THE INTERNATIONAL HUMANITARIAN LAW NOTION OF DIRECT PARTICIPATION IN HOSTILITIES – A REVIEW OF THE ICRC INTERPRETIVE GUIDE AND SUBSEQUENT DEBATE

The phrase "direct participation in hostilities" has a very specific meaning in international humanitarian law (IHL). Those individuals who are clothed with combatant status are authorised to participate directly in hostilities without fear of prosecution, while civilians lose their civilian immunity against direct targeting whilst they participate directly in hostilities. Any civilian activity which amounts to "direct participation in hostilities" temporarily suspends their presumptive civilian protection and exposes them to both direct targeting as a legitimate military target and prosecution for their unauthorised participation in hostilities. Since existing treaty sources of IHL do not provide a definition of what activities amount to "direct participation in hostilities", the ICRC in 2009 released an Interpretive Guide on the Notion of Direct Participation in Hostilities - in the hope of providing a neutral, impartial and balanced interpretation of the longstanding IHL principle of direct participation in hostilities. While not without criticism, the Interpretive Guide aims to respect the customary IHL distinction between "direct participation in hostilities" and mere involvement in the general war effort. The Guide proposes a three-pronged test which establishes a threshold of harm, and requires direct causation together with a belligerent nexus. Collectively, these criteria limit overly-broad targeting policies, while distinguishing occasions of legitimate military targeting from common, criminal activities. Together with these three criteria, the Guide introduces the notion of the revolving door of protection, together with the concept of a "continuous combat function". Both these new concepts have been the subject of criticism, as too the idea that a presumption of non-participation status should apply in cases of doubt. Nevertheless "nothing indicates that the ICRC's interpretive guidance is substantively inaccurate, unbalanced, or otherwise inappropriate, or that its recommendations cannot be realistically translated into operational practice" in a way which will ensure that the fundamental principles of distinction and civilian immunity upon which all of IHL is built are observed.


Introduction
The phrase "direct participation in hostilities" 1 has a very exacting meaning in the realm of international humanitarian law (IHL), and refers generally to those activities normally undertaken by combatants. 2 As a general rule, all those with combatant status are authorised to participate directly in hostilities and are immune from prosecution for their participation. 3Civilians, on the other hand, enjoy immunity against direct attack precisely because they refrain from any such direct participation in hostilities. 4As civilians, they remain protected from any direct targeting for so long as they refrain from participating in combative activities which would otherwise compromise their protected status. 5 Any civilian activity which amounts to direct participation in hostilities temporarily suspends civilians' inherent immunity against direct targeting, and exposes them to As the ICRC Commentary on AP I art 51(3) explains: "... the immunity afforded individual civilians is subject to an overriding condition, namely, on their abstaining from all hostile acts ... thus a civilian who takes part in an armed combat, either individually or as part of a group, thereby becomes a legitimate target, though only for as long as he takes part in hostilities" (Jensen "Direct Participation in Hostilities" 1995-2003).
direct targeting as a legitimate military target, and to prosecution for their unauthorised participation in hostilities. 6is area has been the subject of much controversy, since: neither the Geneva Conventions nor their Additional Protocols provide a definition of what activities amount to "direct participation in hostilities". 7is lacuna in the law is of particular concern in the light of the realities of contemporary international armed conflict, where non-state actors (often dressed as civilians) are playing an increasing role, states are outsourcing military functions to private contractors, and civilians are increasingly active as "farmers by day, fighters by night". 8There is a dire need for a consensus understanding of exactly what constitutes direct participation in hostilities, especially when considering to what extent such activities might be deemed unlawful. 9It is precisely this need that the International Committee for the Red Cross (ICRC) sought to address in the drafting of the Interpretive Guide, which this piece seeks to review.

2
The treaty and customary international law notion of direct participation in hostilities

"Direct participation" in treaty law
Reference is made to the concept of direct participation in hostilities in many treaty provisions of IHL, including GC common article 3(1) 10 and AP I article 51(3). 11In the commentary on AP I article 51 it is explained that: 6 "Once he ceases to participate, the civilian regains his right to the protection under this section… and he may no longer be attacked" (Jensen " … direct participation means acts of war which by their nature or purpose are likely to cause actual harm to the personnel and equipment of the enemy armed forces. 12e commentary goes on to differentiate "direct participation" from general "war effort", which is often simply expected of the whole population, and adopts a more cautious interpretation of the notion of direct participation in hostilities. 13The commentary limits its understanding of "direct participation in hostilities" to: … acts of war which are intended by their nature and purpose to hit specifically the personnel and matériel of the armed forces of the adverse Party. 14ile IHL treaty law makes reference to this concept, the treaty law does not offer a definition of the phrase or specify definitively when an individual's actions might be said to amount to direct participation in hostilities. 15

"Direct participation" in customary IHL
At a national level, the principle that civilians lose their immunity against direct attack when they participate in hostilities is endorsed by several states' military manuals, 16 and is also endorsed by reported state practice, official statements and judicial decisions, 17 even by states that were not party to AP I. 18 According to the ICRC's study of the customary international law status of this provision, there is no evidence of contrary state practice, 19 and on the whole the principle (that civilians lose their immunity from prosecution when they participate in hostilities) was seen Additional Protocols, Mexico stated that AP I art 51 was so essential that it "cannot be the subject of any reservations whatsoever since these would be inconsistent with the aim and purpose of Protocol I and undermine its basis", and in the end there were no reservations made to this provision when states signed up to AP I (Henckaerts and Doswald-Beck Customary International Humanitarian Law 23).
as a "valuable reaffirmation of an existing rule of customary international law". 20For the most part states work on the assumption that assessing whether an activity amounts to direct participation in hostilities or not has to be done on a "case-by-case basis" 21 -although very few actually explain what activities amount to direct participation. 22 a regional level, the Inter-American Commission on Human Rights understands the term "direct participation in hostilities" to mean "acts which, by their nature or purpose, are intended to cause actual harm to enemy personnel and matériel". 23 evidenced by the ICRC's study into customary international law, "a precise definition of the term direct participation in hostilities does not exist" 24 in either state practice or international jurisprudence. 25What is clear, however, is that civilian "use of weapons or other means to commit acts of violence against human or material enemy forces", 26 amounts to direct participation in hostilities.Short of this very obvious occurrence, states are having to interpret: … the notion of direct participation in hostilities ... in good faith in accordance with the ordinary meaning to be given to its constituent terms in their context and in light of the object and purpose of IHL.Initially, the ICRC had sought a unanimous consensus at these expert meetings, but it soon became apparent (when some experts wanted to remove their names from the final report) that seeking unanimity might scuttle the whole project. 30In the end, the ICRC elected to omit all the names of the external experts, and instead had the Assembly of the ICRC adopt the final version of the guide on 26 February 2009. 31e ICRC's Interpretive Guide was not intended to change the existing binding treaty and customary based rules of IHL, but rather to offer a comprehensive guide as to how to interpret the term legally -giving careful consideration to balancing both military necessity and humanitarian concerns. 32The 10 recommendations (supported by commentary) strove to: … reflect the ICRC's institutional position as to how existing IHL should be interpreted in light of the circumstances prevailing in contemporary armed conflicts. 33 an interpretive guide the document is not legally binding -but coming from the ICRC (as the champion of IHL) the guide was intended to have persuasive influence, 34 and some suggest it "may even be viewed as a secondary source of international law ... analogous to writings of the most highly qualified publicists". 35til it becomes binding, or is acknowledged as having crystallised into customary IHL, Fenrick 36 warns that it is not likely that legal advisors to government department will be inclined to adopt it wholesale, unless it can be shown that these recommendations are "well researched, well thought out, relevant and persuasive".

The ICRC's Interpretive Guide's limitations and controversies
The Interpretive Guide explicitly limits its analysis of the notion of direct participation in hostilities to decisions around military targeting.It does not purport to deal with issues of how direct participation impacts on questions around detention, or how this impacts on a combatant's claim to immunity from prosecution. 37ce it is ascertained that an issue of direct participation has an impact on targeting decisions, the first enquiry that the Interpretive Guide directs is to whether or not the particular hostile act (often referred to as the specific hostile act) falls within the ambit of those restricted acts which amount to direct participation in hostilities. 38termining which specific activities amount to direct participation in hostilities is not dependent on one's "status, function, or affiliation", 39 neither does it matter whether the act is carried out by civilians or members of the armed forces: … on a spontaneous, sporadic, or unorganised basis; or as part of a continuous combat function assumed for an organised armed force or group belonging to a party to the conflict.Even prior to the first meeting of the experts, it was apparent that there were divergent opinions on how one should interpret the concept of direct participation in hostilities.Some academics favoured a more restricted interpretation, equating direct participation with actual combat. 41Others believed a more liberal interpretation was appropriate, 42 and favoured: … an approach … which essentially encompasses all conduct that functionally corresponds to that of government armed forces, including not only the actual conduct of hostilities, but also the activities such as planning, organising, recruiting and assuming logistical functions. 43ese competing approaches were not new to the ICRC.Already in the commentary on AP I, the ICRC has noted that: … to restrict this concept to combat and active military operations would be too narrow, while extending it to the entire war effort would be too broad. 44ven this background, it was not surprising that the guide has generated some heated academic debates. 45At the heart of much of the generalised criticism leveled at the Interpretive Guide is its alleged failure to adequately balance humanitarian concerns and military necessity in the manner intended by the Geneva Conventions and Additional Protocols. 46Schmitt and Boothby are critical of what they claim is an overly-restrictive interpretation. 47Boothby 48 argues that "the ICRC interprets the concepts of preparation, deployment, and return too restrictively", and Schmitt 49 is concerned with the fact that the definition "excludes support activities not directly causing harm to the enemy".On the contrary:  As Melzer notes, the liberal approach "stands in contradiction not only to the prevailing opinion in the doctrine, but also to state practice, and to the express distinction drawn in convention law between direct participation in hostilities on the one hand, and work of a military character, activities in support of military operations and an activity linked to the military effort, on the

4
The specific hostile acts which amount to direct participation in hostilities The concept of direct participation in hostilities is the means of determining when civilians' actions compromise their otherwise protected civilian immunity. 57According to the ICRC's Interpretive Guide, before an act amounts to direct participation in hostilities it must meet three cumulative criteria: (1) The 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 direct attack (threshold of harm); and (2) 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 (direct causation); and (3) The act must be specifically designed to directly cause the required threshold of harm in support of a party to the conflict and to the detriment of another (belligerent nexus). 58

The threshold of harm
The first criterion -also called the "threshold of harm" determination -requires that The degree of harm includes "not only the infliction of death, injury, or destruction on military personnel and objects, but essentially any consequence adversely affecting the military operations or military capacity of a party to the conflict" (ICRC 2009 http://www.icrc.org/web/eng/siteeng0.nsf/html/direct-participation-ihl-feature-02060997).
(b) harm ("by inflicting death, injury or destruction") 61 of a protected person or object, must be reasonably expected to result from a civilian's actions before the civilian can be said to be participating directly in hostilities. 62Or to put it another way, in order for civilians to lose their immunity from direct attack: … they must either harm the enemy's military operations or capacity, or they must use means and methods of warfare directly against protected persons or objects. 63l that is required is the likelihood that the act bring about this sort of harm, not that the harm necessarily materialise. 64Moreover, it is not the "quantum of harm caused the enemy" which determines whether or not it reaches the necessary threshold of harm criterion 65 -but rather the nature of the intended harm.I turn now to unpack these two categories of harm in more detail.

Military harm
As Melzer 66 points out, while military harm is commonplace in armed conflicts, the term applies only to objects which "contribute militarily" to the belligerent's success.
The term military harm cannot be used in respect of civilian objects, despite the fact 60 From a cursory examination of the criterion, it is apparent that the test is framed in the alternative, that is, "the harm contemplated may either adversely affect the enemy or harm protected persons or objects" ( 64 This is assessed objectively as "harm which may reasonably be expected to result from an act in the prevailing circumstances" (ICRC 2009 http://www.icrc.org/web/eng/siteeng0.nsf/html/directparticipation-ihl-feature-02060933, 47).As was discussed at the expert meeting, "wherever a civilian had a subjective 'intent' to cause harm that was objectively identifiable, there would also be an objective 'likelihood' that he or she would cause such harm" (Schmitt 2010 NYU J Int'l L & Pol 724).Schmitt concedes that this is a sensible requirement since it would be "absurd to suggest that a civilian shooting at a combatant, but missing, would not be directly participating because no harm resulted" (Schmitt 2010 NYU J Int'l L & Pol 724).65 Schmitt observes that perhaps the choice of the label "threshold", which is a quantitative concept, was "unfortunate", when the substance of the test talks to the "nature of the harm", the performance of a specified act, and not that the act reaches a "particular threshold" (Schmitt 2010 NYU J Int'l L & Pol 716).that these objects may contribute to the military success of a particular belligerent. 67is interpretation, in line with the universally accepted definition of what constitutes a military objective, excludes those political, economic and psychological contributions which might play a role in a military victory but ain isolation are not considered military objects. 68The term military harm includes: … not only the infliction of death, injury, or destruction on military personnel and objects, but essentially any consequence adversely affecting the military operations or military capacity of a party to the conflict. 69

Attacks against protected persons
In accordance with treaty law, a civilian's actions might amount to direct participation in hostilities when their actions constitute attacks specifically aimed at civilians and civilian objects -despite the fact that such actions might cause no specific military harm. 70However, acts which fall short of causing military harm are required at a minimum to "cause at least death, injury, or destruction of these civilians or civilian objects". 71Such acts are distinguished from harm resulting from "political, diplomatic, economic, or administrative measures like for example deportation". 72Where protected persons are the target of an attack, these actions The Interpretive Guide relies on AP I art 49's definition of "attack", which "does not specify the target, but the belligerent nexus of an attack, so that even acts of violence directed specifically against civilians or civilian objects may amount to direct participation in hostilities" (Schmitt 2010 NYU J Int'l L & Pol 723).Legal precedence for this position can be found in the jurisprudence emerging from the ICTY, where it was concluded that "sniping attacks against civilians and bombardment of civilian villages or urban residential areas" constitutes an "attack" in the IHL must not only rise to this minimum degree of harm, but they must also be part of the armed hostilities. 734.1.3Activities which have been cited as satisfying the threshold of harm requirement 74 These include "acts of violence against human and material enemy forces"; 75 sabotaging or causing "physical or functional damage to military objects, operations or capacity"; 76 hindering military "deployments, logistics and communications"; 77 controlling or hindering the use of "military personnel, objects and territory, to the detriment of the adversary"; 78 demining the opposition's mines; "guarding captured military personnel to prevent them being forcibly liberated"; 79 interfering with "military computer networks"; 80 "wiretapping the adversary's high command or transmitting tactical targeting information for an attack"; 81 directly targeting civilians or civilian objects; 82 "building defensive positions at a military base certain to be attacked"; 83 and "repairing a battle-damaged runway at a forward airfield so it can be used to launch aircraft". 8473 For example, a "prison guard may kill a prisoner for purely private reasons" without his actions amounting to direct participation in hostilities, but were he to engage in "a practice of killing prisoners of a particular ethnic group during an ethnic conflict", those actions would meet the

Activities which have been cited as falling short of the threshold of harm requirement
These include "building fences or roadblocks"; 85 disturbing "electricity, water, or food supplies"; 86 assuming control of cars and fuel; 87 manipulating computer networks; 88 arresting or deporting individuals who potentially have a "serious impact on public security, health, and commerce"; 89 declining appeals "to engage in actions that would positively affect one of the parties"; 90 the civilian rescuing of aircrew members; 91 and producing improvised explosive devices. 92

Critique of the threshold of harm requirement
The threshold of harm requirement has been criticised mainly for being underinclusive and unduly difficult to satisfy.Jensen 93 gives expression to this when he comments that the: … actual harm standard from the ICRC commentary is too restrictive in that it fails to address individuals who, although they are not members of an armed group that is party to the conflict, still openly support hostilities by constructing, financing, or storing weapons and materials of warfare.
He is in favour of an interpretation which would see some differentiation between those civilians found financing, storing or assembling weapons, and civilians who steer clear of any association with the hostilities. 94Jensen 95 would also support an interpretation of direct participation, which would: 85 … include not only those who cause actual harm, but those who directly support those who cause actual harm … this would also include those who gather intelligence, or act as observers and supply information to fighters, those who solicit others to participate in hostilities, and those who train them on military tactics.
In his critique Schmitt raises a similar concern: … strict application of the threshold of harm constitutive element would exclude conduct that by a reasonable assessment should amount to direct participation. 96ving said that, Schmitt 97 himself concedes that the treaty definition of a military objective favours limiting the notion of direct participation to refer only to harm of a military nature; in short, "an act of direct participation must impact the enemy's military wherewithal".Nevertheless, Schmitt 98 argues that the military harm requirement is "under-inclusive because it excludes loss of protection for support activities which do not adversely affect the enemy".In respect of attacks which target protected persons Schmitt 99 disputes the ICRC's interpretation, which requires death or destruction, because he argues that such an interpretation will exclude activities such as civilian deportation or civilian hostage taking.Instead, he suggests: … a better standard is one which includes any harmful acts directed against protected persons or objects, when said acts are either part of the armed conflict's war strategy … or when there is an evident relationship with ongoing hostilities, 100 even if such acts do not result in death or destruction.Schmitt101 argues that this strict requirement clearly favours humanitarian concerns over notions of military necessity.
Heaton102 is also critical of this strict interpretation for its failure to include within its ambit the "essential links in the chain immediately preceding that final step".Heaton 103 argues that the final act of the combatant is heavily reliant on the personnel who support and make combative actions possible.
In reply to his critics, Melzer 104 cautions against any suggestion of diluting the required threshold of harm.He fears that such a move would result in a wide range of support personnel losing their immunity against direct targeting, and would result in: … undermining the generally recognised distinction between direct participation in hostilities and mere involvement in the general war effort.

The direct causation requirement
The second requirement, also termed the direct causation test, was included as a response to the controversy traditionally surrounding questions about whether or not a "general war effort" 105 and activities aimed at sustaining war 106 would amount to direct participation in hostilities.While it is certainly true that war-sustaining activities are indispensable to the war effort, which in effect harms the adversary, a line must be drawn between the two degrees of involvement. 107All the experts present at the ICRC's expert meetings were: … agreed on the centrality of a relatively close relationship between the act in question and the consequences affecting the ongoing hostilities.During the expert meetings emphasis was placed on the "idea that direct participation in hostilities is neither synonymous with 'involvement in' or 'contribution to' hostilities, nor with 'preparing' or 'enabling' someone else to directly participate in hostilities, but essentially means that an individual is personally 'taking part in the ongoing exercise of harming the enemy' and personally carrying out hostile acts which are 'part of' the hostilities" (ICRC 2009 http://www.icrc.org/web/eng/siteeng0.nsf/html/direct-participation-ihl-feature-02060953).
… sometimes causation is so direct that the shield of humanitarian considerations must yield in the face of military necessity, while in other situations the causal connection is too weak (or indirect) to overcome humanitarian factors.
As a result, and so as to prevent depriving the civilian population of their protected status, there must be a close causal link between the hostile act and the resulting harm before the action can be said to qualify as direct participation in hostilities. 110cording to the ICRC's Interpretive Guide, this notion of direct causation is understood as meaning that there is only one causal step between the hostile act and the resulting harm. 111This notion of direct causation excludes activities that indirectly cause harm. 112Similarly mere "temporal or geographic proximity" 113 is insufficient to justify a finding of direct participation.Moreover, in cases of collective operations, the ICRC's Interpretive Guide does recognise that: … the resulting harm does not have to be directly caused by each contributing person individually, but only by the collective operation as a whole. 114 short, where a particular activity does not result in the required degree of harm, those individual actions might nevertheless constitute direct participation in hostilities where the actors are "part of a collective operation" 115 that directly causes harm of the required threshold.These include "bearing, using or taking up arms"; 116 "taking part in military or hostile acts, activities, conduct or operations"; 117 "participating in attacks against enemy personnel, property or equipment"; 118 "coordinated tactical operations which directly cause harm"; 119 engaging in sabotage of military installations; 120 manning an antiaircraft gun; 121 supervising the operation of weaponry; 122 "gathering tactical intelligence on the battlefield"; 123 transmitting military information for immediate use; 124 "identifying and marking targets"; 125 "instruction and assistance given to troops for the execution of a specific military operation"; 126 transporting weapons in proximity to combat operations; 127 "transporting unlawful combatants to or from the place where the hostilities are taking place"; 128 "serving as guards, intelligence agents, lookouts, or observers on behalf of military forces"; 129 "capturing combatants or their equipment"; 130 "sabotaging lines of communication"; 131 "performing missionessential work at a military base"; 132 "providing logistical support"; 133 and "delivering ammunition to combatants". 1344.2.2Activities which have been cited as falling short of the direct causation requirement 116 These include "designing, producing and shipping weapons"; 135 "transporting arms and munitions"; 136 purchasing materials in order "to build suicide vests"; 137 purchasing, smuggling, assembling or storing "improvised explosive devices"; 138 gathering and transmitting military information; 139 work undertaken by civilians in military vehicle maintenance depots; 140 work undertaken by civilians in munitions factories; 141 "driving military transport vehicles" where the driver is a civilian; 142 "activities in support of the war or military effort"; 143 "the recruitment and general training of personnel"; 144 "providing specialist advice regarding the selection of military personnel, their training, or the correct maintenance of the weapons"; 145 "general strategic analysis"; 146 "voluntary human shielding"; 147 "expressing sympathy for the cause of one of the parties to the conflict"; 148 distributing war propaganda; 149 "failing to act to prevent an incursion by one of the parties to the conflict"; 150 "contributing funds to a cause"; 151 "economic sanctions"; 152 and "providing an adversary with supplies (for example food and medicine) and services".authors of the interpretive guide settled on direct causation being linked to a physical act causing harm, when in modern warfare "acts that directly enhance the military capacity or operations of a party, without resulting in direct and immediate harm to the enemy" 154 may have a marked effect on the belligerent's capacity to win.
Schmitt 155 argues that: … the key is whether the acts in question are sufficiently causally related to the resulting harm/benefit to qualify as directly caused.
Moreover, Schmitt 156 argues, the effect of the one causal step requirement is that a range of activities aimed at building capacity (which Schmitt concedes are indirect in nature) are excluded from those parameters.Schmitt 157 prefers the "integral part test" which makes it possible to "extend participation as far up and downstream as there is a causal link".In a similar vein, Watkin 158 argues that the role played by logistics in a military sense has not been adequately recognized in the direct causation requirement.Watkin 159 warns that the causal-chain requirement limits responses to a "reactive posture focused on acts rather than on the capacity of an opponent to plan and attack in the future".
Van der Toorn 160 raises a related criticism when he suggests a sound interpretation of direct participation in hostilities should extend beyond the specific activities to "include precursor operational activities that facilitate and are closely connected with the materialisation of harm". 161 the ICRC's interpretation stands at the moment, participation is understood as "single, discrete acts" 162 which in effect allow civilians to interrupt their hostilities with numerous periods during which they continue with their civilian lifestyles.Van der Toorn 163 argues that the ICRC's direct causation requirement needs to balance the needs of military necessity and humanitarian concerns more effectively.In this regard, his proposal would: … permit the targeting of the precursor operational activities that make possible the ultimate infliction of harm. 164lzer 165 warns that Schmitt's "integrated part test" would translate into an unnecessarily permissive understanding of direct causation, and that in effect: … any act connected with the resulting harm through a causal link would automatically qualify as direct participation in hostilities, no matter how far removed the act may be from the final harm caused.
Melzer 166 warns that any such relaxation of the direct causation test would result in "excessively broad targeting policies, prone to error, arbitrariness, and abuse".
According to Melzer, 167 there is no indication on the part of states by way of general opinio juris that would favour Schmitt's integrated part interpretation over the ICRC's direct causation requirement.

The belligerent nexus requirement
According to the ICRC's Interpretive Guide, the final requirement is that the specific harm must have a link to the hostilities. 168The belligerent nexus requirement is there to ensure that those criminal activities which are simply facilitated by the hostilities and not intended to specifically support one party while causing the requisite threshold of harm to the opposing party are excluded from the purview of direct participation in hostilities. 169As Rogers 170 points out, in "the case of children throwing petrol bombs or stones at enemy military patrols", members of the patrol will have to assess carefully whether their actions are just common criminal activities or whether the children have forfeited their inherent civilian immunity from direct targeting through these actions.
In short, this leg of the test requires that: … an act must be specifically designed to directly cause the required threshold of harm, in support of a party to the conflict and to the detriment of another. 171 other words hostile actions (for example looting or other civil unrest) which are not intended to harm a specific party to the conflict while supporting the opposing party do not amount to direct participation in hostilities. 172, for example, if civilians are found causing harm: (c) in individual self-defence or defence of others; 173 (d) in exercising power or authority over persons or territory; (e) as part of civil unrest against such authority; or (f) during inter-civilian violence, their acts will not be regarded as participating in hostilities, since these acts lack the requisite belligerent nexus. 174Moreover: … when civilians are totally unaware of the role they are playing in the conduct of hostilities… or when they are completely deprived of their physical freedom of action... they cannot be regarded as performing an action in any meaningful sense and, therefore, remain protected against direct attack despite the belligerent nexus of the military operation in which they are being instrumentalised.175

Activities cited as satisfying the belligerent nexus requirement
These include the "preparatory collection of tactical intelligence"; 176 "loading explosives in a suicide vehicle"; 177 "transporting personnel"; 178 and "positioning weapons and equipment". 179

Activities cited as falling short of the belligerent nexus requirement
These include "hiding or smuggling weapons"; 180 and "financial or political support of armed individuals". 181

Critique of the belligerent nexus requirement
While Schmitt 182 supports the notion that there must be a link to the hostilities, he favours the belligerent nexus test's being formulated in the alternative: "in support of a party to the conflict or to the detriment of another".Melzer 183 warns that a disjunctive interpretation of these two aspects of the belligerent nexus test would permit a belligerent to respond with military force against criminal elements who had no connection to the armed conflict.IHL does not permit a person to be categorised as a military target until it can be shown that the person has some link to a belligerent party.Instead, as IHL stands at present, such instances will be dealt with as any other threat to security. 184

General comments regarding specific hostile acts which amount to direct participation in hostilities
Schmitt 185 concedes that all three legs of the test for direct participation in hostilities represent legitimate factors which have a valid role to play in assessing when civilians' actions compromise their immunity against direct targeting.However, the thrust of his criticism is that there are insufficiencies to be found in each of the elements, which give rise to what he considers to be an "under-inclusive" 186 notion of direct participation in hostilities.Schmitt's 187 concern is that this pro-humanitarian treatment of the concept of direct participation reveals an ignorance of the realities of the modern battlefield.Rogers 188 is of the opposite view and supports a narrow interpretation of direct participation, which will not risk jeopardising the IHL principle of distinction and civilian immunity from direct targeting.To this end Melzer 189 notes that there were several safeguards built into the three constitutive elements to ensure that the test would not permit the arbitrary or erroneous targeting of civilians.Despite their criticism, many like Schmitt 190 concede that the Interpretive Guide is: … superior to the various ad hoc lists because it provides those tasked with applying the norm on the battlefield with guidelines against which to gauge an action.

The temporal element of loss of protection "for such time as" civilians take a direct part in hostilities
In terms of IHL, civilians normally enjoy complete immunity against attack for such time as they refrain from any direct participation in hostilities.However, as soon as civilians compromise their civilian immunity by electing to participate directly in hostilities, their actions expose other truly innocent civilians to "erroneous or arbitrary attack". 191Consequently, in order to dissuade civilians from abusing their civilian immunity, IHL condones the temporary suspension of their civilian immunity against direct targeting, for so long as they participate directly in hostilities. 192pressed another way: … considerations of military necessity are presumed to override those of humanity for such time as a civilian "directly participates in hostilities". 193ile their civilian immunity is temporarily suspended, this has no effect on their primary IHL status as civilians. 194At no time do they lose their civilian status and assume primary combatant status. 195Moreover, when they cease their participation, they resume full civilian immunity against attack.This temporary suspension of a civilian's immunity against direct attack is afforded only "civilians who participate in hostilities in a spontaneous, unorganised or sporadic basis". 196Consequently, once it has been determined that a civilian has carried out a specific act which amounts to direct participation in hostilities, the next level of enquiry must address determining the beginning and end of the loss of civilian immunity. 197e notion that direct participation has a temporal limitation has a longstanding history in IHL, having been in existence since the mid-nineteenth century. 198The phrase "for such time as", as it appears in AP I, "is binding as a matter of treaty law 191 on ... approximately eighty-five per cent of the world's states". 199Not surprisingly, the ICRC's study into the customary international law status of the phrase "and for such time as" concluded that it was widely recognised as constituting customary international law. 200ile the "for such time as" criterion might reflect customary international law, its practical implementation has not been without controversy.For the most part, the controversy lies in that fact that when civilians are no longer engaged in direct participation, and consequently no longer pose a threat to the opposition, they regain their full civilian immunity from direct attack, giving rise to what is called the "revolving door" of civilian protection. 201The terminology "revolving door" -whereby civilians might vacillate between being a belligerent and reserving the right to reclaim their civilian status -was first coined by Hays Parks 202 in his 1990s commentary on the practical effect of AP I.

The parameters of the "for such time" window: execution, preparation, deployment and withdrawal
The Interpretive Guide expressly recognises that the concept of direct participation in hostilities includes not only the obvious individual armed activities but also those unarmed activities which have an adverse effect on the opposing belligerent. 203turally the execution of a specific hostile act, which amounts to direct participation in hostilities, will fall within the "for such time" window and amount to a temporary loss of immunity from attack. 204Furthermore, given the fact that contemporary military activities are often the result of several collective actions, an interpretation of direct participation in hostilities must include those activities which cause harm only "in conjunction with other acts". 205Consequently, the ICRC's Interpretive Guide includes "measures preparatory to the execution of a specific act ... as well as the deployment to and the return from the location of its execution", 206 as comprising an "integral part of the specific hostile act". 207The ICRC Guide cites as examples: … equipping, instructing, and transporting personnel; gathering intelligence; and preparing, transporting and positioning weapons and equipment, 208 if these are carried out as preparation for the undertaking of a specific hostile act.
These preparations for a specific hostile act are to be distinguished from preparatory activities which merely establish a generalised capability to perpetrate hostile actions, which do not amount to direct participation in hostilities. 209Preparations which are part of a generalised "campaign of unspecified operations" 210 or are merely capacity building do not fall within the scope of the activities for which civilian immunity can be forfeited.Examples of such general preparations include: … the purchase, smuggling, production, and hiding of weapons; recruitment and training of personnel; and financial, political, and administrative support to armed actors. 211ere the particular hostile act does not necessitate any prior deployment, the loss of civilian immunity is limited to the integral preparations and the actual execution of the hostile act. 212On the other hand, where the specific hostile act necessitates preparatory deployment to a particular location, those preparatory deployment activities will also form an integral part of the hostile act -and result in the loss of civilian immunity. 213For an activity to amount to a deployment which will compromise a civilian's immunity, a deploying individual must be seen to be relocating in a physical manner with the intention of carrying out the specific hostile 206 act. 214Similarly, in instances where withdrawal activities remain integral to the hostile action, such withdrawal remains a part of the "for such time window", 215 and full civilian immunity is fully restored only once it is evident that the civilian has physically distanced himself from the hostile operation. 216Civilians can demonstrate their complete disengagement "by laying down, storing or hiding the weapons or other equipment used, and resuming activities distinct from that operation". 217

A critique of the "revolving door" concept
The most commonly cited criticism leveled at the "revolving door" phenomenon is that it gives rise to a profound operational advantage for those civilians who vacillate between hostile acts and peaceful civilian existence. 218It is suggested that this revolving door creates an uneven playing field, in that regular combatants (be they cooks or infantry) are at all times potential targets for attack, while civilians abusing this revolving door can claim full immunity from direct targeting and yet launch "spontaneous, unorganised or sporadic" 219 attacks from behind these protected positions. 220In this regard Watkin 221 agrees that it is difficult to rationalise the tactical advantage which the revolving door affords the civilian, particularly since there appears to be no limitation 222 on the number of times the civilian can walk through the door.Some even suggest that "a civilian can go through the revolving door on a persistently recurring basis". 223

214
Moreover, there is always potential for this interpretation of the "revolving door of protection" to be abused by non-state actors. 224me writers, like Boothby, 225 argue that the notion of a revolving door is not found in customary IHL, and that the way the Guide has interpreted the treaty reference to "participation" excludes the possibility that a civilian might be classified as a "persistent civilian participator". 226Boothby 227 argues that there must be a way to distinguish between genuinely sporadic acts and "repeated or persistent of direct participation in hostilities.Boothby 228 proposes that the time dimension to the rule: … must permit the targeting of those who, whether voluntarily or otherwise, choose to participate on a persistent or regular basis in the conflict, whether they are or are not members of organised armed groups.
Boothby 229 argues that states (like Israel and the US) are unlikely to adopt an interpretation which would afford the benefit of the revolving door to those civiliains who engage in regular (albeit unorganised) direct participation by a civilian, when he suggests they should lose their protected status "while such persistent or repeated involvement in hostilities continues".
The Interpretive Guide warns that it would be too difficult to reliably ascertain whether a civilian had previously carried out a hostile act in a manner which might be said to be "persistently recurring" and whether they might be said to have the requisite intent to do so again. 230reover, as Melzer 231 points out, Boothby fails to provide a reliable tool to assess these notions of "sporadic and repeated" hostile acts in practice.
Furthermore, as Jensen 232 asserts: … any extension of the concept of direct participation in hostilities beyond specific acts would blur the distinction made in IHL 233 between temporary activity-based loss of protection (due to direct participation in hostilities) and continuous, statusor function-based loss of protection (due to combatant status or continuous combat function).
While it does seem bizarre to afford civilians immunity from direct targeting when they have already participated directly in hostilities and show intent to do so again, 234 Melzer 235 maintains that this scenario is unlikely to pose a major problem in reality.In the rare instances where it does occur, 236 these civilians can be dealt with through normal legal enforcement channels during the intermissions between hostile acts. 237other aspect of the temporal approach adopted by the ICRC's Interpretive Guide which has come under criticism is its interpretation of which preparatory, deployment or withdrawal activities amount to unprotected direct participation in hostilities.Jensen 238 proposed that: … a modern view of "for such time" must include the full time that an individual is directly participating, not just the time that results in actual harm.
Consequently, Jensen 239 is of the view that even those instructing or training those intending to take part in hostile acts can lose their immunity from direct targeting.
Boothby 240 is also critical of what he describes as a "restrictive" interpretation of the preparatory activities that amount to direct participation in hostilities.According to Boothby: 241 … "participate" could also refer to individual involvement in "groups or sequences of activity spread over a period," with the effect that the civilian in question would lose protection for the entire period of his involvement, including in the intervals between specific hostile acts.
Boothby 242 favours an interpretation which regards any preparatory acts (including any necessary prior deployment) as constituting direct participation.
The ICRC's Interpretive Guide acknowledges that the net effect of the "revolving door" phenomenon will limit attacks on civilian participants. 243It justifies the revolving door position as being a necessary safeguard rather than a "malfunction" 244 of the Guide aimed at preventing civilians being targeted arbitrarily or in error when they do not constitute a military objective. 245Any interpretation which has the effect of increasing the degree of risk to innocent civilians is squarely at odds with any reading of the treaty provisions read in the light of their object and purpose. 246

6
The "continuous combat function" and its implications for civilians participating directly in hostilities 6.1 The rationale behind the concept of the "continuous combat function" The term "continuous combat function" was first coined at the expert discussions which gave rise to the ICRC's Interpretive Guide. 247During the discussions the view was expressed that, since the revolving door of protection was intended to apply only to those spontaneous and unorganised acts of participation, it should not also be applied to members of organised non-state armed groups, since their activities were oftentimes neither unorganised nor spontaneous. 248It was felt that organised armed groups (who fell short of the requirements for full combatant status) would benefit unfairly from a considerable tactical advantage if the revolving door phenomenon were extended to them -while the state's armed force would be vulnerable to potential attacks on a continuous basis. 249While there is, under IHL, no express provision authorising civilians to participate directly in hostilities, that fact does not necessarily translate into "an international prohibition (or criminalisation) of such participation". 250Nevertheless, at the expert meetings the concern was raised that such inequality between the states' armed forces and organised non-state armed groups would not only undermine any respect for IHL but as a consequence would further endanger innocent civilian lives. 251Rogers 252 agrees that: … there is certainly a case for arguing that a person who becomes a member of a guerrilla group, or armed faction that is involved in attacks against enemy armed forces, forfeits his protected status for so long as he participates in the activities of the group.
As a consequence of these concerns, the general consensus at the expert discussion was that there was a legitimate and defensible 253 need for a special legal regime applicable to organised armed groups who participated in hostilities in a more organised, structured and continuous manner, as compared with those civilians who participated only intermittently in hostilities, and who benefitted from the revolving door of civilian immunity.Rather than apply the revolving door of protection, which limits their protection from attack only for such time as they participate directly in hostilities, as is the case with civilians, this group of participants lose their civilian protection for the duration of their membership of the organised group -by virtue of their "continuous combat function". 254In other words, the "revolving door of protection starts to operate based on membership" 255 in the organised group, and the door revolves again, rendering the individual once again a protected civilian, but only once his or her membership of the group has ceased.The net effect of this regime for non-state actors, who like child soldiers and PMSCs are affiliated with organised armed groups, is that they stand to forfeit their immunity from direct targeting not only during their continuous combative acts but "even when they put down their weapons and walk home for lunch with their family". 256ile this approach does draw on notions of group membership, it is nevertheless different from the regime applicable to those who are members of the regular armed forces.For members of the state's armed forces, their status as combatants is determined by their formal membership of the armed group, regardless of the function the individual might perform, and until the individual leaves the force. 257As Melzer 258 points out, any legal regime aimed at organised armed groups needs to take into consideration the: … more informal and fluctuating membership structures of irregularly constituted armed forces fighting on behalf of state and non-state belligerents.

Activating the loss of protection based upon a "continuous combat function"
The effect of this regime is that, once it is de facto evidenced that individual members of the organised armed group have functioned in a continuous combative matter, their membership results in their loss of civilian immunity against direct targeting for the duration of their membership. 259ccording to the ICRC Interpretive Guide: … membership in an organised armed group begins in the moment a civilian starts de facto to assume a continuous combat function for the group, and lasts until he or she ceases to assume such function. 260ch an assessment requires proof of repeated direct participation in hostilities, along with a degree of integration into an armed group, with indications that: … such conduct constitutes a continuous function rather than a spontaneous, sporadic, or temporary role assumed for the duration of a particular operation. 261ere is no assessment based upon the donning of a uniform or the possession of an identification card -it is determined solely by function. 262

Exclusion and cessation of the "continuous combat function" classification
At their core, these members of organised armed groups still enjoy primary civilian status (ie they do not acquire combatant status).As Melzer 263 points out, "continuous combat function does not, of course, imply de jure entitlement to combatant privilege".Consequently, it is imperative that only those members of the group who actually engage in the continuous combat function stand to lose their otherwise civilian immunity from attack. 264Those who, while affiliated with an organised armed group, fail to participate directly in the hostilities, cannot be said to perform a continuous combat function, and consequently are excluded from the loss of protection on account of their failure to directly participate in hostilities. 265Moreover: 260 … once a member has affirmatively disengaged from a particular group, or has permanently changed from its military to its political wing, he regains his civilian immunity against attack. 266e Interpretive Guide suggests that disassociation from the group "need not be openly declared", 267 but it might be manifested: … through conclusive behaviour, such as a lasting physical distancing from the group and reintegration into civilian life, or the permanent resumption of an exclusively non-combat function. 268nsequently, the Guide favours a reasonable assessment as to whether an individual has disengaged from an organised armed group to be carried out in good faith on the basis of the presumption of civilian status in instances where doubt prevails. 269

A critique of the continuous combat function
The ICRC's "continuous combat function" has not been without criticism.In particular, some academics have raised concerns around the issue that the specific treaty language, which the Interpretive Guide was attempting to interpret, states that civilians lose their immunity from attack for such time as they participate directly in hostilities. 270The ICRC's interpretation of the continuous combat function effectively arrives at a conclusion which makes it permissible to directly target civilians at all times, provided they are engaged in a continuous combat function. 271e potential increased risk to civilians posed by the creation of the continuous combat function category has seen critics of the concept call for the "the other constituent parts of the guidance (ie the threshold of harm, direct causation and belligerent nexus criteria) not [to] be diluted" 272 so as to adequately protect civilians in times of armed conflict.Melzer -and others who defend the proposed continuous combat function category -cite principle XI 273 in the ICRC's Interpretive Guide as providing the necessary counterbalance to prevent the continuous combat function category posing an increased risk to civilians, around whom there might be some doubt as to their degree of involvement in hostilities (ie as a sporadic direct participant or having a continuous combat function).
Watkin is critical that the continuous combat function approach still gives rise to a tactical disadvantage for the state's armed forces, in that the regularly-constituted armed forces can target only those within the organised armed group who exhibit a continuous combat function, while their own non-combatant members can be targeted at all times. 274Watkin is skeptical that, at a split second's notice, a soldier can: … realistically be expected to distinguish between a civilian who participates on a "persistent recurring basis", and a member of an organised armed group who performs a "continuous combat function". 275nrick concurs.In essence he argues that a smaller proportion of members of nonstate organised armed groups will find themselves liable to direct targeting, as compared to their counterparts employed by the state. 276273 "In addition to the restraints imposed by international humanitarian law on specific means and methods of warfare, and without prejudice to further restrictions that may arise under other applicable branches of international law, the kind and degree of force which is permissible against persons not entitled to protection against direct attack must not exceed what is actually necessary to accomplish a legitimate military purpose in the prevailing circumstances".Furthermore, protected immunity against attack is afforded to persons: … who are an integral part of the combat effectiveness of an organised armed group when their regular force counterparts performing exactly the same function can be targeted. 277 response to these criticisms, Melzer 278 points out that this perceived bias is not a fiction developed by the Interpretive Guide, but has its foundations in both treaty and customary IHL, which prohibits the direct targeting of civilians until such time as they participate directly in the hostilities.Melzer 279 concedes that, while notionally more of the regular armed forces might be exposed to direct targeting than the members of their non-state counterparts, … the actual practical effect will have very little consequence, since in organised armed groups many of the so-called non-combative roles are performed by the very individuals who engage in the continuous combat function.280 Moreover, as Melzer 281 points out: … almost all non-combatant members of regular armed forces, with the exception of medical and religious personnel ... are not only entitled, but also trained, armed, and expected to directly participate in hostilities in case of enemy contact and, therefore, also assume a continuous combat function. 282other criticism raised by Watkin 283 is that a restrictive interpretation of which activities amount to a combat function is at odds with interpretations adopted in legal writings.Watkin 284 argues that the criteria for attaining membership in an organised armed group is couched so restrictively as to make the potential unlikely that an otherwise civilian may lose that status and thus be targetable.Watkin 285 prefers to apply the continuous loss of civilian immunity from attack: … not only to fighting personnel of organised armed groups, but essentially to any person who could be regarded as performing a "combat", "combat support", or even "combat service support" function for such a group, including unarmed cooks and administrative personnel.
Van der Toorn 286 shares a similar concern, that the continuous participation requirement "imposes a very high threshold and would likely exclude a large number of individuals" who, for all intents and purposes, are "carrying out substantial and continuing integrated support functions for such groups", but "who fight for the group on a regular but not continuous basis". 287Van der Toorn 288 suggests relaxing the strict continuous combat function requirement in favour regular participation, or to require an individual's "primary function" to be direct participation in hostilities for the group.
Watkin 289 also suggests a similar formulation which would state that: … after the first involvement, any subsequent act demonstrating direct participation would start to provide the basis to believe that there is the beginning of a pattern of conduct that reflects an intention to regularly engage in the hostilities.
So, for Watkin, 290 repeated acts of participation can factor into one's determination of whether an individual's actions amount to continuous combat or not, and when such repetition has taken place "affirmative disengagement would be required in order to establish that such persons are no longer direct participants in hostilities".
These criticisms certainly do give expression to legitimate concerns surrounding the ability of the state's armed forces to deal with an enemy which flouts the principle of distinction and whose actions at time might be perfidious and even in violation of

IHL. 291
That being said, any interpretation which gives rise to overly permissive direct targeting of civilians will result in an unjustifiable level of arbitrary targeting of civilians. 292In response to this critique, Melzer 293 cautions that what Watkin and Van der Toorn refer to as combat support activities would in any event be considered integral to the hostile act, and consequently would result in the loss of immunity from attack.As Melzer 294 explains, any civilian who participates directly in hostilities on a regular basis will very likely have some affiliation to an organised armed group and "thus, may be regarded as a de facto member assuming a continuous combat function for that force or group".According to Melzer 295 this would apply to all participants from the armed full-time combatant to the private security contractors who are employed to defend military objectives, and also to the "farmer by day and fighter by night" who plants "IEDs, mines, or booby-traps, or provid[es] tactical intelligence or logistic support as part of specific attacks or combat operations".
Moreover, Melzer 296 argues that to adopt an over reaching notion of who could be targeted in an organised armed group to include any civilians who are merely accompanying or supporting an armed group (regardless of their specific function) would not respect the distinction between "direct" and "indirect" participation in hostilities, which is well established in both treaty and customary IHL.

Presumptions in assessing direct participation in hostilities
IHL operates on the presumption that in cases of doubt an individual will always be afforded the presumption of civilian status and is to enjoy immunity against direct targeting. 297It is a recognised principle of customary IHL that in instances of doubt, any targeting assessment must be able to show "sufficient indications to warrant an attack". 298Consequently, prior to and during any attack all reasonable precautions should be taken to ensure that the intended target of a planned attack is in fact a legitimate target.It must constitute a military objective and not be entitled to civilian immunity against direct targeting. 299 The rationale behind the principle of distinction and this legal presumption is to prevent civilians being targeted in error. 300The same rationale would make the presumption applicable in instances when an assessment needs to be made as to whether or not an individual has directly participated in hostilities.In the words of the ICRC's Interpretive Guide: ... in case of doubt as to whether a specific civilian conduct qualifies as direct participation in hostilities, it must be presumed that the general rule of civilian protection applies and that this conduct does not. 301

Critique of the presumption's application to assessments of direct participation
Schmitt 302 rejects the ICRC's application of the presumption of civilian status to assessments of direct participation in favour of a presumption in favour of a finding of direct participation.Schmitt 303 argues that once a determination is made that civilians are directly participating in hostilities, they may be legally targeted without further need to justify any resultant injury or death by considerations of proportionality or by taking special precautions in attack. 304Schmitt 305 defends what he concedes may seem like a "counter-intuitive" approach on the grounds that it: … is likely to enhance the protection of the civilian population as a whole, because it creates an incentive for civilians to remain as distant from the conflict as possible.
Melzer 306 cautions that instructing the armed forces that they are justified in directly targeting civilians whose actions are questionable is clearly contrary to the ethos of IHL and in violation of many of its fundamental provisions.As Melzer 307 points out, it is not surprising, given the radical approach that Schmitt proposes, that there is no state practice or legal jurisprudence which supports this reverse presumption of civilian status.I prefer a more nuanced conclusion: that the proportionality and special precautions test be easier to satisfy when doubts are raised regarding the degree of a civilian's involvement in hostilities.In other words, there is still an obligation to assess the proportionate result of the impending attack, as well as to take special precautions during the attack. 308The threshold for justifying these actions is easier to achieve when civilians are playing an active role in the hostilities.
As Melzer 309 correctly points out, if a civilian's actions threaten public law and order while not rising to the level of direct participation in hostilities, then these actions are to be dealt with in terms of civil law enforcement or the regime appropriate to self defence.

8
The legal consequences for civilians found participating directly in hostilities It is also worth noting that at all times -even whilst participating directly in hostilities -civilians retain their primary civilian status.Their actions alone do not reclassify them as combatants.They are, however, exposed to direct attack for so long as they persist with their direct participation in hostilities, despite their primary civilian status.While they lose their civilian immunity against direct attack, they never lose their inherently civilian status.Once they desist from their direct participation or disengage from the group's continuous combat function, they regain their full civilian immunity against direct attack.Civilians, by definition, do not enjoy combatant status, with its attendant authorisation to participate directly in hostilities, associated POW status, and immunity from prosecution.Consequently, when civilians are found to be participating directly in hostilities without the requisite combatant privileges, they are exposed to the potential of criminal prosecution, even if during their participation they observed the laws of war regarding the means and methods of warfare. 310What is particularly problematic for civilians taking a direct part in hostilities or acting with a continuous combat function is that they very often ambush an adversary whilst failing to adequately distinguish themselves from the civilian population and feigning the right to civilian immunity against direct targeting. 311This is considered a serious violation of the IHL prohibition against perfidy. 312

Conclusion
At present the ICRC's Interpretive Guide appears to provide a neutral, impartial and balanced interpretation of the longstanding IHL principle against civilian direct participation in hostilities.In setting a minimum threshold of harm, the Interpretive Guide respects the customary IHL distinction between mere general war effort and true direct participation in hostilities. 313In applying the direct causation requirement, the Interpretive Guide attempts to limit targeting decisions which may be overly broad, arbitrary and simply incorrect. 314The belligerent nexus link distinguishes occasions of legitimate military targeting from common criminal activities.
As for the temporal scope of the loss of civilian immunity from attack, the revolving door phenomenon ensures maximum protection for the civilian population against arbitrary targeting decisions 315 -in line with the fundamental principles of IHL.The concept of a "continuous combat function" distinguishes those "farmers by day and 310 The drafting process and the legal implications of the GuideBetween 2003 and 2008, more than 40 legal experts (drawn from NGOs, academia and governmental bodies, including the military) came together on five occasions at the invitation of the ICRC.28The resultant discussions informed the ICRC's Interpretive Guide on the Notion of Direct Participation in Hostilities under IHL.29

96
Schmitt 2010 NYU J Int'l L & Pol 714.97 Melzer 2010 NYU J Int'l L & Pol 859.98 Melzer 2010 NYU J Int'l L & Pol 859; 861.99 Schmitt 2010 NYU J Int'l L & Pol 723. 100 Melzer 2010 NYU J Int'l L & Pol 861. 107 4.2.1 Activities which have been cited as satisfying the direct causation requirement 110 Schmitt 2010 NYU J Int'l L & Pol 726.111 In short, where an "individual's conduct ... merely builds up or maintains the capacity of a party to harm its adversary, or which otherwise only indirectly causes harm, these actions do not amount /www.icrc.org/web/eng/siteeng0.nsf/html/direct-participation-ihl-feature-02060953, 55; Melzer 2010 NYU J Int'l L & Pol 866). 112 153 4.2.3Critique of the direct causation requirementSchmitt raises a number of technical issues in respect of the ICRC's explanation relating to the direct causation requirement.His first critique questions why the 135 Fenrick 2009 YIHL 293.136 Watkin 2010 NYU J Int'l L & Pol 707.While the "act of driving a munitions truck might not amount to direct participation in hostilities ... the truck itself remains a targetable military objective" (Schmitt 2010 NYU J Int'l L & Pol 710).