Introduction

The problem of seemingly endless war – or “forever war”, in more colloquial terms – has perhaps never received more attention.1 While the problem exists globally, recent decade-long campaigns by Western States have shed new light on it. As lessons are learned about the need to avoid deadly and destructive conflicts at the outset, increased attention has focused on the necessity of putting an end to participation in conflicts currently under way. In the United States, three recent presidents struggled to extricate coalition forces from Afghanistan before a final exit was achieved in September 2021. In France, President Macron has repeatedly suggested his intention to pursue withdrawal of forces from Mali, where his country has been mired for over eight years. January 2021 also saw a near-complete withdrawal of US ground forces from Somalia, where those forces had been supporting counterterrorism operations against Al-Shabaab and the so-called Islamic State, while October 2019 saw President Trump's call for the exit of US forces from Syria. Also in the Middle East, the United Arab Emirates announced in July 2019 its military withdrawal from Yemen, where it had supported Saudi Arabian forces, though reports differ as to that withdrawal's completion.

This article focuses on the international legal obligations relevant to States when withdrawing from situations of armed conflict. When and how States end situations of armed conflict are questions no less vexing and important to the individuals and societies impacted by those decisions than questions of when and how States go to war. At the same time, the obligations applicable to States2 as they wind down their participation in armed conflict have not yet received significant scholarly or practical attention. While a growing literature has focused on precisely when armed conflicts come to a legal end,3 as well as obligations triggered by the cessation of active hostilities,4 comparatively little attention has been paid to the situation of States withdrawing from contexts of armed conflict generally, and indeed the contours of the obligations relevant to States in doing so. Perhaps one explanation for this gap is that, at least for some, the notion of legal regulations attending to the end of armed conflict may not be initially intuitive. Any end to war, looked at from one perspective, will always justify the means.5 At the same time, recent examples nonetheless show that the means by which States end armed conflicts also matter. When States extricate themselves from conflict with little care, significant and unnecessary suffering can result, as well as breaks in even the most basic metrics of accountability. Moreover, in many, though perhaps not all, cases, leaders make deliberate choices about how to withdraw, rendering analysis of the relevant rules and their consequences hardly theoretical.

This article therefore considers States’ obligations when withdrawing from armed conflicts from the perspective of international humanitarian law (IHL). In recent years, just war scholars have aimed to develop the moral principles applicable to the termination of war. Moellendorf has referred to these principles as “jus ex bello”,6 and Rodin has referred to them as “jus terminatio”.7 Picking up on these developments, a few IHL scholars have begun to consider jus ex bello from the perspective of ascertaining “when” a State must justly discontinue prosecuting an armed conflict, a question which may hinge on jus ad bellum and the reasons bringing a State to war in the first place.8 However, IHL scholars have not yet focused on the other core question framed by jus ex bello thinkers – namely, “how” a State should discontinue its involvement in a conflict. For instance, while Moellendorf has suggested minimizing civilian harm as a counterweight to the moral principle of “all due haste” when ending an unjust war,9 such principles, and others deriving from moral philosophy, have not been concretely studied in the context of existing IHL. By engaging, then, the question of “how” States should exit an armed conflict from the perspective of IHL, this article thus begins efforts to divine the legal content of those principles, ideally so as to inform further critical analysis and debate. While international human rights law and other bodies of law may also form part of the jus ex bello applicable to a State's withdrawal, this article limits its scope to IHL.

In distilling these relevant IHL rules, this article focuses on the factual situation of a State's exit from armed conflict, or what we refer to as a State's decision to “end its participation” in armed conflict. Focusing on the factual situation of a State's end of participation in armed conflict pays several dividends. First, it enables an analysis of the various types of obligations that States must consider as they exit an armed conflict – indeed, the IHL requirements corresponding to jus ex bello – as well as the challenges they may face in complying with them. In doing so, the article aims to provide a basic road map for States to consider when withdrawing from armed conflict, as well as to open up discussion about the sufficiency of the current IHL regime with regard to exits from armed conflicts, and the practical difficulties of operationalizing and conceptualizing the boundaries of obligations within that regime.

Second, focusing on States’ obligations at the end of their participation in armed conflict provides needed doctrinal clarification. At one level, considering a State's exit from armed conflict in its entirety shows how a State's withdrawal interacts with multiple legal and temporal triggers that are already well known in the literature, such as the end of armed conflict and the cessation of hostilities. Considering these temporal benchmarks together, and alongside additional under-appreciated obligations that are relevant as States end their participation in conflict, also provides insight into those IHL rules which continue to apply in the post-conflict period, even after a State may have withdrawn from an armed conflict, and the legal mechanics by which such rules continue to apply. Indeed, while it is commonly understood that IHL ceases to apply at the end of armed conflict, this article explains that this is in reality an over-simplification; a number of obligations actually endure. Moreover, while certain other IHL obligations are triggered at the “end of hostilities”, we show that these are not the only obligations which continue to take effect as a conflict winds down, and which continue to apply in the post-conflict period.

The article proceeds as follows. Part 1 begins by explaining contemporary approaches to the end of armed conflict and the applicability of IHL, focusing on the literature concerning the end of armed conflict and the temporal scope of IHL. Part 2 then offers a proposal for analyzing the obligations applicable to States as they exit armed conflict as obligations governing the “end of participation” in armed conflict. Doing so accounts for the fact that the end of a State's participation – for example, a withdrawal of forces – may not always coincide with the legal end of armed conflict, or at least it may not do so immediately. In many instances, whether the conflict continues will be unclear. More importantly, it also reflects the fact that a number of IHL obligations governing withdrawals in effect straddle the in bello and post bellum phases of conflict,10 which exiting States must take into account as they withdraw, regardless of where the conflict stands. Against this background, the article then develops three key end-of-participation obligations: obligations governing detention and transfer of persons, obligations imposed by Article 1 common to the four Geneva Conventions (common Article 1, CA 1), and obligations relating to accountability and the consequences of conflict. In doing so, the article shows that under some rules, States have duties to consider the foreseeable effects of their withdrawal, and potentially take pre-withdrawal precautionary measures. In other circumstances, States will need to take specific post-withdrawal remedial measures as factual situations require. Overall, however, the article suggests that these rules are relatively limited in nature and that further exploration and development are needed.

Enduring obligations and the end of armed conflict

Before turning to the substantive IHL rules relevant to States as they withdraw from situations of armed conflict, this part of the article first considers the applicability of IHL, or what is known as the “temporal scope” of IHL, focusing on the applicability of IHL towards the end of an armed conflict. This part and the next then elaborate on the various IHL rules relevant to a State's withdrawal from an armed conflict. Included in our discussion are those rules with specific temporal triggers, as well as those rules which operate as explicit or implicit exceptions to the general temporal applicability of IHL – that is, by continuing to apply even after an armed conflict has ceased.

Armed conflict and the “end” of IHL's applicability

As is well established in treaty and custom, the existence of an armed conflict triggers the application of the vast majority of IHL rules. In contrast to earlier IHL treaties, the Geneva Conventions and their Additional Protocols consider the existence of an armed conflict a factual matter, meaning that a formal declaration of war, or recognition of a state of war, is not needed in order for the Conventions and Protocols to apply.11 Indeed, with respect to international armed conflicts (IACs), Article 2 common to the four Geneva Conventions provides that the Conventions “shall apply to all cases of declared war or of any other armed conflict which may arise between two or more of the High Contracting Parties, even if the state of war is not recognized by one of them”.12 Article 6 of Geneva Convention IV (GC IV) further specifies that the Convention “shall apply from the outset of any conflict … mentioned in Article 2”. Additional Protocol I (AP I) takes a similar approach, with Article 3 providing that the Protocol “shall apply from the beginning of any situation” constituting an IAC.13

As with IACs, most rules applicable in non-international armed conflicts (NIACs) are also understood to apply from the onset of an armed conflict of this character. In this way, the temporal reach of the IHL rules pertaining to NIACs follows that of IACs in the sense that both sets of rules are understood to typically apply from the outset of a corresponding armed conflict. Notably, Article 3 common to the four Geneva Conventions (common Article 3, CA 3) provides for the application of certain IHL rules to “armed conflict … occurring in the territory of one of the High Contracting Parties”, thus making the presence of a NIAC the key trigger.14 While the identification of NIACs has given rise to some debate, under the prevailing test, which finds expression in the famous Tadić decision of the International Criminal Tribunal for the former Yugoslavia (ICTY) Appeals Chamber, a NIAC exists where there is “protracted armed violence between governmental authorities and organized groups or between such groups within a State”.15 Along similar lines, Additional Protocol II (AP II) applies from the onset of a qualifying armed conflict under that Protocol, specifically one taking place between a State Party and non-State armed groups “which, under responsible command, exercise such control over a part of [the State Party's] territory as to enable them to carry out sustained and concerted military operations and to implement this Protocol”.16

As is relevant in the context of this article, however, there is markedly less consensus with respect to the termination of IHL's applicability. The most express articulation of the end of IHL's application in IACs comes under GC IV Article 6, which states that “[i]n the territory of Parties to the conflict, the application of the present Convention shall cease on the general close of military operations”.17 This formulation is also followed in AP I Article 3, though neither treaty defines the terms “general close” or “military operations”, much less their relationship to “armed conflict”. Indeed, while AP I, like AP II, explicitly refers to the “end of armed conflict”, it does not explain the relevance or meaning of that term in relation to the application of the Protocol's provisions, and specifically the term “general close of military operations”.18 For its part, the International Committee of the Red Cross (ICRC) has described the general close of military operations in its Commentary to the Geneva Conventions as typically “the final end of all fighting between all those concerned”, where “military operations” are defined as “movements, manoeuvres and actions of any sort, carried out … with a view to combat”.19 Commentators have nonetheless come to understand the “general close of military operations” as essentially coincident with the end of “armed conflict”, thus making the end of armed conflict the critical switch-off point for IHL.20

For a variety of reasons, the question of exactly when an armed conflict ends, legally speaking, has in turn attracted significant controversy in recent years. As just noted, determining the end of an armed conflict under prevailing understandings means determining when, in general, IHL does and does not apply. Milanovic mentions the most basic conceptual approach to the end of armed conflict, and its relationship to the application of IHL. As he explains, “if a particular situation can no longer be qualified as an IAC, a NIAC, or an occupation, the application of IHL will end”.21 By contrast, historically, the main approach to determining the end of an IAC and the application of IHL focused on the existence of a peace treaty or declaration among belligerents,22 not, as today, on the presence of facts on the ground, such as the existence of any combat-related manoeuvres. In the Gotovina case, the Trial Chamber of the ICTY embraced such a factual approach to determining the end of armed conflict and the application of IHL. As it explained, it would “consider whether at any point during the Indictment period the international armed conflict had found a sufficiently general, definitive and effective termination so as to end the applicability of the law of armed conflict”, and in particular “whether there was a general close of military operations and a general conclusion of peace”.23 Contemporary commentators have considered similar factors, such as whether hostilities have ended with a degree of “stability and permanence”, when assessing whether an IAC has concluded.24

The challenge of determining the end of armed conflict is even more pronounced for NIACs. In comparison with references to the “general close of military operations” for IACs, the Geneva Conventions and AP II provide virtually no guidance regarding the temporal scope of IHL in NIACs, and in particular, the end of IHL application in NIACs. In the absence of such guidance, commentators have focused on the criteria giving rise to a NIAC, namely the intensity and organization thresholds in CA 3 and AP II.25 Milanovic has suggested that this approach appears “logical from a purely IHL standpoint” as it “take[s] into account the heightened NIAC intensity threshold when compared to IACs”.26 In the international criminal context, tribunals have suggested that IHL should extend in NIACs beyond the cessation of hostilities until a “peaceful settlement” has been achieved,27 although the meaning of “peaceful settlement” remains unclear. Along similar lines, the ICRC has advocated a more generous understanding of the application of NIAC rules, cautioning that NIACs should not be understood to conclude until there is a “complete cessation of all hostilities” and no “real risk of resumption”.28 Accordingly, uncertainty remains over the appropriate standard for determining the end of NIACs and how to apply it. The Commentary to AP II adds further confusion by noting that “rules relating to armed confrontation are no longer applicable after the end of hostilities”.29 Whether this suggestion is better read as meaning that IHL ceases to apply when a situation no longer constitutes a NIAC, as described above, or as its own standard, has been debated.30

Exceptions to the general temporal scope of IHL applicability

As the previous section has recounted, the existence of an armed conflict is commonly understood as the starting point for the onset of the applicability of IHL. For similar reasons, it is equally common to find suggestions with respect to the end of IHL's applicability that IHL ceases to apply upon the end of an armed conflict or the close of military operations, whenever those legal criteria are met in the circumstances.31 Nevertheless, qualification is needed, as in reality this position with respect to IHL's termination is a significant simplification. A number of exceptions, some explicit and others more implicit, extend the applicability of IHL outside of the strict confines of an armed conflict.

At the most basic level, it is well understood that one important exception to the general suggestion that IHL applies in circumstances of armed conflict relates to peacetime obligations. Article 2 common to the four Geneva Conventions (common Article 2, CA 2) makes explicit that even as the Conventions are generally oriented towards the regulation of armed conflict, they also set forth several obligations with which parties must comply during peacetime. In its precise formulation, CA 2 states that the Conventions apply with respect to situations of armed conflict “[i]n addition to the provisions which shall be implemented in peacetime”. The Geneva Conventions do not, however, specify directly what those peacetime obligations are. Several such peacetime obligations are in any case easily identified, including the obligation to disseminate the text of the Conventions and to include study of IHL in military and civilian instruction, as well as obligations concerning restrictions on the use of the emblems of the Red Cross and other markings, identification of medical aid societies, and preparations in relation to hospital zones.32 In relation to each of these, the text of the relevant Convention explicitly states that the rule applies “in time of peace”. However, identifying the full extent of obligations considered to apply in peacetime poses challenges.33 Indeed, not all peacetime obligations contain such explicit language, nor is their complete identification made in either the travaux34 or the Commentaries. The updated Commentaries, in comparison with the Pictet Commentaries, more comprehensively recognize the role of peacetime obligations, particularly in relation to the general rule of IHL applicability.35 According to the updated Commentaries, such obligations include the obligations to implement legislation providing penal sanction for grave breaches and to suppress violations of the Conventions, as well as to implement legislation to prevent misuse and abuse of emblems.36

Identifying the relevant peacetime obligations under the Additional Protocols presents further challenges, specifically in regard to AP I, which makes a caveat in its Article 3 that the Protocol applies from the beginning of armed conflict, “[w]ithout prejudice to the provisions which are applicable at all times” (emphasis added). While AP I, like the Geneva Conventions, explicitly identifies (on a rule-by-rule basis) certain rules as applying during times of “peace”,37 and thus “at all times”, not all rules understood to apply during peacetime are clearly identified in the text. The Commentary suggests a lengthy list of obligations under AP I which are understood to be applicable “at all times”,38 including rules pertaining to the development of new weapons, the activities of the Red Cross, legal advisers, precautions against attacks, the protection of land works and installations containing dangerous forces, and measures for the protection of journalists.39 The Commentary notes that these provisions can be categorized as either applicable from the entry into force of the Protocol, and thus in peacetime, or at least as “grounds” for taking proactive and precautionary measures prior to the outbreak of hostilities in line with obligations spelled out in the Protocol.40

As is particularly relevant to this article, there are two other critical exceptions to the general temporal rule for IHL which extend the applicability of certain IHL provisions even after the termination of an armed conflict. These exceptions are explicitly provided for in the Geneva Conventions and Additional Protocols.41 The first exception relates to protected persons deprived of their liberty. In relation to prisoners of war (PoWs), Article 5 of Geneva Convention III (GC III) states that the Convention shall continue to apply to protected persons “until their final release and repatriation”, a formulation followed in relation to protected persons in GC IV Article 6.42 Article 5 of Geneva Convention I (GC I) provides that for “persons who have fallen into the hands of the enemy”, IHL continues to apply, while AP I Article 3 likewise states that application of the Protocol ceases on the general close of military operations “except” with respect to persons whose “final release, repatriation or re-establishment takes place thereafter”. AP II Article 2(2) also provides protections “[a]t the end of the armed conflict” for “all the persons who have been deprived of their liberty or whose liberty has been restricted for reasons related to such conflict”, or whose liberty is restricted after the conflict for reasons related to it. As will be further explored in Part 3 below, these provisions are critical to guard against gaps in protection, particularly in light of what can often be severe delays in release and repatriation following the cessation of hostilities.

The second exception pertains to occupation. Under GC IV Article 6(3), “application of the present Convention shall cease one year after the general close of military operations; however, the Occupying Power shall be bound, for the duration of the occupation, to the extent that such Power exercises the functions of government in such territory”, by a number of provisions of the Convention. AP I Article 3 omits mention of a one-year extension but nonetheless applies the Protocol until “termination of the occupation”, except in relation to “persons whose final release, repatriation or re-establishment takes place thereafter”.

At the same time, it is clear that these two exceptions – for protected persons deprived of their liberty and for situations of occupation – are not the only exceptions to the general proposition that IHL “ceases” to apply at the end of armed conflict.43 Indeed, the more recent Commentaries to the Geneva Conventions and Additional Protocols appear to recognize the possibility that certain other, often unspecified, IHL obligations may also continue to apply after an armed conflict has terminated. For example, while the Pictet Commentary did not as explicitly suggest that possibility or identify such exceptions, the updated Commentaries to the Geneva Conventions now indicate with respect to the general temporal scope of the Conventions that “[t]he end of an armed conflict does not mean that humanitarian law will cease to apply entirely” and that “[s]ome provisions continue to apply after the end of armed conflict”, though they do not identify such provisions specifically.44 Meanwhile, while the Commentary to AP I provides a more lengthy list of “articles whose application in relation to a conflict may continue beyond the termination of this conflict”, it does not purport to provide a full list.45 It notes that such obligations may include obligations pertaining to missing persons, the remains of the deceased, the reunion of dispersed families, the evacuation of children, the repression of breaches of the Protocol, the duties of commanders, mutual legal assistance, and others.

Outside of the Commentaries, scholars have not yet considered in depth the existence or features of such obligations extending into the post-conflict period. When discussing the end of IHL's applicability, many commentators focus only on the exceptions for detention and occupation, or the relevance of “peacetime” obligations. As will be explored, such a characterization of the rules applicable in the aftermath of armed conflict may be incomplete. A number of the obligations which remain relevant cannot easily be categorized as purely peacetime or conflict rules. Some rules begin to apply during armed conflict and continue to apply into the post-conflict period, while others are more obviously suited to post-conflict application.

Obligations at the “end of participation”

Having recalled that certain IHL rules may extend after the end of armed conflict, the remainder of this article considers the rules which apply in relation to a State's end of participation in an armed conflict. Doing so illustrates the ways in which various IHL obligations relevant to a State's end of participation become relevant during armed conflict, and even after the cessation of hostilities or the end of armed conflict. By considering these obligations together, despite their differing temporal anchors, we aim to identify and analyze the obligations pertinent to a State's end of participation more collectively.

“End of participation” as a factual scenario

First, we must explain what we mean by the “end of participation”. A State's end of participation in a situation of armed conflict may have a number of legal consequences. At the most basic level, considering States’ obligations as they withdraw from situations of armed conflict in relation to the term “end of participation”, as opposed to the end of “armed conflict”, makes sense because many obligations relevant to States as they withdraw are not actually triggered by the legal end of “armed conflict”. Of course, the fact of a withdrawal may in some cases mean the legal end of an armed conflict. In such a case, the withdrawing State will no longer benefit from IHL rules applicable during conflict. Bombing enemy targets and killing enemy soldiers is permitted in times of armed conflict, for example, but not in “peacetime”. At the same time, whether or not an armed conflict has ended may often be unclear in the immediate post-withdrawal period. A close assessment of the facts is required, including of any ongoing operations continued by the withdrawing State. And particularly as many contemporary armed conflicts involve complex constellations of bilateral or trilateral adversarial relationships, whether a withdrawing State continues to support another actor may affect the status of armed conflict.

In any case, divining the precise end of armed conflict is not entirely relevant to the question of how a State must carry out its withdrawal and what obligations it may continue to possess after that withdrawal. Notably, an important category of end-of-participation obligations are instead triggered by the “cessation of hostilities”. While none of the relevant treaties define this term, the “cessation of active hostilities” may often precede the “general close of military operations” and thus the end of armed conflict, as described earlier.46 At the same time, distinguishing between the cessation of hostilities and the end of armed conflict is admittedly often difficult from a factual perspective,47 and in some cases, the two may coincide.48 As will be shown, States ending their participation in armed conflict will therefore need to consider whether their withdrawal causes an end of active hostilities in order to determine their responsibilities. In addition, focusing on a State's “end of participation”, as opposed to the end of conflict, makes sense as some obligations relevant to a State's withdrawal actually arise while a State is still participating in a conflict, and may continue to apply even after it has ended. Still other IHL rules are triggered on account of the potential foreseeable ramifications of the State's end of participation, such as its withdrawal of forces or other forms of support.

Additionally, orienting States’ obligation around this term helps clarify the legal implications of a State ceasing to be a “party” to an armed conflict. A key application comes in relation to “third States”. Where States provide a sufficient level of support to parties to an armed conflict, it has been suggested that IHL applies to those States by extension.49 When such a third State in turn withdraws support to the conflict party, it no longer itself constitutes a “party” to the conflict, and IHL ceases to apply.50 In commenting on this situation, Weizmann has suggested that the third State will at that point be required to release any detainees it has collected as part of its involvement in the conflict, just as if it were a primary party to the conflict.51 At the same time, the question of whether States remain bound by a broader corpus of obligations beyond releasing detainees remains. Accordingly, in referring to “end-of-participation” obligations, our aim is to address those obligations which any State incurs by having been a party to an armed conflict, and which that State owes at the temporal benchmarks specified in relation to each obligation.

Extension of end-of-participation obligations into the post-conflict period

Concretely, there are several types of rules which are relevant to a State's end of participation in an armed conflict. While these categories are not necessarily exclusive, it is helpful to identify key features of these obligations, particularly in relation to when they take effect, and the ways in which they apply in the post-conflict period.

As noted, the first and perhaps most obvious type of end-of-participation obligations are those triggered by the “cessation of active hostilities” or the “close of hostilities”. For example, GC III Article 118(1) states that “[p]risoners of war shall be released and repatriated without delay after the cessation of active hostilities”.52 Similarly, Article 10(1) of amended Protocol II to the Convention on Certain Conventional Weapons (CCW) calls for minefield removal “without delay after the cessation of active hostilities”.53 Such rules tethered to the cessation of hostilities are relevant not only at that point in time, it is argued, but instead can only reasonably be interpreted as continuing to apply after a conflict has concluded, particularly as provisions of this type do not contain any express temporal limitation. Indeed, as mentioned previously, reading each as not applying after the termination of armed conflict essentially leads to an absurd result,54 particularly as the cessation of hostilities can, in some instances, coincide with the end of armed conflict.55

Another category of end-of-participation obligations which are similar to those triggered by the cessation of hostilities includes those which take effect during an armed conflict, but which, often under a functional interpretation, continue to apply even after it ends. These rules most clearly straddle the in bello and post bellum phases of armed conflict: States are clearly required to attempt to carry them out during hostilities, and before the cessation of hostilities, though implementation may sometimes prove most feasible in the immediate post-conflict period. Examples include obligations pertaining to investigating and prosecuting breaches of IHL, searching for missing persons, and the repatriation and transfer of persons.56 In some cases, these rules make no mention of the end of armed conflict or hostilities (for example, obligations relating to investigation and prosecution), while in other cases they reference the cessation of hostilities as the latest point at which a State may undertake efforts to fulfil them (for example, the obligation to search for missing persons under AP I). A related, third type of end-of-participation obligations are those which are in effect always applicable, as is the case with CA 1, but which may become increasingly relevant as a State undertakes preparations for withdrawal on account of the foreseeable consequences of doing so. Other examples include rules relating to assisting children who are separated from their families and reuniting dispersed families, which States incur by virtue of participating in an armed conflict, but which are said to be owed by States in all circumstances. Such end-of-participation obligations thus may be distinguished from obligations more genuinely considered “peacetime” in nature or focus, such as dissemination.57 That is to say, in contrast to rules which always apply regardless of armed conflict, States incur such end-of-participation obligations, it is submitted, precisely because of their involvement in an armed conflict, and these rules then in turn continue into post-conflict peacetime.

Finally, another type of obligation falling under the “end of participation” umbrella and applying even after the end of armed conflict are those IHL rules which might be described as inherently “post-conflict” or “end-of-conflict” in nature. Such obligations include those regarding perfidy and surrender, reparations, gravesite maintenance, and obligations specifically tethered to the conclusion of “peace”.58 Each of these obligations logically extends into the post-conflict period by virtue of the issues that it pertains to.

Parts 3, 4 and 5 of this article discuss three substantive areas of such end-of-participation obligations, focusing on the ways by which they become relevant during, and after, the end of armed conflict.

IAC and NIAC Rules

Before moving on, one aside is needed with regard to conflict classification, which is particularly important for determining which rules apply. As is well appreciated, apart from CA 3 and AP II, IHL treaties are addressed primarily to States and specifically to IACs. Nevertheless, efforts have been made to develop those rules which apply in both IAC and NIAC – for example, through the comprehensive study of customary IHL undertaken by the ICRC.59 Accordingly, where a treaty obligation relevant at the end of participation qualifies as customary, as many discussed in this article do, it could regulate a State withdrawing from a NIAC.

A related issue is the extent to which IHL obligations bind non-State armed groups (NSAGs). This article focuses on the duties incumbent on States, though we highlight as an issue for further scholarship the extent to which certain end-of-participation obligations may apply to NSAGs. CA 3's application to NSAGs is indicated by the textual reference to “each Party to the conflict”, and although the text of AP II is not as explicit in this regard, its provisions are generally understood by scholars and the ICRC as applying to NSAGs as well.60 Various legal theories have been suggested as to why such groups may be bound by NIAC law, such as through the binding effect of customary rules or general principles.61 At the same time, whether NSAGs retain the capacity to carry out a particular duty designed for a State has been regarded as relevant in determining the scope of their obligations.62 This latter issue has been the subject of focus in discussions of NSAG duties relating to prosecution and internment, both of which are relevant at the end of participation.63

Rules Relating to Detention and Internment

A first set of obligations relevant as a State ends its participation in armed conflict pertains to persons deprived of their liberty, including both detainees and civilians. As discussed below, the Geneva Conventions and Additional Protocols set forth a number of obligations in relation to detainees and internees which provide a minimum standard of treatment across the in bello and post bellum phases of armed conflict. In IHL, “detention” is commonly defined as the custodial deprivation of liberty, whereas “internment” is the non-criminal deprivation of liberty by order of the executive branch, often referred to as “administrative detention”. While most detainee- and internee-related obligations have been articulated in treaties only in respect to IACs, the principles and rules in these treaties have been understood as providing guidance in determining State obligations in certain situations of NIAC as well.64

Duties regarding protection and release

An important issue which withdrawing States may need to first consider accordingly concerns the obligation to release and repatriate detainees and internees, as well as obligations to continue providing protections to such persons deprived of their liberty regardless of conflict status. In the first instance, obligations to release and repatriate such persons arise during the course of the armed conflict. They in turn become particularly relevant to withdrawing States at the cessation of hostilities, and are subsequently not extinguished by the termination of armed conflict.

With respect to civilian internees, GC IV Article 132 provides that interned persons “shall be released by the Detaining Power as soon as the reasons which necessitated [their] internment no longer exist”.65 At the latest, civilian internees must be released “as soon as possible” following the close of hostilities.66 Nevertheless, as GC IV Articles 132 and 133 make clear, the cessation of hostilities is an outer bound, and States should “endeavour during the course of hostilities … to conclude agreements” (emphasis added) for the release and repatriation of “certain classes of internees”, such as children, pregnant women, and the wounded and sick, as well as “internees who have been detained for a long time”. AP I Article 75(3) similarly requires that with respect to persons “arrested, detained or interned for actions related to the armed conflict”, such persons “shall be released with the minimum delay possible and in any event as soon as the circumstances justifying the arrest, detention or internment have ceased to exist”, except in cases of penal offences. Moreover, under AP I, the failure to release and repatriate persons is a serious matter, with unjustifiable delay amounting to a grave breach.67

With respect to PoWs, the cessation of active hostilities marks a critical temporal benchmark for the obligation to release and repatriate.68 Under GC III Article 118, prisoners “shall be released and repatriated without delay after the cessation of active hostilities”. Although there is less certainty regarding the laws relating to persons deprived of their liberty in NIAC as compared to IAC, such persons must be released “as soon as the reasons for the deprivation of their liberty cease to exist” under the customary rule.69 And while somewhat less clear, the close of hostilities is often taken as being the latest trigger point for release as well.70

For a withdrawing State, a critical first question, then, is whether as a result of the withdrawal, or related circumstances, the reasons justifying internment no longer exist. As Pejic observes, “[i]n view of the rapid progression of events in armed conflict, a person considered to be a threat today might not pose the same threat after a change of circumstances on the ground”.71 A second critical question for a withdrawing State is whether an act of withdrawal contributes to or coincides with the end of hostilities, thus prompting release obligations.72 And as a third point, a withdrawing State should also be aware that the obligations to release and repatriate remain in force after the end of hostilities, or even after the end of armed conflict.73 As mentioned previously, obligations tethered to the cessation of hostilities extend even into the post-conflict period, in part because interpreting them otherwise would not make sense in light of the close relationship between the cessation of hostilities and the end of armed conflict.74

While the rules governing detainees and internees call for release no later than the end of hostilities, they explicitly contemplate that some persons will not be released at that point, or even immediately thereafter. Indeed, the fact that such persons are afforded protections until the point of release, regardless of conflict status, is perhaps the clearest manifestation of the end-of-conflict (and even post-conflict) orientation of detainee- and internee-related rules. As noted previously, under the Geneva Conventions, persons deprived of their liberty remain protected until repatriation, release or re-establishment.75 AP I Article 75(6) similarly specifies that certain “fundamental guarantees” apply until “release, repatriation, or reestablishment”, even if this occurs “after the end of the armed conflict”. The extension of these protections therefore safeguards persons in cases where it might not be clear whether the conflict has continued, and thus, generally, whether IHL applies. It also ensures protection in light of challenges in the post-conflict application of these rules. As Crawford notes, for example, some detainees held in the aftermath of the Iran–Iraq War were not ultimately repatriated for decades.76 And while States’ concern with ensuring reciprocal release has led to significant delays by States in returning PoWs, the ICRC has emphasized that a State's obligation to release does not depend on the compliance of the other party.77

Tethered to the obligation to release is the obligation to repatriate,78 which is understood as prohibiting States from simply releasing PoWs without assisting in their return home or to the State in whose forces they served.79 Under GC III Article 118, repatriation must take place pursuant to a plan that typically specifies the means of repatriation, the detained persons involved, and a timeline.80 Krähenmann points to Iraq's release of Kuwaiti PoWs at the end of the Second Gulf War, in which it simply sent prisoners home through the desert, as a clear breach of this obligation.81 In discussing this obligation, the Eritrea–Ethiopia Claims Commission has suggested that “repatriation cannot be instantaneous” and that “[p]reparing and coordinating adequate arrangements for safe and orderly movement and reception … may be time-consuming”.82 Accordingly, a withdrawing State should be prepared to consider how it will ensure safe return in practice, and how it will abide by these obligations.

Transfer obligations

Another rule that may become particularly relevant as a State withdraws from armed conflict, and which also continues to apply in the post-conflict period, pertains to the transfer of detained persons. Under IHL, States are permitted during hostilities to transfer PoWs to co-belligerent, non-belligerent or neutral parties under certain circumstances,83 and when ending participation, States commonly seek to relieve themselves of detention-related duties by transferring detainees to other powers. Under GC III Article 12(2), PoW transfers may only be performed where the State “has satisfied itself of the willingness and ability of such transferee Power to apply the Convention”.84 Compliance with this provision accordingly requires withdrawing States to undertake, at a minimum, a “prior assessment” of how prisoners will be treated by the receiving State.85 While States often approach this obligation by seeking to conclude agreements regarding detainee treatment, such agreements and assurances, though important, may not be sufficient to prove compliance, particularly where “there is a systematic practice of non-compliance in the receiving country”.86 GC IV Article 45 imposes a similar obligation with respect to civilians, as well as adding that “[i]n no circumstances shall a protected person be transferred to a country where he or she may have reason to fear persecution for his or her political opinions or religious beliefs”. The GC IV Commentary further elaborates that States may not transfer persons where they “ha[ve] serious reason to believe that economic difficulties will prevent the receiving Power from providing for [the detainees’] maintenance” or where detainees “may be subjected to discriminatory treatment”.87

Transfer-related obligations apply during armed conflict and may continue to apply even afterwards. At the simplest level, the post-conflict continuation of such obligations can be said to stem from the fact that the Geneva Conventions and Additional Protocols make an exception to the general scope of IHL applicability for protected persons deprived of their liberty, as explained above.88 While the text of the relevant treaties does not directly link transfer obligations to this exception, it makes sense that States cannot simply discharge their obligations in relation to protected persons at the end of armed conflict by transferring them without the safeguards afforded by GC III Article 12(2). Other aspects of transfer obligations extending the reach of these obligations after the end of armed conflict are more subtle. For instance, one feature of GC III Article 12 which notably continues to apply post-conflict is found in Article 12(3), which provides that if a receiving State “fails to carry out the provisions of the Convention in any important respect” following the transfer, the transferring State, upon being “notified”, shall “take effective measures to correct the situation or shall request the return of the prisoners of war”. This rule thus applies even after a transfer has taken place, regardless of whether the transfer occurs during or after armed conflict. Another instance concerns the transferee State, which continues to be bound to afford GC III obligations to persons transferred to it, even if the State which made the transfer no longer remains in a state of armed conflict.89 An interesting discussion of this rule, involving its potential relevance in the post-conflict period, can be found in the case of General Manuel Noriega, who was captured during the US invasion of Panama in 1989, classified as a PoW, convicted in US federal court, and extradited to France in 2010.90 In approving his extradition, the US Court of Appeals for the Eleventh Circuit considered the requirements of Article 12 – well after the termination of the IAC in Panama – among other Convention provisions, in response to arguments raised by General Noriega's defense team.91

Lest liability for violation of these obligations with respect to transfer appear purely academic (particularly in the context of withdrawals), the Vukovar Hospital case before the ICTY encourages caution. The judgment paints in sharp relief armed forces’ obligations to consider the impact of troop withdrawal on the risk of harm to protected persons. The facts can be briefly summarized as follows. The indictment charged Colonel Mile Mrkšić and Major Veselin Šljivančanin of the Yugoslav People's Army (Jugoslovenska Narodna Armija, JNA) with war crimes, including murder, for their role in the 1991 evacuation of 194 Croat prisoners from the Vukovar Hospital to a hangar in Ovčara, where the prisoners were transferred to the custody of Serb paramilitary forces. After JNA forces departed from Ovčara, the paramilitaries mistreated and killed the prisoners. While the Trial Chamber acquitted Šljivančanin on the charge of murder on the basis that he ceased to be responsible for the prisoners after Mrkšić's withdrawal order, the Appeals Chamber held that Šljivančanin had aided and abetted murder through his failure to act, clarifying that his obligations had not concluded with the withdrawal order. Indeed, according to the Appeals Chamber, citing Article 12(2–3), Šljivančanin remained obliged “not to allow the transfer of custody of the prisoners of a war to anyone without first assuring himself that they would not be harmed”.92 While the case did not involve the issue of whether the armed conflict had ended, the Appeals Chamber's characterization of Šljivančanin's duty as “an ongoing duty to protect the [PoWs] at Ovčara” irrespective of the withdrawal order and JNA troop presence, suggests that transfer duties, while taking effect prior to withdrawal, persist after it.93

Similarly significant is the distinction drawn by the Appeals Chamber between Šljivančanin's knowledge prior to the withdrawal order and after it.94 The mens rea requirement was satisfied post-withdrawal order because Šljivančanin “knew that following the withdrawal … the killing of the prisoners of war was probable and that his inaction assisted the TOs [territorial defence forces] and paramilitaries”.95 In the case against Mrkšić, the Appeals Chamber likewise upheld the Trial Chamber's determination that Mrkšić knew the withdrawal order “left the TOs and paramilitaries with unrestrained access to the prisoners of war and that by enabling this access, he was assisting in the commission of their murder”.96 Accordingly, decision-makers engaged in prisoner transfer during a withdrawal must consider not only the likelihood of harm but also the effect of withdrawal on prisoner treatment.

In addition to clarifying the content of transfer obligations, Vukovar Hospital calls into focus the relevance of other international law doctrines – in particular, aiding and abetting and omission liability – in determining responsibility at the end of participation. Although the scope of omission liability remains underdeveloped in international criminal law, an omission can give rise to liability where an underlying duty to act is not fulfilled, and indeed an omission can also constitute a wrongful act for purposes of State responsibility.97 Many of the duties discussed in this article could be violated by an omission, and it is possible for third States to be responsible for complicity on the basis of them.

In assigning liability based on complicity, knowledge regarding the possibility of harm is often determinative. Whereas Vukovar Hospital concerned criminal mens rea, States may be responsible under Article 16 of the Draft Articles on State Responsibility where, among other requirements, they provide aid or assistance while being “aware of the circumstances making the conduct of the assisted State internationally wrongful”, and where the aid or assistance was “given with a view to facilitating the commission of the act”.98 Though uncertainty remains as to the precise content of this standard, consistent with the plain text of Article 16, it may be satisfied by “knowledge of the circumstances of the internationally wrongful act”.99 Along similar lines, depending on the facts and precise test adopted, a withdrawing State's awareness of impending harm precipitated by withdrawal could supply a basis for responsibility. For instance, as Jackson has suggested, complicity by omission can arise in circumstances where a State is provided with a clear and strong warning of impending harm to persons under its control.100 While we do not aim here to provide a full account of the relevance of omission liability at the end of participation, it is submitted that, consistent with Vukovar Hospital, omission liability could work to regulate the manner by which States engage in withdrawal efforts.

Human rights law and enduring obligations

Another manifestation of States’ enduring detention and internment obligations comes via certain IHL rules’ reference to international human rights law (IHRL). Through these references, such rules in effect anticipate the problem that, for persons affected, the end of armed conflict often will not immediately result in a return to safety, as well as IHL's unclear application towards the end of armed conflict.

It is well established that IHL and IHRL are not mutually exclusive in application,101 and commentators have indicated that, particularly with respect to detainees and internees, IHRL may serve a complementary and protective role.102 In the present context, the link to IHRL drawn by certain IHL provisions operates to provide a minimum standard of protection that bridges in bello and post bellum regimes, in effect reducing the importance of conflict status for certain State obligations. As an example, AP I Article 72 specifies that obligations pertaining to the “treatment of persons in the power of a party” are “additional to … other applicable rules of international law relating to the protection of fundamental human rights during international armed conflict”.103 In addition, AP I Article 75 provides certain “fundamental guarantees” that exist “as a minimum” and which continue to apply “even after the end of the armed conflict”.104 With respect to NIACs, CA 3 imposes obligations aimed at ensuring a minimum standard of humane treatment “at any time and in any place whatsoever”, and the Commentary to CA 3 refers to human rights law in describing the content of CA 3's prohibitions.105 More expressly, preambular paragraph 2 of AP II links IHRL and the Protocol, stating that “international instruments relating to human rights offer a basic protection to the human person”.106 As the Commentary notes, AP II is designed to include those “irreducible rights” which cannot be derogated from, and which accordingly “[correspond] to the lowest level of protection which can be claimed by anyone at any time”.107

These rules thus recognize that certain IHRL principles supply a minimum standard of protection that is always applicable, including where IHL may not otherwise govern. Moreover, because States’ obligations may be assessed prior to withdrawal under human rights and State responsibility regimes, withdrawing States may be obligated to consider, as part of their positive human rights duties, certain foreseeable effects on internees and detainees.108

The duty to ensure respect

A second source of obligations relevant as States end their participation in conflict is CA 1, which requires States to “undertake to respect and to ensure respect” for the Geneva Conventions. While CA 1 obligations are understood to in some sense always apply, thus including in peacetime, they are likely to become particularly relevant as a State discontinues its involvement in an armed conflict, particularly where fighting remains ongoing and where withdrawal may precipitate certain abuses. CA 1 is commonly regarded as imposing obligations on States in respect to NIACs as well as IACs.109

As indicated by a plain reading, CA 1 contains both a negative aspect – a requirement “neither [to] encourage, nor aid or assist” IHL violations – and a positive aspect – a requirement to “do everything reasonably in [the State's] power to prevent and bring such violations to an end”, i.e., to stop ongoing breaches and prevent future ones.110 In terms of whose IHL compliance States must ensure, CA 1 applies to a State's armed forces, civilian authorities and the general population. More debated is the extent to which CA 1 requires States to take steps to ensure IHL compliance by other entities, under what is referred to as the article's “external dimension”.111 While the ICRC has suggested that under the “prevailing view” States are indeed required to take measures in relation to other forces112 (a view supported by scholarly commentary,113 legal developments and subsequent practice114 ), the precise scope and content of the obligation remains subject to some uncertainty.115 In practice, positive obligations under CA 1 have been circumscribed based on a State's degree of influence over other parties, the foreseeability of violations, the gravity of the breach, and “the means reasonably available to the State”.116 Often focused on is a State's control over such forces, including ties in the form of “financing, equipping, arming or training”, which place a State in a “unique position to influence the [other forces’] behaviour … and thus to ensure respect”.117

Viewed from this perspective, it might seem that because withdrawal naturally coincides with the loss of traditional indicators of control (fewer troops on the ground and less resource expenditure, for example), exiting an armed conflict means cutting off responsibility. However, a full reading of CA 1 cautions against that view. Indeed, particularly relevant to withdrawing States is the obligation to “prevent violations when there is a foreseeable risk that they will be committed” or recommitted.118 The International Court of Justice (ICJ) has described similar duties in regard to prevention in the Bosnian Genocide case as dependent on essentially a “knew or should have known” standard of knowledge.119 Scholars have emphasized the likelihood of harm,120 the relevance of foreseeable risk,121 and the “certain degree of predictability” of the “prospective inobservance of IHL”.122 In addition, in regard to the negative obligation, the ICRC Commentary on GC III refers to the ICJ's recognition in the Nicaragua case of the obligation “not to encourage” other parties to commit violations,123 where the ICJ emphasized the circumstances of the alleged breach and particularly whether “the commission of [unlawful acts] was likely or foreseeable”.124 Accordingly, in order to comply with CA 1, a departing State may well need to consider the prospective effects of its withdrawal and undertake efforts to minimize particularly foreseeable and severe harms.

In terms of the nature and extent of measures required by the withdrawing State, the positive aspect of CA 1 is a duty of due diligence, meaning an obligation of means. A State is not required to achieve specific outcomes.125 In this respect, as indicated in the ICRC Commentary, the duty has been understood as deriving content from the ICJ's holding that a due diligence obligation is “one of conduct and not one of result, in the sense that a State cannot be under an obligation to succeed, whatever the circumstances”.126 Rather, States are obliged “to employ all means reasonably available to them, so as to prevent [a breach] so far as possible”.127 Commentators have also carefully circumscribed the external obligation's scope, emphasizing the importance of a State's capacity to prevent or halt the breach, and the seriousness of the breach.128 Kessler has noted, for instance, that positive action may only be required in “rare” cases to stop “severe” IHL breaches.129

In the context of withdrawals from armed conflict, exactly what is foreseeable may be difficult to determine. At the same time, where a withdrawal obviously precipitates substantial and clear risks of severe breaches, certain measures could be required. One example where due diligence obligations could be incumbent upon a withdrawing State as a result of the foreseeability of impending IHL violations comes in the context of joint detention or internment operations. In such circumstances, where a withdrawing State A abruptly removes support for partner State B, with the foreseeable result being the inability of State B to provide protections to persons previously held under both States’ jurisdiction, withdrawing State A may have an obligation to undertake measures to prevent offences, depending on the degree of predictability and severity. Another, more challenging case is potentially posed by the United States’ abrupt removal of support for the Kurdish-led Syrian Democratic Forces (SDF) in Northern Syria in October 2019.130 To the extent that the severe deterioration of SDF detention centres, which the United States had been supporting, and the related release of ISIS prisoners were foreseeable consequences of withdrawal, an appropriate step under CA 1 may have been for the United States to ascertain at the time of departure the minimum amount of assistance needed to help prevent such breaches.131

Other obligations stemming from CA 1 that might be regarded as more generally relevant at withdrawal concern the transfer of arms. When exiting a situation of armed conflict outside of its own territory, a State may aim to sell its weapons or provide them to a local partner State or NSAG that remains engaged in conflict. In such circumstances, CA 1 obligates the withdrawing State to make an assessment of the recipient's past compliance with IHL and its intention and capacity to ensure use of the weapons consistent with IHL.132 As the ICRC has explained, “States that transfer weapons can be considered particularly influential in ‘ensuring respect’ for international humanitarian law owing to their ability to provide or withhold the means by which violations may be committed”.133 It has even been suggested that where concerns remain after an examination of the risk, CA 1 creates a presumption against transfer authorization.134 Less well established, but perhaps equally important in terms of real-world consequences, is the situation where an exiting State hastily abandons weapons or leaves them to an entity that will predictably lose them to an IHL offender, as has become an issue in withdrawals.135 Given the recognized role of CA 1 in the context of transfer, it could potentially be argued that a State's CA 1 obligations extend to the issue of abandoned weapons. Any such obligation would of course be in addition to relevant treaty duties, in particular those pertaining to removal or destruction of mines and explosive remnants, as mentioned in Section 2.B of this article.

Following from the previously-identified limitations inherent to CA 1, it is important to mention that in light of the wide range of compliance measures available to States to satisfy CA 1,136 military commitment is not required.137 Indeed, States “remain in principle free to choose between different possible measures, as long as those adopted are considered adequate to ensure respect”.138 Under CA 1, commonly identified measures include diplomatic pressure, public denunciation, and restrictions on commercial relations or public aid, all of which could be relevant in certain withdrawal scenarios.139 The Commentaries also suggest specific measures – namely “addressing questions of compliance within the context of a diplomatic dialogue”, “offering legal assistance” and supporting “training” – that are specifically geared towards promoting compliance via partners, which could be relevant to withdrawing States to the extent that their partner forces remain engaged in conflict.140 Cooperation with international organizations and humanitarian efforts offer additional means of enforcement,141 as of course does financial assistance to relevant parties. Accordingly, in our view, compliance with end-of-participation obligations does not require States to continue to prosecute a conflict, and indeed the Commentaries indicate that the need to comply with CA 1 obligations cannot constitute a justification for a use of force otherwise not permitted by international law.142

The array of options available to States in relation to ensuring respect for IHL is particularly important in the context of withdrawals, where States have decided (and for good reason) to depart from another territory and end a conflict. France, during its attempted wind-down of military operations in Mali, might consider, for example, supporting various international organizations and local partners that remain involved in stabilization and anti-terrorism-related efforts, as part of its compliance with CA 1.143 Similarly, in Afghanistan, the United States might have considered continued financial, intelligence and other support to the Afghan security forces, and particularly those holding detainees, to support CA 1 compliance.

Obligations to investigate and prosecute and other accountability-related rules

A third set of obligations that States need to consider when withdrawing from armed conflict consists of rules that can be categorized, generally, as pertaining to accountability for the consequences of armed conflict and actions taken during it. Such obligations are often inherently “end-of-conflict” in orientation, as they include duties to search for missing persons, account for the dead and provide information to families, investigate war crimes, prosecute offenders and make reparation. Although their continued applicability is not always specified expressly by the relevant treaties, these duties take effect during an armed conflict and continue to be owed after a conflict has ended. Indeed, the purpose of many of these rules is to address the consequences of armed conflict and offer redress for wartime conduct. This goal is often implementable at the near-end of conflict, or even after it,144 and is thus of particular relevance for withdrawing States. In other cases, as with, for instance, the removal of materials such as mines and explosive remnants, the obligation is more explicitly tethered to the “cessation of hostilities”.

The duty to investigate and prosecute

A first duty of this type incumbent on withdrawing States is the duty to investigate and prosecute war crimes committed during the conflict. Under the Geneva Conventions, States must search for persons alleged to have committed grave breaches of the Conventions and bring such persons before their national courts.145 Rule 158 of the ICRC Customary Law Study, described as applying to both NIACs and IACs, similarly provides that “States must investigate war crimes allegedly committed by their nationals or armed forces, or on their territory, and, if appropriate, prosecute the suspects”, and that they must investigate other war crimes over which they have jurisdiction.146 Unlike the exceptions to the general temporal scope of IHL specified for occupation or the continued protection of detainees, the obligation to investigate and prosecute is not mentioned as an exception to the end of IHL's application in the provisions circumscribing the Conventions’ temporal scope; instead, its extension into the post bellum context is implicitly provided by the rule itself and supported by its underlying purpose.

While in one sense this obligation is a “peacetime” obligation, in that it requires a State not party to a conflict to prosecute persons on its territory who it learns have committed IHL offences, the obligation applies with particular force to parties which have taken part in a conflict. Indeed, conflict parties are likely to have more extensive knowledge of the events that transpired and better access to relevant evidence. As the GC I Commentary explains, a State “should take action when it is in a position to investigate and collect evidence, anticipating that either it itself at a later time or a third State, through legal assistance, might benefit from this evidence”.147 The principle of complementarity underlying international criminal law is in a related sense tethered to this idea that States which are closest to the action triggering a violation should be the ones to carry out an investigation and prosecution in the first instance.148 The duty to investigate also applies with particular force to States ending their participation in a conflict on account of its connection to in-conflict duties related to violation prevention. As an example, in the field, AP I Article 87(3) requires commanders who are aware of breaches “to initiate such steps as are necessary to prevent such violations of the Conventions or this Protocol, and, where appropriate, to initiate disciplinary or penal action against violators thereof”. In a similar vein, IHL obligations to take precautions in attack, for example, are not limited “to the avoidance of war crimes” but are instead “governed by the need to take constant care to spare the civilian population” and to avoid collateral damage – considerations which are specific to States participating in conflict.149

As for the outer temporal scope of the obligation to investigate and prosecute, it is generally understood that the obligation remains even after a State has ended its participation in a conflict. While the AP I Commentary identifies the duty to investigate and prosecute as one that “applies at all times”,150 the GC I Commentary suggests that the obligation to investigate and prosecute grave breaches “does not contain a specific time frame for the performance of the [obligation]”, but instead that “it is implicit in the text that States Parties should act within a reasonable time”.151 In terms of methods of implementation, provisions in peace agreements are one common way by which States effectuate the duty to investigate and prosecute – a practice reflective of how such obligations, in some cases, might realistically be expected to take effect once a conflict has abated. For instance, the 2016 peace deal between the Colombian government and the Revolutionary Armed Forces of Colombia (Fuerzas Armadas Revolucionarias de Colombia, FARC) established the Special Jurisdiction for Peace tribunal, in which FARC leaders are being prosecuted for their role in killings, sexual violence, kidnappings and other crimes.152

Other clues to the post-conflict application of this obligation come in relation to IHL rules on amnesties and statutes of limitations for war crimes. With respect to grave breaches, all four Geneva Conventions provide that a State may not absolve itself or any other State Party “of any liability incurred by itself or by another High Contracting Party in respect of [such] breaches”.153 It has also become clear in recent years that amnesty provisions in peace agreements cannot preclude prosecution for war crimes, as such would amount to an abrogation of IHL obligations. As the ICRC has articulated, “amnesties … would be incompatible with States’ obligation to investigate and, if appropriate, prosecute alleged offenders”.154 Indeed, the ICRC Customary Law Study includes various examples of States issuing amnesties for crimes committed in NIACs, with the ICRC emphasizing that “these have often been found to be unlawful by the [States’] own courts or by regional courts and were criticized by the international community”.155 Along similar lines, statutes of limitations for war crimes are also understood to be customarily prohibited.156 Moreover, in terms of implementation, commissions of inquiry are often set up during or even after a State ends its participation in a conflict. As Cohen and Shany point out, after serious allegations of war crimes were made against Canadian forces participating in the Unified Task Force in Somalia between 1992 and 1993, an independent public inquiry was concluded in 1997.157 Similarly, after the war in Yugoslavia reached an end in 1995, the Netherlands commissioned a study of Dutch involvement in the Srebrenica massacre.158

Nevertheless, as indicated by recent practice, compliance challenges remain. The United States’ withdrawal of the majority of troops from Somalia in early 2021, with armed conflict still under way, has prompted questions about whether the United States will continue to investigate and address the consequences of certain air strikes.159 And in relation to the conflict in Afghanistan, in March 2020 the International Criminal Court (ICC) Appeals Chamber authorized the commencement of an investigation into alleged war crimes by US, Afghan and Taliban forces.160 In the wake of the ICC's Afghanistan inquiry, which began in 2006, as well as public pressure, other States have undertaken investigations into the conduct of their own forces.161

Duties related to missing persons, the dead and others

In addition to the duty to investigate and prosecute, other end-of-participation obligations which function similarly are the obligations to search for missing persons, assist with the reunion of families and maintain gravesites. Indeed, as with the duty to investigate and prosecute, these obligations are not explicitly mentioned within the general articles describing the temporal scope of the Geneva Conventions and Additional Protocols. Instead, their post-conflict relevance appears to stem from the application of each rule in light of its underlying purposes. That said, some related obligations contain more explicit references to their temporal scope. For example, GC IV Article 24 requires States to take measures “in all circumstances” to assist children who are separated from their families, while AP I Article 74 obligates States to “facilitate in every possible way” the reunion of dispersed families.

Compared with such provisions, the obligations to search for missing persons, as set forth in multiple provisions of the Geneva Conventions and the Additional Protocols, are more implicitly ongoing in nature as no temporal restriction as to their application is explicitly provided in relation to them. Under the Geneva Conventions, the obligation to search for missing persons is understood to be provided somewhat indirectly, as the Conventions call on parties to, for instance, establish an information bureau to collect and relay knowledge about protected persons, as well as to respond to enquiries.162 A clearer formulation of the obligation to search for missing persons is found in AP I Article 33(1), which specifies that parties must commence searches for missing persons “[a]s soon as circumstances permit, and at the latest from the end of active hostilities”.163 While AP I Article 33(1) thus provides a beginning point for the obligation to search, the provision is not limited in temporal scope, and nor is the ICRC's corresponding formulation of the customary obligation on States to search for missing persons.164 Indeed, as Petrig has explained in describing missing persons obligations, such obligations “regulate phenomena originating in or resulting from an armed conflict or occupation, but the effects of which extend beyond the conclusion of these situations”.165 Notably, with respect to AP I Article 33(1), a draft proposal stating that search activities must continue without temporal limit was rejected because such a provision was believed to be “implicit in the paragraph”.166

While they are sometimes characterized as “peacetime” obligations, it is important to recall that such rules in relation to missing persons contemplate State action as soon as feasible during the armed conflict, even if that action is likely to be carried out towards the end of hostilities or a conflict. In this way, these rules constitute a clear type of obligation that straddles the in bello and post bellum boundary. For instance, as indicated by the phrase “as soon as circumstances permit”, the search rules effectively anticipate that compliance will only be feasible at certain times due to the challenges associated with search activities. Nevertheless, States cannot delay indefinitely, and must periodically assess when these activities can be undertaken. The AP I Commentary reiterates that although States retain “great latitude” with respect to the feasibility of search efforts during hostilities, they must assess feasibility at regular intervals and must not delay until hostilities have ceased.167 Hence, a withdrawing State should be aware of the continuing nature of search-related duties and, in view of these paradigms, may need to consider whether its withdrawal will facilitate or hinder the feasibility of its searches, as well as whether withdrawal also marks the end of hostilities.

While the rules with respect to searching for missing persons thus reflect a preference to undertake efforts before hostilities have ended, they remain applicable afterwards, to the extent that such efforts have not yet been completed. Other rules related to searches also continue to apply post-conflict, with their post-conflict orientation being more expressly suggested. For example, GC IV Article 133, which concerns searching for displaced internees, directly contemplates that committees may be established by agreement “after the close of hostilities, or of the occupation of territories, to search for dispersed internees”.168 While AP I Article 74, which obliges parties to facilitate the reunion of families “dispersed as a result of armed conflicts”, does not directly specify its temporal applicability, the Commentary to AP I recognizes its continued application after the termination of conflict.169 In comparison, other search-related rules, particularly in relation to the wounded and dead, apply “[w]henever circumstances permit, and particularly after an engagement”, a formulation that may also become relevant where a State's withdrawal coincides with the end of an engagement.170

The obligations related to gravesites are also not limited in their temporal scope. Article 34 of AP I requires States to facilitate access and to conclude agreements to protect and maintain the gravesites “permanently”.171 GC I Article 17(3) likewise requires that graves be respected, marked and maintained “so that they may always be found” (emphasis added). As explained in the Pictet Commentaries, the key point about the marking of gravesites is “that it should always be possible to find the grave of any combatant”.172 Accordingly, where a State's withdrawal prompts the cessation of hostilities, or where the withdrawal would impact the ability to search for missing persons or account for the dead, the withdrawing State may be required to increase such efforts.

In addition, withdrawing States should be aware that they may be expected to continue abiding by these obligations in the post bellum period. For instance, following ceasefires reached in the Nagorno–Karabakh conflict, the ICRC has recalled the need to identify missing persons and to assist in their return from conflict areas, in accordance with IHL.173 And even decades after the conflict in Yugoslavia, Balkan states have undertaken efforts regarding missing persons, such as multilateral commitments to update missing persons records and ensure public access to key information.174 Such obligations in relation to the missing and care of the dead are thus some of the more obvious examples of conflict-incurred obligations which States must consider at withdrawal, and even after.

Conclusion

As frequently as States enter armed conflicts, they must also exit them. Withdrawals from conflicts in recent years have called attention to the need to consider how States do so. This article has shown that the laws of armed conflict are relevant to this question, imposing a series of enduring obligations aimed at avoiding unnecessary suffering and ensuring certain minimal measures of accountability. Some of these duties are tethered to explicit temporal triggers set forth in IHL treaties, while others are only more implicitly relevant during, and even after, withdrawal. In all cases, such rules supply the minimum content of States’ IHL obligations when transitioning between the in bello and post bellum phases of armed conflict. The goal of this article has therefore been to offer guidance on how these IHL rules relevant to a State's end of participation operate, and in doing so, to invite discussion as to the IHL content corresponding to jus ex bello notions.

One area open for further analysis concerns how States are to reasonably comply with their end-of-participation obligations. To a certain extent, the precise point at which the various end-of-participation obligations discussed in this article no longer remain relevant remains open to debate, particularly given the potentially very long-lasting application of certain obligations, and the potential difficulties associated with interpreting concepts like foreseeable effects too broadly. Accordingly, for each of the rules discussed here and any others, further development of just what minimal efforts are required for compliance is appropriate. Importantly, the literature on CA 1, as explained above, has highlighted the myriad ways – diplomatic, economic and political – by which States can achieve compliance, providing a starting point for determining which means are appropriate in the facts of any particular withdrawal effort. At the same time, achieving compliance with certain such rules may well be difficult, and leaders must accordingly internalize that risk when entering conflict.

Further research might also focus on the relevance of these rules to NSAGs, including under what circumstances a State could be considered responsible for NSAG breaches. This issue is particularly important given that contemporary State withdrawals from armed conflict often involve the continuation of the armed conflict by other actors, including NSAGs, many of whom may have partnered with withdrawing States.

Finally, and most importantly, this article invites further reflection on the adequacy of the IHL rules corresponding to jus ex bello principles. The rules discussed in this piece pertaining to the end of participation are of relatively limited nature and indeed address only a fraction of considerations relevant to exiting armed conflicts, let alone those relevant to building post bellum peace and respect for human rights. Recent events have nonetheless shown both the necessity of putting an end to situations of armed conflict (as well as the grave consequences that can result in the course of doing so) and the need for measures of accountability. By beginning to operationalize the IHL rules pertaining to the end of participation, this article has accordingly sought to better understand their application to challenging factual situations, and by doing so, encourage future efforts aimed at their improvement.

• 1The research in this article was finalized for publication in September 202. In the literature, see, generally, the thematic issue of the International Review of the Red Cross on “Protracted Conflict”, Vol. 0, No. 92, 209.
• 2This article focuses on the obligations owed by States when withdrawing from international and non-international armed conflicts. As indicated in Section .C, some of the obligations discussed in this article may also apply to non-State armed groups (NSAGs), although we leave this issue for future scholarship.
• 3See Section 1.A.
• 4See, for example, Nathalie Weizmann, “The End of Hostilities Versus the End of Armed Conflict”, Lawfare, 28 May 2015, available at: www.lawfareblog.com/end-active-hostilities-versus-end-armed-conflict-%C… (all internet references were accessed in November 2021); Nathalie Weizmann, “The End of Armed Conflict, the End of Participation in Armed Conflict, and the End of Hostilities: Implications for Detention Operations under the 2001 AUMF”, Columbia Human Rights Law Review, Vol. 7, No. 3, 2016; Bettina Scholdan, “‘The End of Active Hostilities:’ The Obligation to Release Conflict Internees Under International Law”, Houston Journal of International Law, Vol. 38, No. 1, 2016.
• 5See Darrel Moellendorf, “Two Doctrines of Jus ex Bello”, Ethics, Vol. 12, No. 3, 201, pp. 63–69 (explaining the potential relationship between an unjustly begun armed conflict and its continued prosecution).
• 6Moellendorf defines jus ex bello as “the set of considerations that govern whether a war, once begun, should be brought to an end and if so how”. Darrel Moellendorf, “Jus ex Bello”, Journal of Political Philosophy, Vol. 1, No. 2, 2008, p. 123. Moellendorf and Rodin distinguish between two jus ex bello considerations, namely “whether” and “how” a conflict should be brought to an end. See ibid.; David Rodin, “Ending War: A Response to Richard W. Miller”, Ethics & International Affairs, Vol. 25, No. 3, 2011, p. 30. The focus of the present article, however, is only on the “how” question. See also D. Moellendorf, above note 5; Darrel Moellendorf, “Jus ex Bello in Afghanistan”, Ethics & International Affairs, Vol. 25, No. 2, 2011.
• 7Rodin describes jus terminatio as pertaining to questions of “when it is obligatory to terminate a state of war and how this can be done in the morally best way”. D. Rodin, above note 6, pp. 359–360. See also David Rodin, “The War Trap: Dilemmas of Jus Terminatio”, Ethics, Vol. 125, No. 3, 2013.
• 8See Gabriella Blum and David Luban, “Unsatisfying Wars: Degrees of Risk and the Jus ex Bello”, Ethics, Vol. 125, No. 3, 2015. See also Michael Walzer, “The Aftermath of War: Reflections on Jus Post Bellum”, lecture, Berkeley Center for Religion, Peace and World Affairs, 22 April 2010.
• 9D. Moellendorf, above note 5, p. 670.
• 10For more on jus post bellum, see, for example, Jens Iverson, Jus Post Bellum: The Rediscovery, Foundations, and Future of the Law of Transforming War into Peace, Brill, Leiden, 2021. As is relevant in the context of this article, Moellendorf distinguishes jus ex bello from jus post bellum by explaining that jus post bellum is primarily concerned with post-conflict order and responsibilities, whereas jus ex bello concerns the conditions for ending a conflict. D. Moellendorf, “Jus ex Bello”, above note 6, pp. 130–131. Rodin similarly distinguishes his term jus terminatio from jus post bellum, noting that the latter “governs conduct in the subsequent postwar state” while the former “governs the transition from a state of war back into a state of peace”. D. Rodin, above note 6, p. 360.
• 11See, for example, International Committee of the Red Cross (ICRC), Commentary on the Second Geneva Convention: Convention (II) for the Amelioration of the Condition of Wounded, Sick and Shipwrecked Members of Armed Forces at Sea, 2nd ed., Geneva, 2017 (2017 Commentary on GC II), paras 231–233.
• 12See Geneva Convention (I) for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field of August 1949, 75 UNTS 31 (entered into force 21 October 1950) (GC I); Geneva Convention (II) for the Amelioration of the Condition of Wounded, Sick and Shipwrecked Members of Armed Forces at Sea of August 1949, 75 UNTS 85 (entered into force 21 October 1950) (GC II); Geneva Convention (III) Relative to the Treatment of Prisoners of War of August 1949, 75 UNTS 135 (entered into force 21 October 1950) (GC III); Geneva Convention (IV) relative to the Protection of Civilian Persons in Time of War of August 1949, 75 UNTS 287 (entered into force 21 October 1950) (GC IV).
• 13See Protocol Additional (I) to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of International Armed Conflicts, 1125 UNTS 3, 8 June 1977 (entered into force 7 December 1978) (AP I).
• 14See Gabriella Venturini, “The Temporal Scope of Application of the Conventions”, in Andrew Clapham, Paola Gaeta and Marco Sassòli (eds), The 1949 Geneva Conventions: A Commentary, Oxford University Press, Oxford, 2015, p. 59.
• 15ICTY, The Prosecutor v. Duško Tadić, Case No. IT-94-1-AR72, Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction, 2 October 1995, para. 70. AP II Article 1 sets a similar, albeit higher, threshold.
• 16AP II, Art. 1(1).
• 17Emphasis added. The other Geneva Conventions are less explicit as to the end of their general applicability. See GC III, Art. 5; GC II, Art. 2; GC I, Arts 2, 5.
• 18See AP I, Arts 75(6), 99(1); AP II, Arts 2(2), 25(1).
• 19ICRC, 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 (2016 Commentary on GC I), paras 278–279; see also N. Weizmann, “The End of Armed Conflict”, above note 4, p. 221 (discussing the definition provided in the AP I Commentary).
• 20While the 1958 Commentary on GC IV interpreted the general close of military operations as the “final end of all fighting between all those concerned”, the updated Commentary explains that “evidence that there has been a ‘general close of military operations’ is the only objective criterion to determine that an international armed conflict has ended in a general, definitive, and effective way”. 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, 1958 (1958 Commentary on GC IV), p. 62; 16 Commentary on GC I, above note 19, paras 277–278. See also Marko Milanovic, “The End of Application of International Humanitarian Law”, International Review of the Red Cross, Vol. 96, No. 893, 15, p. 174. For more on the temporal applicability of IHL, see Julia Grignon, L'applicabilité temporelle du droit international humanitaire, Schulthess Éditions Romandes, Geneva, 14, pp. 7–408.
• 21See M. Milanovic, above note 20, p. 170. Milanovic notes the general rule on IHL applicability as well as a few of the exceptions to this general position.
• 222016 Commentary on GC I, above note 19, para. 275.
• 23ICTY, Prosecutor v. Gotovina, Case No. IT06-90-T, Judgment (Trial Chamber), 15 April 2011, para. 1694 (emphasis added).
• 24M. Milanovic, above note 20, p. 171. See also Robert Kolb and Richard Hyde, An Introduction to the International Law of Armed Conflicts, Hart, Oxford, 2008, p. 102.
• 25M. Milanovic, above note 20, pp. 178–179; N. Weizmann, “The End of Armed Conflict”, above note 4, p. 223; Nathan Derejko, “A Forever War? Rethinking the Temporal Scope of Non-International Armed Conflict”, Journal of Conflict and Security Law, Vol. 26, No. 2, 2021, p. 14. See also Rogier Bartels, “From Jus in Bello to Just Post Bellum: When Do Non-International Armed Conflicts End?”, in Carsten Stahn, Jennifer S. Easterday and Jens Iverson (eds), Jus Post Bellum: Mapping the Normative Foundations, Oxford University Press, Oxford, 2014, pp. 303–309.
• 26M. Milanovic, above note 20, p. 180.
• 27ICTY, Tadić, above note 15, para. 70; International Criminal Court (ICC), Prosecutor v. Jean-Pierre Bemba Gombo, Case No. ICC-01/05-01/08, Judgment Pursuant to Article 74 of the Rome Statute (Trial Chamber), 21 March 2016, para. 141.
• 28See ICRC, International Humanitarian Law and the Challenges of Contemporary Armed Conflict, October 2015, pp. 10–11. See also Dustin A. Lewis, Gabriella Blum and Naz K. Modirzadeh, “Four Theories on the End of Non-International Armed Conflict”, in Indefinite War: Unsettled International Law on the End of Armed Conflict, legal briefing, Harvard Law School Program on International Law and Armed Conflict, February 2017, Chap. 8, available at: https://pilac.law.harvard.edu/indefinite-war (discussing approaches).
• 29Yves Sandoz, Christophe Swinarski and Bruno Zimmermann, Commentary on the Additional Protocols, ICRC, Geneva, 1987, paras 4491–4492 (ICRC Commentary on AP I/AP II) (emphasis added).
• 30The former view, that IHL ceases to apply at the end of armed conflict, is more consistent with prevailing understandings, particularly given the absence of such language in the text of AP II, and the Commentary's note that the Protocol applies to qualifying NIACs from the “moment when the criteria laid down in [Article 1] are objectively fulfilled”, thus meaning that application of the Protocol is tethered to the existence of armed conflict. Ibid., para. 4491. For more on this issue, see J. Grignon, above note 20, pp. 334–338.
• 31See, for example, Emily Crawford, “The Temporal and Geographic Reach of International Humanitarian Law”, in Ben Saul and Dapo Akande (eds), The Oxford Guide to International Humanitarian Law, Oxford University Press, Oxford, 2020, pp. 60–65; R. Kolb and R. Hyde, above note 24, pp. 73 (“[T]he LOAC [law of armed conflict] is designed to apply only in a time of armed conflict and not in peacetime”), 74 (“[T]he applicability of the LOAC depends on the existence of an armed conflict. This is true for [IACs and NIACs] even if the definition of ‘armed conflict’ is not absolutely identical in both branches of the LOAC”); Jann K. Kleffner, “Scope of Application of International Humanitarian Law”, in Dieter Fleck (ed.), The Handbook of International Humanitarian Law, Oxford University Press, Oxford, 2021, p. 69, para. 3.23; Sandesh Sivakumaran, The Law of Non-International Armed Conflict, Oxford University Press, Oxford, 2012, p. 253 (“Ultimately, the applicability of the law of non-international armed conflict turns on whether or not a non-international armed conflict continues to exist at the relevant time. … Just as the law of non-international armed conflict commences upon the existence of certain facts (an armed conflict) so too should it cease to apply upon the non-existence of certain facts (the lack of an armed conflict)”). These sources recognize exceptions to the general position on IHL termination to varying degrees.
• 32See GC I, Arts 23, 26, 44, 47; GC II, Arts 44, 48; GC III, Art. 127; GC IV, Arts 14, 144.
• 33See Anna Petrig, “The War Dead and their Gravesites”, International Review of the Red Cross, Vol. 91, No. 874, 2009, pp. 363–364.
• 34Ibid.
• 352017 Commentary on GC II, above note 11, paras 221–222 (identifying various rules); ICRC, Commentary on the Third Geneva Convention: Convention (III) relative to the Treatment of Prisoners of War, 2nd ed., Geneva, 2020 (2020 Commentary on GC III), paras 232–233; 2016 Commentary on GC I, above note 19, paras 199–200.
• 362017 Commentary on GC II, above note 11, para. 221.
• 37AP I, Arts 6, 18, 60, 66, 83 (concerning, for example, training qualified personnel, emblems, demilitarized zones and dissemination of IHL).
• 38ICRC Commentary on AP I, above note 29, para. 149.
• 39AP I, Arts 36, 56, 58, 79, 81, 82.
• 40ICRC Commentary on AP I, above note 29, para. 149.
• 41See E. Crawford, above note 31, pp. 63–64.
• 42See also AP I, Art. 75(6).
• 43See M. Milanovic, above note 20, p. 170 (recognizing exceptions to the general statement that IHL ceases to apply at the end of armed conflict); N. Derejko, above note 25, pp. 10–11 (explaining that “neither the ‘end of hostilities’ nor the ‘end of NIAC’ necessarily equate to the termination of IHL's applicability”).
• 442017 Commentary on GC II, above note 11, para. 306; 2020 Commentary on GC III, above note 35, para. 317; 2016 Commentary on GC I, above note 19, para. 284. See also 2016 Commentary on GC I, above note 19, para. 941 (noting that while “some provisions of the Conventions and Protocols may cease to be applicable at the end of a conflict”, “many continue to apply even after that time”).
• 45ICRC Commentary on AP I, above note 29, para. 149.
• 46Ibid., para. 153; see also J. Grignon, above note 20, pp. 245–275 (discussing the cessation of hostilities in relation to both IACs and NIACs).
• 47See M. Milanovic, above note 20, pp. 172–174.
• 48Ibid., pp. 173–174; N. Weizmann, “The End of Armed Conflict”, above note 4, p. 233 (explaining that the cessation of hostilities and the end of a NIAC may coincide if the test for determining the end of a NIAC is taken to be the inverse of the test for its existence). Cf. J. K. Kleffner, above note 31, pp. 70, 77–78 (commenting on the relationship between ceasefires and the end of armed conflict).
• 49See N. Weizmann, “The End of Armed Conflict”, above note 4, pp. 225–226.
• 50See ibid.; Raphael Van Steenberghe and Pauline Lesaffre, “The ICRC's ‘Support-Based Approach’: A Suitable but Incomplete Theory”, Questions of International Law, May 2019, available at: www.qil-qdi.org/the-icrcs-support-based-approach-a-suitable-but-incompl….
• 51N. Weizmann, “The End of Armed Conflict”, above note 4, p. 232.
• 52GC III, Art. 118(1); see also Jean-Marie Henckaerts and Louise Doswald Beck (eds), Customary International Humanitarian Law, Vol. 1: Rules, Cambridge University Press, Cambridge, 2005 (ICRC Customary Law Study), Rule 128, available at: https://ihl-databases.icrc.org/customary-ihl/eng/docs/v1.
• 53CCW Protocol (II) on Prohibitions or Restrictions on the Use of Mines, Booby-Traps and Other Devices, 2048 UNTS 133, Geneva, 10 October 1980 (amended 3 May 1996), Art. 10(1). See also CCW Protocol (V) on Explosive Remnants of War, UN Doc. CCW/MSP/2003/2, Geneva, 28 November 2003 (entered into force 12 November 2006), Arts 3(1–3) (imposing obligations regarding clearance, removal and destruction of explosive remnants “after the cessation of active hostilities”), 4(2).
• 54For a similar take on this point, see J. Grignon, above note 20, p. 367.
• 55See the references at above note 48.
• 56See Parts 3 and 5 of this article. For more on a functional approach to the temporal scope of IHL rules applicable in NIAC, see N. Derejko, above note 25, pp. 11–29.
• 57See, for example, GC I, Art. 47; GC III, Art. 127; GC IV, Art. 144; AP I, Art. 83.
• 58See, for example, Hague Convention (IV) respecting the Laws and Customs of War on Land and Its Annex: Regulations Concerning the Laws and Customs of War on Land, 205 CTS 227, 18 October 1907 (entered into force 26 January 1910), Art. 54 (“Submarine cables connecting an occupied territory with a neutral territory shall not be seized or destroyed except in the case of absolute necessity. They must likewise be restored and compensation fixed when peace is made”).
• 59See, generally, ICRC Customary Law Study, above note 52. See also ICTY, Tadić, above note 15, para. 134; S. Sivakumaran, above note 31, pp. 101–152; Marco Sassòli, “Taking Armed Groups Seriously: Ways to Improve Their Compliance with International Humanitarian Law”, Journal of International Humanitarian Legal Studies, Vol. 1, No. 1, 2010, pp. 16–17; Marco Sassòli and Marie-Louise Tougas, “International Law Issues Raised by the Transfer of Detainees by Canadian Forces in Afghanistan”, McGill Law Review, Vol. 56, No. 4, 2011, p. 970.
• 60ICRC Commentary on AP II, above note 29, paras 4442–4444; M. Sassòli, above note 59, p. 12.
• 61See M. Sassòli, above note 59, pp. 12–14 (summarizing predominant approaches); Andrew Clapham, “Focusing on Armed Non-State Actors”, in Andrew Clapham and Paola Gaeta (eds), The Oxford Handbook of International Law in Armed Conflict, Oxford University Press, Oxford, 2014, pp. 771–782.
• 62See A. Clapham, above note 61, pp. 781–782; M. Sassòli, above note 59, p. 16. See also Ezequiel Heffes, “Generating Respect for International Humanitarian Law: The Establishment of Courts by Organized Non-State Armed Groups in Light of the Principle of Equality of Belligerents”, Yearbook of International Humanitarian Law, Vol. 18, 2015 (focusing on the issue of equality of belligerents in determining NSAG IHL obligations).
• 63See, for example, A. Clapham, above note 61, pp. 781–782; Hannes Jobstl, “Bridging the Accountability Gap: Armed Non-State Actors and the Investigation and Prosecution of War Crimes”, Journal of International Criminal Justice, Vol. 18, No. 3, 2020, pp. 569–570, 590–593.
• 64As examples, see the discussion relating to release obligations and transfer duties in Sections 3.A and 3.B.
• 65See also ICRC Customary Law Study, above note 52, Rule 128(B).
• 66GC IV, Art. 133(1); see also GC IV, Art. 46(1).
• 67AP I, Art. 85(4)(b).
• 68At the same time, some prisoners, in particular certain wounded or sick prisoners, may need to be released and repatriated during hostilities. See GC III, Arts 109, 110, 114. In addition, an exception to the release rule exists for prisoners facing criminal proceedings for an indictable offence or serving a sentence. See, for example, GC III, Art. 119(5); ICRC Customary Law Study, above note 52, Rule 128.
• 69ICRC Customary Law Study, above note 52, Rule 128(C).
• 70See, for example, N. Weizmann, “The End of Armed Conflict”, above note 4, p. 248; ICRC Commentary on AP II, above note 29, para. 4493 (“In principle, measures restricting people's liberty, taken for reasons related to the conflict, should cease at the end of active hostilities … except in cases of penal convictions”).
• 71Jelena Pejic, “Procedural Principles and Safeguards for Internment/Administrative Detention in Armed Conflict and Other Situations of Violence”, International Review of the Red Cross, Vol. 87, No. 858, 2005, p. 382. See also GC IV, Arts 42 (specifying that a person may only be interned if “the security of the Detaining Power makes it absolutely necessary”), 43 (regarding review of internment decisions); AP I, Art. 75(3).
• 72Cf. N. Weizmann, “The End of Armed Conflict”, above note 4, p. 232.
• 73Cf. Sandra Krähenmann, “Protection of Prisoners in Armed Conflict”, in D. Fleck (ed.), above note 31, pp. 445–446.
• 74See the references at above note 48.
• 75GC I, Art. 5; GC III, Art. 5; GC IV, Art. 6; see also 1958 Commentary on GC IV, above note 20, pp. 64, 271, 515; 2020 Commentary on GC III, above note 35, para. 1093.
• 76E. Crawford, above note 31, p. 64.
• 772020 Commentary on GC III, above note 35, para. 4451; see also Marco Sassòli, “Release, Accommodation in Neutral Countries, and Repatriation”, in A. Clapham, P. Gaeta and M. Sassòli (eds), above note 14, pp. 1049–1050.
• 78GC III, Art. 118; ICRC Customary Law Study, above note 52, Rule 128(A); see also GC IV, Art. 134 (regarding civilian repatriation).
• 79S. Krähenmann, above note 73, pp. 445–446; M. Sassòli, above note 77, pp. 1040, 1057.
• 80See 2020 Commentary on GC III, above note 35, paras 4474–44. In NIACs, AP II Article 5(4) requires that where a State decides to release persons deprived of liberty, the State must take “necessary measures to ensure [the] safety” of these persons. See ICRC Commentary on AP II, above note 29, para. 4596 (suggesting that such safety measures should remain in place “until the released persons have reached an area where they are no longer considered as enemies, or otherwise until they are back home, as the case may be”). See, generally, Cordula Droege, “Transfers of Detainees: Legal Framework, Non-Refoulement and Contemporary Challenges”, International Review of the Red Cross, Vol. 90, No. 871, 2008, pp. 675–676.
• 81S. Krähenmann, above note 73, p. 445.
• 82Eritrea–Ethiopia Claims Commission, Partial Award: Prisoners of War – Eritrea's Claim 17, 1 July 2003, para. 147.
• 83See Keiichiro Okimoto, “Evacuation and Transfer of Prisoners of War”, in A. Clapham, P. Gaeta and M. Sassòli (eds), above note 14, p. 968.
• 84The ICTY, in the Vukovar Hospital case, considered this article in relation to obligations arising in NIAC. In addition, some States involved in NIACs have expressly assumed transfer obligations based on special agreements or declarations applying GC III. See M. Sassòli and M Tougas, above note 59, pp. 965, 975–981; K. Okimoto, above note 83, p. 973; see also C. Droege, above note 80, p. 675 (“[T]he humanitarian principle underlying these provisions, namely that a detaining power should ensure that the ally to whom it transfers detainees treats them according to the standards of the Geneva Conventions, should also be taken into account in non-international armed conflict”). In NIACs, transfer of a detainee of a non-State party to its own party would more properly qualify as a repatriation, and different rules, for example non-refoulement, may apply. See M. Sassòli and M. Tougas, above note 59, p. 979.
• 852020 Commentary on GC III, above note 35, para. 1535; see also K. Okimoto, above note 83, pp. 968–969; M. Sassòli and M. Tougas, above note 59, pp. 980, 998.
• 862020 Commentary on GC III, above note 35, para. 1537. Canada and Afghanistan, for instance, entered into agreements in 2005 and 2007 to abide by GC III PoW obligations, but these did not “foreclose the knowledge of Canadian forces members that the transfer would lead to the commission of the crime of torture by Afghan authorities”. See M. Sassòli and M. Tougas, above note 59, p. 1008.
• 871958 Commentary on GC IV, above note 20, p. 268.
• 88See Section 1.B of this article.
• 89See K. Okimoto, above note 83, p. 970 (“[E]ven when an IAC between state A and state B has ended, state C, which holds POWs transferred by state A or B, must apply GC III until their final release and repatriation”). See also 2020 Commentary on GC III, above note 35, paras 1546 (temporal scope of duties for transferee State), 1550 (temporal scope of duties for transferring State).
• 90See K. Okimoto, above note 83, p. 968.
• 91“Eleventh Circuit Approves General Noriega's Extradition to France”, American Journal of International Law, Vol. 103, No. 3, 2009, p. 5.
• 92ICTY, Prosecutor v. Mile Mrkšić and Veselin Šljivančanin (Vukovar Hospital Case), Case No. IT-95-13/1-A, Judgment (Appeals Chamber), 5 May 2009, paras 71–75. See M. Sassòli and M. Tougas, above note 59, pp. 1005–1008 (summarizing key facts). The Appeals Chamber appears to have understood the customary international law applicable in NIACs, reflected in CA 3, as including obligations in relation to transfer similar to those expressed in Article 12. See ibid., pp. 980–981.
• 93ICTY, Vukovar Hospital, above note 92, para. 75 (emphasis added).
• 94Ibid., paras 57–63.
• 95Ibid., para. 101.
• 96Ibid., paras 333–334.
• 97See, generally, Jessie Ingle, “Aiding and Abetting by Omission before International Criminal Tribunals”, Journal of International Criminal Justice, Vol. 14, No. 4, 2016 (summarizing international criminal tribunal case law); International Law Commission (ILC), Draft Articles on Responsibility of States for Internationally Wrongful Acts, with Commentaries, UN Doc. A/56/10, November 2001 (Draft Articles), p. 32, para. 1 (“An internationally wrongful act of a State may consist in one or more actions or omissions or a combination of both”). Omission liability also finds roots in the Geneva Conventions. For instance, GC III Article 13(2) prohibits “any unlawful act or omission … causing death or seriously endangering the health of a prisoner of war”, and the war crime of “wilful killing” has been taken to cover omission. See Michael Duttwiler, “Liability for Omission in International Criminal Law”, International Criminal Law Review, Vol. 6, No. 1, 2006, pp. 10–11.
• 98Draft Articles, above note 97, Art. 16, and Commentary on Art. 16, p. 66, paras 3–4.
• 99Ryan Goodman and Miles Jackson, “State Responsibility for Assistance to Foreign Forces (aka How to Assess US-UK Support for Saudi Ops in Yemen)”, Just Security, 31 August 2016, available at: www.justsecurity.org/32628/state-responsibility-assistance-foreign-forc….
• 100Miles Jackson, Complicity in International Law, Oxford University Press, Oxford, 2015, p. 157.
• 101See International Court of Justice ICJ), Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory, Advisory Opinion, 9 July 2004, para. 106.
• 102See, for example, J. Pejic, above note 71, pp. 378–379.
• 103See ibid., p. 378 (suggesting that AP I Article 72 therefore “allows recourse to human rights law as an additional frame of reference in regulating the rights of internees”, and that the “minimum” mentioned in Article 75, read in light of Article 72, “is supplemented by other provisions of humanitarian and human rights law”).
• 104AP I, Arts 75(1), 75(6).
• 105See, for example, 2020 Commentary on GC III, above note 35, paras 548–551, 654, 657–658, 709, 715, 745–746.
• 106The Commentary specifies that such instruments include UN human rights treaties as well as regional instruments. ICRC Commentary on AP II, above note 29, paras 4428–4430.
• 107Ibid., para. 4430.
• 108See, generally, Human Rights Committee, General Comment 36, “Article 6: Right to Life”, UN Doc. CCPR/C/GC/36, 30 October 2018, paras 7, 21–22 (explaining State obligations in relation to foreseeable harms caused by actors whose conduct is otherwise not attributable to the State); European Court of Human Rights, El Masri v. The Former Yugoslav Republic of Macedonia, Appl. No. 39630/09, Judgment, 12 December 2012, paras 215–223, 228.
• 109See, for example, ICJ, Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), Judgment, ICJ Reports 1986, paras 220, 255; Laurence Boisson de Chazournes and Luigi Condorelli, “Common Article 1 of the Geneva Conventions Revisited: Protecting Collective Interests”, International Review of the Red Cross, Vol. 82, No. 837, 2000, pp. 68–69; Clause Kreß, “The International Court of Justice and the Law of Armed Conflicts”, in Christian J. Tams and James Sloan (eds), The Development of Law by the International Court of Justice, Oxford University Press, Oxford, 2013, p. 275; ICRC Customary Law Study, above note 52, Rule 139.
• 110See 2020 Commentary on GC III, above note 35, para. 187 (illustrating positive and negative aspects of CA 1 in relation to the duty to ensure respect by others). See also Knut Dörmann and Jose Serralvo, “Common Article 1 to the Geneva Conventions and the Obligation to Prevent International Humanitarian Law Violations”, International Review of the Red Cross, Vol. 96, No. 895, 2014, pp. 728–731.
• 1112020 Commentary on GC III, above note 35, paras 186–206.
• 112Ibid., para. 153.
• 113See, e.g., Oona A. Hathaway, Emily Chertoff, Lara Domínguez, Zachary Manfredi and Peter Tzeng, “Ensuring Responsibility: Common Article 1 and State Responsibility for Non-State Actors”, Texas Law Review, Vol. 95, No. 3, 2017, pp. 540, 574; L. Boisson de Chazournes and L. Condorelli, above note 109; Birgit Kessler, “The Duty to Ensure Respect under Common Article 1 of the Geneva Conventions: Its Implications on International and Non-International Armed Conflicts”, German Yearbook of International Law, Vol. 44, 2001, pp. 498, 504, 516.
• 114See 2020 Commentary on GC III, above note 35, para. 189; K. Dörmann and J. Serralvo, above note 110, pp. 719–721.
• 115See 2020 Commentary on GC III, above note 35, paras 202–206.
• 116Ibid., paras 198–199.
• 117Ibid., para. 200.
• 118Ibid., para. 197.
• 119See ICJ, Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro), Judgment, ICJ Reports 2007, para. 431 (“[A] State's obligation to prevent, and the corresponding duty to act, arise at the instant that the State learns of, or should normally have learned of, the existence of a serious risk that genocide will be committed”). See also 2020 Commentary on GC III, above note 35, para. 199 (comparing the positive obligation under CA 1 to the duty to prevent genocide under Article 1 of the Genocide Convention).
• 120Hans-Peter Gasser, “Ensuring Respect for the Geneva Conventions and Protocols”, in Hazel Fox and Michael A. Meyer (eds), Armed Conflict and the New Law, Vol. 2: Effecting Compliance, British Institute of International and Comparative Law, London, 1993, pp. 30–32.
• 121Alexandre Devillard, “L'obligation de faire respecter le droit international humanitaire”, Revue Québécoise de Droit International, Vol. 20, No. 2, 2007, pp. 96–97. See also K. Dörmann and J. Serralvo, above note 110, pp. 729–730 (analyzing the approaches of Gasser and Devillard).
• 122K. Dörmann and J. Serralvo, above note 110, pp. 729–730.
• 1232020 Commentary on GC III, above note 35, para. 191; ICJ, Nicaragua, above note 109, para. 220.
• 124ICJ, Nicaragua, above note 109, paras 255–256.
• 1252020 Commentary on GC III, above note 35, paras 198–199.
• 126ICJ, Bosnian Genocide, above note 119, paras 430, 438.
• 127Ibid.
• 128See, for example, K. Dörmann and J. Serralvo, above note 110, pp. 724–725. See also B. Kessler, above note 113, p. 506 (suggesting that “[t]he intensity of the treaties’ violations is another element that is important for obliging the States to take further steps to ‘ensure respect’”).
• 129B. Kessler, above note 113, p. 506.
• 130See Beatrice Walton and Paul Strauch, “Three Lingering Question about the Legality of Withdrawal from Syria: Parts I and II”, Opinio Juris, 7–8 January 2020, available at: http://opiniojuris.org/2020/01/07/three-lingering-questions-about-the-l…; Charlie Savage, “The Kurds’ Prisons and Detention Camps for ISIS Members, Explained”, New York Times, 13 October 2019, available at: www.nytimes.com/2019/10/13/us/politics/isis-prisoners-kurds.html.
• 131See, generally, O. A. Hathaway et al., above note 113 (discussing the 2016 ICRC Commentary and State obligations to prevent violations of CA 3 by non-State actors).
• 132See 2020 Commentary on GC III, above note 35, paras 189, 195 (noting the reference to the duty to ensure respect in the 2013 Arms Treaty, and explaining that CA 1's negative obligation requires States “to refrain from transferring weapons if there is an expectation, based on facts or knowledge of past patterns, that such weapons would be used to violate the Conventions”). CA 1's relevance to arms transfer also has a positive dimension. See K. Dörmann and J. Serralvo, above note 110, pp. 732–734.
• 133ICRC, Arms Transfer Decisions: Applying International Humanitarian Law Criteria, Geneva, May 2007, p. 9.
• 134Ibid.
• 135See, for example, “U.S. Military Left ‘Thousands’ of Weapons ‘Vulnerable to Loss or Theft’ During Fight Against ISIS in Syria”, Newsweek, 2 February 2020, available at: www.newsweek.com/us-military-thousands-weapons-vulnerable-loss-theft-fi…; “Planes, Guns, Night-Vision Goggles: The Taliban's New U.S.-Made War Chest”, Reuters, 19 August 2021, available at: www.reuters.com/business/aerospace-defense/planes-guns-night-vision-gog….
• 136See, for example, 2020 Commentary on GC III, above note 35, para. 214; Umesh Palwankar, “Measures Available to States for Fulfilling Their Obligation to Ensure Respect for International Humanitarian Law”, International Review of the Red Cross, Vol. 34, No. 298, 1994; B. Kessler, above note 113.
• 137Indeed, CA 1 is not capable of providing a legal justification for military action. See 2020 Commentary on GC III, above note 35, para. 207.
• 138Ibid., para. 198.
• 139See U. Palwankar, above note 136, pp. 9, 12, 16–17; 2020 Commentary on GC III, above note 35, paras 211, 214–215.
• 1402020 Commentary on GC III, above note 35, para. 214.
• 141U. Palwankar, above note 126, pp. 12, 17–23; see also K. Dörmann and J. Serralvo, above note 110, p. 726.
• 142See 2020 Commentary on GC III, above note 35, para. 207. The IHL rules may therefore deviate from, or at least not provide explicit support for, some jus ex bello principles debated in just war theory, where authors have expressed support for the continuation of military action in certain contexts. Cf. D. Rodin, above note 6, pp. 359–361; M. Walzer, above note 8 (suggesting that parties to a conflict may have an obligation to continue fighting in order to prevent mass murder and atrocity).
• 143See, generally, “After Barkhane: What France's Military Drawdown Means for the Sahel”, European Council on Foreign Relations, 2 July 2021, available at: https://ecfr.eu/article/after-barkhane-what-frances-military-drawdown-m….
• 144See M. Milanovic, above note 20, pp. 174–175 (discussing in relation to IACs). Some have expressed doubt regarding the ability of obligations to extend past the end of a NIAC specifically. Nevertheless, other commentators emphasize a more functional approach, for instance by taking into account the triggering of the obligations during the conflict. See, for example, N. Derejko, above note 25, p. 11. Further, it is notable that the ICRC Customary Law Study rules relevant to the obligations in this section have no express temporal restriction.
• 145See GC I, Art. 49; GC II, Art. 50; GC III, Art. 129; GC IV, Art. 146; AP I, Arts 86, 88.
• 146ICRC Customary Law Study, above note 52, Rule 158.
• 1472016 Commentary on GC I, above note 19, para. 2871 (emphasis added).
• 148See, for example, ICC Office of the Prosecutor, “Informal Expert Paper: The Principle of Complementarity in Practice”, 2003, para. 1 (“The principle of complementarity is based both on respect for the primary jurisdiction of States and on considerations of efficiency and effectiveness, since States will generally have the best access to evidence and witnesses and the resources to carry out proceedings”).
• 149Mark Lattimer, “The Duty to Investigate Civilian Deaths in Armed Conflict: Looking Beyond Criminal Investigations”, EJIL: Talk!, 22 October 2018, available at: www.ejiltalk.org/the-duty-to-investigate-civilian-deaths-in-armed-confl….
• 150ICRC Commentary on AP I, above note 29, para. 149 (referring to Arts 85–87).
• 1512016 Commentary on GC I, above note 19, para. 2868.
• 152“Colombia Tribunal Begins Mammoth Task of Investigating, Trying War Crimes”, Reuters, 15 March 2018, available at: www.reuters.com/article/us-colombia-peace-court/colombia-tribunal-begin….
• 153GC I, Art. 51; GC II, Art. 52; GC III, Art. 131; GC IV, Art. 148.
• 154ICRC, “Amnesties and International Humanitarian Law: Purpose and Scope”, legal fact sheet, Geneva, 4 October 2017.
• 155ICRC Customary Law Study, above note 52, Rule 158.
• 156Ibid., Rule 160.
• 157Amichai Cohen and Yuval Shany, “Beyond the Grave Breaches Regime: The Duty to Investigate Alleged Violations of International Law Governing Armed Conflicts”, Yearbook of International Humanitarian Law, Vol. 14, 2011, p. 79.
• 158Ibid.
• 159Amnesty International, “Somalia: U.S. Must Not Abandon Civilian Victims of its Air Strikes After Troop Withdrawal”, press release, 7 December 2020, available at: www.amnesty.org/en/latest/news/2020/12/somalia-us-must-not-abandon-civi….
• 160ICC, Judgment on the Appeal against the Decision on the Authorisation of an Investigation into the Situation in the Islamic Republic of Afghanistan, Case No. ICC-02/17-138, Decision, 5 March 2020.
• 161See, for example, “Dutch Prosecutors Asked to Check Whether Dutch Soldiers Killed Afghan Civilians”, Reuters, 23 December 2020, available at: www.reuters.com/article/us-netherlands-afghanistan-prosecutors/dutch-pr….
• 162See GC I, Art. 16; GC II, Art. 19; GC III, Arts 122–123; GC IV, Arts 136–140. Further, although neither CA 3 nor AP II contain explicit rules regarding missing persons, many obligations relating to missing persons and the dead have been understood to apply in NIACs by virtue of customary law. See Anna Petrig, “Search for Missing Persons”, in A. Clapham, P. Gaeta and M. Sassòli (eds), above note 14, p. 272; ICRC Customary Law Study, above note 52, Rule 117. Obligations to search for and collect the dead are also understood to be customary in nature and applicable in NIACs. See ICRC Customary Law Study, above note 52, Rule 112; A. Petrig, above note 33, pp. 343–344.
• 163Emphasis added. Notably, the UN General Assembly in Resolution 3220 (XXIX) called upon conflict parties to provide information about missing persons and undertake related obligations “during and after the end of hostilities”. See UNGA Res. 3220 (XXIX), 6 November 1974 (emphasis added). In addition, Article 33(2) of AP I suggests an obligation to search for persons “if they have died in other circumstances [than detention] as a result of hostilities or occupation”.
• 164ICRC Customary Law Study, above note 52, Rule 117.
• 165A. Petrig, above note 162, p. 269.
• 166ICRC Commentary on AP I, above note 29, para. 1239; A. Petrig, above note 162, p. 269. The AP I Commentary identifies Article 33 as being among those articles “whose application in relation to a conflict may continue beyond the termination of this conflict”. ICRC Commentary on AP I, above note 29, para. 149.
• 167ICRC Commentary on AP I, above note 29, paras 1235, 1237.
• 168See also A. Petrig, above note 162, p. 270. GC III Article 119 also calls for agreements to be made for the purpose of searching for dispersed PoWs. See also GC I, Art. 15 (calling for the arrangement of “an armistice or a suspension of fire … to permit the removal, exchange and transport of the wounded left on the battlefield”).
• 169ICRC Commentary on AP I, above note 29, para. 149. GC IV Article 26, similarly pertaining to facilitating enquiries made by members of dispersed families, arguably also continues to apply after the conclusion of armed conflict.
• 170AP II, Art. 8; see also GC I, Art. 15; GC II, Art. 18. Somewhat less certain is the extent to which obligations pertaining to searching for the wounded and sick also continue to apply after the end of armed conflict. See, for example, GC I, Art. 15; GC II, Art. 18; AP II, Art. 8. The Geneva Conventions stress that the obligation to search for the wounded arises after each engagement, and that searches in relation to this obligation must be undertaken “without delay”. See also 2016 Commentary on GC I, above note 19, paras 1487 (“The obligation to act without delay is strict, but the action to be taken is limited to what is feasible, in particular in the light of security considerations”), 1486 (“[W]henever there is an indication that there may be wounded or sick people in an area, and circumstances permit, a reasonable commander should commence search and rescue activities”). Notably, under GC I Article 15, States must “[a]t all times … take all possible measures to search for and collect the wounded and sick” (emphasis added), whereas GC IV Article 16(2) provides that States should search for the dead and wounded “[a]s far as military considerations allow”. According to the Commentary on GC II, it was a conscious decision by the drafters to omit the phrasing “at all times” found in GC I Article 15 from GC II Article 18 on account of the challenging conditions of warfare at sea. See 2017 Commentary on GC II, above note 11, para. 1653. As such, while “the obligations of Article 18(1) remain applicable and are to be continued for as long as there is a reasonable chance of such persons being found”, even after one search operation has been conducted (ibid., para. 1657), the Commentary suggests that the obligation only exists during “conflict”.
• 171See also GC III, Art. 120(4) (stating that PoW graves must be “suitably maintained and marked so as to be found at any time”). It has been noted that the duty to conclude agreements regarding gravesite access and maintenance may not apply in NIACs, out of concern that the conclusion of such agreements in the form of a treaty could be interpreted as affording recognition to an NSAG. A. Petrig, above note 33, p. 359.
• 172See A. Petrig, above note 33, p. 360, citing Jean Pictet (ed.), Commentary on the Geneva Conventions of 12 August 1949, Vol. 3: Geneva Convention relative to the Treatment of Prisoners of War, ICRC, Geneva, 1960, p. 566. An alternative is that the obligation is not ad infinitum but instead follows the approach of AP I Article 34, which provides a system in the event that the parties are unable to enter into agreements on the maintenance of gravesites. Ibid.
• 173See, for example, ICRC, “Nagorny Karabakh: ICRC Submits Updated List of Missing Persons”, news release, Geneva, 15 December 2015, available at: www.icrc.org/en/document/nagorny-karabakh-icrc-submits-updated-list-mis….
• 174See International Commission on Missing Persons, “EU and Western Balkan Countries Sign Joint Declaration Citing Efforts to Account for Missing Persons from 1990s”, press release, 10 July 2018, available at: www.icmp.int/press-releases/eu-and-western-balkan-countries-sign-joint-…; Joint Declaration on Missing Persons in the Framework of the Berlin Process, Western Balkans Summit, London, July 2018, available at: https://assets.publishing.service.gov.uk/government/uploads/system/uplo….