Armed escorts to humanitarian convoys: An unexplored framework under international humanitarian law
The use of armed escorts to humanitarian convoys delivering humanitarian assistance potentially increases the targeting of these convoys, yet so far this use has not been examined from the perspective of international humanitarian law (IHL). This article attempts to determine whether the resort to armed escorts is in line with the principle of passive precautions under IHL, how the principle of proportionality could apply in cases of attack against the escort, and whether the convoy turns into a military objective when escorted. Finally, the article tackles the limitations of such a framework in order to define the situations it covers.
* The views expressed in this article are the author’s own and do not necessarily reflect those of the organization with which she is affiliated.
The legal framework on humanitarian assistance and humanitarian access is clear under international humanitarian law (IHL). When referring to relief actions, Additional Protocol I to the Geneva Conventions (AP I) implicitly defines humanitarian assistance as food, medical supplies, clothing, bedding, means of shelter, other supplies essential to the survival of the population, and objects necessary for religious worship.1 In occupied territory, the Occupying Power must supply such goods if the territory is inadequately supplied.2 Further, it must consent to the free passage of the goods.3 In all other cases, the delivery of relief consignments is subject to the consent of the State in an international armed conflict (IAC)4 and that of the High Contracting Party concerned in a non-international armed conflict (NIAC),5 and such consent cannot be withheld arbitrarily.6 In a NIAC, the consent of an armed group controlling or operating in the territory where the relief action is carried out may also be necessary from a practical point of view.7 Moreover, parties to the conflict must allow and facilitate rapid and unimpeded passage of relief consignments.8
In practice, however, the delivery of humanitarian assistance is increasingly difficult in light of current concerns on humanitarian security.9 Faced with insecure environments of opportunistic criminality or targeting by a party to the conflict, humanitarian organizations have resorted to armed escorts from the military, United Nations (UN) peacekeeping missions10, private military and security companies (PMSCs) and non-State armed groups.11
Yet, IHL is largely silent on the resort to armed escorts in the specific case of relief convoys. Only the Commentaries refer to an authorization to use these escorts in “zones where the control is disputed by the parties [when the relief organization] does not have the benefit of the protective emblem”.12 The Geneva Conventions expressly refer to the practice of armed escorts only in the context of medical units and establishments.13 In this context, the Commentaries define an escort as “a person, vehicle, or group accompanying another to provide protection”14 and to defend the medical unit or establishment “against attacks by rioters or pillagers and unlawful attacks by enemy soldiers, … or ensure the maintenance of order”.15
In the case of armed escorts to humanitarian convoys, various international organizations have provided non-binding guidelines. For example, the UN Inter-Agency Standing Committee (IASC) has issued a non-binding framework regulating the resort to armed escorts.16 According to these guidelines, armed escorts may only be used as an exception, in last resort, if four cumulative criteria are met: firstly, “[t]he level of humanitarian need is such that the lack of humanitarian action would lead to unacceptable human suffering”;17 secondly, “State authorities or local non-State actors are unable or unwilling to permit the movement of humanitarian supplies or personnel without the use of armed escorts”;18 thirdly, “[t]he armed escorts utilised are capable of providing a credible deterrent necessary to enhance the safety of the humanitarian personnel”; and lastly, “[t]he use of an armed escort will not irreversibly compromise the humanitarian operating environment or the longer-term capacity of the organisation(s) to safely and effectively operate in the future”.19 In 1995, the International Committee of the Red Cross (ICRC) provided guidelines20 stating that the use of armed escorts is generally strictly prohibited, and it has maintained this position over time.21 In the same guidelines, the ICRC also took the view that such use shall be permitted only on an exceptional basis where “the refusal of such an escort would lead to the paralysis of humanitarian activities”.22 The guidelines also indicated that when permitted, the use of armed escorts should also respect the principles of humanity, independence, impartiality and neutrality.23 The guidelines further stressed that this use should consider cumulatively and as a minimum the pressing nature of humanitarian needs, the ability of the ICRC to fulfil those needs to the exclusion of alternative organizations, the security of the beneficiaries, the function of deterrent of the armed escort, the consent of the party or authority through which the convoy will travel to the passage of the convoy, and the nature of the actor against whom the escort is defending the convoy.24
In spite of the existence of this non-binding framework, “in insecure contexts, the use of one-size-fits-all armed escorts for humanitarian actors often becomes the rule, instead of the exception”.25 Despite the risks26 that this use may entail, the legal framework on the resort to armed escorts protecting humanitarian convoys against targeting by a party to the conflict or opportunistic criminality remain largely unexplored. Hence the question: what is the legal framework regulating armed escorts to humanitarian convoys under IHL?
This article analyzes the rules applying to the parties to the conflict regarding the use of, or attacks against, armed escorts to humanitarian convoys, and the status of escorted convoys under the existing rules of IHL. It does not examine armed escorts to medical units and establishments because this practice is specifically regulated under Article 22(2) of Geneva Convention I (GC I).27 Due to space constraints, the article focuses on relief supplies of the convoy rather than the entire convoy, which usually also contains relief personnel. More specifically, the article argues that although IHL does not tackle the practice of armed escorts to humanitarian convoys expressly and specifically, the principle of passive precautions, the principle of proportionality and the principle of distinction provide a legal framework for this practice. In arguing so, the article aims to analyze an unexplored angle of the protection of humanitarian aid.
The protections afforded to escorted convoys under IHL
The principle of passive precautions: A protection for escorted convoys rather than an obstacle to the legality of using armed escorts
Intuitively, armed escorts by a party to the conflict could qualify as a measure covered by Article 58(c) of AP I, as they protect humanitarian convoys from potential attacks. Indeed, Article 58(c) refers to “other necessary precautions to protect … civilian objects under [the party's control] against the dangers resulting from military operations”.28 However, a closer look at the use of armed escorts can reveal that such escorts could themselves be contrary to Article 58(c) as they could qualify as military operations endangering civilian objects by increasing the risk that the convoys could be harmed as collateral damage. Thus, the question of whether the practice of armed escort by a party to the conflict is even allowed under IHL, given the application of the principle of passive precautions as interpreted below, is raised. Each element of the material scope of Article 58(c) is examined below to address this question.
Regarding the notion of military operations under Article 58(c) of AP I, “the use of the term ‘military operations’ rather than ‘attacks’ (as in the general rule) may well mean that subparagraph (c) applies to a broader range of activities”29 such as armed escorts. Some challenge this interpretation, arguing that “the scope of the application is no different, regardless of the different language used”.30 But this last interpretation is against the letter of Article 58(c) and the understanding in the Conventions of the notion of military operations, which are understood in their ordinary meaning in this provision as “all movements and acts related to hostilities that are undertaken by armed forces”.31 As such, the first interpretation should be retained, and armed escorts should fall under the notion of military operations as understood in Article 58(c).
As to the notion of danger, the application of this concept is unclear32 in the context of Article 58 of AP I. Its ordinary meaning refers to “the possibility of harm or death to someone”.33 As such, danger may be equated with a risk of a conduct arising from military operations, such as targeting in the case of armed escorts. This is supported in light of the temporal scope of Article 58, whereby passive precautions are to be taken before the actual attack occurs.34
With regard to the notion of control, interpretations vary. On the one hand, Sandoz, Swinarski and Zimmerman interpret the notion of control as control over territory.35 On the other hand, Jensen asserts that the notion of control should be understood as control over persons or objects.36 In the case of escorted convoys, both thresholds overlap in practice. Indeed, escorting a convoy indicates that the escorting party has control over the delivery of the goods, as it is a modality of delivery that the party can accept, refuse or impose. Further, the party controlling the territory is the one consenting to this modulated delivery and providing the escort in the territory that it controls. Since both thresholds overlap, an escort other than the forces of a State that is a party to the conflict may also be considered as controlling the relief consignments in the sense of the customary rule equivalent to Article 58(c).37
In light of the above, it can be argued that the use of armed escort constitutes in itself a dangerous military operation because it increases the risk that the convoy will be attacked as collateral damage. As such, this use can be contrary to Article 58(c), which requires that the parties to the conflict avoid military operations endangering civilians or civilian objects under their control. While this is true in IACs, difficulties can arise in NIACs. Indeed, the ICRC Customary Law Study identifies the obligation to take passive precaution as customary.38 This customary nature has further been clearly stated by the International Criminal Tribunal for the former Yugoslavia (ICTY) in the Kupreskic case.39 However, the ICRC fully equates the scope of the customary rule with the treaty rule only in IACs.40 In NIACs, the scope is narrower in that the sole obligation of a party to the conflict is to take all feasible precautions to protect civilian objects under its control against the effects of attacks rather than the wider notion of military operations.41 In order to overcome this difficulty, it may be argued that the obligation of precautions is an overarching principle stemming from the obligation of constant care to “spare the civilian population, civilians and civilian objects” traditionally imposed on the attacker,42 which applies to a broader range of operations.43
Nonetheless, it is flawed to consider that the principle of passive precautions expressed in Article 58(c) implies that each and every resort to armed escort by parties to the conflict is prohibited. Indeed, this obligation should be fulfilled to the extent of what is “reasonably practicable” or “practically possible”.44 There is a consensus that the assessment of the feasible measures is extremely context-dependent. It is “a matter of fact”45 that is assessed in light of the “circumstances ruling at the time”.46 Traditionally, this formulation is understood to allow for legitimate military requirements to be taken into account in the requirement of feasibility of the measures.47 Beyond this traditional understanding, however, humanitarian concerns may also play a role in the feasibility assessment. For example, the US Law of War Manual includes in the circumstances encompassed by the feasibility of passive precautions the humanitarian benefits arising from taking the precaution.48 In the case of armed escorts, taking the said precautionary measure consists in abstaining from escorting the convoy – but then the humanitarian benefit stemming from taking the precaution is null as the goods do not reach the population if targeted or harmed as collateral damage. The UK also takes the view that humanitarian considerations should be taken into account in the feasibility requirement.49
Further, regarding the availability of “other viable alternatives”50 in order to assess the feasible character of a precaution, there are indeed alternatives to armed escorts. In practice, they include “cultivation of greater acceptance with local stakeholders, Parties to the Conflict and other relevant stakeholders, humanitarian negotiations, de-conflation arrangements, humanitarian pause, [and] humanitarian corridors”.51 The fact that armed escorts were resorted to after considering these kinds of alternatives may justify that taking passive precautions as encompassed under Article 58 was not feasible. This conclusion concurs with the requirements of the IASC non-binding framework, which insists on the use of armed escorts as a means of last resort, after the exhaustion of these available alternatives.52
In sum, in the author's view, the practice of armed escort falls under Article 58(c) of AP I applicable in NIACs under Rule 22 of the ICRC Customary Law Study, and it violates this provision only when used in a systematic manner, without considering alternatives. As such, Article 58 does not constitute an obstacle to the legality of the use of armed escorts but rather sets strong requirements to their use. Those requirements further protect humanitarian convoys, in line with the existing non-binding framework addressed to organizations on the resort to armed escorts.53
Convoys stopped as a result of attacks on an armed escort which is a party to the conflict: Application of the principle of proportionality
When an armed escort to a humanitarian convoy is attacked, the convoy might be incidentally damaged, stopped from proceeding, or both if partially damaged. Considering the reverberating effects on the civilian population of the attack on the escort is essential for assessing the lawfulness of the attack.
Taking into account the reverberating effects on the civilian population of an attack on the armed escort is especially relevant when the convoy is incidentally halted. This is because contrarily to incidental destruction of relief supplies54, incidental halting of a convoy is not necessarily covered by the prohibition on the use of starvation of the civilian population as a method of warfare.55 Indeed, the halting does not constitute an attack on, destruction of or removal of objects indispensable to the civilian population in the sense of the prohibition. In some cases, relief supplies could fall under the scope of the prohibition because they are rendered useless – for example, foodstuffs may perish because of the prolonged incidental halting of a convoy. Outside such cases, the incidental halting does not fall under the specific prohibition against attacking, destroying, removing or rendering useless objects indispensable to the civilian population. Neither does it fall under the general prohibition on starvation of the civilian population because it is the consequence of an attack that is purported to pursue a military advantage.56
Moreover, when the convoy is halted, the obligation to facilitate the unimpeded passage of the convoy still applies.57 Provided that this obligation also applies to the attacking party,58 it may be questioned whether the incidental impediment of the passage of the convoy by the attack is in breach of the obligation. Indeed, Article 70(2) of AP I does not specify the type of impediment covered by this rule. The Commentaries refer to administrative impediments, denial of access and harassment as illustrations,59 and customary IHL requires the impediment to be “deliberate”.60 As such, with the purpose of an attack on the escort being the pursuit of a military advantage, one may wonder whether the incidental impediment falls under the obligation to facilitate the unimpeded passage of the convoy, at least under the customary rule and potentially under the treaty rule.
Because the incidental halting is covered neither by the prohibition on starvation nor the impediment of the passage of the convoy, correctly assessing the proportionality of the attack on the escort by considering its reverberating effects on the civilian population is extremely important. The proportionality rule applies to an attack, in the sense of an “act of violence against the adversary, whether in offence or in defence”,61 against a legitimate target escorting a humanitarian convoy. If the convoy is merely stopped because of an attack on an escort, it appears that it does not fall under the scope of Article 51(5)(b) of AP I because there is no “loss of life, injury to civilians, damage to civilian objects, or a combination thereof”.62 However, this conclusion is flawed as it does not take into account the actual harm to civilians caused by the halting of the transportation of the objects concerned. This harm is reflected in the reverberating effects of the attack, which are those that are not directly and immediately caused by the attack but are nevertheless the product thereof.63 The author examines an application of the reverberating effects doctrine in the following paragraphs.
On the side of the expected military advantage to be gained, this advantage is the annihilation of the armed escort.64 If the attacking party seeks the advantage of stopping the convoy, this advantage is merely political, as part of the politicization of humanitarian aid.65 The military advantage remains the annihilation of the escorting party, and even this small military advantage can render the attack lawful where only minimal incidental harm to property is expected.66 Moreover, the context of the conflict may change the value of the military advantage.67 In a highly asymmetric conflict, a few soldiers may have a higher value as a military advantage to the weaker party, such as a non-State armed group, than the same amount of escorting military in another context.68
On the side of the incidental harm, the attacker should take into account the reverberating effects of the attack on the civilian population. In contemporary practice, the question of reverberating effects in the proportionality assessment has been posed inter alia in the case of the use of explosive weapons in urban areas,69 cyber warfare70 and the targeting of dual-use objects71 because of the interconnectivity of the military objective with the civilian infrastructure incidentally harmed. This question is controversial in multiple aspects, but the controversy can be mitigated when applied to convoys incidentally halted by an attack on the armed escort.
First, on the very existence of an obligation to take those effects into account, the proponents of this obligation emphasize that the use of “the qualifiers ‘expected’ and ‘incidental’ do[es] not limit the relevant incidental harm to direct effects only, unlike the relevant military advantage which must be ‘concrete and direct’”.72 Moreover, they also hold that the purposive interpretation of Article 51(5)(b) of AP I leads to the inclusion of indirect effects in the proportionality assessment73 and provides for a legal basis to this obligation.74 By contrast, opponents of this obligation argue that as the anticipated military advantage must be concrete and direct, so must the expected incidental harm. To them, only the direct effects of the attack on civilian objects or the civilian population must be taken into consideration in the proportionality assessment.75 To this author, the fact that States increasingly accept the need to factor reverberating effects into the proportionality assessment76 strengthens the argument in favour of the existence of such an obligation.
The temporal and geographical scope of this obligation is less controversial. The travaux preparatoires of AP I indicate that States explicitly rejected the attempts to limit incidental harm to the areas surrounding the military objective.77 Similarly, there is no requirement that the effects be limited in time.78 Consequently, “provided that the harm falls into one of the categories identified in Additional Protocol I, the geographic or temporal proximity of the harm to the attack is not determinative”.79 This means that even an injury taking place away from the location of the attack on the armed escort and not immediately after the attack can be factored into the proportionality assessment.
Further, regarding the material scope of application of the obligation, the most controversial issue is the fact that the incidental harm expected from the reverberating effects of the attack must be foreseeable. It is particularly difficult to determine whether the effects accounted for should be restricted to those brought about in one causal step. Some argue that “the number of causal steps between the attack and the harm” is not determinative.80 However, others opine that this relationship is limited to one causal step.81 This aspect of the controversy can be mitigated in the case of a humanitarian convoy stopped as a result of an attack on an armed escort to the convoy. Indeed, contrarily to objects targeted in cyber warfare and urban warfare, and to dual-use objects for which reverberating effects are traditionally discussed, a humanitarian convoy is not integrated into an extensive network of services for the civilian population. When a humanitarian convoy is stopped as a result of an attack on the escort, the deprivation of humanitarian assistance already brings the injury in one causal step, although it may occur later in time. In addition, in many cases, the population in question is already inadequately supplied with essential goods,82 to the extent that the injury is probable.83 For example, the deprivation of food supplies likely leads to hunger, especially if the civilian population is food insecure.84 If hunger is an injury within the meaning of Article 51(5)(b) of AP I, it should not be excessive in relation to the military advantage.
Hence, the consequences of stopping the convoy are more directly foreseeable than those arising in a complex network of essential services. This mitigates the traditional controversy surrounding the foreseeability of reverberating effects. In other terms, when humanitarian convoys are incidentally halted, it is the delineation of the notion of injury brought about in the first causal step that would narrow down the foreseeability criteria. By contrast, in the case of attacks on essential services integrated into a complex network, it is the question of the number of causal steps which defines the contours of the notion.
Moreover, with regard to foreseeability, opponents of the obligation to take into account reverberating effects of attacks argue that “there are too many potential variables outside of the attacker's control that make it practically impossible to consider these effects as ‘expected’”.85 Among those, the willingness of the enemy to repair the harm may arise in the case of a humanitarian convoy incidentally halted as a result of an attack on the armed forces escorting it. Even some of the proponents of such a doctrine have expressed that this willingness is relevant in the assessment.86 However, this issue is controversial; it does not have a clear answer.87 To answer these concerns, the author suggests that the politicization of humanitarian aid comes into play to infer the willingness of the enemy to mitigate the effects of an attack on an armed escort to a humanitarian convoy. For example, a pattern of denial of access to humanitarian aid may indicate a lack of willingness on the part of the attacker to mitigate the reverberating effects of an attack in the future.
Furthermore, there is a strong tendency against “the idea that humanitarian assistance [as a mitigating measure] could be viewed as broadening the possibilities of attack”.88 The author agrees with this statement. In the case of a humanitarian convoy, the fact that the attacker of its escort is aware that the relief convoy is not the only relief convoy destined to civilians or sends another convoy instead to compensate for the reverberating effects of the attack cannot create an open door for collaterally impeding humanitarian assistance. Similarly, the fact that the attacker expects the convoy to be halted only temporarily, or that the attacker knows that the other party is also under an obligation to facilitate the unimpeded passage of the convoy after the attack on the escort,89 cannot open such a door.
A final difficulty which may arise is that it is controversial whether the obligation to account for reverberating effects in the proportionality assessment is present when the assessment is undertaken under the customary rule of proportionality. However, practice is evolving in this direction.90
All in all, without solving the question of reverberating effects in abstracto, this author suggests that the specific case of humanitarian convoys incidentally stopped from proceeding offers an opportunity for the application of the reverberating effects concept, and does not pose the same controversial questions on the matter as those arising from contemporary practice regarding cyber warfare, explosive remnants of war and the targeting of dual-use objects. As such, it is an avenue for enhancing the protection of humanitarian convoys and correctly assessing the proportionality of an attack against an armed escort which is a party to the conflict.
It appears that a categorization of the nature of the collateral damage would operationalize the criteria of foreseeability. Approaches to classifying the reverberating effects depending on the nature of the objects surrounding the attack have been made in the area of the use of explosive weapons.91 While for civilian lives such an approach could “lead to a slippery slope towards considering that the lives of some civilians ‘weighed less’”92, these ethical considerations are less strong for objects.93 Further, this approach is in line with the principle of humanity underlying IHL.94
Attacks on convoys: The status of escorted convoys under IHL
In practice, humanitarian convoys are targeted in order to prevent access of the relief consignments into enemy-controlled areas, for instance.95 In addition, in the specific case of the use of armed escorts, humanitarian convoys are at higher risk of being targeted due, inter alia, to confusion as to the impartiality of humanitarian organizations vis-à-vis the parties to the conflict.96
Yet, IHL is clear on the fact that humanitarian relief objects must be respected and protected.97 Contrary to medical units,98 however, humanitarian convoys are not specially protected under IHL. For targeting purposes,99 relief objects are considered civilian objects protected under Article 52 of AP I and Rule 7 of the ICRC Customary Law Study.100 Consequently, relief objects cannot be targeted unless they turn into military objectives.
When relying on the applicable targeting rules, it appears that the mere presence of the armed escort is not sufficient to turn the convoy into a military objective. Formally, the aid is no longer perceived as neutral when escorted by a party to the conflict because the relief convoy does not maintain “an absolute distance from the different parties to the conflict”.101 For targeting purposes, however, what matters is whether this situation is more than an absence of formal neutrality and rather a situation falling under Article 52 of AP I. In the case of escorted humanitarian convoys, the most relevant situation in which the relief objects would turn into a military objective would be a use of these convoys which fulfils the requirements of Article 52 and its customary equivalent. For example, the mixed transport of weapons along with humanitarian assistance and the case of diversion of humanitarian goods to combatants could amount to such a use.
As such, humanitarian convoys remain protected under IHL,102 even when joined by an armed escort: they must be respected in that they may not be attacked unless they fulfil the requirement of a military objective under Article 52 of AP I. The mere presence of the escort does not automatically lead to the convoy's fulfilment of those requirements.
The limits to the protections afforded to escorted humanitarian convoys under IHL
It is essential to identify some limitations that could nuance the application of the framework identified in the present article. First, the definitions of armed escorts laid out in both the Commentary to GCI, in the specific context of medical units, and the UN Office for the Coordination of Humanitarian Affairs (OCHA) Civil-Military Coordination Field Handbook state that the escort has a defensive function. According to the latter, armed escorts are a “security measure that serves as a visible deterrent to a potential attack and, if necessary, acts in self-defence against an attack”.103 In the author's view, the fact that a non-State armed group or the armed forces of a State which is a party to the conflict escorting the convoy is acting in defence of the convoy against an attack on the convoy from another party to the conflict or in self-defence against an attack on their person may lead to the question of whether the law enforcement paradigm instead prevails. In addition, the nature of some escorting parties, such as peacekeeping missions and PMSCs, may beg similar questions. Lastly, regardless of the nature of the escort, some situations other than an attack against the escort or against the convoy would also exclude the application of an IHL framework. This is the case with opportunistic criminality by an organized armed group against the convoy or the resort to an escort to resist searches of the convoy.
Non-State armed groups or armed forces of a State acting in self-defence or in defence of the convoy against attacks by another party to the conflict
From the perspective of the armed escort, the first question to highlight is whether this “security measure” amounts to a military operation against which the convoy under its control should be protected to the maximum extent feasible as per Article 58(c) of AP I, or to a law enforcement operation subject to the law enforcement paradigm as a paradigm which “continues to govern all exercise by parties to the conflict of their authority or power outside the conduct of hostilities”.104 The answer to such a question is not necessarily clear-cut. On the one hand, the resort to the armed escort clearly occurs in the context of the conduct of hostilities, because it is an attempt to protect the convoy from the effects of such hostilities. In addition, as previously explained, provided that the convoy remains a civilian object, if used systematically, the practice of armed escorts may in itself violate Article 58(c) because it qualifies as a military operation endangering the civilian convoy in the sense of this provision. Moreover, “the resort to means and methods of warfare between parties to an armed conflict is governed by the legal paradigm of hostilities even if the ultimate purpose of [the] military operations is to maintain, restore, or otherwise impose public security, law, and order”.105 On the other hand, however, it remains uncertain whether the resort to an armed escort qualifies as a method of warfare in that while it may contribute to the war effort, it does not necessarily directly or actually inflict harm on the enemy.106 Lastly, these uncertainties may be strengthened if the escorting party is a non-State armed group, since whether or how the law enforcement paradigm applies to such groups is controversial.107
The same question regarding the applicable paradigm from the perspective of the escort – who can be targeted given their status as armed forces of a State or non-State armed group which is a party to the conflict – arises in cases of the use of armed force in self-defence against an attack on their person by the other party to the conflict. In the author's view, the defensive reaction of the escort can be qualified as conduct of hostilities because of the pre-existing conflict; the fact that, at least from the perspective of the attacker, the attack occurs in the context of the conduct of hostilities; the definition of “attack” in IHL as potentially occurring not only in offence but also in defence;108 and the fact that the escort actually and directly inflicts harm109 on a legitimate target110 who is a party to the conflict.
Similarly, in the author's view, in cases of the use of force in defence of the convoy by the military or an armed group which is a party to the conflict against an attack on the convoy which is unlawful, the unlawful character of the attack does not affect the prevailing paradigm for such a use of force on both the attacker and the escort's side. Indeed,
in practical terms, the conduct of hostilities certainly includes all attacks, that is to say, offensive or defensive operations involving the use of violence against the adversary, whether (lawfully) directed against legitimate military targets or (unlawfully) against protected persons or objects.111
Application of the IHL framework to peace forces
Peace forces often have a mandate to protect civilians.112 In this context, they use force in “defence of the mandate”113 with a gradual response ranging from deterrence, such as resorting to armed escorts, to direct confrontation when civilians are at risk.114 When peace forces are escorting humanitarian convoys in situations of armed conflict, questions related to the applicability of the IHL framework delineated in the present article may arise.
One may wonder how UN peace forces may be bound by the principle of precautions as interpreted in the present article. The now widely cited Secretary-General's Bulletin on “Observance by United Nations Forces of International Humanitarian Law” famously sets the engagement of UN forces in armed conflicts as combatants as a prerequisite for the application of the fundamental principles and rules of IHL.115 Those principles include the obligation to protect the civilian population from the dangers resulting from military operations.116 Given the lack of a clear definition117 of the situations in which peace forces are engaged as combatants,118 it is difficult to assess whether the Bulletin can serve as a legal basis to infer that peace forces should avoid the systematic use of armed escorts. A more solid legal basis would be Article 1 common to the four Geneva Conventions, which the updated Commentary of 2016 interprets as imposing an obligation on national contingents put at the disposal of international organizations to respect and ensure respect for IHL in the same way as States, to the extent that the organization retains command and control over the operation.119 In that sense, and if peace forces can be considered as a party to the conflict,120 the same question as to whether the escorting operations falls within the purview of Article 58(c) of AP I under the conduct of hostilities paradigm which arises for State forces escorting a convoy can be posed here. An additional difficulty is that the principle of passive precautions would apply as matter of customary law, which is not the same in IAC as in NIAC, as previously examined.
Moreover, if peace forces are a party to the conflict and are considered legitimate targets, in case of a use of force in self-defence or in defence of the convoy, the same questions as those that arise for parties to an armed conflict can be raised here mutatis mutandis and answered accordingly. Moreover, an additional layer from the attacker's perspective compared to the armed forces of a State or a non-State armed group that is party to the conflict escorting the convoy is whether the attack falls within the conduct of hostilities or not if UN peace forces are not considered legitimate targets. If not, it means that the obligation to consider the reverberating effects of the attack does not apply to the attacker since the attack does not even fall under the conduct of hostilities paradigm.
Application of the IHL framework to private military and security companies
When considering attacks on convoys, the central question is whether the a priori unlawful character of the attack against the civilian convoy gives rise to the application and precedence of the conduct of hostilities paradigm or not. When PMSCs guard or defend a civilian object against unlawful attacks, they are considered as not directly participating in hostilities121 because the defensive reaction lacks a belligerent nexus.122 Therefore, it appears that a PMSC escorting a humanitarian convoy generally does not need to respect Article 58(c) of AP I as interpreted in the present paper, since it does not prevail in the first place. However, this author has examined the conditions under which humanitarian convoys turn into military objectives, which are scarce but do exist. As such, it is not always a given that the attack against a humanitarian convoy is not legitimate and therefore that the PMSC is defending the convoy in self-defence.123 In light of this, a PMSC might be precluded from systematically escorting humanitarian convoys under Article 58(c) through the obligation of States – as parties to the conflict – to ensure respect for IHL,124 national laws and their contract.125
When an attack on the escort falls under the conduct of hostilities paradigm, difficulties which blur the application of the IHL framework considered in the present article arise. PMSC members may not know that the convoy has turned into a military objective and thus that the practice of armed escort falls under the conduct of hostilities paradigm.126 Moreover, the fact that PMSC members are in the vicinity of a convoy which has turned into a military objective does not mean that they are legitimate targets.127 Yet, this assessment is crucial to the attacker even before envisaging the proportionality principle considered in the present article. In light of such uncertainties, this author agrees with Sassòli that PMSC staff should not be put in such ambiguous situations,128 in order to enhance protection surrounding humanitarian access.129
Looting convoys and convoys resisting searches
The nexus requirement has been understood to require that a conduct took place “in the context of and was associated with” an armed conflict.130 In the author's view, when a convoy is halted and looted by a party to the conflict in order for it to proceed further into the territory under that party's control, the belligerent nexus is sufficiently clear and the situation falls under the conduct of hostilities paradigm: it is covered by Rule 32 of the ICRC Customary Law Study, and if force is used by the escort in order to proceed, that force needs to comply with IHL. However, when the convoy is looted as a result of opportunistic criminality by an organized armed group, the conduct lacks a belligerent nexus, it is covered by international human rights law and domestic law, and the force used by the escort as a result must comply with the law enforcement paradigm.131 This is clear when the armed escort is comprised of State military forces. However, in a NIAC, where a non-State armed group which controls a territory requires members of its own group to escort a convoy which is then looted by an organized criminal group, the situation might not be as clear-cut. Indeed, the position according to which “the way in which non-State armed groups exercise control over, and interact with, persons living in territory under their de facto control is inherently linked to the conflict in question”132 has been put into question and can be opposed to the position where an armed group acting against ordinary criminals lacks a belligerent nexus.133
Regarding escorted convoys resisting searches, the use of an armed escort does not indicate an intent to resist a search of the consignments to which the parties are entitled.134 This is supported by the fact that the arrival of the relief consignments is anticipated. Indeed, in both IACs and NIACs, technical arrangements may be concluded between the parties concerned. Lattanzi notes:
Concrete modalities for carrying out relief activities are left in the hands of impartial organizations, albeit under a certain level of supervision by the state concerned and any non-state entity, sometimes in accordance with special agreements between states/armed groups and relief personnel.135
These agreements may include the requirement not to be escorted.136 Therefore, the parties concerned already consent to the armed escort. When a party did not consent to an armed escort – in such arrangements, for instance – it may indicate an intent to resist search, but such resistance triggers at best a denial of entry and “raise[s] suspicion and may justify the use of force according to a law enforcement paradigm”.137
Far from offering a one-size fits all solution, this article has aimed to explore the framework for armed escorts to humanitarian convoys under IHL because they may be at higher risk of being damaged or halted due to attacks by a party to the conflict on the convoy or on the escorting party. To this end, the author has analyzed whether resorting to armed escorts is in line with the principle of passive precautions, by which the escorting party may be bound. Further, this article has suggested that the proportionality assessment of attacks on the escort from a party to the conflict halting the convoy should take into account the reverberating effects of the attack on the civilian population. Lastly, this article has confirmed that the presence of an armed escort does not necessarily turn the escorted convoy into a legitimate target. As such, this observation clarifies how to assess current attacks on humanitarian convoys – caused, for instance, by a perceived lack of impartiality of the humanitarian organization supplying humanitarian assistance. This author has also defined the limitations of this framework in order to identify the situations that it covers, given the diversity both of the actors escorting the convoy, which range from the military to PMSCs, and of the use of force against the convoy, which can range from attacks to organized crimes. In doing so, the author hopes to shed some light on a novel and important aspect of the protection of humanitarian aid under IHL. This article will hopefully foster further reflections both on the legal implications of the practice of armed escorts to humanitarian convoys and on the way in which these implications can be refined to meet the realities on the ground.
- 1. Protocol Additional (I) to the Geneva Conventions of 2 August 949, and relating to the Protection of Victims of International Armed Conflicts, 25 UNTS 3, 8 June 977 (entered into force 7 December 978) (AP I), Art. 69() applicable in occupied territories and Art. 70() applicable in non-occupied territories, which refers to Art. 69() for an enumeration of the objects it covers. See also Marco Sassòli, International Humanitarian Law: Rules, Controversies and Solutions to Problems Arising in Warfare, st ed., Edward Elgar, Cheltenham, 209, p. 575.
- 2. Geneva Convention (IV) relative to the Protection of Civilian Persons in Time of War of 1 August 1949, 75 UNTS 87 (entered into force 1 October 1950) (GC IV), Arts 55, 60; AP I, Art. 69.
- 3. GC IV, Arts 59, 61; AP I, Art. 69.
- 4. GC IV, Art. 23; AP I, Art. 70.
- 5. Protocol Additional (II) to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of Non-International Armed Conflicts, 112 UNTS 609, 8 June 1977 (entered into force 7 December 1978) (AP II), Art. 18; Jean-Marie Henckaerts and Louise Doswald-Beck (eds), Customary International Humanitarian Law, Vol. 1: Rules, Cambridge University Press, Cambridge, 200 (ICRC Customary Law Study), Rule , available at: https://ihl-databases.icrc.org/customary-ihl/eng/docs/v1 (all internet references were accessed in November 2021).
- 6. M. Sassòli, above note 1, pp. 579, 580.
- 7. Ibid., p. 58. Also note that such consent would not confer any the legal status to the armed group: see, for example, Art. 3(4) common to the four Geneva Conventions.
- 8. GC IV, Arts 23, 59, 61; AP I, Art. 70(2); ICRC Customary Law Study, above note 5, Rule 55.
- 9. Abby Stoddard, Paul Harvey, Monica Czwarno and Meriah-Jo Breckenridge, Aid Worker Security Report 2020: Contending with Threats to Humanitarian Health Workers in the Age of Epidemics, Humanitarian Outcomes, London, August 2020, pp. 1–21, available at: www.humanitarianoutcomes.org/sites/default/files/publications/awsr2020_….
- 10. For example, this is the case with MINUSCA in the Central African Republic and MONUSCO in the Democratic Republic of the Congo. See, respectively, UN Office for coordination of Humanitarian Action (OCHA), Country Specific Guidelines for Civil-Military Coordination between Humanitarian Actors and MINUSCA in the Central African Republic, 2017, p. 7, available at: https://ccorca.org/wp-content/uploads/2018//country-specific-guidelines…; “La premiere aide alimentaire arrive dans l'Est du pays”, France 24, 14 November 2011, available at: www.france24.com/fr/20081114-premiere-aide-alimentaire-arrive-lest-pays….
- 11. For example, “the ICRC resorted to an armed escort by clan representatives for its food convoys in circumstances where State consent could not be obtained”: see Stephen Tully, “Soldiers Bearing Gifts”, Australian Defence Force Journal, No. 125, 1997, p. 26. Tully further differentiates between unilateral State-initiated convoys, UN convoys, convoys of the Office of the UN High Commissioner for Refugees, and convoys composed of local personnel hired by aid agencies: ibid., p.26.
- 12. Yves Sandoz, Christophe Swinarski and Bruno Zimmerman (eds), Commentary on the Additional Protocols, ICRC, Geneva, 1987 (ICRC Commentary on AP I/AP II), p. 834, para. 2888.
- 13. Geneva Convention (I) for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field of 12 August 1949, 75 UNTS 31 (entered into force 21 October 1950) (GC I), Art. 22(2).
- 14. ICRC, Commentary on the First Geneva Convention: Convention (I) for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the field, 2nd ed., Geneva, 2016 (ICRC Commentary on GC I), para. 1871.
- 15. Ibid., para. 1874.
- 16. IASC, Non-Binding Guidelines on the Use of Armed Escorts for Humanitarian Convoys, 2013, available at: www.unocha.org/sites/unocha/files/Armed%20Escort%20Guidelines%20-%20Fin…. See, by contrast, for civil–military cooperation other than armed escorts to humanitarian convoys, OCHA, Guidelines on the Use of Military and Civil Defence Assets to Support United Nations Humanitarian Activities in Complex Emergencies, 2003 (revised in 2006), available at: www.unocha.org/publication/guidelines-use-military-and-civil-defence-as….
- 17. IASC, above note 16, p. 6.
- 18. Ibid., p. 6.
- 19. Ibid., p. 6.
- 20. ICRC, Report on the Use of Armed Protection for Humanitarian Assistance, Geneva, 1995, available at: www.icrc.org/en/doc/resources/documents/report/57jneg.htm.
- 21. ICRC, “The ICRC and the Use of Armed Guards”, News Release 97/04, Geneva, 2009, available at: www.icrc.org/en/doc/resources/documents/news-release/2009-and-earlier/5….
- 22. ICRC, above note 20.
- 23. Ibid.
- 24. Ibid.
- 25. OCHA, United Nations Civil-Military Coordination: Field Handbook, 2018, p. 162, available at: https://tinyurl.com/9ut8ctmv.
- 26. Kate Mackintosh, “Beyond the Red Cross: The Protection of Independent Humanitarian Organizations and their Staff in International Humanitarian Law”, International Review of the Red Cross, Vol. 89, No. 864, 2007, p. 122.
- 27. GC I, Art. 22(2).
- 28. AP I, Art. 58(c).
- 29. Eric T. Jensen, “Precautions against the Effects of Attacks in Urban Areas”, International Review of the Red Cross, Vol. 98, No. 1, 2016, p. 161.
- 30. Ibid., p. 161.
- 31. ICRC Commentary on AP I, above note 12, p. 600, para. 1875. The Commentary gives an additional definition of military operations as “the movements, manoeuvres and actions of any sort, carried out by the armed forces with a view to combat” (emphasis added) in the context of Article 3 of AP I: see Ibid., p. 66, para. 152. It can be argued that if used in this sense, the practice of armed escorts would not fall under the qualification of military operations because such escorts are not per se carried out with a view to combat. However, the definition given under the Commentary on Article 48 of AP I should be preferred when it comes to the protection of the civilian population as expressed under the Commentary itself: “this term is used in several articles in this Section … and it may be useful to refer to the commentary thereon”. See ibid, p. 600, para. 1875.
- 32. E. T. Jensen, above note 29, pp. 61, 156.
- 33. “Danger”, Cambridge Dictionary, available at: https://dictionary.cambridge.org/dictionary/english/danger.
- 34. Marco Sassòli, “The Obligation to Take Feasible Passive Precautions and the Prohibition of the Use of Human Shields: Can Military Considerations, including Force Protection, Justify Not to Respect Them?”, Collegium, No. 46, 2016, p. 78.
- 35. ICRC Commentary on AP I, above note 12, p. 692, para. 2239.
- 36. E. T. Jensen, above note 29, p. 162; see also William H. Boothby, The Law of Targeting, Oxford University Press, Oxford, 2012, p. 132.
- 37. ICRC Customary Law Study, above note 5, Rule 22.
- 38. Ibid., Rules 22–24.
- 39. ICTY, Prosecutor v. Kupreskic et al., Case No. IT-95-16-T, Judgment (Trial Chamber), 14 January 2000, para. 524.
- 40. The customary nature of the obligation to avoid locating military objectives in densely populated areas can, however, be doubted: see M. Sassòli, above note 34, p. 78.
- 41. ICRC Customary Law Study, above note 5, Rule 22.
- 42. AP I, Art. 57.
- 43. This suggestion reflects the approach of Geoffrey Corn and James A. Schoettler, “Targeting and Civilian Risk Mitigation: The Essential Role of Precautionary Measures”, Military Law Review, Vol. 223, No. 4, 2015.
- 44. W. H. Boothby, above note 36, p. 131.
- 45. Ibid., p. 132.
- 46. This interpretation is accepted by the Netherlands, the UK and “at least eight other States”: see E. T. Jensen, above note 36, p. 165. The United States is also included: see M. Sassòli, above note 34, p. 83.
- 47. G. Corn and J. A. Schoettler, above note 43, p. 830.
- 48. US Department of Defense, Law of War Manual, Office of General Counsel, 2015, p. 190, cited in E. T. Jensen, above note 36, p. 156.
- 49. W. H. Boothby, above note 36, p. 131.
- 50. G. Corn and J. A. Schoettler, above note 43, p. 830.
- 51. OCHA, above note 25, p. 62.
- 52. IASC, above note 16, p. 6.
- 53. It is interesting to note that this reasoning based on passive precautions applied to armed escorts to humanitarian convoys is also coherent with the way in which escorts must be used in other contexts, such as medical units. In the context of medical units, armed escorts shall be resorted to in exceptional circumstances. Moreover, the resort to armed escorts is subsidiary in that it only occurs when the lightly armed medical personnel are not sufficiently numerous to ensure the protection of the medical unit or where the medical personnel are not armed. Further, the use of force must be made for defensive purposes. Lastly, the escort may not use heavy weapons. The Commentary on GC I also specifies that the armed escort must have been assigned to guard the medical unit to the exclusion of any military objective surrounding it. See ICRC Commentary on GC I, above note 14, para. 1874.
- 54. Article (2) of AP I prohibits the destruction of such goods “whatever the motive”. Although Article 14 of AP II is drafted differently, a similar reasoning applies. See ICRC Commentary on AP II, above note 12, para. 4807; ICRC Customary Law Study, above note 5, Rule .
- 55. AP I, Art. 54; AP II, Art. 14; ICRC Customary Law Study, above note 5, Rule 54.
- 56. Compare with incidental starvation of the civilian population stemming from sieges pursuing a lawful military purpose: see ICRC Customary Law Study, above note 5, Rule 54. See also Emanuela-Chiara Gillard, “Seventy Years of the Geneva Conventions”, Chatham House, 24 March 2020, available at: www.chathamhouse.org/2020/03/seventy-years-geneva-conventions/sieges-st…; Gloria Gaggioli, “Are Sieges Prohibited under Contemporary IHL?”, EJIL: Talk!, 30 January 2019, available at: www.ejiltalk.org/joint-blog-series-on-international-law-and-armed-confl….
- 57. See GC IV, Art. 23 on the free passage of specific consignments; GC IV, Art. 59 on the obligation to permit free passage of supplies on their way to occupied territories and guarantee their protection; AP I, Art. 70(2) on the obligation to facilitate unimpeded passage of relief consignments; and ICRC Customary Law Study, above note 5, Rule 55 as the latter's customary equivalent.
- 58. Strictly speaking, the Commentary on Article 70(2) of AP I envisages the addressees of the obligation as the signatories of AP I who control a territory through which the relief supplies are passing (ICRC Commentary on AP I, above note 12, paras 2824, 2828). It is silent on the impediment of humanitarian aid by a State Party in another State Party's territory through means other than direct attacks on the supplies – these attacks are already covered by other rules of IHL. In the case of an attack on an armed escort to a humanitarian convoy, it may be that a party to the conflict is impeding the passage of the convoy in another State Party's territory without directly attacking the consignments. If it is considered that the attacking party is not an addressee of the obligation to facilitate the unimpeded passage of the convoy because it does not control the territory where the relief action is taking place, this would lead to absurd results. In addition, the letter of Article 70(2) refers to “Parties to the conflict” and “Each High Contracting Party”. This wording, combined with an interpretation of the provision in good faith, would lead to the inclusion of this attacking party as an addressee of this provision. This interpretation in good faith also applies to Rule 55 of the ICRC Customary Law Study, above note 5, which refers to the facilitation of the unimpeded passage of relief consignments by the parties controlling the areas through which the supplies are passing. Article 59 of GC IV applicable in situations of occupation is interpreted more broadly in the Commentaries and refers to “states concerned” without specifying whether they are those through which relief supplies are passing. See Jean Pictet (ed.), Commentary on the Geneva Conventions of 12 August 1949, Vol. 4: Geneva Convention relative to the Protection of Civilian Persons in Time of War, ICRC, Geneva, 19, p. 321.
- 59. See ICRC Commentary on AP I, paras 2805, 2829.
- 60. ICRC Customary Law Study, above note 5, Commentary on Rule 55.
- 61. AP I, Art. 49.
- 62. Ibid., Art. 51.
- 63. Michael N. Schmitt, “Wired Warfare: Computer Network Attack and the Jus in Bello”, International Law Studies, Vol. 76, 2002, p. 204.
- 64. The author retains the definition of a military advantage as annihilating or weakening enemy forces. See ICRC Commentary on AP I, p. 685, para. 2218.
- 65. See International Institute for Strategic Studies, “Humanitarian Aid as a Weapon of War”, Armed Conflict Survey, Vol. 5, 2019, pp. 14–20.
- 66. Laurent Gisel, The Principle of Proportionality in the Rules Governing the Conduct of Hostilities under International Humanitarian Law, ICRC, Geneva, 2018, p. 18.
- 67. Ibid., p. 61.
- 68. Ibid., p. 61.
- 69. ICRC, Explosive Weapons in Populated Areas: Humanitarian, Legal, Technical and Military Aspects, Report of the Expert Meeting, Chavannes-de-Bogis, 24–25 February 2015.
- 70. Hensey A. Fenton III, “Proportionality and Its Applicability in the Realm of Cyber-Attacks”, Duke Journal of Comparative and International Law, Vol. 29, No. 2, 2019.
- 71. Lindsey Cameron and Marco Sassòli, “The Protection of Civilian Objects: Current State of the Law and Issues De Lege Ferenda”, in Natalino Ronzitti and Gabriella Venturini (eds), The Law of Air Warfare: Contemporary Issues, Eleven International, The Hague, 2006, p. 67.
- 72. L. Gisel, above note 66, p. 44.
- 73. Isabel Robinson and Ellen Nohle, “Proportionality and Precautions in Attack: The Reverberating Effects of Using Explosive Weapons in Populated Areas”, International review of the Red Cross, Vol. 98, No. 901, 2016, p. 114.
- 74. Ian Henderson and Kate Reece, “Proportionality under International Humanitarian Law: The ‘Reasonable Military Commander’ Standard and Reverberating Effects”, Vanderbilt Journal of Transnational Law, Vol. 51, No. 3, 2018, p. 852.
- 75. Ian Henderson, The Contemporary Law of Targeting: Military Objectives, Proportionality and Precautions in Attack under Additional Protocol I, Martinus Nijhoff, Leiden and Boston, MA, 2009, p. 210.
- 76. L. Gisel, above note 66, p. 44.
- 77. Ibid., p. 44.
- 78. I. Henderson, above note 75, p. 128.
- 79. Emanuela-Chiara Gillard, Proportionality in the Conduct of Hostilities: The Incidental Harm Side of the Assessment, Chatham House, 2018, p. 18, available at: www.chathamhouse.org/sites/default/files/publications/research/2018-12-….
- 80. Ibid., p. 18.
- 81. L. Gisel, above note 66, p. 47.
- 82. Most provisions on humanitarian assistance reflect this reality: see, for example, AP I, Art. 70(1); AP II, Art. 18; GC IV, Art. 59.
- 83. The standard of foreseeability in considering this first causal step remains open. Even proponents of the obligation to account for reverberating effects in the assessment of proportionality disagree on this meaning. In particular, the different interpretations of foreseeability vary from likelihood, probability and reasonable causality to a subjective criteria consisting of an open inquiry or even a criteria encompassing effects which would not have happened but for the attack. See L. Gisel, above note 66, pp. 118–119. This author interprets the notion of “expected” in Article 51(5)(b) of AP I as “likely” or “probable” because this is the ordinary meaning of the concept. As such, it is not necessary to be able to trace back the injury to the civilian population to the specific incidental halting of the convoy in question.
- 84. As previously examined, this deprivation is not necessarily covered by the prohibition on starvation of the civilian population, if the food supplies are not yet rendered useless in the sense of the prohibition for instance.
- 85. L. Gisel, above note 66, p. 44. Here, a difficulty that remains is that in an attack of opportunity, the attacker of the escort to the convoy may not know the nature of the specific consignment being transported and may thus be unable to correctly assess the reverberating effects on the civilian population. This is a shared difficulty with the traditional debate surrounding reverberating effects of attacks on the civilian population.
- 86. Ibid., p. 49.
- 87. Ibid., p. 49.
- 88. Ibid., p. 51.
- 89. In addition, the obligation that the passage of the convoy remains unimpeded is an obligation of means that the other party may be considered as having fulfilled after the attack in light of security concerns, for example. See ICRC Commentary on AP I, above note 12, para. 2829. As such, the attacking party may not even have certainty that the convoy would be halted only temporarily.
- 90. I. Robinson and E. Nohle, above note 73, p. 116.
- 91. Michael Talhami and Mark Zeitoun, “The Impact of Explosive Weapons on Urban Services: Direct and Reverberating Effects across Space and Time”, International Review of the Red Cross, Vol. 98, No. 901, 2016.
- 92. L. Gisel, above note 66, p. 61.
- 93. Ibid., p. 61. See also I. Henderson, above note 75, p. 192.
- 94. ICTY, Kupreskic, above note 39, para. 524.
- 95. For example, in Syria. See UN Security Council, UN Summary by the Secretary-General of the Report of the United Nations Headquarters Board of Inquiry into the Incident involving a Relief Operation to Urum Al-Kubra, Syrian Arab Republic on 19 September 2016, UN Doc. S/2016/1093, 21 December 2016, para. 19.
- 96. For example, in Nigeria. See OCHA, above note 25, p. 163.
- 97. ICRC Customary Law Study, above note 5, Rule 32.
- 98. GC I, Arts 19, 21.
- 99. Ibid., Arts 21, 22; AP I, Art. 13; ICRC Customary Law Study, above note 5, Rule 25.
- 100. ICRC Customary Law Study, above note 5, Commentary on Rule 32.
- 101. Flavia Lattanzi, “Humanitarian Assistance”, in Andrew Clapham, Marco Sassòli and Paola Gaeta (eds), The 1949 Geneva Conventions: A Commentary, Oxford University Press, Oxford, 2015, p. 240.
- 102. ICRC Customary Law Study, above note 5, Rule 32.
- 103. OCHA, above note 25, p. 6.
- 104. Nils Melzer and Gloria Gaggioli, “Conceptual Distinction and Overlaps between Law Enforcement and the Conduct of Hostilities”, in Terry D. Gill and Dieter Fleck (eds), The Handbook of the International Law of Military Operations, 2nd ed., Oxford University Press, Oxford, 2015, p. 75.
- 105. Ibid., p. 75 (emphasis added).
- 106. Ibid., p. 72.
- 107. Gloria Gaggioli, “The Use of Force in Armed Conflicts: Conduct of Hostilities, Law Enforcement, and Self-Defense”, in Winston S. Williams and Christopher M. Ford (eds), Complex Battlespaces: The Law of Armed Conflict and the Dynamics of Modern Warfare, 1st ed., Oxford University Press, New York, 2018, p. 103.
- 108. AP I, Art. 49.
- 109. Or at least affects its military capacity – see N. Melzer and G. Gaggioli, above note 104, p. 74.
- 110. Gloria Gaggioli, The Use of Force in Armed Conflicts: Interplay between the Conduct of Hostilities and Law Enforcement Paradigms, Report of the Expert Meeting, ICRC, Geneva, 2013.
- 111. N. Melzer and G. Gaggioli, above note 104, p. 73.
- 112. Report of the High-level Independent Panel on Peace Operations on Uniting our Strength for Peace: Politics, Partnerships and People, UN Doc. A/70/95–S/2015/446, 17 June 2015, para. 90.
- 113. Ibid., para. 128.
- 114. Ibid., para. 128.
- 115. Secretary-General's Bulletin, “Observance by United Nations Forces of International Humanitarian Law”, UN Doc. ST/SGB/1999/13, 6 August 1999, Section 1.
- 116. Ibid., Section 5.4.
- 117. Mathew Happold, “Comment: Obligations of States Contributing to UN Peacekeeping Missions under Common Article 1 of the Geneva Conventions”, in Heike Krieger (ed.), Inducing Compliance with International Humanitarian Law: Lessons from the African Great Lakes Region, Cambridge University Press, Cambridge, 2015, p. 369.
- 118. Which may be the case even in non-forcible peacekeeping mandates: see S. Tully, above note 11, p. 26.
- 119. ICRC Commentary on GC I, above note 14, para. 140. See also Eve Massingham and Annabel McConnachie, “Common Article 1: An Introduction”, in Ensuring Respect for International Humanitarian Law, 1st ed., Routledge, London, 2020, pp. 1–11.
- 120. M. Sassòli, above note 1, p. 471. On criteria for the support-based theory, see Tristan Ferraro, “The Applicability and Application of International Humanitarian Law to Multinational Forces”, International Review of the Red Cross, Vol. 95, No. 891–892, 2015, p. 584.
- 121. M. Sassòli, above note 1, p. 551.
- 122. Lindsey Cameron and Vincent Chetail, Privatizing War: Private Military and Security Companies under Public International Law, Cambridge University Press, Cambridge, 2013, p. 466.
- 123. Ibid., p. 467.
- 124. Ibid., p. 262.
- 125. Ibid., p. 264.
- 126. M. Sassòli, above note 1, p. 552.
- 127. L. Cameron and V. Chetail, above note 122, pp. 467, 432. On the status of PMSCs, see Ibid., p. 386. It is also worth noting that the presence of a PMSC in the vicinity of the convoy does not per se turn the convoy into a military objective.
- 128. M. Sassòli, above note 1, p. 552.
- 129. Even when paradigms other than the conduct of hostilities prevail, uncertainties arise as to whether a right to self-defence would by granted to private individuals, and whether such a right would also be granted when private individuals are defending property, such as a humanitarian convoy in our case. See L. Cameron and V. Chetail, above note 122, pp. 467, 463; N. Melzer and G. Gaggioli, above note 104, p. 86.
- 130. Elvina Pothelet, “Life in Rebel Territory: Is Everything War?”, Armed Groups and International Law, 20 May 2020, available at: https://armedgroups-internationallaw.org/2020/05/20/life-in-rebel-terri….
- 131. See G. Gaggioli, above note 107, p. 88.
- 132. See also ICRC, International Humanitarian Law and the Challenges of Contemporary Armed Conflicts, Geneva, 2019, p. 53, available at: www.icrc.org/sites/default/files/document/file_list/challenges-report_i….
- 133. E. Pothelet, above note 130.
- 134. See GC IV, Art. 59; AP I, Art. 70; ICRC Customary Law Study, above note 5, Rule 55.
- 135. F. Lattanzi, above note 101, p. 241. On the possibility of making technical arrangements, see GC IV, Art.59; AP I, Art. 70; ICRC Customary Law Study, above note 5, Rule 55.
- 136. This is the case in the Democratic Republic of the Congo, where the rare humanitarian organizations that have access to the desired areas must be unescorted, pursuant to the requirements of non-State armed group commanders. See OCHA, above note 25, p. 163.
- 137. M. Sassòli, above note 1, p. 405 (emphasis added).