1. IntroductionFootnote *
Over the past two decades, scholars of international humanitarian law (IHL) have increasingly turned their attention to non-international armed conflicts (NIACs) occurring beyond the territorial boundaries of a single state. This shift has prompted important questions regarding the appropriate legal characterisation of such conflicts. Are these armed conflicts a modern evolution of the NIAC concept? How do IHL instruments apply in these cross-border contexts? These questions stem from the fact that the application of NIAC law in such cases necessarily extends beyond the territory of a single state. This extraterritorial dimension introduces an international character to a conflict type that has traditionally been defined by its internal nature. As a result, it raises concerns about whether the existing NIAC framework within IHL is sufficiently equipped to address and regulate this expanded scope.
In this light, NIACs occurring outside state territories have been variously labelled as ‘extraterritorial’, ‘transnational’, ‘hybrid’ or ‘external’ NIACs.Footnote 1 These terms do not fully capture the nature of the armed conflict in question. For instance, ‘extraterritorial’, ‘transnational’ and ‘external’ may signal that there is an international armed conflict (IAC).Footnote 2 Such terms arguably imply that states are engaged in the use of force against one another, including a state controlling another state’s territory. They therefore tend to obfuscate the constitutive elements of NIAC and signal that the armed conflict is primarily international and then non-international. As for the hybrid form of conflict, it suggests containing elements of both NIAC and IAC, conceptually merging the two types of armed conflict into one, so that such armed conflict would be the offspring of both types of armed conflict.Footnote 3 Nor do these terms appear in the legal framework governing NIACs. IHL, therefore, applies specifically to situations of IAC and NIAC; it does not recognise a third category of armed conflict within its framework.Footnote 4 When NIACs occur beyond state borders, it is important to assess whether this dynamic alters the applicable jus in bello framework envisaged for NIACs.Footnote 5 Specifically, should NIACs beyond state borders be classified as IACs simply because they take place beyond state borders, despite involving non-state actors and the opposing state? Alternatively, should classification hinge on the identity of the fighting parties regardless of the geographical location of the fighting, an example in point being two armed groups fighting one another from different states? Or could such contexts represent a modern form of classical NIAC – one that occurs outside a state’s territory and is referred to as cross-border NIAC, ultimately considered as a subtype of NIAC? What would the applicable legal framework be under IHL for such forms of violence? Some of the practical examples that prompted the above questions include, but are not limited to Israel and Hezbollah engaging in hostilities, including through rocket attacks,Footnote 6 Houthis and Israel firing rockets against each other, or Türkiye’s operations against the so-called Islamic State (ISIS) and the Kurdistan Workers’ Party (PKK).Footnote 7 In 2006, Israel’s High Court of Justice, in the Targeted Killings case, held that an IAC existed between Israel and armed groups operating from the occupied West Bank and Gaza.Footnote 8 On 4 May 2025, Houthis from Yemen fired a ballistic missile into Israel, hitting Ben Gurion Airport.Footnote 9 As a response, on 5 May 2025, Israel carried out airstrikes in the Houthi-held port city of Hodeidah in Yemen.Footnote 10
Building on the above, the present study suggests two points. Firstly, it puts forward the idea of the term ‘cross-border NIAC’ and considers the armed violence occurring beyond a state border between the state and the armed group, or armed groups engaging one another from different states as cross-border NIACs. While ‘cross-border NIAC’ is not a legal term as such,Footnote 11 this work suggests from the IHL standpoint that the latter captures more clearly the nature of the violence that occurs outside the state territory involving states and armed groups. In other words, this term allows the focus to be maintained on the parties involved in the fighting, which ultimately determine the nature of the armed conflict, rather than on the international aspect of the violence taking place. At the same time, with this term the possibility of a parallel IAC occurring alongside a NIAC is not ruled out. This, in turn, would be dependent on whether the intervening state has obtained the host state’s consent (explicit or tacit) to operate on its territory against an armed group. If there is no consent provided by the host state, then IHL of IAC would apply; thus, two parallel armed conflicts would exist, an IAC and a NIAC. Secondly, while this work supports the understanding that in such contexts states are inherently bound by common Article 3 and customary IHL,Footnote 12 it develops and highlights the potential for Additional Protocol II 1977 (AP II) to apply extraterritorially to cross-border NIACs. Such a claim is based on state jurisdiction arising from actions of its armed forces acting outside its borders, much like in the extraterritorial application of human rights law (IHRL), while all along acknowledging IHL’s lack of a fully developed concept of jurisdiction akin to that in IHRL. It is argued that when a state acts outside its territory and exercises authority and control over individuals in the context of cross-border NIAC, the application of AP II could extend beyond that state’s own territory to encompass actions of its organs – that is, its armed forces. It should be emphasised, however, that the analogy with personal jurisdiction under IHRL is advanced as an interpretive tool. It is not intended to suggest, nor should it be read as implying, any modification of the AP II provisions. Any formal extension or alteration of AP II would require explicit state consent through the established mechanisms of treaty amendment and modification.Footnote 13 This suggestion, therefore, remains within the bounds of permissible interpretive reasoning, exploring the analytical insights that such an analogy may provide.Footnote 14
This article is structured into three parts and a conclusion. The first part (Section 2) introduces the term ‘cross-border NIAC’ as a potential alternative to terms such as ‘extraterritorial’ or ‘transnational NIAC’. Nevertheless, the article acknowledges that the terms ‘cross-border’ and especially ‘extraterritorial’ are closely related in their substantive meaning. This section also examines the significance of the territorial state’s absence of consent and the use of force by an external state as criteria for triggering the parallel existence of an IAC.Footnote 15 Thus, the issue of state explicit and tacit consent is also inspected. Sections 3 and 4 then explore the applicable legal framework governing cross-border NIACs. Specifically, Section 3 analyses the cross-border NIAC considering the geographical reach of common Article 3 of the Geneva Conventions 1949 (GCs 1949) and of Article 1 AP II. Section 4 turns to the geographical scope of Article 1 AP II, testing the argument for its applicability in cross-border NIACs by drawing an analogy with the extraterritorial application of IHRL. In doing so, it focuses on the concept of state obligations arising through the conduct of its agents – particularly where authority or control is exercised over individuals. The article concludes by summarising the main findings of this study (Section 5).
2. Going beyond semantics: The term ‘cross-border NIAC’
Neither the Geneva Conventions of 1949 (GCs 1949) nor AP II explain what cross-border NIAC is or how many variants of NIAC there can be. However, the classical understanding of NIAC forms the basis for interpreting the nature and legal implications of cross-border NIAC. Before actually defining the cross-border NIAC, some words are in order regarding what a NIAC is and, particularly, how common Article 3 GCs 1949 conceptualises a NIAC.Footnote 16 In its chapeau, this article reads: ‘In the case of armed conflict not of an international character occurring in the territory of one of the High Contracting Parties, each Party to the conflict shall be bound to apply, as a minimum, the following provisions …’.Footnote 17 The parties to a NIAC are presumed to be state armed forces pitted against non-state armed groups, or non-state armed groups fighting one another.Footnote 18 The level of organisation of the non-state armed group and the intensity of violence between the adversary parties needs to reach a minimum level.Footnote 19 Instead of the adjective ‘minimum’, one can also find the descriptors ‘certain’ or ‘sufficient’ denoting the requisite threshold of non-state armed group organisation and the level of violence.Footnote 20 In sum, in order to determine whether a situation of violence has morphed into a NIAC and the threshold of common Article 3 is crossed, two criteria must be met: (i) there should be confrontations occurring between state Delta’s armed force and an armed group Z that is organised to a minimum level, or between two organised armed groups, Z and Leon, fighting each another on state Delta’s territory; and (ii) the violence occurring between the opposing parties must reach a minimum level of intensity.Footnote 21 These elements remain central to the definition of cross-border NIAC.
Continuing from this point, in April 2024, the International Committee of the Red Cross (ICRC) issued its opinion paper ‘How Is the Term “Armed Conflict” Defined in International Humanitarian Law’. As variants of NIAC,Footnote 22 spillover NIAC and extraterritorial NIAC were suggested, whereas the idea of transnational violence (i.e., a global NIAC) was not supported.Footnote 23 A spillover NIAC is defined as a NIAC ‘spilling over into the territory of the neighbouring state’,Footnote 24 whereas an extraterritorial NIAC is viewed as one ‘where states or multinational armed forces are fighting against one or more NSAGs in another host state’.Footnote 25 Both variants involve an extraterritorial dimension: a territory of another state, but with a caveat. In the case of spillover conflicts, extraterritoriality could be conceptualised in the very near proximity of the country where the NIAC originated and depicted in the fact that it crossed over to the territorially adjacent state.Footnote 26 In the other variant, it is implied that the state on the territory of which the violence is ongoing may be rather far off geographically. Conceptually, however, both variations involve foreign state territories and may arguably trigger a parallel IAC. Examples are the United States conducting drone strikes on the territory of PakistanFootnote 27 against Al-Qaida from 2004 without the government’s consentFootnote 28 (see Section 3 for further discussion on consent), or when the United States invaded Afghanistan in 2001 to fight Al-Qaida and the Taliban.
That said, between the two proposed terms, ‘extraterritorial NIAC’ appears problematic for the following two reasons. It automatically suggests that it is an IAC that exists between the intervening state and the host state, as ‘extraterritorial’ is best understood as necessarily including a state actor. The notion of extraterritoriality, therefore, weakens the relevance of host state consent, obscuring the crucial question of whether the host state has consented to such operations. Second, this descriptor tends to imply that a state engaging an armed group on the territory of another state has a significant geographical distance from the host state. This may not always be the case. In this regard, if state Delta conducts operations against armed group Z on the territory of state Arcadia, which is not directly adjacent to Delta but is still geographically close, the legal characterisation depends on whether Arcadia has consented to Delta’s presence and actions on its territory. If state Arcadia has not provided such consent to state Delta, an IAC exists in parallel with the NIAC between state Delta and armed group Z.
Based on the foregoing, the key criterion under IHL for classifying armed conflicts, including those that take place outside a state’s own territory, is identifying the parties involved rather than focusing on territorial descriptors. It is contended, therefore, that the term ‘cross-border NIAC’ is preferable, as it more clearly emphasises the actors involved in the conflict and the presence or absence of the host state’s consent. The term ‘cross-border NIAC’ is also regarded as a more neutral descriptor, as it avoids implying geographical remoteness and does not suggest the existence of an IAC with the territorial state. It should nonetheless be acknowledged that the proposed term remains closely aligned in substance with the qualifier ‘extraterritorial’, as both concern situations involving the crossing of international borders.
Be that as it may, ‘cross-border’ involves two or more states (cross-country)Footnote 29 and implies activities that traverse national frontiers. In a narrow sense, the term typically refers to adjacent territories. In a broader sense, it involves movement across at least one international border, but the states of origin and of destination do not necessarily need to be adjacent. It is in this broader sense that the term ‘cross-border’ is used here.Footnote 30 In the context of trade and business activities, cross-border trade spans countries that share a physical border, as well as covering services across national boundaries, including between countries that do not directly border one another. Why, then, should the term ‘cross-border’ not also be employed in this broader sense to refer to NIACs occurring outside a state’s territory, near or far?
The term ‘cross-border NIAC’ combines two elements: the involvement of an international border and violence amounting to a NIAC. Through the presence of these two elements, this term maintains the focus on the parties involved in the nature of the violence, emphasising that the territorial state may or may not be a party to the fighting, and that its consent has either been granted or not to the intervening state to act on its soil.Footnote 31 A cross-border NIAC can therefore be explained as a type of NIAC in which states or multinational armed forcesFootnote 32 engage non-state armed groups on the territory of another state, with or without that state’s consent, as well as hostilities between non-state armed groups across state borders.Footnote 33 The idea behind this term is to encompass NIACs that occur not only on geographically proximate territory of the host state but also on state territory that is further away.Footnote 34 This term proposes the inclusion of such scenarios where armed groups are fighting each other across international borders.Footnote 35 From this perspective, the cross-border NIAC can manifest itself in the following factual constellations:
(a) State Aragon (with or without a coalition of other states) is pitted against an armed group Z on the territory of state Delta, with or without Delta’s consent and participation. Four examples can fit this constellation. The first is seen in Israel fighting Houthi rebels who control part of the Yemeni territory but do not represent the government of Yemen.Footnote 36 On 4 May 2025, the Houthi rebels fired a rocket from Yemen towards Israel, which landed at Israel’s Ben Gurion Airport.Footnote 37 Following this incident, Israeli jets struck Yemen.Footnote 38 The internationally recognised Yemeni government has not condemned the attack on Ben Gurion Airport.Footnote 39 The Israeli strikes have been criticised by the UN-recognised Yemeni government based in Aden, which said it had not been consulted and airstrikes alone were not an integrated plan to remove the Houthis from power.Footnote 40 The second example is Türkiye’s long-running military operations against the PKK, conducted in Iraq and Syria.Footnote 41 Neither the Iraqi nor the Syrian government had given its support or consent for Türkiye’s military operations on their respective territories. However, this position has recently evolved, as the current administrations in Iraq and Syria have moved towards a framework of strategic security coordination with Türkiye to address mutual concerns.Footnote 42 The third example is the US-led coalition that fought (along with the Afghan government) the Taliban in 2001–21.Footnote 43 The fourth example is the US-led coalition against ISIS from 2014 onwards, when the former started carrying out air strikes against ISIS.Footnote 44 The US-led coalition was defending Iraq from the ISIS attacks and trying to put an end to ISIS safe havens and sanctuaries in Syria.Footnote 45 Syria did not provide consent for such actions, nor did anything to put an end to ISIS safe havens on its territory.Footnote 46 Where there is no host-state consent (either explicit or tacit), the intervening state and the host state would be in a separate IAC.
(b) Armed group Z, based in state Bravo, is engaged in hostilities against armed group Gray, located in state Delta. In this scenario, neither of the territorial states is directly involved with the armed groups, nor is there any third-state involvement. A pertinent example, were it not for the engagement of foreign states,Footnote 47 would be the conflict between ISIS and the PKK, which involved cross-border fighting across Iraq and Syria.Footnote 48 Another example fitting this hypothetical scenario would be if Hezbollah and Hamas were to turn against one another and engage in hostilities from the respective territories in which they operate, free from any outside support.Footnote 49
Based on the foregoing, the term ‘cross-border NIAC’ appears to constitute a broader conceptual category encompassing all forms of NIAC that occur beyond the territorial boundaries of a single state, whether involving state forces and organised armed groups or exclusively non-state armed actors.Footnote 50 A cross-border NIAC, therefore, is but one subtype of NIAC. The next section examines the legal significance of host-state consent in the classification of such conflicts.
2.1. The significance of host-state consent in the classification of cross-border NIAC
This section addresses the issue of host state Delta’s consent for allowing a foreign state Aragon to fight armed group Z on its territory. It further examines the relevance of state consent in situations where two armed groups, Z and Gray, are engaged in a cross-border NIAC. The broader question addressed in this section is how consent may play out in cross-border NIAC and whether the absence of consent is an indicator of IAC to exist in parallel with that of cross-border NIAC.
The ability of a state to consent in international law is an important safeguard against encroachment on its sovereign spheres and political priorities.Footnote 51 There is no definition of the phenomena of state consent in conventional IHL, nor is there any reference to the role that consent plays in conflict classification. In IHL, consent may be linked to the provision of humanitarian assistance,Footnote 52 the ICRC requesting access to visit in times of NIAC to persons deprived of their liberty.Footnote 53 In terms of conflict classification, consent can transform a prima facie IAC (as well as end the situation of military occupationFootnote 54) into a NIAC, or show that there is an armed conflict taking place in parallel with that of a NIAC.Footnote 55
The consent obtained by Aragon to act on Delta’s territory against the armed group Z should ‘have been previously expressed or established (explicitly or tacitly)’Footnote 56 by the latter. Such consent would maintain the nature of the armed conflict as a NIAC. Thus, explicit or tacit forms of consent would have the same classificatory effect to preclude a parallel IAC. The ICRC Commentary on the First Geneva Convention states that consent stricto sensu must be given by a valid authority.Footnote 57 However, determining whether such consent exists can be difficult, as states often do not publicise this.Footnote 58 The intervention of a third state also may be met with no protest; it may elicit contradictory statements from government authorities, or may prompt only symbolic objections intended to appease the domestic audiences.Footnote 59 From two forms of consent, explicit and tacit, establishing the latter would pose a further challenge. Here, it is contended that there would be no denouncing of the intervening state’s actions, nor would it be expressed in words or made verbally known through statements. In fact, it may be seen in the state’s actual behaviour or cooperation with the intervening state on its territory, for instance, in the form of its organs assisting the intervening state in achieving its goal.Footnote 60 Hence, the key would be to interpret the host-state behaviour. One such example in point would be if the internationally recognised government of Yemen were not to object to Israeli strikes on its territory; nor, having made a single official comment over them, instead only denounced the Houthi actions, saying they cause instability in the region.Footnote 61
The example of an IAC becoming a NIAC based on the host state’s explicit consent is that of the intervention by the US and its coalition forces in Afghanistan. In October 2001, the US, with the support of Britain, started bombing the Taliban in Afghanistan. This armed conflict was classified initially as an IAC, involving the US and its coalition armed forces against the Taliban.Footnote 62 After having ousted the Taliban from Afghanistan in 2001 December, Afghan groups, in Bonn (Germany), agreed to an arrangement for forming the interim government.Footnote 63 Hamid Karzai was subsequently sworn in as head of an Afghani interim power-sharing government.Footnote 64 Shortly after the Bonn Agreement, UN Security Council Resolution 1386 followed, establishing the International Security Assistance Force (ISAF).Footnote 65 This resolution attests that the ‘Afghan authorities welcome the deployment of such force to Afghanistan’. The armed conflict based on such welcome (consent) was subsequently regarded as a NIAC. The consent of the host state for the intervening state’s operations on its territory is therefore crucial and affects the classification of the entire conflict. In this manner, a NIAC existed between the government of Afghanistan, supported by the ISAF, and the Taliban, the non-state armed group.Footnote 66 The consent of the government of Afghanistan was given to the US-led coalition forces to be on Afghan territory and fight the adversary. Therefore, no IAC existed between the coalition members and the Afghan government. If no such welcome had been extended or if the consent had been withdrawn,Footnote 67 arguably there would have been a parallel IAC taking place between the government of Afghanistan and the US-led coalition. A separate NIAC would ensue between the US, its allies and the Taliban.
The question, then, is how should the absence of explicit and tacit consent by the host state be interpreted under IHL? A non-consensual state intervention raises the wider question whether to have a single (entirely IAC or entirely NIAC) or double (parallel IAC and NIAC) classification of such a cross-border scenario.Footnote 68 What is meant here is when a host state does not give its consent to the actions of the intervening state and use of force on its soil to fight the armed group Z, would such an intervention amount to an IAC pursuant to common Article 2 (1) be classified as a cross-border NIAC, or would there be two parallel armed conflicts running, that of IAC and NIAC?Footnote 69 To address this question, let us suppose state Aragon is fighting armed group Z on Delta’s territory. Aragon and Delta are not neighbours and do not share a border. Aragon’s state armed forces crossed into Delta’s territory and began fighting with armed group Z. Z is not affiliated with Delta in any way and Delta has provided neither explicit nor tacit consent for Aragon to engage in the fight on its territory. The criteria of common Article 3, however, are met. The question is whether Aragon would be in a parallel IAC with Delta because it is fighting armed group Z on its territory without Delta’s consent. It is submitted that the nature of the armed conflict between Aragon and Delta would depend on whether the latter gave consent to Aragon to act on its soil. In other words, Delta must either explicitly or tacitly give its consentFootnote 70 for Aragon’s actions on its soil and, importantly, Delta’s consent for Aragon to act on its territory and use force against the armed group should by no means be obtained by force, coercion or duress.Footnote 71 However, if there is no explicit or tacit consent given, how would the classification of the violence play out?Footnote 72
There are at least three approaches in this case: (i) the conflict is classified in its entirety as an IAC irrespective of the fact that Aragon is fighting the armed group Z, as this fighting is taking place on the foreign state territory without its consent;Footnote 73 (ii) view the entire armed conflict as (a cross-border) NIAC because the armed conflict is primarily between the state of Aragon and armed group Z and the host state is not involved in any way, nor was force used against its civilians; or (iii) employ the dual classification framework and regard this cross-border scenario as an IAC between Aragon and Delta, and a NIAC between Aragon and armed group Z.
Each mode of conflict classification, whether single or double, entails both advantages and disadvantages. A single classification offers functional simplicity by avoiding the need to continually shift between the legal frameworks governing IAC and NIAC.Footnote 74 In contrast, a dual classification approach is more attuned to the factual dynamics of the conflict. It requires an assessment of the identity of the parties involved and necessitates engagement with both IAC and NIAC legal regimes, all the while maintaining due regard for the territorial sovereignty of the host state. Based on the initial appearance, adopting a single classification as an IAC has the benefit of triggering the comprehensive legal framework of IAC. This is particularly relevant in cases where host state consent is absent. Because IAC law contains more detailed provisions governing the conduct of hostilities and detention, more protective rules would be put into operation. According to this approach, the entire conflict would be regarded as an IAC. The counter argument, however, to this approach overlooks the fact that the armed group against which the intervening state is fighting would not be affiliated with the territorial state, and that IAC law was not designed to regulate state actions against an organised armed group, or for the violence that can come about between these two sides.Footnote 75
In the Targeted Killing case, the Israeli High Court of Justice, with regard to the nature of the armed conflict between Israel and armed groups active in occupied Palestinian territory, opined that it was an IAC. It reasoned in the following manner. The normative framework governing armed conflict between Israel and armed groups operating in the occupied territory was complex. At its core stood the law applicable to IACs. As Antonio Cassese argued – on whose analysis the Court relied – hostilities between an occupying power and armed groups operating within occupied territory amounted to an IAC. Accordingly, for the Court, the law of IAC applied whenever hostilities occurred across state borders, regardless of whether the territory in question was under occupation.Footnote 76
In the context of this article, the Court’s reasoning raises two main concerns. Firstly, it regards any armed conflict between the occupant and the non-state armed groups as international, just because it is a situation of military occupation.Footnote 77 With such a stance, the Court is sidestepping the understanding – given the binary classification of conflicts under IHL – that NIACs may still occur in situations of military occupation. Therefore, one cannot conflate or merge the distinct bodies of law applicable to these different situations. Secondly, just because the armed group is coming from the occupied area, it does not signify that there is no parallel NIAC with an IAC.Footnote 78 The Court, in its analysis, should either have clearly stated that the armed groups represented Palestine,Footnote 79 hence there was an IAC, or the armed conflict between Israel and the armed groups was a NIAC within the context of a broader military occupation by Israel of Palestine, which is governed by IAC.Footnote 80 Overall, in this case, the Court’s reasoning is to disregard the premise that the nature of the armed conflict in IHL is to be determined based on the identity of the fighting parties.Footnote 81
Turning now to the situation solely of a cross-border NIAC despite the absence of host state consent, on the one hand, such classification is beneficial because it revolves directly around the relationship between the foreign state and the armed group, precisely the parties envisaged by NIAC law. It allows the established body of NIAC rules to be brought into operation in coherency, however imperfect they may be.Footnote 82 It provides legal clarity for intervening states regarding the applicable legal regime, increasing the likelihood of compliance and contributing to building a more protective legal environment in cross-border NIAC situations.Footnote 83 On the other hand, however, a significant jus ad bellum concern would eventually arise: it may imply violation of the host state’s territorial sovereignty, as the intervening state operates without its consent and ultimately uses force on its soil. This arguably could further suggest that the intervening state does not view the territorial state as an equal under the principles enshrined in the UN Charter and Declaration on Principles of International Law concerning Friendly Relations and Cooperation among States in accordance with the Charter of the United Nations.Footnote 84
Moving on to the double classification method, as previously noted this aligns more closely with the factual dynamics of the armed conflict.Footnote 85 This approach was recently adopted by the International Criminal Court (ICC) in the context of the Palestine–Israel armed conflict. The ICC stance also offers a degree of clarity regarding the dicta in the Targeted Killing case mentioned above. On 20 May 2024, the ICC Prosecutor KAA Khan KC filed applications for warrants of arrest before Pre-Trial Chamber I in the situation in the State of Palestine. The prosecutor’s office submitted that the war crimes alleged in these applications were committed in the context of an IAC between Israel and Palestine, and a NIAC between Israel and Hamas ‘running in parallel’.Footnote 86 This approach also suggests that, ultimately, non-consensual intervention and use of force create two distinct armed conflicts: a NIAC between the intervening state and the organised armed group, and an IAC between the intervening and territorial states.Footnote 87
Nonetheless, there are some critical points to be kept in mind. One aspect is that when employing a dual classification method, there is the potential for conflation of IAC and NIAC legal regimes, both regimes having different temporal scopes of application, which may result in operational uncertainty regarding the rules on the ground. There is also the aspect of hostile intent. In this light, Pauline Lesaffre argues that double classification may create uncertainties concerning the intervening state’s hostile intent towards the territorial state. Lesaffre explains that hostile intent does not derive from territorial state lack of consent only; it requires that a state deliberately uses force against another state.Footnote 88 To prevent misuse of classification, multiple factual indicators – such as who or what is targeted, where and when attacks occur, and the state’s prior relations and its official statements – must also be considered. The double classification approach overlooks these aspects and wrongly assumes that the lack of host-state consent alone proves hostile intent, which is ultimately insufficient and politically subjective.Footnote 89 The author argues that the ‘double classification’ approach oversimplifies things by treating lack of consent as proof of hostility.
It is asserted here that dual classification is triggered primarily by the absence of the host state’s consent to the use of force or other operations conducted on its territory, broadly speaking. Identifying the absence of consent should not be limited to attacks against the armed forces of the state.Footnote 90 It also need not only involve actual use of force against host state citizens or civilian and military objectives.Footnote 91 In other words, because there is no consent given by the host state, there is already a double classification framework engaged as non-consensual force was used on the territory of the host state.Footnote 92 It is suggested that the absence of consent is a broad enough element. It encompasses a hostile intent, political or diplomatic tensions between states, as well as breach of the territorial state’s sovereignty, manifested in foreign state acts carried out on the territory of a third state without the latter’s approval.Footnote 93 This could result in a parallel IAC, because there was the use of force by one state on another state’s territory, irrespective of the targets engaged, broadly speaking.Footnote 94
To conclude this section, this article adopts the position that the dual classification framework should be invoked in situations where the territorial state has not given its consent, explicit or tacit, to the actions of a foreign state within its territory. In other words, a cross-border NIAC may arise in parallel with an IAC, namely, where a state conducts military operations on the territory of another state without that state’s consent. In such circumstances, the absence of consent generates an IAC between the states, while the hostilities between the intervening state and the non-state armed group simultaneously amount to a NIAC. This results in a ‘double classification’ scenario, with both conflict frameworks applying concurrently but to different relationships.
Thus, where a cross-border NIAC arises in the absence of such consent, the application of a dual classification, recognising both a NIAC and an IAC, is supported. This framework allows for a more accurate legal characterisation of the conflict by differentiating between the distinct legal relationships at play. It upholds the sovereign rights of the territorial state under both jus ad bellum and jus in bello. Moreover, it facilitates a clearer attribution of legal responsibility for the conduct of the foreign state, both in relation to the non-state armed group and the territorial state, notwithstanding the potential complexity of the conflict’s operational dynamics. After all, ‘depending on the parties involved in the hostilities, it is apparent … that contemporaneous conflicts of a different nature may take place on a single territory and therefore that international and non-international conflicts may coexist’.Footnote 95
2.2. The relevance of state consent in situations where two armed groups are engaged in a cross-border NIAC
It must be said that IHL was not originally designed for NIACs going beyond a single territorial state. However, what IHL did envisage was a NIAC arising exclusively between armed groups. Such a scenario would be regulated by common Article 3 and customary IHL,Footnote 96 but not by AP II. This recognition provides a basis for asserting that cross-border NIACs between such groups should not be excluded from the scope of IHL application and from the variation of cross-border NIACs. The other question, then, is to what extent state consent becomes relevant in the context of cross-border NIACs involving only non-state armed actors, when the territorial state is not directly involved. This section explores the legal relevance of a territorial state’s consent, or lack thereof, in situations where two non-state armed groups engage in hostilities across national borders. It considers why consent is required, and on whose behalf. It is submitted that state consent becomes relevant in situations where one of the armed groups seeks external state support, triggering the need to determine whether the territorial state has authorised or welcomed such third-state intervention.
This matter may be approached from both IHL and public international law (PIL) angles. From the perspective of IHL, when two non-state armed groups are engaged in hostilities across borders, the issue of state consent does not directly affect the classification of the conflict under IHL, as states are not themselves parties to the hostilities.Footnote 97 Rather, the relevance of state consent in this context pertains to the principle of territorial sovereignty and whether the state is a failed one in the light of PIL. Specifically, the extent to which a state maintains effective control over its territory and performs core governmental functions may serve as an indicator of whether it is a failed or functioning state and allows such actions of the armed group to take place.Footnote 98 Further, from the PIL perspective, if the territorial state is unable or unwilling to prevent armed groups from using its territory to launch cross-border attacks, neighbouring states may argue the necessity of acting in self-defence to protect themselves and secure their borders.Footnote 99 Within this context, should the territorial state knowingly tolerate or even support the actions of the armed group, this could raise the question of whether the group is acting on behalf of the state. In such cases, under the Articles on State Responsibility, the group could potentially be deemed to be the state organ, thereby transforming the legal nature of the conflict into an IAC from the IHL perspective.Footnote 100
For further clarification from an IHL perspective, let us consider a scenario in which an armed group, Z, operating from the territory of state Delta, seeks assistance from state Bravo to fight a rival armed group, Gamma, located in state Utopia. Initially, hostilities between Z and Gamma constitute a cross-border NIAC, as the parties are non-state actors and the fighting crosses international borders. If state Bravo intervenes on the territory of Delta with the territorial state’s consent to support Z, it becomes a co-party to the NIAC against armed group Gamma, provided that Bravo does not conduct operations in Utopia without its consent. In this case, the conflict retains its cross-border NIAC character. By contrast, if Bravo were to operate in Utopia without Utopia’s consent, a parallel IAC would arise between Bravo and Utopia, while the original cross-border NIAC involving armed group Z, state Bravo and armed group Gamma would continue. This example illustrates the central role of territorial state consent in determining the legal classification of hostilities and highlights how cross-border NIACs can coexist with or occur alongside IACs. In this manner, territorial state consent matters chiefly when an external state actor intervenes. Whether such consent is validly given determines the proper legal classification of the resulting hostilities under IHL.
The analysis below next turns to the determination of the applicable legal framework governing cross-border NIACs. As will be shown, this inquiry presents several complexities, particularly with regard to the territorial scope of AP II. Before addressing that issue, however, the interpretation of the scope of application of common Article 3 in the context of cross-border NIACs has to be undertaken.
3. Common Article 3: Geographical reach and scope of application in cross-border NIACs
Having examined the concept of cross-border NIACs, it is now essential to consider the geographical scope and territorial limitations of common Article 3 GCs 1949. What is the precise geographical reach of common Article 3? How did the drafters of the Geneva Conventions 1949 envisage the territorial application of IHL in the context of NIACs? Can its protection be extended to cross-border NIACs? The central element of a cross-border NIAC is that this type of NIAC is unfolding on the territory of another state, which may or may not be geographically adjacent to the external state involved.
The NIAC that is conceived in common Article 3 appears seemingly to be tied to one geographical location, a state territory, with hostilities taking place on that state’s territory.
Marco Milanovic considers that there are two ways in which reference to conflicts ‘occurring in the territory of one of the high contracting parties’ can be interpreted. The first would be to view the territory referred as ‘that of a state actually taking part in the conflict, i.e. to conflicts which are purely internal in scope’.Footnote 101 The second way would be ‘to say that the reference is to the territory of any High Contracting Party, not necessarily that of a state actually a party to the conflict’.Footnote 102 The latter interpretation allows common Article 3 to encompass situations of NIACs that are not confined to a single state, and also occur beyond the borders of a state. There is, however, a contrary view to the above elaborated position. Common Article 3 is understood by some to apply exclusively to such internal armed conflicts – that is, to NIACs that concern only a specific high contracting party to the Geneva Conventions – and there is nothing in the negotiating record to suggest an extensive scope of NIACs.Footnote 103 According to such a view, the drafters of common Article 3 had been focused on maintaining states’ sovereign prerogatives, ‘hence the notion that they sought to protect persons who were beyond the sovereign authority of the territorial State is illogical in light of their overarching goal’.Footnote 104 This comes through as a narrow interpretation of the drafting history and scope of application of common Article 3. The scope of application of this Article should not be viewed in a vacuum. Common Article 3 has the material capacity to apply outside state territory.Footnote 105 It mentions the territory in a following manner: ‘in the territory of one of the High Contracting Parties’. That territory could be any high contracting party territory in the world.Footnote 106 Further, the Article uses the formulation ‘conflict not of an international character’ rather than ‘internal’ armed conflict; hence, it is believed that the Article does not exclude its application to NIACs taking place outside a state party’s own territory.Footnote 107 It also refers to parties to the armed conflict (‘each Party to the conflict’), which means there must be parties identifiable for a NIAC scenario commonly known in IHL. Such an interpretation and reasoning could also be inferred from the travaux préparatoires to common Article 3 GCs 1949. The travaux préparatoires to common Article 3 convey that states understood NIACs as being primarily civil wars, resistance wars, and wars of colonial domination.Footnote 108 These examples all comprise events that can unfold both on and outside the state’s own territory.Footnote 109 The understanding of the concept of territory in common Article 3 thus appears to be broad,Footnote 110 extending its application to such state territories that may or may not be involved in the armed conflict stricto sensu. That is to say, it can apply to any high contracting party territory where the armed conflict unfolds with its literal meaning:Footnote 111
Common Article 3, by contrast, affords some minimal protection, falling short of full protection under the Conventions, to individuals associated with neither a signatory nor even a non-signatory ‘Power’ who are involved in a conflict ‘in the territory of’ a signatory. The latter kind of conflict is distinguishable from the conflict described in Common Article 2 chiefly because it does not involve a clash between nations (whether signatories or not). In context, then, the phrase ‘not of an international character’ bears its literal meaning.
Territory would consequently have to be understood as comprising the territory of any high contracting party to the 1949 Geneva Conventions.Footnote 112 Here, again, the concept of territory could be broadly assumed: any state territory on the world map where an armed conflict may take place. The draft text submitted by the ICRC to the XVII International Red Cross Conference in Stockholm suggests the concept of territory could be conceived as not confined to the territory of a single state:Footnote 113
In all cases of armed conflict not of an international character, especially cases of civil war, colonial conflicts or wars of religion, which may occur on the territory of one or more of the high contracting parties, the implementing principles of the present Convention shall be obligatory on each of the adversaries. The application of the Convention in these circumstances shall in no wise depend on the legal status of the Parties to the conflict and shall have no effect on that status.
The reading of common Article 3 should thus be premised on the understanding that not only ‘it will apply so long as a non-international conflict originated in the territory of one of the High Contracting Parties’,Footnote 114 but also to any high contracting party where the armed conflict may be taking place with or without host-state consent or involvement. This leaves us to advance, irrespective of the territory, the determinant factor of whether the NIAC exists would be identifying who is fighting whom.Footnote 115 Ultimately, a NIAC is distinct from an IAC based on the legal status of the entities opposing each other.Footnote 116 The next section examines the territorial scope of AP II. It will become evident that its geographical reach differs from that of common Article 3. The discussion will explore alternative legal arguments that could support a broader application of AP II. However, as was mentioned in the Introduction to this work, the analogy with personal jurisdiction under IHRL is advanced as an interpretive tool rather than as a proposal to modify the application of AP II, thus exploring the analytical insights that such an analogy may provide (see Section 4).
3.1. Article 1 Additional Protocol II 1977: Geographical reach and scope of application in cross-border NIACs
Additional Protocol II (AP II) applies also in times of NIAC. This instrument holds a different threshold and different criteria of operation, as opposed to common Article 3 GCs 1949. It develops and supplements common Article 3, but neither supersedes nor modifies its conditions for application. There can thus be a NIAC triggering the application of common Article 3, but not an armed conflict meeting the threshold of Article 1, AP II.Footnote 117 AP II applies to:Footnote 118
[an armed conflict] which takes place in the territory of a High Contracting Party between its armed forces and dissident armed forces or other organized armed groups which, under responsible command, exercise such control over a part of its territory as to enable them to carry out sustained and concerted military operations and to implement this Protocol.
Article 1 sets out stringent criteria for the application of the Protocol. The armed conflict should be taking place in the territory of a high contracting party. On the one side, there should be government armed forces and, on the other, an organised armed group. This Protocol, by default, excludes a NIAC in which two armed groups are pitted against each other. It may happen that the state armed forcesFootnote 119 face a mutiny and a part of the state armed forces turns against it, thereby potentially making it a party to the armed conflict. Obviously, this occurrence should not be a mere hit-and-run event.Footnote 120 Further, the armed group should be organised ‘under responsible command’. The element of organisation hinges on the existence of a responsible command over the armed group,Footnote 121 but it is not necessary that there is a hierarchical system of military organisation similar to that of regular armed forces; rather, the armed group must be capable ‘on the one hand, of planning and carrying out sustained and concerted military operations, and on the other, of imposing discipline in the name of a de facto authority’.Footnote 122 The armed group needs to ‘exercise such control over a part of its territory’ that there are concerted military operations. The control should be of a level that enables the armed group to hold the state territory in its power.Footnote 123 The control of the territory by the armed group should be sustained, as this would be necessary for it to effectively apply the rules of AP II.Footnote 124 Applying AP II is important because it establishes more detailed and comprehensive rules to protect individuals affected by NIACs. For instance, AP II prohibits collective punishment, acts of terrorism, slavery and slave trading in all their forms, pillage (Article 4(2)). It protects objects that are indispensable for the civilian population’s survival (Article 14), as well as cultural objects and places of worship (Article 16). Extending the application of AP II beyond state boundaries would arguably strengthen the protection of the civilians who are caught up in NIACs.
Given the preceding outlined criteria, can AP II apply to cross-border NIACs? The main obstacle to its applicability to cross-border NIACs is the fact that the state involved in the fighting not only should have ratified the Protocol, but chiefly the fighting between its armed forces and the non-state armed group should be on its own territory, and the non-state armed group itself should control part of the state territory.Footnote 125 It is submitted that these factors are the main reason why AP II would not be suitable for application beyond the state’s own territory.Footnote 126 Notwithstanding this line of argument, the approach may be contested by advancing a broader application of AP II. The following section elaborates on this possibility and further examines the ways in which the Protocol’s territorial reach can be extended.
3.2. Routes to a broader interpretation of the AP II geographical reach
The proceedings at the Conference of Government Experts on the Reaffirmation and Development of International Humanitarian Law Applicable in Armed Conflicts, held in 1971, addressed the potential for a broad application of AP II. However, the clause for a wider application would subsequently not be included in the final wording of Article 1 AP II. In its commentary section, the Conference delegates of 1971 considered, among the material elements of a NIAC, also its territorial aspect. Through such consideration, the application of AP II was not to be confined to the high contracting party territory, as the hostile action must occur on the territory of one of the contracting parties.Footnote 127 In other words, the hostile action must happen within the geographical borders of one of the states that were parties to AP II. The drafting history of Article 1 AP II substantiates the argument that the instrument was intended for broad application:Footnote 128
Article 1. – Material field of application
1. The present Protocol shall apply to all armed conflicts not covered by Article 2 common to the Geneva Conventions of August 12, 1949, taking place between armed forces or other organised armed groups under responsible command.
2. The present Protocol shall not apply to situations of internal disturbances and tensions, inter alia, riots, isolated and sporadic acts of violence and other acts of a similar nature.
3. The foregoing provisions do not modify the conditions governing the application of Article 3 common to the Geneva Conventions of August 12, 1949.
Upon observation, the proposed version of the provision is notably broad in that it does not specify that it has to be taking place in the territory of a high contracting party and encompasses all NIACs where state armed forces are pitted against armed groups.Footnote 129
However, at its 29th meeting of 17 March 1975 (the Diplomatic Conference on the Reaffirmation and Development of IHL), Article 1 AP II was adopted by consensus in its current form.Footnote 130 The current version of Article 1 AP II states the conflict must be taking place ‘in the territory of the high contracting party between its armed forces and …’.Footnote 131 Based on the foregoing, Article 1 AP II contains a narrower territorial reach. Arguably, if there were a formulation of ‘shall apply to all armed conflicts not covered’, or ‘in the territory of one of the high contracting parties’, or ‘the high contracting party is a party to the armed conflict’, then it would have been possible to advocate that AP II should apply also to situations of cross-border NIAC.Footnote 132 As currently worded, Article 1 AP II sets geographical limitations on the application of the instrument to the territory of the state party, where its armed forces are engaged in a conflict with an armed group.Footnote 133
However, the application of AP II so that it covers cross-border NIACs can also be done through focusing on the formulation ‘armed conflicts’ rather than ‘in the territory’ and ‘its armed forces’.Footnote 134 Such a focus signifies that the location of the fighting is not to be determinative of the instrument’s application.Footnote 135 Maintaining focus on such a formulation would ultimately pave the way for the use of the Protocol outside the state party’s territory, both in a neighbouring sense and possibly beyond.Footnote 136 Sandesh Sivakumaran allows AP II to apply to that part of the NIAC that is not taking part on the state’s own territory, provided the other part of the same conflict does indeed take place on the state’s own territory and crosses the geographical border.Footnote 137 Hence, AP II should rather apply broadly, and there should not be a restrictive territorial interpretation of its reach. Sivakumaran suggests that when a non-state armed group from state A engages in hostilities in state B – where state B is a party to the Protocol and the conflict meets the required conditions – AP II should apply to the entire conflict, even if some hostilities occur outside state B’s territory. Sivakumaran finds it unreasonable that the legal framework would change simply because the conflict crosses a geographical border. Ultimately, the author believes the focus should be on the nature of the armed conflict, and not the geography where it unfolds.Footnote 138 Such an interpretation may eventually be problematic.Footnote 139 One of the chief barriers to such an application of AP II could be the requirement of the armed group holding territorial control. The armed group has to continue to maintain territorial control on the state soil that it was fighting initially.Footnote 140 This would mean that the stringent criteria of Article 1 AP II have to be sustained. More specifically, let us suppose state C’s armed force is fighting armed group X, and this armed conflict has spilled over to close neighbouring state D. Initially, the armed conflict started on the territory of state C – that is, the armed conflict was already in existence, falling within the criteria of both common Article 3 and of AP II.Footnote 141 Again, the key to the application of this Protocol lies in paragraph 1 of Article 1, in which the cumulative criteria are spelled out. In the context that the armed group still manages to retain control over part of the territory in state C, because of the retained territorial control the parties to this armed conflict would be bound by AP II even on the neighbouring state’s soil, along with common Article 3 and customary IHL rules on the protection and conduct of hostilities.Footnote 142 In another scenario, it could also be because the fighting went outside the territorial boundaries of state C, the armed group opposing state C lost territorial control and fled into the neighbouring state; or that state C pushed them out from the part of the territory they were once controlling. This would be one of the reasons why AP II would stop its application at the border.
At least five additional arguments have been advanced in support of a broader application of AP II, concerning specifically the extension of its territorial scope. In the long run, such modes of application provide far more rules of protection for armed conflict victims than does common Article 3. These are: (i) applying AP II on humanitarian grounds and considerations; (ii) relying on the fact that states are parties to the Protocol; (iii) as a matter of policy; (iv) by considering the armed forces of an intervening state as those of the territorial state; and (v) through a good faith interpretation of Article 31(1) of the VCLT and, by extension, Article 1(1) AP II.Footnote 143
Bringing the protective measures of AP II to the most vulnerable in times of armed conflict based on a humanitarian perspective is advanced by Andrew Clapham. This approach suggests that legal protection should be dependent on humanitarianism and not on the geography. According to Clapham, it would be inconsistent from a humanitarian standpoint to limit the application of AP II only to conflicts occurring within a state’s own borders, especially when that state is a party to the Protocol. The author highlights a contradiction using the example of Afghanistan, where AP II would apply to the conflict between the Afghan government and the Taliban, but not to the conflict involving foreign states (who are also parties to AP II) fighting the same armed group on Afghan territory. All this ultimately would create an unjustified legal gap in rendering protection to the vulnerable.Footnote 144 While the geographic location of a conflict, in essence, should not determine whether civilians receive the protection of AP II – especially when the nature and intensity of the conflict are the same on both the state’s own territory and on the territory of another state – there remains a significant concern. Applying AP II on humanitarian grounds in cross-border NIACs may create unrealistic expectations for states to live up to. In parallel, a state engaged in such conflicts may be unwilling or doubtful whether its treaty obligations under AP II still apply beyond its borders, even for humanitarian purposes. The fact that common Article 3 of the GCs already applies to all NIACs – alongside customary IHL, which provides important forms of protection – in the understanding of the states this legal framework would be enough. Hence, why would the states apply AP II on humanitarian grounds? After all AP II, Article 1 sets out criteria for its application. Whether AP II applies hinges solely on the nature of the armed conflict and not on humanitarian considerations.
Applying AP II because states are party to AP II, or applying it as matter of policy, would be other modes of its application outside state borders. These modes of application were put forward by Annyssa Bellal, Giles Giacca and Stuart Casey-Maslen. When, in 2001, an international coalition led by the United States invaded Afghanistan to take down Al-Qaida,Footnote 145 one of the questions was whether AP II applied to the US-led coalition. Later, supported by the NATO International Security Assistance Force (ISAF), the Afghan National Security Forces (ANSF) fought against the Taliban.Footnote 146 The new Afghan government forces were fighting side by side with the foreign military personnel.Footnote 147 The question was would the Afghan forces apply AP II but not the foreign military?Footnote 148 Bellal, Giacca and Casey-Maslen are of the view that because the forces of the states were party to AP II they should formally have been bound by its provisions in their military operations in Afghanistan, as they were engaged in the armed conflict in which Afghanistan government forces were fighting against at least one armed group meeting the AP II criteria for application.Footnote 149 This, then, would avert interoperability concerns, as well as avoid a lack of clarity in operations between the different parties to the conflict.Footnote 150 According to the authors, the other way in which AP II could also have been applied by the states present in Afghanistan was ‘at the very least as matter of policy, if not law’.Footnote 151 The latter would arguably depend on the goodwill of the states and humanitarian aspirations.Footnote 152
Both approaches raise concerns. First, the fact that a state is party to AP II does not automatically mean it would accept the application of the treaty beyond its own territorial boundaries. Article 1 AP II clearly limits its scope to conflicts occurring ‘in the territory’ of a high contracting party and requires that the conflict unfolds between that state’s armed forces and an organised armed group. Second, even if states find themselves involved in cross-border NIACs and are parties to AP II, they may argue that the Protocol does not apply if they are not directly engaged in combat with the armed group. Third, there are drawbacks to the policy application of AP II. When applied voluntarily on policy grounds, the Protocol may be used selectively, granting states discretion to define what constitutes policy and ethical considerations. This can result in certain armed conflicts being governed by higher standards than others, making the application of AP II contingent on the political will of the state rather than on consistent legal obligation. Moreover, in coalition settings, it is doubtful that the application of AP II would ensure a consistent or unified approach of application among other coalition members. Not all coalition members may be parties to AP II, leading to differing standards and legal obligations among partners. For instance, the United States and Türkiye are not party to AP II. This inconsistency may ultimately contribute to the fragmentation of the legal framework governing NIACs under IHL, especially when the coalition leader is a country that is not the party to the instrument.
Another possible solution to the question of the Protocol’s application outside state borders could be to view the armed forces of the intervening state as the armed forces of the territorial state. However, according to Dapo Akande, while this would be a desirable approach to extend the humanitarian protections of AP II, such a test finds no support in IHL, because the ‘forces of a co-belligerent are not usually regarded as part of the armed forces of a party’.Footnote 153 Each state is ultimately responsible for the actions of its own armed forces.Footnote 154 Treating state forces as legal extensions of the other state is not consistent with the treaty language either of Article 1 AP II, common Article 3 GCs 1949 or customary IHL.Footnote 155 It is thus highly unlikely in practice that states would assume responsibility for violations by the armed forces of another state. It is also highly unlikely that states legally become a single actor in a coalition or in a joint operation. Ultimately, state obligations and liabilities would remain separate.
Carrie McDougall, too, questions whether AP II had an extraterritorial effect for its state parties that intervened in the NIAC in Afghanistan, Australia being a case of point.Footnote 156 McDougall dismisses the interpretation of a restrictive territorial scope of AP II and instead advocates its extraterritorial application. The author bases the analysis on the good faith interpretation of Article 1 AP II, considering Article 31(1) VCLT. In McDougall’s view, a good faith interpretation of Article 1(1) provides that the Protocol applies to all armed conflicts that meet the threshold requirements set out in Article 1(1). In other words, ‘the requirement that an armed conflict involve the armed forces of the High Contracting Party in whose territory the conflict takes place should be viewed as a threshold requirement, and not as a factor that excludes the application of the Protocol to the armed forces of foreign States participating in the NIAC alongside the armed forces of the host State’.Footnote 157 McDougall’s argument ensures that when an armed conflict meets the Article 1(1) AP II threshold, the obligations of the Protocol extend to all participants in that conflict, including foreign states intervening on the government’s side. While the good faith interpretation is plausible, especially with regard to considering ‘a high contracting party’ to mean any state that has ratified the Protocol, there is a potential problem with regard to considering a threshold requirement involving ‘the armed forces of the High Contracting Party in whose territory the conflict takes place’. For instance, the territorial state with its troops may not join the coalition states on its territory in actual fighting and keep its involvement at a minimum level. Such involvement can be shown through the provision of food and running canteens for the foreign forces, letting the third-state forces stay in territorial state military barracks, providing them with medical care or guiding them through geographically unknown and difficult areas. If the territorial state’s armed forces are involved only in such a way, can it be said the threshold of involvement is still met as it is subsumed under the classical criteria of AP II? In the understanding of this article such an involvement would not meet the threshold of involvement as understood in AP II.
Advocating a broader application of AP II – out of humanitarian aspirations, or because the states were initially party to the instrument, or as a matter of policy, or viewing all the forces to belong to the territorial government, or because of the good faith interpretation in Article 1 AP II – is certainly not without grounds. All these modes of broader application require states to be willing, showing their goodwill to apply this instrument on foreign state soil.Footnote 158 I, for my part, advance that AP II could apply in cross-border NIACs based on an analogy with the extraterritorial application of human rights law treaties, particularly where state jurisdiction arises based on state agent authority and control over an individual (personal jurisdiction). What is meant here is that invoking the extraterritorial jurisdictional trigger of IHRL as a basis for extending the protective obligations of AP II beyond state territorial confines is not without grounds.Footnote 159 This is because, although the concept of jurisdiction in IHL is not defined in the same way as in IHRL, a functionally similar mechanism exists within IHL; hence, the analogy with IHRL. In IHL, both in IACs and NIACs, such a mechanism is reflected in the conduct of a state acting through its armed forces and exercising control over individuals, whether civilians or members of armed groups, upon coming into contact with them, on its own territory or on that of another state.Footnote 160 In other words, both IHL and IHRL share the state agent authority and control test over an individual. The following section proceeds to explore the extraterritorial applicability of AP II to cross-border NIACs, building upon the preceding analysis. As was mentioned before, the analogy with personal jurisdiction under IHRL is put forward as an interpretive tool with a view to explore the analytical insights such an analogy may provide. The mechanism through which the above-argued analogy could acquire legal force in cross-border NIACs is through Article 31(3c) VCLT, which permits reliance on ‘any relevant rules of international law applicable in the relations between the parties’.
4. Applying AP II beyond borders: Analogy with the extraterritorial reach of international human rights law
This section seeks to develop the preceding argument by engaging with the concept of extraterritorial jurisdiction under IHRL and drawing analogies with operational precepts found in IHL. The focus here is not on state territorial jurisdiction per se, but rather on the notion of personal jurisdiction: specifically, the circumstances under which a state incurs legal responsibilities in cross-border NIACs by virtue of its capacity to exercise authority or control over individuals, thereby making AP II applicable.
The application of IHRL during armed conflicts, especially its extraterritorial application in the context of IAC and military occupation, has been the subject of significant academic discussion and research. Whether states’ human rights obligations are triggered in cross-border NIAC continues to be a matter of discussion.Footnote 161 Nonetheless, in the view of the present author, given that IHRL applies in both peacetime and situations of armed conflict,Footnote 162 cross-border NIACs should not be considered beyond its scope. Accordingly, states engaged in such conflicts remain bound by their obligations under IHRL, particularly through the lens of personal jurisdiction, whereby state legal responsibilities in the light of human rights conventions arise from the exercise of authority or control over individuals. Being subject to state jurisdiction in human rights law should be construed as the state using its power of control either over a certain space or over certain individuals.Footnote 163 Human rights bodies consider the notion of ‘being within the jurisdiction’ in two ways: firstly, when the state in fact exercises jurisdictional authority over a territory, even if this is not sovereign territory, as in the case of military occupation; secondly, by exercising authority and control over an individual, thus bringing the person within the jurisdiction of the state both in armed conflict and peace times.Footnote 164 The case law and other authoritative statements on such international instruments as the International Covenant on Civil and Political Rights, the European Convention on Human Rights, the American Convention on Human Rights and the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or PunishmentFootnote 165 view ‘jurisdiction’ in the extraterritorial context in terms of the existence of a factual connection between the state and either the territory in which the relevant acts took place (a spatial connection) or the individual affected by these acts (a personal connection).Footnote 166 Unlike IHRL, IHL does not employ the concept of jurisdiction as a basis for its applicability. A peculiarity of IHL is that it is inherently designed to operate in situations that are often extraterritorial by nature, such as IACs and situations of occupation.Footnote 167 In this sense, one could argue that IHL presumes a form of implied jurisdiction, as it assumes that states are in a position to affect individuals and objects through armed conflicts, occupation, targeting or detention outside their territorial borders. Armed groups, too, can affect individuals and objects through targeting, detention or even occupation when they act as proxy for a third party.Footnote 168 All these actions require control demonstrated by the organs or de facto agents over the affected individuals.
In the framework of NIACs, the territorial jurisdiction of the state is generally presumed to persist as well as its laws continue to apply. This presumption may be challenged when an organised armed group contests state authority and asserts control over portions of territory, as envisaged under Article 1 AP II. In such contexts, the state engages in hostilities against the non-state actor through its armed forces. This foundational understanding of AP II provides a basis for exploring whether the concept of jurisdiction, particularly personal jurisdiction, can serve to trigger the applicability of AP II in cross-border NIAC scenarios. Where state organs or agents exercise authority or control over individuals beyond national borders in a cross-border NIAC, a plausible argument arises that the state’s obligations under AP II continue to apply in the same manner as they do under IHRL instruments. Accordingly, state responsibility for the conduct of its armed forces, considering AP II, may be engaged based on the analogy with the logic of the state agent personal jurisdiction link. The question that ensues is where is such a link seen between IHL and IHRL? Reverting to Article 1 AP II, where the formulation ‘in the territory of a high contracting party between its armed forces and dissident armed forces’ is found, it is suggested that the link lies in the armed forces of the high contracting party – namely, the state’s official military apparatus, through which the state’s obligations under AP II are carried out. In armed conflicts, be they international or non-international, states fight their adversaries and act through their armed forces – that is, through their organs.Footnote 169 A state’s obligations under IHL are therefore engaged through the conduct of its agents, including when they exercise power or authority over individuals.Footnote 170 Accordingly, if a state that is a party to AP II becomes involved in, or is a party to a NIAC beyond its territorial borders, there is a compelling argument that it should remain bound by the provisions of AP II when acting through its organs, namely, its armed forces. As states find themselves in cross-border NIACs through their armed forces, and these forces often come into contact with local populations or opposing armed groups, the obligations enshrined in AP II should arguably follow them across borders. The ensuing question is whether the state must implement all provisions of AP II in such cross-border contexts, or only those over which it has effective control. In these situations, the state acting through its armed forces on foreign territory exercises factual authority over certain rights protected by AP II, thus potentially triggering a duty to uphold those forms of protection that fall within the IHL ambit:Footnote 171
[I]n a situation of armed conflict, the test for assessing the observance of a particular right, such as the right to liberty, may, under given circumstances, be distinct from that applicable in a time of peace. For that reason, the standard to be applied must be deduced by reference to the applicable lex specialis.
For instance, Article 4 (fundamental guarantees), Article 5 (persons whose liberty has been restricted), Article 13 (protection of the civilian population), Article 14 (protection of objects indispensable for the survival of the civilian population), Article 17 (prohibition of forced movement of civilians), all from AP II, would be considered to apply in times of cross-border NIAC. Ultimately, the approach developed above allows for filling any gaps left by common Article 3 GCs.
Advocating the broad application of AP II, even when supported by analogy with the extraterritorial application of IHRL and the concept of state personal jurisdiction, may, however, present certain challenges. For instance, states that have ratified AP II could reject such an interpretation,Footnote 172 contending that common Article 3 GCs, along with customary IHL rules on the protection and conduct of hostilities, already govern all situations of NIAC,Footnote 173 and that AP II has a defined territorial scope of application, thereby rendering the extension of AP II unnecessary or legally unfounded. Under the lex lata of NIAC, cross-border NIACs would be regulated by common Article 3 GCs along with customary IHL, transcending state boundaries and regulating any NIAC on a foreign state territory with or without the involvement of that state. Nevertheless, avenues for the extraterritorial application of AP II should be further probed and not ruled out, especially in such contexts where armed groups control the territory.
To conclude this section, while the analysis provided above is primarily theoretical, the proposed interpretive framework may nonetheless carry implications for practice. In the long run, by encouraging a more coherent understanding of the extraterritorial application of AP II, the approach presented could prompt states to better account for the consequences of their operations in cross-border NIACs. These practical considerations, however, remain ancillary to the central aim of this article, which was to illuminate the doctrinal contours of IHL as it could apply to cross-border NIACs, and to identify interpretive avenues that warrant further scholarly and institutional reflection.
5. Conclusion
Armed conflicts, sadly, remain an enduring feature of international relations. States and non-state armed groups alike continue to advance their strategic objectives through warfare, with devastating consequences for civilian populations. It is, therefore, of utmost importance to explore effective avenues for the implementation of IHL, ensuring that its protective framework extends to all those in need, even in the most volatile and unconventional conflict situations. This article has therefore examined the legal framework applicable to cross-border NIACs and unconventional armed conflict within the scope of IHL, a variant of NIAC that has gained significant attention, particularly because of the US intervention in Afghanistan and the rise of armed groups such as Al-Qaida, ISIS, Hezbollah and Houthi.
The principal objective of this work was to put forward the notion of cross-border NIACs and to evaluate the extent to which existing IHL norms apply to such conflicts. It was argued that hostilities involving a state (or a multinational force) and a non-state armed group operating on the territory of another state, or between non-state groups across international borders, are regarded as cross-border NIACs. The presence or absence of the territorial state’s consent determines whether the intervening state’s actions generate an IAC between the states involved. Where valid consent is given, no parallel IAC arises between the territorial state and the intervening state. However, consent does not preclude the existence of a separate cross-border NIAC between the intervening state and the non-state armed group, provided the requisite intensity of the violence and organisation of the armed group are met. Similarly, a NIAC between armed groups operating across borders can exist independently of any interstate consent dynamics. This approach allows for a more nuanced application of IHL norms in complex conflict scenarios that transcend state territorial boundaries. The discussion identified common Article 3 GCs 1949 and customary IHL as the core legal framework applicable to cross-border NIACs. Notably, common Article 3, with its broad and flexible language, provides a baseline standard capable of adapting to cross-border dimensions of NIAC. In contrast, Article 1 AP II was found to have a narrower material and geographical scope. However, this article has argued that the applicability of AP II to cross-border NIACs may be extended by analogy with IHRL, particularly in cases where a state exercises effective control or authority over individuals beyond its national territory.
Acknowledgements
This article is a result of my visiting research stay at the Geneva Academy of International Humanitarian Law and Human Rights between August and December 2022.
I wish to thank the anonymous reviewers of this article and Professor Yaël Ronen for their constructive and very helpful feedback. Gratitude is also extended to Professors Gloria Gaggioli, Marco Sassòli and Marten Zwanenburg for their insightful comments that the author received at various stages of the development of this article as a visiting Research Fellow at the Geneva Academy of International Humanitarian Law and Human Rights. I also thank Elizabet Kakabadze (Erasmus Mundus Scholar in International Law of Global Security, LLM) for her valuable feedback on an earlier version of the manuscript. Any errors or omissions that may be found are my sole responsibility.
Funding statement
Not applicable.
Competing interests
The author declares none.
Open access
