The Fight Against the Islamic State and Jus in Bello

Abstract

This article examines several questions relating to international humanitarian law (jus in bello) with respect to the conflicts against the Islamic State. The first question explored is the classification of conflicts against the Islamic State and the relevant applicable law. The situation in Iraq is a rather classic non-international armed conflict between a state and a non-state actor with third states intervening alongside governmental forces. The situation in Syria is more controversial, especially with respect to the coalition's airstrikes against the Islamic State on Syrian territory. If the Syrian government is considered as not having consented to the coalition's operations, then, according to this author's view, the coalition is involved in two distinct armed conflicts: an international armed conflict with the Syrian government and a non-international armed conflict with the Islamic State. The second question analyzed in the article bears on the geographical scope of application of international humanitarian law. In this context, the article examines whether humanitarian law applies: in the entire territory of the state in whose territory the hostilities take place, in the territories of the intervening states, and in the territory of a third state.

1. Introduction

Aside from all the relevant jus ad bellum issues, the rise of the Islamic State (or the Islamic State of Iraq and the Levant, ISIL) ¹ and the military operations conducted against it by foreign states in both Iraq and Syria also raise several interesting questions in relation to the application of jus in bello (international humanitarian law, IHL). As always when it comes to the application of IHL, the first crucial point revolves around the classification of conflicts. The transnational character of the conflict seems to challenge the established dichotomy between international and non-international armed conflicts. ² The foreign interventions against ISIL in Iraq and Syria add to the complexity of the classification. This question is not theoretical; it has concrete consequences in terms of applicable law. The treaty rules applicable in international armed conflicts (IACs) are more developed than those applicable in non-international ones (NIACs). ³ The gap between the legal regimes regulating the two types of armed conflicts has been bridged to a certain extent through customary international law. ⁴ However, this extension has not gone completely unchallenged and has been met with some scepticism by a few states, the most prominent among them being the United States. ⁵ Thus, at least with respect to these states, the determination of the restraints imposed by IHL on their operations will depend to some extent on the classification of conflicts.

Moreover, classification of conflicts may have a direct impact on international criminal responsibility for war crimes committed during military operations against ISIL. In the Rome Statute of the International Criminal Court (ICC), the war crimes applicable in an IAC are more numerous than those applicable in a NIAC. ⁶ Violations of IHL rules as fundamental as the prohibition of attacks directed against civilian objects and the prohibition of disproportionate attacks constitute war crimes only when committed in the context of an IAC. ⁷ Thus, since some of the states conducting airstrikes against ISIL in Iraq and Syria are parties to the Rome Statute, ⁸ whether certain conduct will come under the material jurisdiction of the ICC may depend on the classification of the conflict the intervening states are involved in. ⁹ This distinction may be relevant even for states that are not parties to the Statute, in case of scrutiny of their operations by international bodies ¹⁰ or non-governmental organizations, if these bodies use the ICC Statute as a point of reference.

The second crucial question raised in the context of the fight against ISIL relates to the geographical scope of application of IHL. This question is exemplified by the attacks linked to ISIL and conducted outside the territory of Iraq and Syria, for example in Tunisia, ¹¹ Yemen, ¹² and Libya. ¹³ The attacks were claimed either by ISIL or by groups pledging allegiance to ISIL. The most prominent example for Western countries may be the 13 November 2015 attacks in Paris, claimed by ISIL, in which more than 100 people were killed and many more were wounded. ¹⁴ In a blog post published on 11 January 2016, a Judge Advocate in the US Air Force raised the salient question:

if “Paris” happened in the United States, would the rules of engagement center on law enforcement principles or the laws of war? . . . [T]here are significant differences between past isolated terror attacks and those like Paris that make answering this question more difficult than it may seem. When it comes to ISIS in the US, an argument may be made that either legal regime — or both — actually apply. ¹⁵

This statement hints at the difficulties of identifying the geographical scope of application of IHL in the fight against ISIL. Deciding in favour of the application of IHL in situations such as the Paris attacks means that the lawfulness of the use of lethal force will be judged first and foremost according to IHL rules regulating the conduct of hostilities. These rules appear, at least in some cases, to be more permissive than those of international human rights law (IHRL) regulating resort to force. ¹⁶

The present article will explore these questions. First, it will examine the classification of conflicts and the applicable law in relation to the foreign interventions against ISIL in Iraq and Syria (Section 2). Second, it will address the issue of the geographical scope of application of IHL in the context of the fight against ISIL (Section 3).

2. Classification of conflicts and applicable law

This section assesses the classification of armed conflicts in the fight against ISIL in Iraq and Syria. As noted above, the main focus is the conflicts involving the foreign states intervening in both countries, in particular the operations conducted by the US-led coalition named ‘Operation Inherent Resolve’. ¹⁷ We will thus not deal with the NIACs opposing ISIL to other armed groups operating in these countries. ¹⁸ After analysing the coalition's operations in Iraq, whose classification does not present particular difficulties (Section 2.1), we will turn to the military operations of the coalition in Syria, whose classification is somewhat more complicated (Section 2.2).

2.1. Conflict against the Islamic State in Iraq

At the beginning of 2014, after days of intense fighting with the Iraqi security forces, ISIL seized the city of Fallujah, in the Anbar province. ¹⁹ Hostilities spread rapidly throughout the first half of 2014 and ISIL took control of part of the Iraqi territory including several cities. ²⁰ On 29 June 2014, ISIL proclaimed the establishment of a caliphate in the territories under its control in Iraq and Syria. ²¹ Following an appeal by the Iraqi government, ²² the US launched airstrikes in Iraq in August 2014. ²³ Soon after, a coalition was formed in order to coordinate the military intervention against ISIL, codenamed Operation Inherent Resolve. ²⁴ Several states joined the US in the airstrikes. ²⁵ Military operations continue to this day.

2.1.1. Classification of conflicts

As is the case with all NIACs, it is very hard to establish with precision the beginning of the NIAC opposing the Iraqi governmental forces to ISIL. According to established case law by the International Criminal Tribunal for the Former Yugoslavia (ICTY), two criteria are relevant for establishing the existence of a NIAC: (i) the intensity of the conflict and (ii) the organization of the parties. ²⁶ The Tribunal has proposed a list of indicative factors relevant in determining whether the required threshold of intensity and organization has been met. ²⁷ Also, the ICRC Commentary to the 1977 Second Additional Protocol defines the notion of internal disturbances, ²⁸ which serves as ‘the lower threshold of the concept of armed conflict’. ²⁹ However, despite these elements, the moment when a situation of internal disturbances transforms itself into a NIAC is notoriously difficult to identify. In any case, it seems clear that in 2014, there was a NIAC between the Iraqi governmental forces and ISIL. ³⁰ The conflict cannot be viewed as international because, despite its name, the Islamic State is not a state under international law, ³¹ nor is it a non-state actor acting under the (overall) control of another state. ³² Therefore, the conflict does not oppose the Iraqi governmental forces to the forces of another state. ³³

The intervention of the coalition of foreign states alongside the Iraqi governmental forces does not change this classification. Indeed, according to state practice and international case law, the intervention by a foreign state or states in a NIAC on the side of the government does not internationalize the pre-existing NIAC. ³⁴ For example, Mrs Claire Landais, Director of Legal Affairs of the French Ministry of Defence, noted that France views the French military operations in Mali (operation Serval launched in January 2013 at the request of the Malian government ³⁵ ), and in Iraq against ISIL as non-international. ³⁶ This also reflects the position adopted by the ICRC, which confirms that such ‘multinational’ NIACs remain NIACs. ³⁷

It should be noted that in such cases (foreign intervention alongside a state in a NIAC opposing the host state and a non-state armed group operating within the host state's territory) there is no new NIAC between the intervening states and the armed group. The NIAC is the same as the one that existed before the intervention. The only difference is that, while before the intervention the adversary of the armed group was only the host state, now the group has to fight against more than one state. Thus, no separate evaluation is required in order to determine whether the hostilities between the intervening states and the armed group rise to the level of intensity required for a NIAC to exist. Indeed, the criterion of intensity is considered as established due to the previous confrontations between the armed group and the host state government. The same goes for the criterion of the organization of the parties. As a result, the military operations of the intervening states are integrated into the pre-existing NIAC and are regulated by IHL since the very first strike. This has been confirmed in the context of the coalition's airstrikes against ISIL in Iraq: we are not aware of any state having argued that IHL does not apply to the initial phase of its military operations claiming that the hostilities between its forces and ISIL had not yet reached the required level of intensity. ³⁸ This classification is the logical consequence of the invitation addressed by Iraq to the intervening states. The opposite solution would result in IHL not being able to regulate the initial phase of the operations of invited states, thus allowing the host state to circumvent the application of IHL by inviting foreign states to fight a non-state actor in its stead. ³⁹

One case needs to be distinguished from the aforementioned analysis: Turkey's intervention in Iraq in December 2015. At the beginning of December 2015, several hundred Turkish soldiers entered Northern Iraq without Iraq's consent. ⁴⁰ These actions were considered by Iraq as a hostile act ⁴¹ and Turkey was asked to withdraw its troops from Iraqi territory within 48 hours. ⁴² Iraq complained that Turkey's actions were ‘in flagrant violation of the provisions and principles of the Charter of the United Nations’ and were ‘an act of aggression under the Charter and the relevant provisions of international law’. ⁴³ The invocation of Turkey's jus ad bellum violation by Iraq is indicative of the absence of consent by Iraq with respect to the presence of Turkish forces in its territory. ⁴⁴ Although Turkey did withdraw part of its troops during the following days, ⁴⁵ on 30 December 2015, the Iraqi Prime Minister denounced the continued presence of Turkish soldiers in Iraqi territory. ⁴⁶ Given the absence of Iraq's consent concerning the presence of the Turkish troops within its territory ⁴⁷ this situation constitutes an IAC, despite the absence of direct hostilities between the forces of the two states, according to Common Article 2, paragraph 2 of the Geneva Conventions. ⁴⁸

2.1.2. Applicable law

What is the applicable law to the NIAC in Iraq? In view of the universal state participation to the Geneva Conventions, there is no doubt that Common Article 3 to the Geneva Conventions applies to this conflict, along with customary IHL. ⁴⁹ Also, conventions imposing an obligation on states parties to ‘never, under any circumstances’ use certain weapons ⁵⁰ are also applicable independently of whether Iraq is a party to these treaties or not. Things get more complicated with other IHL conventions specifying that they apply to NIACs occurring in the territory of a high contracting party. This is for example the case with the 1954 Hague Convention for the protection of cultural property in the event of armed conflict, ⁵¹ the 1999 Second Protocol to this Convention, ⁵² or the 1977 Second Additional Protocol to the Geneva Conventions. ⁵³ The point of reference in this case is Iraq, as the state on whose territory the conflict takes place. The fact that Iraq is not party to some of these Conventions, such as the 1977 Second Additional Protocol and the 1999 Second Protocol to the 1954 Hague Convention, means that the NIAC against ISIL does not take place ‘in the territory of’ a high contracting party’. Therefore, these instruments do not apply to operations against ISIL in Iraq, ⁵⁴ even when these operations are conducted by states parties to the two Protocols. ⁵⁵

It should also be noted that Iraq acceded to the 1980 Certain Conventional Weapons Convention and to all its Protocols on 24 September 2014. Following the amendment adopted in 2001, the Convention and its Protocols apply to NIACs ‘occurring in the territory of one of the High Contracting Parties’. ⁵⁶ After the entry into force of the Convention and its Protocols with respect to Iraq (six months after 24 September 2014 ⁵⁷ ), the NIAC will be ‘occurring in the territory of’ a High Contracting Party and, thus, both Iraq and the intervening states will be bound to apply the Convention and its Protocols. ⁵⁸

2.2. Conflict against the Islamic State in Syria

The evolution of ISIL in Syria has been parallel to the one in Iraq. According to the fifth report of the UN Commission of inquiry on Syria, from May to July 2013, ISIL developed its own stronghold in the north of Syria. ⁵⁹ Since August 2013, ISIL is reported as having reinforced its control over areas in northern Syria, running them through local administrations. ⁶⁰ In the eighth report of the UN Commission of inquiry on Syria (covering the period from 20 January to 15 July 2014), ISIL is described as being ‘far better organised and financed owing to the seizure of considerable resources and military equipment in Iraq’ and as having ‘consolidated its control over large areas in northern and eastern governorates’ of Syria. ⁶¹ Some direct confrontations with Syrian governmental forces are reported, although it seems that the main bulk of hostilities was directed towards other opposition groups. ⁶²

After the proclamation of the caliphate on 29 June 2014 and in view of the atrocities reported, a US-led coalition of states launched a military campaign against ISIL and other terrorist groups in Syrian territory in September 2014. ⁶³ However, contrary to Iraq, in this case there was no formal request from the Syrian government addressed to intervening states. The only state that was invited to conduct military operations in Syria was Russia. ⁶⁴ Moreover, although the coalition has occasionally coordinated its operations with the ones by rebel groups fighting ISIL and offered assistance to such groups, there does not seem to be sufficient evidence to consider that these groups operate under the overall control of the intervening foreign states. ⁶⁵

2.2.1. Classification of conflicts

Identifying with precision the beginning of the NIAC opposing the Syrian governmental forces and ISIL is subject to the same difficulties as the ones already exposed with respect to the NIAC in Iraq. According to the UN Commission of Inquiry, ISIL committed a series of war crimes during the second half of 2013. ⁶⁶ This means that the Commission considers IHL to be applicable to ISIL as a belligerent party to a NIAC. However, it is not clear whether, in the eyes of the Commission, the war crimes committed at the relevant period took place only in the context of NIACs opposing ISIL to other rebel groups, or whether there was also a NIAC between ISIL and the Syrian governmental forces. Be that as it may, in 2014, in view of the direct (albeit not numerous) confrontations between the Syrian army and ISIL, the number of casualties, the extent of material destruction, the involvement of the UN Security Council and, most importantly, the exercise of control by ISIL over part of the Syrian territory, it appears clear that a NIAC existed between the Syrian government and ISIL. ⁶⁷

In terms of classification of conflicts, Russia's intervention in Syria is similar to the intervention of the US-led coalition in Iraq: Russia joins in the pre-existent NIAC between the Syrian armed forces and ISIL and becomes a party to the conflict alongside the Syrian government. ⁶⁸ The developments made above apply here mutatis mutandis. ⁶⁹

The classification of the military operations conducted by the US-led coalition in Syrian territory against ISIL is much more challenging. There is no doubt that IHL applies to the coalition's airstrikes in Syria. ⁷⁰ The question is whether we are before a conflict that is international, non-international or both. In this author's view, the crucial element in this respect is the consent to the foreign intervention on behalf of the government of the state on whose territory the intervening states are conducting their military operations. This question has been dealt with in detail in a previous contribution to this issue. ⁷¹ If the consent exists, then the classification of the conflict involving the US-led coalition is similar to the classification of the coalition's intervention in Iraq or of Russia's intervention in Syria: a NIAC opposing the states of the coalition to ISIL. There is no IAC because there is no opposition between two or more states. ⁷²

What if no consent exists on behalf of the Syrian government? Then the US-led coalition conducts hostilities in the territory of Syria against ISIL without the consent of the Syrian government but also without any open hostilities between the coalition and Syrian governmental forces. The classification of such conflicts has been the subject of extensive debate. Some scholars consider such conflicts as purely non-international: the absence of military confrontation between the intervening state(s) and the state on whose territory the intervention takes place implies that no conflict can be considered as existing between them; thus, the only conflict which does exist is the one between the intervening state(s) and the non-state armed group – in this case, between the US-led coalition and ISIL. Since one of the belligerent parties is not a state, this conflict can only be a NIAC. ⁷³ Another view is that such conflicts are purely international because they cross the borders of a state. ⁷⁴ However, if one accepts, as the majority of state practice and scholars do, that it is the identity of the belligerent parties and not territory which is decisive for the classification of conflicts, ⁷⁵ it is difficult to see how such conflicts may be classified as purely international. A third view considers these conflicts as sui generis or ‘hybrid’, neither IAC nor NIAC. ⁷⁶ This proposition did not find any support in state practice, and thus remains a proposal de lege ferenda. ⁷⁷

A fourth view considers that military interventions such as the one conducted by the US-led coalition in Syria trigger an IAC between the intervening state(s) and the state on whose territory the hostilities take place. ⁷⁸ The absence of consent on behalf of the latter state is the crucial element in classifying the conflict as international, in other words, as opposing two (or more) states. In our view, this is indeed the most plausible classification under existing IHL.

Three main arguments may be raised against this conclusion. First, it may seem counterintuitive to assert that an IAC exists between two states despite the absence of active military confrontation between them. However, the absence of hostilities is not decisive for the purposes of the classification of a conflict as international. In this respect, Common Article 2, paragraph 2 to the four Geneva Conventions recognizes the applicability of the Conventions (and thus the classification of the situation as an IAC) to an occupation which meets with no armed resistance on behalf of the occupied state. ⁷⁹ This paragraph confirms the applicability of the Conventions to situations where the states involved do not actually engage in open hostilities, ⁸⁰ for example because the targeted state is not in a position to react to the military operations of another state. This may be because it has no army, or because – as may very well be Syria's current position – its armed forces are already involved in another armed conflict and the state is reluctant or unable to open a new front of active hostilities against militarily powerful states. When one or more states resort to multiple airstrikes ⁸¹ affecting another state's territory and population without its consent, it is logical to conclude that such strikes trigger an IAC, irrespective of whether the targeted state reacts militarily or not. Moreover, if military reaction by the targeted state were a necessary condition for the existence of an IAC, this would lead to the absurd result that, in every IAC, the very first attack conducted by the attacking state(s) would not be regulated by IHL. This author is not aware of any such claim having been put forth by states.

Second, those denying the existence of an IAC in situations like the coalition's intervention in Syria invoke the absence of an animus belligerendi on behalf of the intervening state(s) against the state in whose territory the intervention takes place. This argument was raised, for example, with respect to the classification of the conflict triggered by Colombia's intervention against the FARC in the territory of Ecuador, without the latter's consent: it has been asserted that no IAC existed between Colombia and Ecuador, because the use of force by Colombia was directed against the FARC not against Ecuador. ⁸² In this author's view, this argument is unconvincing. Firstly, it is highly doubtful whether the subjective intentions of a party to the conflict may be allowed to dictate the application of IHL. ⁸³ After all, this influence was precisely what the states that negotiated the Geneva Conventions sought to prevent by replacing ‘war’ with the notion of ‘armed conflict’, a replacement which was meant to make the application of IHL dependent only on the facts on the ground. ⁸⁴ The relevance of the animus of the parties for the classification of conflicts is a minor one: it serves to exclude cases where force is used by mistake or accident (such as the targeting of the Chinese embassy during the 1999 NATO bombings in Kosovo) or violent acts by individual members of armed forces, not ordered or approved by their hierarchy. In other words, it excludes cases of unintentional resort to force by a state against another state. ⁸⁵ Secondly, from the moment a state conducts military operations within the territory of another state without its consent, these operations are, by definition, not only directed against the targeted non-state actor but also against the territorial state itself. Thus, it is difficult to see why, in order for an IAC to exist, it is necessary for a state to actively wish to target another state with the specific purpose of destroying its forces. When the US-led coalition conducts its 3500 strikes in Syria, even if the final aim of the strikes is ISIL, the damage resulting from the strikes on population and infrastructure is damage suffered by Syria. In pursuing its ultimate objective of fighting ISIL, the coalition accepts such damages as an unavoidable consequence of its actions. In this sense, the coalition's actions are also directed against Syria. When a man fathers a baby, he does not need to be actively seeking to do so in order to be recognized as the father and be bound by the legal obligations flowing from this action. Similarly, when a state conducts hostilities and destroys targets within the territory of another state without the latter's consent, it does not need to be actively seeking to target that state in order to be recognized as a party to an international conflict with that state; the conflict exists and the intervening state is bound by the legal obligations flowing from this factual reality.

Three, it has been suggested that placing consent at the centre of the debate on classification of conflicts conflates jus ad bellum and jus in bello. ⁸⁶ This argument is grounded on the principle that the application of jus in bello should not be influenced by the legality of the use of force under jus ad bellum (the so-called principle of the equality of belligerents). ⁸⁷ It is of course correct that the question whether a resort to force violates jus ad bellum is distinct from the question of whether the said resort to force triggers the application of jus in bello, and that the two questions should be examined separately. ⁸⁸ It is however difficult to see why ‘[t]he underlying question for classification must be that of identifying the parties to the conflict, rather than consent’. ⁸⁹ In this case, consent is a crucial element in the identification of the parties to the conflict. Furthermore, consent is not exclusively a jus ad bellum rule to begin with. It is rather a concept of general international law, as is attested by its inclusion among the circumstances precluding wrongfulness in the ILC draft articles on state responsibility. ⁹⁰ As such, consent may be examined, on the one hand, in order to determine whether jus ad bellum has been violated and, on the other hand, in order to establish whether an armed conflict between two states exists. There is a case of conflation only if the IAC is considered as existing because jus ad bellum has been violated; if, in other words, the classification of the conflict as an IAC is considered as being the legal consequence of the violation of jus ad bellum. ⁹¹ With the exception of such reasoning, consent may influence both the application of jus ad bellum and that of jus in bello. The fact that the same concept of general international law may be taken into account in the analysis of jus ad bellum and jus in bello does not mean that this analysis conflates the two sets of rules. ⁹²

The ICJ seems to have followed the same approach in its 2005 judgment on the Armed Activities in the Territory of the Congo case. The Court established that the Ugandan armed forces were present inside Congolese territory since September 1997, following an invitation by the Congolese President, which was withdrawn in August 1998. ⁹³ The Court engaged in a detailed analysis in order to determine when the consent by the Democratic Republic of the Congo was withdrawn, ⁹⁴ and on the basis of the results held that Uganda had violated jus ad bellum since August 1998. ⁹⁵ Turning to the violations of IHL, the Court did not dwell at all on establishing the date of the beginning of the conflict. Significantly, both the states parties to the dispute and the Court seem to have taken for granted that the relevant facts for the application of IHL (including the determination of the existence of an occupation) were actions by Ugandan armed forces that took place after the Democratic Republic of the Congo withdrew its consent. ⁹⁶

Aside from the above, by virtue of the equality of the belligerents principle, the legality of the coalition's intervention under jus ad bellum does not affect the classification of conflicts and the application of jus in bello. In other words, even if the coalition's intervention in Syria does not violate jus ad bellum (for example, because the right to self-defence against non-state actors is accepted), this does not affect the existence of an IAC between the intervening states and Syria.

In conclusion, the airstrikes of the US-led coalition trigger an IAC between the coalition and Syria. This means that there are two distinct conflicts currently taking place in Syria: a NIAC between the coalition and ISIL and an IAC between the coalition and Syria. ⁹⁷

One final question that needs to be addressed in this context relates to the identification of the parties to the two conflicts. In the context of the IAC between the coalition and Syria, any state that has conducted airstrikes in Syrian territory should be considered party to the conflict. Turning to the NIAC against ISIL, a distinction should be made between those states of the coalition that also participate in the NIAC against ISIL in Iraq (e.g., the US, Canada, or Australia), and the states who participate only in the strikes against ISIL in Syria (e.g., Saudi Arabia, the United Arab Emirates, Bahrain, etc.). ⁹⁸ For the first group of states, there is little doubt that they are parties to a NIAC. In their case, the strikes in Syria can be seen as the mere continuation of their NIAC against ISIL in Iraq. The jus ad bellum argument according to which the coalition states intervene in Syria on the basis of the – extraterritorial – exercise of Iraq's right to collective self-defence ⁹⁹ is also an indication pointing in this direction.

What about the states of the coalition intervening against ISIL for the first time in Syria? Taken individually, the hostilities between them and ISIL do not immediately rise to the level of intensity required for a NIAC to exist. This would mean that the very first military operations by Saudi Arabia, or Bahrain conducted in Syria would not be regulated by IHL. To the best of our knowledge, none of the coalition states has invoked this argument in order to refuse the application of IHL. In reality, applying this reasoning would result in a very complex situation in terms of the temporal – not to mention territorial – scope of application of IHL. Imagine an airstrike in Syria, in October 2014, conducted jointly by US and Saudi aircrafts: while US aircrafts would operate under IHL rules, the Saudi aircrafts would not. This means, for example, that while US would be allowed to invoke the principle of proportionality in order to justify collateral damage, this argument could not be invoked by Saudi Arabia. Along the same lines, the commission of war crimes would only be conceivable for the US pilots but not for the Saudi ones. To conclude, it seems clear that, instead of evaluating the existence of the NIAC separately for each member of the coalition, the coalition's operations should be viewed as a whole, both in Iraq and Syria. In this context, all airstrikes against ISIL are carried out under the same military operation, Operation Inherent Resolve. ¹⁰⁰ This analysis of the parties to the conflict reflects the facts on the ground and is thus in conformity with the ‘principle of effectiveness’, ¹⁰¹ which is central in applying IHL.

2.2.2. Applicable law

Common Article 3 and customary IHL apply to the NIAC between the coalition states and ISIL, as do the conventions that prohibit the use of certain weapons in all circumstances. ¹⁰² As was the case with foreign interventions in Iraq, ratification by Syria is crucial with respect to conventions that apply to NIACs occurring in the territory of a high contracting party. ¹⁰³

The IAC between the coalition states and Syria triggers the application of all the relevant IHL instruments to which the states involved are parties, as well as of customary IHL. The parallel existence of an IAC and a NIAC raises interesting questions in terms of the applicable law. It is submitted that, in the vast majority of cases, the military operations of the coalition will simultaneously be part of both conflicts. For example, on 24 September 2014 one of the first airstrikes in Syria targeted ISIL-controlled oil refineries. ¹⁰⁴ Being directed against ISIL, the strike was undoubtedly part of the NIAC between the coalition and ISIL. However, the destruction of oil refineries constitutes destruction of Syrian infrastructure and as such the strike is also part of the IAC between the coalition and Syria. In general, when the coalition strikes destroy ISIL-held refineries, roads, bridges or buildings, they destroy Syrian infrastructure. When they result in civilian casualties, they kill or injure the Syrian civilian population. Leaving aside attacks against ISIL with no impact whatsoever on Syrian territory, infrastructure or population (for example, a sniper taking out an ISIL fighter), in all other cases, it is impossible to distinguish between the two conflicts. Therefore, the relevant airstrikes will necessarily be considered as part both of the NIAC against ISIL and of the IAC with Syria. This implies that, in order for these operations to be lawful under IHL, they have to respect the rules applicable in IACs. Given the current nature of the coalition's military operations (airstrikes), this parallel application of IHL rules regulating IACs and NIACs does not change much in terms of applicable law, since most of the rules relating to the conduct of hostilities have acquired customary status and are applicable to both conflicts anyway. ¹⁰⁵ It may however prove important in terms of (a) the application of some rules whose customary status is contested by coalition states, ¹⁰⁶ and (b) war crimes. Thus, for example, an attack by France or the UK against ISIL in Syria violating the principle of proportionality is part of both a NIAC and an IAC and, having a nexus with an international conflict, may constitute a war crime under Article 8, paragraph 2(b)(iv) of the Rome Statute. ¹⁰⁷ The same goes for attacks directed against civilian objects, which are also among the Statute's war crimes applicable only to IACs. ¹⁰⁸ It should be noted that the parallel existence of an IAC and NIAC is only valid for the intervening states and Syria. ISIL on the other hand is only involved in a NIAC. This means that, should the coalition states decide to put forces on the ground in Syria, if ISIL fighters launch a disproportionate attack against them, this attack will not be considered as a war crime under the Rome Statute since it takes place in the context of a NIAC.

Having set out the main aspects with respect to the classification of conflicts and the applicable law to the coalition's military operations against ISIL, we will now turn to one of the most interesting parts of the application of IHL to the fight against ISIL: the geographical scope of application of IHL.

3. The geographical scope of application of international humanitarian law in the context of the fight against ISIL

The territory controlled by ISIL already spreads in two different countries. Moreover, as was already noted in the introduction, attacks by or linked to ISIL have taken place in several countries, including Tunisia, Libya, and France. ¹⁰⁹ The coalition fighting ISIL in Iraq and Syria is also composed of states from around the world. ¹¹⁰ Finally, ISIL itself is an organization of transnational character since thousands of nationals by several countries have joined its ranks, ¹¹¹ while 34 groups from all around the world have reportedly pledged allegiance to it. ¹¹² How should the geographical scope of application of IHL be appreciated in this case?

At this stage of the hostilities, where military operations basically consist of airstrikes conducted by the coalition, this question translates into whether IHL is applicable to an airstrike targeting ISIL fighters independently of where they are found. In other words, can the coalition invoke IHL in order to justify a strike against ISIL fighters which is conducted anywhere in the Syrian or Iraqi territory (Section 3.1), in the territory of a state party to the coalition such as the US, France or Belgium (Section 3.2), and in a third state, such as Tunisia (Section 3.3)? As explained in the introduction, the interest in invoking IHL is that, in some cases, its rules appear to be more permissive than the IHRL rules regulating resort to force. ¹¹³

3.1. Is IHL applicable to an airstrike against an ISIL fighter anywhere in Iraq and Syria?

The first question is whether IHL applies to the targeting of ISIL fighters anywhere in Iraqi or Syrian territory, namely also in an area far from the battlefield or ‘the actual theatre of combat operations’. ¹¹⁴ There are no specific provisions in IHL conventional instruments dealing with the geographical scope of application of NIACs. Thus, the starting point of the analysis is the following dictum by the ICTY in 1995 Tadić Appeals Chamber decision:

the temporal and geographical scope of both internal and international armed conflicts extends beyond the exact time and place of the hostilities. . . . [I]nternational humanitarian law appl[ies] in the whole territory of the warring States or, in the case of internal conflicts, the whole territory under the control of a party, whether or not actual combat takes place there. ¹¹⁵

Confirmed by subsequent case law, ¹¹⁶ and shared by the ICRC ¹¹⁷ and many scholars, ¹¹⁸ this statement points towards the application of IHL to the entire territory of Iraq and Syria.

The opposite view is that the application of IHL rules relating to the conduct of hostilities does not extend beyond the ‘actual context of hostilities’. ¹¹⁹ Rather surprisingly, this view seems to find some support in ICTY case law. Indeed, in reaching its conclusion on the broad geographical scope of application of IHL in NIACs, the Appeals Chamber apparently had in mind the provisions relating to the treatment of persons deprived of their liberty, or persons hors de combat in general. ¹²⁰ In analysing the geographical scope of application of the Geneva Conventions, the Appeals Chamber held that ‘some of the provisions are clearly bound up with the hostilities and the geographical scope of those provisions should be so limited’. ¹²¹ This statement was made in the context of the analysis of IHL provisions applicable in IACs and not in NIACs. Be that as it may, the reasoning behind the distinction put forth by the Appeals Chamber – namely that there are some IHL rules which are ‘bound up with hostilities’ and whose geographical scope of application should be limited to areas of active hostilities as opposed to other IHL rules dealing, for example, with detainees – is equally valid for NIACs.

This view has met with the objection that it would preclude a state from attacking a rebels’ munitions store far from the area of active hostilities, ¹²² or, in general, from striking the rebels in its rear area. ¹²³ Moreover, in most conflicts, it would be difficult to clearly identify a stable area of ‘active hostilities’ in the first place. In its case law, when determining the geographical scope of application of IHL, the ICTY does not seem to have distinguished between rules relating to the conduct of hostilities and rules relating to the protection of victims of war. For example, there is no indication that the Tribunal appreciates the ratione loci applicability of IHL differently depending on whether the rule is one regulating the conduct of hostilities or the protection of persons hors de combat. ¹²⁴

Therefore, ratione loci, IHL would be applicable to an airstrike launched against an ISIL member anywhere in the territory of Iraq or Syria. This position seems to be confirmed by state practice in the context of the NIAC in Afghanistan: ISAF troop contributing countries do not appear to limit the application of IHL only to areas of active hostilities. ¹²⁵

Whether IHL may successfully be used to justify such an attack as a legal regime more permissive than IHRL is a different story. The ICRC, for example, distinguishes between ‘collective hostilities opposing identifiable units’ and the ‘targeting of isolated individuals’:

[i]n situations of collective hostilities opposing identifiable units, IHL/LOAC clearly applies even to engagements taking place away from the main battlefields (think for instance of a one-off attack launched against a military installation or troop convoy in a location with no prior history of military confrontations). But it certainly gets more complicated when on contemplates the targeting of isolated individuals taking place outside of the active battle zones. . . . Within the territory of the country where the NIAC is taking place – and this is a hotly debated issue – the resort to lethal force under the IHL/LOAC principle of military necessity may not always be acceptable depending on the surrounding circumstances, taking also into consideration the restraints imposed on specific means and methods of warfare, and without prejudice to further restrictions that may arise under other applicable branches of international law. ¹²⁶

Few would disagree with the proposition that IHL applies to collective hostilities opposing identifiable units away from the main theatre of operations. As for the IHL/IHRL conundrum, in such ‘battlefield-like’ situations, IHL and IHRL do not seem to lead to substantially different results anyway and the use of lethal force would probably be legal under both sets of rules. ¹²⁷

In reality, the IHL/IHRL controversy revolves mostly around the targeting of isolated individuals far from the battlefield. In this last case, several considerations come into play: what is the interplay between IHL and IHRL? What are the limits (if any) placed by military necessity on the use of force under IHL? Is the targeted person a member of the armed group (does he/she assume a continuous combat function) ¹²⁸ or is he/she directly participating in hostilities? ¹²⁹

We will examine the concrete example of an attack claimed by ISIL that killed at least 10 people near the city of Basrah, in Southern Iraq, in early October 2015. ¹³⁰ Basrah is far from any area controlled by or subject to hostilities involving ISIL. Assuming that the attack was indeed conducted by an ISIL fighter assuming a continuous combat function, the question is whether the Iraqi government could launch an airstrike against the author of the attack. Having determined that both IHL and IHRL ¹³¹ apply in this case, the question is which of the two legal regimes takes precedence over the other. Also, is the answer the same if the ISIL fighter is attacked (a) while he/she is driving towards the place where the bomb will be planted and (b) while he/she is sleeping at a hotel in Basrah, three days before the attack?

To be sure, the appreciation of the legality of such operations is very difficult to make in the abstract and will necessarily involve ‘a complex assessment based on a wide variety of operational and contextual circumstances’. ¹³² With this caveat in mind, launching an airstrike against an ISIL fighter on a deserted road while he/she is driving towards the place of the intended attack may very well be legal under both IHL and IHRL as the only way to prevent the explosion from taking place. In this scenario, the death of the fighter and the potential civilian victims of the airstrike, for example the passengers of another car that happens to be driving by the same road at the moment of the strike, would be justified under IHL as collateral damage and may equally be considered as necessary under IHRL. ¹³³ However, if the two sets of rules are found to result in different outcomes regarding the legality of the strike – if the strike is considered to be illegal under IHRL – it has been suggested that the circumstances of the case may very well justify IHL being considered as lex specialis:

the impossibility of arresting a fighter, the danger inherent in an attempt to do so, the danger the fighter represents for government forces and civilians and the immediacy of this danger may cause the conclusion to be reached that humanitarian law is the lex specialis in that situation. ¹³⁴

It is probably the scenario under (b) that would prove to be more problematic. Indeed, in this case, it is doubtful that an airstrike could be covered by the required degree of necessity under IHRL. However, it is equally debatable whether IHL itself would justify such an attack. According to recommendation IX of the ICRC's interpretative guidance on the notion of direct participation in hostilities, ‘the kind and degree of force which is permissible against persons not entitled to protection against direct attack must not exceed what is actually necessary to accomplish a legitimate military purpose in the prevailing circumstances.’ ¹³⁵

The ICRC's position is that the combination of the principles of humanity and military necessity in IHL requires that ‘no more death, injury, or destruction be caused than is actually necessary for the accomplishment of a legitimate purpose in the prevailing circumstances’. ¹³⁶ Although, here as well, everything depends on the appreciation of the relevant circumstances, the commentary to the recommendation IX asserts that this restriction ‘may become decisive where armed forces operate against selected individuals in situations comparable to peacetime policing . . . where a party to the conflict exercises effective territorial control’. ¹³⁷ One of the examples given is the scenario of a civilian directly participating in hostilities via his /her mobile phone while sitting in a restaurant. The commentary states that:

‘[s]hould the restaurant in question be situated within an area firmly controlled by the opposing party it may be possible to neutralize the military threat posed by the civilian through capture or other non-lethal means without additional risk to the operating forces or the surrounding civilians.’ ¹³⁸

This interpretation does seem to reflect the views of some states. ¹³⁹ Among scholars, recommendation IX has proven controversial, being contested by some ¹⁴⁰ and receiving support from others. ¹⁴¹ Be that as it may, under this view, even if IHL were applied as lex specialis, it would not justify the targeted killing of the ISIL fighter in the restaurant or in his hotel room. All in all, the controversies relating to the interpretation of IHL and IHRL show that the debate over the geographical scope of application of IHL may not be as decisive an element in determining the legality of an airstrike as it is sometimes thought to be.

3.2. Is IHL applicable to an airstrike against an ISIL fighter in the territory of a state party to the coalition?

The question that lies in the heart of this subsection is whether an ISIL fighter can lawfully be targeted under IHL while he/she is found in the territory of a state party to the coalition operating in Iraq and Syria, such as France or Belgium. ¹⁴² Considering that the geographical scope of application of IHL extends to the territory of the states intervening in the fight against ISIL means that, in these states as well, an attack against an ISIL fighter will have to be examined both under IHL and IHRL. This again raises the question of the interplay between the two sets of rules. As in the scenario examined in the previous section, what is at stake here is the possibility to invoke IHL as a justification for what may otherwise be a violation of IHRL.

The question of the application of IHL to the territory of states parties to an extraterritorial NIAC is unregulated by conventional IHL. States members of the coalition conducting airstrikes against ISIL in Iraq and Syria certainly qualify as belligerent parties in the NIAC against ISIL. Thus, it is plausible to consider attacks by ISIL in the territory of one of these states as falling within the context of the on-going armed conflict between the coalition and ISIL and, therefore, as regulated by IHL. As suggested in legal scholarship, this would be in line with the ‘widely accepted territorial interpretation of the scope of IHL that is independent from the concept of hostilities and extends to the geographical borders of the relevant state(s)’. ¹⁴³

On the other hand, state practice with respect to such cases indicates that attacks by ISIL outside the territory of Iraq and Syria have been treated under a law enforcement paradigm rather than under IHL. France's reaction to the 13 November 2015 attacks is a telling example in this respect. Despite the bellicose rhetoric adopted by French authorities in the aftermath of the attacks, ¹⁴⁴ the reaction to the attacks on behalf of the French and Belgian authorities was a typical law enforcement operation of search for and arrest of suspected criminals.

However, some degree of caution is necessary when evaluating such practice, since it may very well be reflective of a political choice rather than a legal conviction that IHL is not applicable and cannot be invoked at all. In deciding how to deal with suspected terrorists hiding in an apartment in Paris the French authorities chose a law enforcement action (raid by police forces) ¹⁴⁵ instead of open hostilities tactics, such as destroying the apartment with a missile and then justifying the related deaths and damage by invoking direct participation in hostilities and IHL's principle of proportionality. Even if France considered IHL as applicable in its territory, this approach is easy to understand. With this caveat in mind, state practice up to now tends to show that states have not applied IHL to attacks by ISIL inside their territory.

The ICRC's view is that IHL is applicable to the territory of belligerent states involved in an extraterritorial NIAC. Although the ICRC acknowledges that state practice is inconclusive, it invokes the principle of the equality of belligerents as an argument pleading in favour of the application of IHL to the territory of all the states involved in the NIAC, because ‘assisting States involved in an extraterritorial NIAC should not be able to shield themselves from the operation of the principle of equality of belligerents under IHL once they have become a party to this type of armed conflict beyond their borders.’ ¹⁴⁶

It should be noted that the principle of equality of belligerents does not benefit much the rebel groups (in this case, the ISIL fighters) operating in the territory of an enemy state. Even when the attack by the rebels is directed against a military objective (a French military compound or Belgian armed forces in Brussels), lawful as this attack may be under IHL, it will still constitute a crime under national law, since IHL rules applicable in NIACs do not give to non-state actors a right to participate in hostilities. ¹⁴⁷ In other words, the state does not really suffer any severe consequences from ‘exposing’ itself to IHL and the equality of belligerents principle rather than ‘shielding’ itself from it. This is not to say that the application of IHL may not have any consequence whatsoever. For example, if it is accepted that incidents such as the Paris attacks amount to hostilities in the context of a NIAC and are regulated by IHL, this may impose some limits on the weapons that may lawfully be used in response to such incidents. ¹⁴⁸ Along the same lines, the application of IHL may prevent actions by the rebels from being considered as terrorist offences. ¹⁴⁹

In any case, it is doubtful whether the application of IHL in the territory of states parties to the conflict with ISIL would grant much leeway to the states involved. Indeed, from a legal point of view, the situation analyzed here is not different from the one examined above of an ISIL attack in an area under the control of the Iraqi government, for example in Southern Iraq. Therefore, at least if one follows the ICRC's approach as set out in recommendation IX, applying IHL in France or Belgium does not offer significant leeway with respect to what would be authorized under IHRL.

3.3. Is IHL applicable to an airstrike against an ISIL fighter in the territory of a third State?

The third and final situation, that of an airstrike against an ISIL fighter in the territory of a state non-party to the NIAC with ISIL, for example Tunisia, is rather more clear-cut.

The argument in favour of applying IHL in these cases has been only marginally invoked, and was developed mainly in the so-called ‘global war on terror’, a concept put forth by the US in the context of its fight against al-Qaeda. This argument views the law of armed conflict as applicable wherever a member of an armed group is found. ¹⁵⁰ According to this view the ratione personae scope of application of IHL rules defines their ratione loci scope of application; the fighter thus carries the conflict with him/her wherever he/she goes. Emphasis is placed on the ‘need to move away from geographic-based ideas of applicability of the law’ and to establish the nexus between the conduct at issue and the conflict as the decisive factor in determining the geographical application of IHL. ¹⁵¹

The ICRC has explicitly rejected this view, pointing to the ‘essentially territorial focus of IHL’ and asserting that the argument of a territorially unhinged application of IHL is not supported by state practice:

[a] territorially unbounded approach would imply that a member of an armed group or an individual civilian directly participating in hostilities would be deemed to automatically “carry” the “original” NIAC wherever they go when moving around the world. Thus, based on IHL, they would remain targetable within a potentially geographically unlimited space. With very few exceptions, State practice and opinio juris . . . do not seem to have accepted this legal approach and the great majority of States do not appear to have endorsed the notion of a “global battlefield.” In addition, in practical terms it is disturbing to envisage the potential ramifications of the territorially unlimited applicability of IHL if all States involved in a NIAC around the world were to rely on the concept of a “global battlefield.” ¹⁵²

Turning to the concrete geographical scope of application of the fight against ISIL outside Iraq, Syria and the territories of coalition states participating in the airstrikes, it is crucial to determine the actual reach of ISIL. As it was set out in the introduction, several attacks in territories of third states such as Libya or Tunisia have been claimed by ISIL or one of the 34 armed groups allegedly related to it. ¹⁵³

It is highly doubtful whether the mere ideological affiliation and the act of pleading allegiance to ISIL is enough to expand the NIAC existing in Iraq and Syria to the territory of the states in which these groups operate. Unless it is proven that, in reality, the various factions operating in third countries form part of the same armed group as ISIL, such factions will not be considered as parties to the NIAC against ISIL. ¹⁵⁴ Consequently IHL will not be applicable in the territory of their respective states on that basis. Thus, if IHL is applicable at all, it will be because the two conditions necessary for the existence of a NIAC (organization of the parties and intensity of hostilities) are fulfilled independently of the NIAC involving ISIL. ¹⁵⁵

4. Concluding remarks

As the previous developments have made clear, the armed conflict against ISIL is not an unprecedented situation with respect to the application of IHL. Indeed, at least some aspects of the conflict are far from exceptional: the armed conflicts between Syria and ISIL, or between Iraq and ISIL with the participation of third states in the conflict alongside Iraq following an invitation by the Iraqi government, are all classic examples of rather commonly experienced NIACs. In this regard, the precedent of the conflict against ISIL merely confirms and reinforces the already established approach as to classification of conflicts and applicable law.

The coalition's intervention in Syria is much more controversial in terms of its classification under IHL, albeit, again, not unprecedented. In this case, the intervention is the most recent example of a new type of conflicts that legal scholars have termed ‘transnational’. ¹⁵⁶ As it was observed above, until now, this precedent has not generated enough state practice in terms of views relating to classification and applicable law. On the basis of the existing IHL definitions, this author considers that – unless the consent of the Syrian government is proven – the coalition's intervention in Syria gives rise to an international conflict between the coalition states and Syria which is distinct from the NIAC between the coalition and ISIL. In turn, this means that, for the coalition's operations in Syria to be in conformity with IHL, they have to respect the IHL rules applicable in IACs.

Turning to the geographical scope of application of IHL in relation to the fight against ISIL, three specific questions were examined: first, whether IHL applies to the entire territory of Iraq and Syria; second, whether it applies to the territory of the intervening states which are parties to the conflict(s) with ISIL; and, third, whether it applies anywhere an ISIL fighter may be found, that is without any territorial limitation. International practice points to a positive answer with respect to the first question and to a negative one with respect to the third. As to the second question, there is a good argument to be made for applying IHL to the territories of the intervening states in the conflict against ISIL. However, answering the question of the application of IHL is only the first step in determining the legality of an attack against ISIL fighters. The application of IHL does not imply that such an attack will necessarily be lawful at all times. The decisive element in this respect depends to a large extent on the interpretation of the substantive rules of IHL and IHRL regulating resort to force. And, despite all the relevant studies, there is sufficient controversy on this matter to justify further research, and, by the same token, to urge caution in giving any definitive answers.