm. Schmitt, for example, has argued that limiting belligerent nexus to conduct that harms an opposing party to the conflict is too restrictive as it would exclude activities that only strengthen one party’s military capacity or operations in a manner that is detrimental to another party to the conflict.39 Schmitt argues that a more useful distinction would be to categorize activities as “directly related to the armed conflict” or as activities “that are merely criminal in nature”.40 For the purpose of this paper, it is assumed that all acts referred to in Part II satisfy the belligerent nexus requirement of direct participation. Conclusion and Stated Assumptions As has been discussed above, the concept of direct participation in hostilities, while critical to determining whether a civilian has lost his protected status and can be targeted by direct attack, is neither clear nor straightforward. To activate a direct participation in hostilities analysis, two requirements must be met. First, there must be an armed conflict. Second, the person engaged in the act must be a civilian. For the purpose of this paper, it is assumed that these two requirements are always satisfied. Once direct participation in hostilities is at issue, the act in question must satisfy the three constitutive elements of direct causation. It must attain the necessary threshold of harm, it must directly cause or be integral to the activities causing the harm, and it must have a belligerent nexus to the armed conflict. If these three requirements are met, then the civilian has engaged in 39 Schmitt, supra note 4 at 27-28. Schmitt agrees with the Interpretive Guidance that a refusal to engage in activities that might positively affect military operations – for example, refusing to provide information on the location of military forces – does not constitute direct participation, even though the refusal would positively benefit the military forces. 40 Ibid at 28. 9 direct participation in hostilities and is no longer entitled to protection under international humanitarian law. While there is much debate about each of these constitutive elements, the purpose of this paper is examine the requirement of direct causation through an in-depth analysis of the nature of an act. As such, it will be assumed that all acts discussed in Part II of this paper satisfy the threshold of harm and belligerent nexus requirements. Part II: Unpacking the Nature of the Act Direct or Indirect, That is the Question: The “Conduct-ual” Nature of the Act The ICRC Interpretive Guidance, which attempts to clarify the concept of direct participation, suggests that there are both direct and indirect forms of participation in hostilities. 41 The Interpretive Guidance also equates ‘direct’ with ‘active’ by referring to Common Article 3 of the Geneva Convention, which provides protection to “persons taking no active part in the hostilities”.42 This implies that in addition to active or direct participation in hostilities, there are not only indirect forms but also inactive ones.43 Many alternative classification criteria have been suggested. Boothby argues that rather than focusing on direct and indirect participation in hostilities, a distinction should be made between acts that are “directly linked to the causation of harm to an adverse party to the conflict and acts that facilitate the creation of capacity to cause harm in general.” He would include acts that provide positive impacts to one side of the conflict – such as the provision of intelligence, the installation of military command and control equipment in a forward operating base and the 41 Interpretive Guidance, supra note 3, at 51. Ibid at 43; Common Article 3, supra note 5. For further discussion on this point, see Boothby, “Discussion of the Interpretive Guidance”, supra note 4 at 158. 43 Ibid (emphasis added). 42 10 loading of defensive aids onto an attack aircraft – even though they do not translate into an “immediate loss” for the opposing party.44 This perspective can be described as “evaluating the value added to war”, which has generally been rejected.45 Similarly, Schmitt has suggested that an analysis of direct participation should consider how critical the act is to the direct application of violence against an enemy.46 This interpretation closely parallels the Interpretive Guidance’s specification that the direct causal link can be satisfied by acts that are integral to a larger military operation that results in harm against an opposing party to an armed conflict. Finally, the Interpretive Guidance implicitly proposes a third classification system when it distinguishes between the actual conduct of hostilities, general war effort activities and warsustaining activities. General war effort activities are defined as “all activities objectively contributing to the military defeat of the adversary”, whereas war-sustaining activities include “political, economic or media activities supporting the general war effort”.47 If this classification system were to be used, direct participation would most likely be satisfied by activities within the categories of actual conduct of hostilities and general war effort activities, but not by those within the category of war-sustaining activities. The classification of acts becomes further complicated by the inclusion of measures preparatory or conclusory to the execution of an act if they constitute an integral part of the act.48 This suggests that both the deployment to and return from the location of the execution of a hostile act could be classified as direct participation in hostilities. Whether a preparatory measure 44 Ibid. van Engeland, supra note 19 at 103. 46 Supra note 4 at 26. 47 Interpretive Guidance, supra note 3 at 51 (emphasis added). See ibid for specific examples of general war effort and war-sustaining activities. 48 Ibid at 65. 45 11 constitutes direct participation will depend on “a multitude of situational factors”. 49 The Interpretive Guidance suggests that preparatory acts constitute direct participation in hostilities when “they are of a specifically military nature” and when they are “closely linked to the subsequent execution of a specific hostile act”.50 This would, however, exclude preparatory measures that establish general capacity building. Irrespective of the classification system that is employed, the literature on direct participation frequently provides examples of the type of conduct that is being discussed. These examples have been compiled in Annex 1 and have been organized using three classification systems: (1) direct or indirect, generally; (2) the Interpretive Guidance’s three-pronged system of conduct of hostilities, general war effort activities or war-sustaining activities; and, (3) Boothby’s system of direct link to the causation of harm or capacity contribution. The purpose of this table is to determine whether any overlap exists between these classification systems and whether any commonalities can be extracted from them to identify specific acts or categories of acts as either direct or indirect participation. What Does It All Mean? Findings and Discussion Some preliminary observations can be drawn from Annex 1. First, there is significant overlap between the categories of war-sustaining and capacity contribution activities. These activities have been classified elsewhere in literature as indirect forms of participation. Second, with regards to preparatory acts, only those acts that are linked to a specific operation or future act can be classified as direct participation. If preparatory acts, such as routine maintenance, are 49 50 Ibid. Ibid at 65-66. 12 undertaken without a specific purpose in mind,51 then they do not constitute direct participation in hostilities. It is not clear whether these activities would be classified as either war-sustaining or general war effort activities, the former of which are indirect forms of participation and the latter of which could be classified as either direct or indirect, depending on the facts specific to each situation. Generally, acts that fell under the category of “hostile acts”52 – such as the killing or wounding of military personnel, the capture or killing prisoners, the destruction of military equipment or engaging in sabotage – were classified as direct participation in hostilities. This occurred irrespective of whether the act targeted civilians, organized armed groups or the armed forces of a State. To the exception of two activities – returning from the location of an act and undertaking observatory acts in support of a party to the conflict – these acts always contained an element of violence. Regardless, all activities in this category are “active”, in that the participant actually undertakes an action. This is in contrast to a “passive” act, where the individual would either refuse to undertake an action or would merely stand by and allow an act to occur. The distinction between direct and indirect participation was less clear for acts classified as either “weapons-related activities” or “service and support provision”. Generally, those acts that merely supported the armed conflict – such as financing the war effort or engaging in propaganda – were classified as indirect participation. Similarly, activities that incidentally benefitted an armed group or the armed forces of a State – such as the provision of electricity, 51 It is not necessary that the participant has an awareness of the purpose of the act. As long as the act has a link to a specific future operation, the specific link requirement will be satisified. 52 For a definition of hostile acts, see e.g. APV Rogers, “Direct Participation in Hostilities: Some Personal Reflections” (2009) 48 Military Law and the Law of War Review 143 at 145 [Rogers, “DPH: Some Personal Reflections”]. Note that Rogers included preparatory and conclusory acts in his definition of hostilities, the former of which has been isolated in this paper for analytical purposes. 13 the production and sale of agricultural goods and the repair of infrastructure – did not constitute direct participation. Conclusions and Limitations Based on the findings of Annex 1, five conclusions can be drawn that will assist in the future classification of acts as either direct or indirect participation: (1) To directly participate in hostilities, a civilian must play an active-intangible or activephysical role in the activity or operation at issue.53 This means that mere acquiescence or a refusal to engage in an activity will not amount to direct participation.54 (2) Assuming that they meet the threshold of harm and belligerent nexus requirements, activities that can be classified as “hostile” or violent will amount to direct participation in hostilities. (3) Preparatory and conclusory acts must be linked to a specific operation to be classified as direct participation in hostilities. A link to a specific operation should not be confused with the causal link requirement for direct participation proposed by the Interpretive Guidance. Instead, the specific link requirement suggests that a preparatory act is only direct participation when it is undertaken with a specific operation in mind. This would 53 The use of the term “active-intangible” is meant to include activities that do not in and of themselves have a physical presence, such as the provision of intelligence. The key feature of this term remains, however, the ‘active’ component, which involves the civilian actually undertaking an active role in the conduct at issue, in contrast to a passive one. 54 This finding is consistent with some of the criticism of the Interpretive Guidance. See e.g. Schmitt, supra note 4 at 27-28. 14 therefore exclude routine maintenance or intelligence gathering as forms of direct participation. (4) The act at issue should have some adverse effect on a party to the conflict. Adverse effects should not, however, be limited to the direct or purposive causation of harm. They should include bolstering the military capacity of one party to the conflict, which often has an ancillary – and perhaps unintended – effect that is detrimental to the opposing party of the conflict.55 (5) Acts that merely support an armed conflict do not constitute direct participation in hostilities. Similarly, acts which by their nature have the consequence of benefitting an armed party to a conflict but that are not undertaken with the primary intention of doing so – for example, repairing local infrastructure – do not constitute direct participation in hostilities. Annex 1 and the five conclusions drawn from it provide some insight and guidance into the types of conduct that can be classified as either direct or indirect participation in hostilities. However, they represent merely a guide to this classification. In some instances, especially those that involve “weapons-related” or “service and support provision” activities, it may be difficult to assess prima facie whether the activity in question is direct or indirect participation. As such, it remains important to assess each act on its facts and in consideration of its particular 55 See e.g. Boothby, “Discussion of the Interpretive Guidance”, supra note 4 at 158. 15 circumstances.56 Two additional factors that should be included in any assessment of direct participation in hostilities are the temporal and contextual nature of the act. Each of these will be discussed in further detail below. Is Time on Your Side? : The Temporal Nature of Direct Participation The Additional Protocols stipulate that civilians lose their protected status only for such time as they are directly participating in hostilities. This temporal clause raises two primary issues: (1) whether civilians lose and regain their protected status when they engage in multiple acts of direct participation (the revolving door of civilian protection) and (2) when direct participation begins and ends (the temporal boundaries of direct participation). Each of these issues will be examined below in further detail. The Issue of the Revolving Door A strict interpretation of the temporal clause “for such time” results in a loss of protection only “for the duration of each specific act”.57 The Interpretive Guidance, in fact, argues that it is not an actual loss of protection that occurs but only a suspension.58 Thus, once a civilian ceases to directly participate in hostilities, he regains his protected status and can no longer be the target of a direct attack. This interpretation of “for such time” has been criticized for creating a “revolving door” of protection for civilians. In the Interpretive Guidance, the ICRC explicitly recognizes that its interpretation creates intervals of civilian protection. It notes that the “suspension of protection lasts exactly as long as 56 For an example of a fact-specific analysis, see Rogers, “DPH: Some Personal Reflections”, supra note 52 at 15153. 57 Interpretive Guidance, supra note 3 at 70. 58 Ibid. 16 the corresponding civilian engagement in […] hostilities. This necessarily entails that civilians lose and regain protection”.59 It has been argued that this revolving door of protection is not only difficult to apply in practice but also provides civilians who directly participate in hostilities on a repeated basis with an unfair advantage.60 Schmitt further challenges the ICRC’s interpretation on the grounds that a civilian who loses his protected status does so not because he is a threat to an organized armed group or State armed force, but because he has elected to participate in the conflict. From this perspective, the protection of civilians should be suspended until they are no longer participating in the conflict rather than to a point in time when they are not a danger to a party to the conflict.61 Boothby proposes a similar test, arguing that a civilian should be understood to be directly participating in hostilities from their “first act to [their] last one”.62 However, the ICRC dismisses criticism of the revolving door of protection on a number of grounds. First, the Interpretive Guidance distinguishes between directly participating in hostilities and engaging in a continuous combat function by separating civilians who undertake “merely spontaneous, sporadic or unorganized” 63 acts from “individuals whose continuous combat function involves the preparation, execution, or command of acts or operations [amounting] to direct participation in hostilities”.64 59 Ibid. See e.g. Boothby, “Discussion of the Interpretive Guidance”, supra note 4 at 147, 154-55; Schmitt, supra note 4 at 38. See also Trevor A Keck’s discussion of the liberal interpretation approach to direct participation in hostilities, in “Not All Civilians are Created Equal: The Principle of Distinction, The Question of Direct Participation in Hostilities and Evolving Restraints of the Use of Force in Warfare” (2012) 211 Military Law Review 115 at 130-131 as well as Keck’s discussion of the imposition of the burden of proof on states at 151. 61 Schmitt, supra note 4 at 37. 62 “Discussion of the Interpretive Guidance”, supra note 4 at 161. As part of his analysis, Boothby uses the example of a civilian gunman who has undertaken a series of hostile acts and who cleans, prepares and conceals his fireman in between attacks. He argues that the period of participation should include not only the hostile acts but also the acts in between because they benefit the civilian’s future acts, which are part of ongoing hostilities. See ibid. 63 Interpretive Guidance, supra note 3 at 35. 64 Ibid at 34. 60 17 This distinction between civilians and members of organized armed groups is, incorrectly, based on membership rather than function. It is equally as plausible to encounter a member of an organized armed group who undertakes spontaneous, sporadic or unorganized acts as it is to encounter a civilian who has repeatedly engaged in spontaneous, sporadic and unorganized acts to such an extent that he is, for all intents and purposes, engaged in a continuous combat function – albeit independently from an organized armed group. Why is it then that civilians who consistently re-engage in the direct participation of hostilities should regain their protected status during periods of inactivity when members of an organized armed who are not always engaged in a continuous combat function do not? The ICRC argues that international humanitarian law purposely distinguishes between “temporary, activity-based loss of protection” and “continuous, status- or function-based loss of protection”,65 but provides little evidence to support this distinction. The Interpretive Guidance states that members of organized armed groups have a “continuous function […] to conduct hostilities on behalf of a party to the conflict” whereas “the behaviour of individual civilians depends on a multitude of constantly changing circumstances and, therefore, is very difficult to anticipate”.66 That the Interpretive Guidance rejects that a civilian’s future conduct cannot be predicted by examining their past participation in hostilities is both counterintuitive and contradictory. In its discussion on the continuous combat function of members of organized armed groups – the Interpretive Guidance states that continuous combat function can be “identified on the basis of conclusive behaviour, for example, whether a person has repeatedly directly participated in hostilities […] indicating that such conduct constitutes a continuous function 65 66 Ibid at 44-45. Ibid at 70-71. 18 rather than a spontaneous, sporadic or temporary role”.67 The position of the ICRC has been challenged in the literature on direct participation,68 perhaps as a result of the absence of evidence demonstrating that membership in an organized armed group has greater predictive value than repeated patterns of behaviour. Lastly, the ICRC further justifies the existence of a revolving door of civilian protection on the grounds that a continuous loss of protection would result in “erroneous or arbitrary attacks against civilians”.69 It, once again, fails to provide evidence to support this claim. When Does Direct Participation Begin and End? A second issue that arises irrespective of whether a continuous or specific act analysis is used is how to identify the beginning and end of direct participation in hostilities. As has been discussed in extensive detail above, preparatory measures that are executed with a view to a specific operation constitute direct participation. However, there is no requirement that a preparatory measure take place immediately before the execution of an act.70 The beginning of a preparatory act – and by extension any other act of direct participation – occurs when a civilian undertakes a physical act “with a view to carrying out a specific operation”.71 The Interpretive Guidance provides that an act ends when the civilian has “physically separated from the operation” and lists “laying down, storing or hiding weapons and resuming activities unrelated to the operation” as an example of how this separation can be achieved.72 This may, however, temporally limit the inclusion of injuries or harms inflicted by delayed 67 Ibid at 35. See e.g. Boothby, “Discussion of the Interpretive Guidance”, supra note 4 at 162. 69 Interpretive Guidance, supra note 3 at 45. 70 Ibid at 66. 71 Ibid at 67. 72 Ibid. 68 19 explosive devices such as IEDs. A preferred approach in such situations would be to examine the chain of causation. This would allow for the expansion of “the period of participation [to include] as far before and after a hostile action as a causal connection existed”.73 This approach is consistent with that proposed by Boothby, who suggests that – in the context of ‘continuous’ direct participation in hostilities – some concluding acts, such as laying down, storing and hiding weapons, should be viewed as preparing for the next instance of direct participation.74 Conclusions Three conclusions can be drawn from the literature with regards to the temporal nature of direct participation: (1) It is not unreasonable to assume that a civilian who has repeatedly engaged in the direct participation of hostilities will do so again. As such, it is permissible to apply a broader interpretation of the temporal limits of direct participation to such civilians if there is a link – either causal or specific – between their intermediary activities and their hostile acts. (2) Direct participation begins when a civilian undertakes a positive action with a view to carrying out a specific operation, regardless of whether that action is the entirety of the operation or whether it is merely one component of it. 73 74 Schmitt, supra note 4 at 36. “Time Dimension in DPH, supra note 4 at 751. 20 (3) Generally speaking, direct participation ends when a civilian is no longer actively participating in hostilities. For specific, sporadic, unorganized and spontaneous acts, this refers to the moment when the civilian visibly and physically separates from an operation. In situations where it is difficult to determine the precise ending of an act – such as in the use of delayed explosive devices – assessing the chain of causation might provide a clearer indication of when the act is complete. What the Heck are Hostilities? : The Contextual Nature of the Act A final consideration that can be used to assess whether an act constitutes direct participation in hostilities is the context in which it is undertaken. First and foremost, as discussed under Part I above, the act must take place in an armed conflict setting. However, it is important to recall that not all acts undertaken during an armed conflict with constitute direct participation in hostilities. Hostilities are defined as “the sum total of all hostile acts carried out by individuals directly participating in hostilities”.75 The use of the term ‘hostilities’ in this definition, however, does little to provide clarity on the concept. It is likely that the term ‘hostilities’ was merely intended to be a synonym for armed conflict. Despite this shortcoming, very little has been written on the concept of hostilities. As such, it is unlikely that, in its current state of development, this contextual factor will be useful in the classification of acts as either direct or indirect forms of participation in hostilities. Conclusion This paper attempts to clarify the concept of direct participation in hostilities by establishing – in a preliminary manner – a classification system that can be used to assess whether a civilian has 75 Interpretive Guidance, supra note 3 at 44. 21 undertaken an act that should result in the loss of their protected status. In particular, this paper focuses on direct causation and the nature of an act. To do so, it has been assumed that all acts have satisfied many of the requirements for direct participation. Based on an analysis of existing classification systems, to constitute direct participation in hostilities the activity at issue must have an active component. This means that refusing to engage in an activity would not be a form of direct participation even if it unintentionally benefitted one party to the armed conflict to the detriment of another. Furthermore, violent or hostile acts that satisfy the belligerent nexus and threshold of harm requirements will always constitute direct participation. As well, all preparatory acts undertaken with a view to a specific operation will meet the requirements of direct participation. With regards to weapons-related and service or support provision activities, an in-depth analysis of the circumstances and facts specific to each situation should always be undertaken. To assist in the classification of an act, it may therefore be necessary to consider the temporal and contextual nature of that act. With regards to the former, examining the past behaviour of a civilian could provide support for the classification of an activity as direct participation in hostilities. Generally speaking, civilians who have previously engaged in the direct participation of hostilities can and may do so again. While further research is required into the contextual nature of an act and the concept of ‘hostilities’, the compilation of a list of activities or categories of activities that constitute direct participation may nevertheless be instructive to military practitioners. Such a list would, of course, require additional guidance for the classification of activities that sit in the grey area between direct and indirect participation. It would also require sufficient flexibility to allow for the realities and nuances of modern warfare. 22 Annex 1: Comparing and Classifying Civilian Acts Classifications of Act Description of Act Preparatory Acts Target designation, including • Identifying and marking of targets • Loading coordinates into a target navigation system Planning an attack Recruitment and training of personnel • For a specific pre-determined hostile act76 Intelligence gathering, including77 • Collecting tactical intelligence • Gathering information in an area of operations • Transmitting information about targets • Gathering strategic intelligence Deployment to an active firing position Attacks Against Civilians and Protected Objects Attacks against civilians and protected objects, including • Sniper attacks against civilians • Bombardment or shelling of civilian villages or urban residential areas “Hostile” Acts Observatory acts, including • Serving as a lookout during an ambush • Acting as a spotter for a combatant Killing or wounding military personnel Killing or taking prisoners Destroying military equipment and communication facilities X X X X X X X X X X X X X X 76 The Interpretive Guidance provides that the recruitment and training of personnel will generally be characterized as an indirect act unless the individuals are “specifically recruited and trained for the execution of a predetermined hostile act”. See supra note 3 at 53. 77 The literature has distinguished between the gathering of tactical intelligence and the gathering of strategic intelligence. The former is understood to have a causal link to a specific hostile act, whereas the latter does not. 23 Classifications of Act Description of Act “Hostile” Acts (Continued) Assisting troops for the execution of a specific military operation Voluntarily and deliberately creating a physical obstacle to military operations Blowing up a bridge carrying enemy war material Acts of sabotage78, including • Electronic interference with enemy military communications or computer networks79 Returning from the location of direct participation in hostilities Weapons-related Activities Manufacturing and assembly, including80 • Manufacturing a bomb cartridge Clearing mines placed by an adversary Installation of weapons, including • Planting a bomb, mine, booby-trap or IED • Loading explosives in a suicide vehicle • Loading a gun • Loading bombs onto an aircraft for a specific attack Detonation or use of a weapon, including • Detonation of a bomb, mine, booby-trap or IED • Aiming or firing a weapon at an opposing combatant Purchasing, smuggling and performing maintenance of weapons Designing, producing or shipping of weapons and military equipment81 X X X X X X X X X X X X X X X X 78 Omitting an act in order to prevent a military operation does not constitute either sabotage or direct participation in hostilities. 79 For further discussion of electronic interference and other forms of cyber warfare as a form of direct participation, see David Turns, “Cyber Warfare and the Notion of Direct Participation in Hostilities” (2012) 17:2 Journal of Conflict and Security Law 279 at 294-95. 80 The literature has distinguished between the manufacture of weapons for use in a specific operation and for use outside of a specific operation. If the weapon is to be used as part of a specific, known, operation, then manufacturing it constitutes direct participation in hostilities. If it is not, then it constitutes indirect participation. 24 Classifications of Act Description of Act Services and Support Provision Intelligence gathering and communication Banking Financing of activities Propaganda Some logistics activities82, including • Transport of military personnel • Installation of military command and control equipment in a forward operating base • Loading of defensive aids onto attack aircraft Provision of food or electricity to the armed forces Construction or repair of infrastructure Production of agricultural and non-military goods Delivering mail (including maps) to the military Showing sympathy for the cause, including • Taking part in civil disobedience X X X X X X X X X X X X X X X X X 81 According to the Interpretive Guidance, the act of producing or transporting weapons and equipment are indirect activities unless they are “carried out as an integral part of a specific military operation designed to directly cause” harm. Supra note 3 at 53. 82 See Boothby, “Discussion of the Interpretive Guidance”, supra note 4 at 158. Boothy suggests that acts which “positively impact” the fighting position of one party to a conflict “while not necessarily adversely resulting in [an] immediate loss to the opposing party” would be direct participation. 25