Introduction
Although international humanitarian law has as its aim the limitation of the effects of armed conflict, it does not include a full definition of those situations that fall within its material field of application. While it is true that the relevant conventions refer to various types of armed conflict and therefore afford a glimpse of the legal outlines of this multifaceted concept, these instruments do not propose criteria that are precise enough to determine the content of those categories unequivocally. A certain amount of clarity is nonetheless needed. In fact, depending on how the situations are legally defined, the rules that apply vary from one case to the next. The legal regimes that need to be taken into account are thus not always the same and depend on whether the situations constitute, for example, an international or a non-international armed conflict. Similarly, some forms of violence, referred to as ‘internal tensions’ or ‘internal disturbances’, do not reach the threshold of applicability of international humanitarian law and therefore fall within the scope of other normative frameworks.
This article proposes a typology of armed conflicts from the perspective of international humanitarian law. It sets out, first, to show how the different categories of armed conflict anticipated by that law can be interpreted in the light of recent developments in international legal practice. In that respect, it is appropriate to refer to the conceptualization efforts relating firstly to the law of international armed conflict and secondly to the law of non-international armed conflict. This article then goes on to examine various controversial cases of application. The reality of armed conflict is actually more complex than the model described in international humanitarian law – to the extent that today some observers question the adequacy of the legal categories.
The law of international armed conflict
The history of the law of international armed conflict shows that the field of application of this legal regime has been progressively extended as treaty law developed. Whereas a narrow formalistic concept of war was predominant initially, the reform of the system with the revision of the Geneva Conventions in 1949 gave precedence to a broader approach, based on the more objective concept of armed conflict. Moreover, that extension was subsequently taken up with the adoption of Additional Protocol I in 1977. That instrument added another type of conflict to the field of the law of international armed conflict, that of wars of national liberation. This legal regime also comprises a specific body of rules whose field of application is determined on the basis of an autonomous concept, that of occupation.
War and international armed conflict
By virtue of common Article 2(1), the 1949 Geneva Conventions apply to ‘all cases of declared war or of any other armed conflict which may arise between two or more of the High Contracting Parties, even if the state of war is not recognized by one of them’.Footnote 1 The situations referred to here are conflicts between States. The ‘High Contracting Parties’ mentioned in this text are sovereign entities. Depending on the case in question, these situations may take the form of a direct conflict between States or of intervention in a previously existing internal conflict. In the latter hypothesis, the conflict is ‘internationalized’. That is the case if a foreign Power sends troops into a territory to support a movement opposing the local government. Intervention may also take place by proxy when that Power merely supports and guides the uprising from a distance.Footnote 2 In that case, it is then vital to determine the level of control that makes it possible to classify the armed conflict as international. Not every form of influence necessarily leads to the conflict becoming internationalized. On that point, the International Criminal Tribunal for the former Yugoslavia (ICTY) pointed out that ‘control by a State over subordinate armed forces or militias or paramilitary units may be of an overall character (and must comprise more than the mere provision of financial assistance or military equipment or training). This requirement, however, does not go so far as to include the issuing of specific orders by the State, or its direction of each individual operation’.Footnote 3 The criterion of ‘overall control’ is achieved when the foreign State ‘has a role in organising, co-ordinating or planning the military actions of the military group, in addition to financing, training and equipping or providing operational support to that group’ (emphasis added).Footnote 4 Involvement must therefore go beyond mere logistical support, but that involvement does not imply that everything done by the group concerned is directed by the State taking part from a distance.
The situations referred to in Article 2(1) common to the 1949 Geneva Conventions are viewed from the twin viewpoints of formalism and effectiveness. First, there are declared wars, implying that the state of war is recognized officially by the parties concerned. Second, there are other forms of inter-State armed conflict, whose existence does not depend on how the parties define them. While the concept of war already exists in the oldest treaties of international humanitarian law,Footnote 5 the 1949 Conventions introduced the concept of armed conflict into this legal regime for the first time. Through this semantic contribution, those who drafted those instruments wanted to show that the applicability of international humanitarian law was henceforth to be unrelated to the will of governments. It was no longer based solely on the subjectivity inherent in the recognition of the state of war, but was to depend on verifiable facts in accordance with objective criteria. Thanks to that contribution in 1949, international armed conflict thus became established as a concept governed by the principle of effectiveness. The relevant rules apply when certain specific factual conditions are met.Footnote 6
As for the nature of those conditions, it is generally acknowledged that it must be evaluated freely, as the level of intensity required for a conflict to be subject to the law of international armed conflict is very low.Footnote 7 Situations envisaged by the relative instruments merely need to exist. Thus ‘as soon as the armed forces of one State find themselves with wounded or surrendering members of the armed forces or civilians of another State on their hands, as soon as they detain prisoners or have actual control over a part of the territory of the enemy State, then they must comply with the relevant convention’.Footnote 8 It is, however, not necessary for the conflict to extend over time or for it to create a certain number of victims.Footnote 9 In other words, an international armed conflict exists, as recalled by the ICTY, ‘whenever there is a resort to armed force between States’.Footnote 10 To be more precise, it might be said that that is the case when the circumstances are characterized by hostility between the parties. The attack must be motivated by the intention to harm the enemy, thus ruling out cases in which the use of force is the result of an error (involuntary incursion into foreign territory, wrongly identifying the target, etc.). Similarly, an international armed conflict does not exist when the targeted State has given its consent for a third State to take action in its territory (for example, to fight a non-governmental armed group).Footnote 11
Since the adoption of Additional Protocol I of 1977, the field of application of the law of international armed conflict has ceased to be limited to inter-State conflicts stricto sensu and also encompasses conflicts between government forces and some non-governmental groups, i.e. peoples fighting in the exercise of their right of self-determination. The Protocol stipulates that the situations targeted by Article 2 common to the 1949 Geneva Conventions include ‘armed conflicts in which peoples are fighting against colonial domination and alien occupation and against racist regimes in the exercise of their right of self-determination, as enshrined in the Charter of the United Nations and the Declaration on Principles of International Law concerning Friendly Relations and Co-operation among States in accordance with the Charter of the United Nations’.Footnote 12
The scope of this provision raises a number of questions of interpretation, beginning with the precise definition of the ‘peoples’ concerned and the criteria which make it possible to distinguish those situations of armed conflict from that covered by Article 3 common to the 1949 Geneva Conventions and their Additional Protocol II.Footnote 13 The two instruments referred to in Article 1(4) of Additional Protocol I are actually couched in terms that are too general to allow fully satisfactory answers to be derived from them. Moreover, it is difficult to find additional clarification in actual practice because the scenario referred to in that Article has never been officially recognized, particularly as the States that might be concerned did not ratify Additional Protocol I. The interested reader can make useful reference to the commentaries already devoted to that particular type of armed conflict.Footnote 14
Occupation
When one of the belligerents succeeds in gaining the upper hand over his adversary, an international armed conflict may take the form of occupation.Footnote 15 In the words of Article 42 of the 1907 Hague Regulations, ‘territory is considered occupied when it is actually placed under the authority of the hostile army’ (our emphasis). For occupation in the meaning of this provision to exist, two conditions must be fulfilled: (a) the occupier is able to exercise effective control over a territory that does not belong to it; (b) its intervention has not been approved by the legitimate sovereign.Footnote 16 Effective territorial control, which is at the heart of the concept of occupation, implies that a substitution of powers may take effect. That condition is fulfilled when, first, the overthrown government is unable to exercise its authority and, second, the occupying Power is in a position to fill that gap by exerting its own power.Footnote 17 This condition implies in principle that enemy troops should be deployed in the territory concerned and succeed in imposing the minimum stability that will allow them to exercise their responsibilities deriving from the law of occupation. As for the second criterion, the absence of consent, it must be understood in fairly broad terms. In particular, it is not limited to cases in which power is seized as a result of hostilities. Article 2(2) of the Fourth Geneva Convention of 1949 complements the 1907 definition by clarifying that the relevant rules apply even if the occupation ‘meets with no armed resistance’.
In some cases, territorial control is not exercised directly by the occupation forces but via a puppet government or another form of subordinate local power.Footnote 18 However, it is difficult to evaluate on a case-by-case basis the degree of influence required for this scenario to actually constitute occupation, as any interference in the affairs of a third State does not necessarily mean that occupation exists. Relations between the local authorities and the foreign forces vary in intensity depending on the circumstances and always reveal a certain reciprocal influence – or even a degree of consultation – in the decision-making process. To resolve this question, the ICTY retains – in this case, too – the criterion of ‘overall control’. Occupation exists when a State has ‘overall control’ of the local agents actually exercising ‘effective control’ over the territory in question.Footnote 19 This is, for example, the pattern of the present situation in Nagorno-Karabakh. Azerbaijan has no longer been able to exercise its sovereignty in that area since the war with its secessionist forces (supported by the Armenian army) which ended in its defeat in 1994. That territory is governed in practice (effective territorial control) by the government of the ‘Nagorno-Karabakh Republic’, as it proclaimed itself on 6 January 1992 following a locally held referendum. To date, however, no State has recognized the sovereignty of the new authority. Moreover, several aspects indicate that the republic is actually in a relation of subordination to Armenia (overall control). The military structures in the region include, in particular, a sizeable number of conscripts and officers from that country. The role of Armenia is not therefore restricted to mere logistic support but implies that it has a hand in the organization, co-ordination and planning of the power established in Nagorno-Karabakh. It may be therefore considered that that territory is de facto in a situation of occupation.Footnote 20 Therefore, the test that determines whether the situation is one of occupation by proxy, namely that of overall control, is the same as the test that needs to be carried out to determine whether an internal conflict is internationalized. In both situations, it is a case of evaluating the intensity of the control exercised by a State over a group or authority in the territory of another State.
The law of non-international armed conflict
The concept of non-international armed conflict in humanitarian law must be analysed on the basis of two main treaty texts: Article 3 common to the 1949 Geneva Conventions and Article 1 of Additional Protocol II of 1977. This section will shed light on the criteria in each of these provisions and will show how these criteria may be interpreted in the light of practice. Moreover, the concept of non-international armed conflict is discussed in connection with the determination of the jurisdiction of the International Criminal Court (ICC). It is appropriate to refer briefly to the terms of that discussion by examining the relevant provisions of the Court's Statute.
Article 3 common to the 1949 Geneva Conventions
Article 3 common to the 1949 Geneva Conventions applies in the case of ‘armed conflict not of an international character occurring in the territory of one of the High Contracting Parties’.Footnote 21
This provision begins with a negative expression, dealing with armed conflict ‘not of an international character’. It thus refers back implicitly to common Article 2, which, as stated above, deals with conflicts between States. Armed conflicts that are not of an international character are those in which at least one of the parties involved is not governmental. Depending on the case in question, hostilities take place either between one (or more) armed group(s) and government forces or solely between armed groups.Footnote 22
Common Article 3 also assumes that an ‘armed conflict’ exists, i.e. that the situation reaches a level that distinguishes it from other forms of violence to which international humanitarian law does not apply, namely ‘situations of internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature’.Footnote 23 The threshold of intensity required in that case is higher than for an international armed conflict. Actual practice, in particular that of the ICTY, reveals that this threshold is reached every time that the situation can be defined as ‘protracted armed violence’.Footnote 24 This condition needs to be assessed against the yardstick of two fundamental criteria: (a) the intensity of the violence and (b) the organization of the parties.Footnote 25 These two components of the concept of non-international armed conflict cannot be described in abstract terms and must be evaluated on a case-by-case basis by weighing up a host of indicative data.Footnote 26 With regard to the criterion of intensity, these data can be, for example, the collective nature of the fighting or the fact that the State is obliged to resort to its army as its police forces are no longer able to deal with the situation on their own. The duration of the conflict, the frequency of the acts of violence and military operations, the nature of the weapons used, displacement of civilians, territorial control by opposition forces, the number of victims (dead, wounded, displaced persons, etc.) are also pieces of information that may be taken into account.Footnote 27 However, these are assessment factors that make it possible to state whether the threshold of intensity has been reached in each case; they are not conditions that need to exist concurrently. As for the second criterion, those involved in the armed violence must have a minimum level of organization. With regard to government forces, it is presumed that they meet that requirement without it being necessary to carry out an evaluation in each case.Footnote 28 As for non-governmental armed groups, the indicative elements that need to be taken into account include, for example, the existence of an organizational chart indicating a command structure, the authority to launch operations bringing together different units, the ability to recruit and train new combatants or the existence of internal rules.Footnote 29
When one or other of these two conditions is not met, a situation of violence may well be defined as internal disturbances or internal tensions. These two concepts, which designate types of social instability that do not pertain to armed conflict, have never been defined in law, despite the fact that they are referred to explicitly in Additional Protocol II.Footnote 30 In its background documents in preparation for the drafting of that instrument, the ICRC considered that internal disturbances are situations in which ‘there is no non-international armed conflict as such, but there exists a confrontation within the country, which is characterized by a certain seriousness or duration and which involves acts of violence. These latter can assume various forms, all the way from the spontaneous generation of acts of revolt to the struggle between more or less organized groups and the authorities in power. In these situations, which do not necessarily degenerate into open struggle, the authorities in power call upon extensive police forces, or even armed forces, to restore internal order’.Footnote 31 As for internal tensions, they cover less violent circumstances involving, for example, mass arrests, a large number of ‘political’ detainees, torture or other kinds of ill-treatment, forced disappearance and/or the suspension of fundamental judicial guarantees.Footnote 32
Lastly, common Article 3 applies to armed conflicts ‘occurring in the territory of one of the High Contracting Parties’. The meaning of this element may be controversial. Is it to be understood as a condition excluding non-international armed conflicts taking place in two or even more State territories, or rather as a simple reminder of the field of application of common Article 3? According to the latter hypothesis, it is argued that this specific point was included in order to make it clear that common Article 3 may only be applied in relation to the territory of States that have ratified the 1949 Geneva Conventions. We shall go on to see that it is probably best to tend towards that interpretation.Footnote 33
Some observers add a further condition to the notion of non-international armed conflict. They suggest that account needs to be taken of the motives of the non-governmental groups involved. This type of conflict would thus cover only groups endeavouring to achieve a political objective. ‘Purely criminal’ organizations such as mafia groups or territorial gangs would thus be eliminated from that category and could in no way then be considered as parties to a non-international armed conflict.Footnote 34 However, in the current state of humanitarian law, this additional condition has no legal basis. The ICTY had occasion to recall this when considering the nature of the fighting that took place in 1998 between Serbian forces and the Kosovo Liberation Army (UCK). In the Limaj case, the defence had challenged the idea that the fighting could constitute an armed conflict, arguing that the operations carried out by the Serbian forces were not intended to defeat the enemy army but to carry out ‘ethnic cleansing’ in Kosovo. The Tribunal rejected that argument by pointing out, in particular, that ‘the determination of the existence of an armed conflict is based solely on two criteria: the intensity of the conflict and organization of the parties, the purpose of the armed forces to engage in acts of violence or also achieve some further objective is, therefore, irrelevant’ (our emphasis).Footnote 35 The reverse position would, moreover, raise problems that it would be difficult to resolve in practice. The motives of armed groups are never uniform and cannot always be clearly identified. Many of them often carry out criminal activities such as extortion or drug-trafficking, while at the same time pursuing a political objective. Conversely, on occasion criminal organizations also exercise a power pertaining to the political sphere or at the very least to the management of populations.
Article 1 of Additional Protocol II
Additional Protocol II applies to non-international armed conflicts ‘which take 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’. However, this instrument does not apply to wars of national liberation, which are equated with international armed conflicts by virtue of Article 1(4) of Additional Protocol I.
As in the case of common Article 3, a non-international armed conflict within the meaning of Additional Protocol II can only exist if the situation attains a degree of violence that sets it apart from cases of internal tensions or disturbances.Footnote 36 That instrument nonetheless defines a more limited field of application than that of common Article 3. It requires non-governmental forces to have a particularly high level of organization, in the sense that they must be placed ‘under responsible command’ and exercise territorial control, allowing them ‘to carry out sustained and concerted military operations and to implement this Protocol’.Footnote 37 Although common Article 3 also presumes that armed groups are able to demonstrate a degree of organization, it does not stipulate that these groups should be able to control part of a territory. In practice, a conflict may therefore fall within the material field of application of common Article 3 without fulfilling the conditions determined by Additional Protocol II. Conversely, all the armed conflicts covered by Additional Protocol II are also covered by common Article 3.
In practice, it is often difficult to identify situations that meet the criteria of application established by Additional Protocol II. The required degree of territorial control, in particular, may be perceived differently from one case to another. If a broad interpretation is adopted, the concept of non-international armed conflict within the meaning of that instrument comes close to that of common Article 3. Even temporary control that is geographically limited would suffice in that case to justify the application of Additional Protocol II.Footnote 38 Conversely, if Article 1(1) is interpreted strictly, the situations covered are restricted to those in which the non-governmental party exercises similar control to that of a State and the nature of the conflict is similar to that of an international armed conflict.Footnote 39 In its Commentary on the Additional Protocols, the ICRC seems to adopt an intermediate position on this issue, accepting that territorial control can sometimes be ‘relative, for example, when urban centres remain in government hands while rural areas escape their authority’.Footnote 40 It nonetheless adds that the very nature of the obligations presented in Protocol II implies that there is ‘some degree of stability in the control of even a modest area of land’.Footnote 41
Additional Protocol II also restricts its field of application to armed conflict between governmental forces and dissident armed forces or other organized armed groups. That means that – contrary to common Article 3, which does not provide for that restriction – it does not extend to conflicts solely between non-governmental groups.Footnote 42
Lastly, Additional Protocol II repeats the ratione loci criterion already formulated in common Article 3, i.e. that it only covers non-international armed conflicts ‘occurring in the territory of one of the High Contracting Parties’. The previous comments on this subject also apply here. The Protocol also stipulates that the conflicts concerned are those taking place on the territory of a High Contracting Party between ‘its’ armed forces and opposition movements. A narrow reading of this passage would make this instrument inapplicable to the troops of a government intervening abroad in support of the local authorities. The forces involved in that case are not those of the State in which the conflict is taking place. An interpretation in keeping with the spirit of humanitarian law indicates, however, that the expression ‘its armed forces’ should in this case cover not only the troops of the territorial State, but also those of any other State intervening on behalf of the government.
As for the scope of the new points introduced in Additional Protocol II, it should be recalled that that instrument expands and supplements common Article 3 but that it does not change its conditions of application.Footnote 43 The additional restrictions provided for in Article 1(1) therefore only define the field of application of the Protocol and do not extend to the entire law of non-international armed conflict. Common Article 3 thus preserves its autonomy and covers a larger number of situations.Footnote 44
The Rome Statute of the International Criminal Court (ICC)
The Rome Statute of the ICC distinguishes between two categories of crimes that occur during ‘armed conflicts not of an international character’: (a) serious violations of common Article 3, and (b) other serious violations of the laws and customs of war that are applicable in those situations.Footnote 45 In both cases, the Statute indicates the lowest level of applicability of the relevant provisions by stipulating that they do not apply to ‘situations of internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature’.Footnote 46 Moreover, whereas this instrument does not give a more precise definition of the material field of application of the rules pertaining to ‘serious violations of common Article 3’ (Article 8(2)(d)), it clarifies the notion of non-international armed conflict in the case of ‘other serious violations’. Article 8(2)(f) stipulates in that case that the rules must apply ‘to armed conflicts that take place in the territory of a State when there is protracted armed conflict between governmental authorities and organized armed groups or between such groups’.Footnote 47 The question that then arises is whether, in referring explicitly to the criterion of duration (‘protracted armed conflict’), paragraph (2)(f) merely clarifies the terms of paragraph (2)(d), without creating a separate category of conflict, or whether it proposes a different type of non-international armed conflict, thus defining a new field of application. That question is the subject of controversy and has not yet been finally resolved.
Some observers consider that the two paragraphs deal with one and the same situation.Footnote 48 They consider, in particular, that the intention of those negotiating the Statute was not to create a separate category of non-international armed conflict. Rather, the reference to duration in paragraph (2)(f) was intended to prevent the restrictive notion in Additional Protocol II from being incorporated into the Statute. It was, in a way, an effort to achieve a compromise between the original draft, which made no distinction between paragraphs (2)(d) and (2)(f), and the desire of some States to include the restrictions of Additional Protocol II in that second paragraph.Footnote 49 Moreover, those who maintain that position claim that their interpretation is the only one that is in keeping with the evolution of customary law, which makes no distinction between different types of non-international armed conflict.
By contrast, other authorsFootnote 50 consider that if the concept of non-international armed conflict in paragraph (2)(d) refers directly to that of common Article 3, the notion in paragraph (2)(f) adds a time criterion. A non-international armed conflict within the meaning of paragraph (2)(f) exists when that conflict is ‘protracted’. Whereas from the point of view of paragraph (2)(d), duration is a factor that may perhaps be taken into account when evaluating the situation but does not constitute a compulsory criterion, it is nonetheless an integral part of the very concept of paragraph (2)(f). This notion does not therefore seem to constitute an extension of the field of application of paragraph (2)(d) but creates a separate category of non-international armed conflict with a view to criminalizing, within the context of the Statute of the ICC, additional violations of international humanitarian law, i.e. violations of rules in Additional Protocol II. The need to achieve a compromise when preparing that provision seems to indicate that the intention was to reach agreement on a different category from that referred to in paragraph (2)(d).
Case law tends to provide support for the second interpretation. In the Lubanga Dyilo case, the ICC Pre-Trial Chamber referred to Additional Protocol II in order to interpret paragraph (2)(f) of the Statute. It thus apparently wanted to confer a distinct meaning on this provision, defining a specific threshold of applicability. The Chamber made it clear that this threshold is characterized by two conditions: (a) the violence must achieve a certain intensity and be protracted; (b) an armed group with a degree of organization, particularly the ‘ability to plan and carry out military operations for a prolonged period of time’ must be involved.Footnote 51 Worded like that, this definition therefore seems to define a field of application that is stricter than that of common Article 3, as it requires the fighting to take place over a certain period of time. It is, however, broader than that of Additional Protocol II as it does not require the armed group(s) concerned to exercise territorial control.Footnote 52 The category of conflict targeted here is therefore half way between the categories referred to in common Article 3 and in Additional Protocol II.
In sum, the Rome Statute of the ICC seems to identify two types of non-international armed conflict: firstly, conflicts within the meaning of common Article 3 (paras (2)(c)–(d)); and secondly, ‘protracted’ non-international armed conflicts (paras (2)(e)–(f)). It should nonetheless be recalled that this innovation in the Statute does not create a new concept of non-international armed conflict in international humanitarian law, but simply aims at determining the ICC's jurisdiction. It therefore applies only to the exercise of that jurisdiction and does not establish a category that is more generally applicable.
Controversial classification of certain armed conflicts
Armed conflicts are in reality not as clearly defined as the legal categories. Some of them may not tally with any of the concepts envisaged in international humanitarian law. This raises the question of whether those categories need to be supplemented or adapted with a view to ensuring that these situations do not end up in a legal vacuum. Without claiming to be exhaustive, this chapter will examine some dilemmas encountered in practice by referring to three types of situation whose qualifications are controversial: control of a territory without military presence on the ground; foreign intervention in non-international armed conflict; and non-international armed conflicts on the territory of several States.
Control of a territory without military presence on the ground
Despite the clarifications contributed by the 1907 Hague Regulations and the 1949 Geneva Convention to the notion of occupation, it is not always easy in practice to identify the situations that are covered by that concept. As Roberts points out, ‘the core meaning of the term is obvious enough; but as usually happens with abstract concepts, its frontiers are less clear’.Footnote 53
The example of the Gaza Strip following the Israeli withdrawal illustrates those difficulties with particular acuity. On 12 September 2005, the last Israeli troops finished withdrawing from that region in which they had maintained a continuous presence since the Six-Day War in 1967. In doing so, they were helping to implement a ‘Disengagement Plan’ adopted by the Israeli government on 6 June 2004 and endorsed by parliament on 25 October of that same year.Footnote 54 By virtue of that plan, the authorities' intention was to put an end to their responsibilities vis-à-vis the people living in that territory.Footnote 55 Should it therefore be concluded that those measures marked the end of the occupation of the region in question? In other words, was the physical withdrawal of the Israeli forces enough to admit that effective territorial control characteristic of occupation did not exist any longer at that time?
Some observers answer that question in the negative. It was thus recalled that Israel retained substantial control over the Gaza Strip, although its troops were no longer physically deployed in that area.Footnote 56 The Disengagement Plan clearly stated that Israel was to continue to exercise control over the borders of that territory, as well as over its air space and coastal region.Footnote 57 Moreover, Israel has the advantage of being able to enter Palestinian territory at any time in order to maintain public order.Footnote 58 This power is made greater by the small size of the territory of Gaza and the military means available. That interpretation would also find some support in Article 42(2) of the 1907 Hague Regulations, which makes it clear that occupation exists when the authority of the hostile army ‘has been established and can be exercised’ (our emphasis). That ‘ability’ could be interpreted as meaning that potential authority would suffice as confirmation of the reality of occupation. The United Nations Secretary General thus considered that ‘the actions of IDF in respect of Gaza have clearly demonstrated that modern technology allows an occupying Power to effectively control a territory even without a military presence’.Footnote 59 According to that position, occupation of the Gaza Strip would therefore not have ceased with the withdrawal of troops in 2005, as Israel could be said to continue to exercise from a distance a power equivalent to the ‘effective control’ required under the law of occupation.
However, other observers consider that a closer study of the treaty texts shows that the ability of an occupier to impose its authority cannot be separated from its physical presence in the territory under its control.Footnote 60 While Article 42 of the 1907 Hague Regulations accepts that occupation exists when the adversary's authority ‘can be exercised’, it makes it clear that that authority must first be ‘established’. It thus forges an indissociable link between the establishment of authority, implying the deployment of a presence in the territory in question, and the ability to extend that authority to the entire territory. As was recalled by the International Court of Justice, effective control becomes apparent as a result of a substitution of powers.Footnote 61 Obviously, a similar threshold of application cannot be achieved if the foreign forces are located outside the region in question. Moreover, it is impossible to conceive of the implementation of most of the rules of occupation unless there is a presence in the territory.Footnote 62 It is actually impossible to ensure public order and life in a territory, as required by Article 43 of the 1907 Hague Regulations, from outside. It would thus be paradoxical to require a State to fulfil its international obligations if it is unable to do so because it is not present in the area concerned. A similar interpretation would run counter to the basic tenets of the law of occupation.
The example of Gaza shows to what extent the concept of occupation poses difficulties of interpretation when it comes to applying it in practice. It would be impossible within the limited framework of this article to deal with all the issues associated with that concept. The ICRC is currently carrying out a consultation process that will help to clarify a number of still controversial points.
Foreign intervention in non-international armed conflict
Two different forms of intervention may be distinguished here: (a) when one (or more) third State(s) become involved in a non-international armed conflict in support of one or other of the parties to the conflict; (b) when multinational forces become involved in a non-international armed conflict in the course of a peace-keeping operation.
The intervention of one or more third States in a non-international armed conflict
This scenario, which is sometimes referred to as a ‘mixed conflict’, combines characteristics which may derive from both international armed conflicts and non-international armed conflicts. Depending on the configuration of the parties involved, fighting in the field may be between the forces of the territorial State and those of an intervening State, between intervening States taking action on both sides of the front line, between government forces (of the territorial State or of a third State) and non-governmental armed groups or between armed groups only. This raises the issue of the legal definition of those situations that do not fit into the standard categories of conflicts established by international humanitarian law.
In its work, the ICRC considers that, depending on the warring parties, the law that applies in such situations varies from one case to the next. Inter-State relations are governed by the law of international armed conflict, whereas other scenarios are subject to the law of non-international armed conflict.Footnote 63 Thus intervention by a third State in support of a non-governmental armed group opposed to State forces results in the ‘internationalization’ of the existing internal conflict. This fragmented application of international humanitarian law was implicitly favoured by the International Court of Justice in the Case concerning Military and Paramilitary Activities in and against Nicaragua: in its analysis of the conflict, the Court differentiated between, on the one hand, the conflict between the Nicaraguan government and the contras, and, on the other, the conflict between that same government and the government of the United States.Footnote 64 However, this differentiated approach also raises certain practical problems. In many cases, the distinction between conflicts deriving from one or other of the two types of armed conflict is artificial or leads to results that are difficult to accept. When there is an alliance of foreign government forces and rebel troops, the following questions are raised, for example: What status needs to be given to civilians taken captive by foreign forces and then handed over to the local group? Are the relevant rules of the Fourth Geneva Convention to be applied to them (to the extent that there is an armed conflict between the intervening State and the territorial State) or the rules stemming from the law of non-international armed conflict (since they are held by a non-governmental armed group)? In other words, does a different set of rules need to be applied depending on whether those persons were arrested by the foreign forces or directly by the local group?
In the light of these difficulties, the question is then raised of whether it is desirable to envisage an adaptation of international humanitarian law as applicable to non-international armed conflicts characterized by foreign military intervention. Some observers suggest as much, requiring the law of international armed conflict to be applicable in every case in which a foreign Power takes action on behalf of one or other of the parties.Footnote 65 This was the nature, in particular, of one of the proposals made by the ICRC in its 1971 Report on the Protection of Victims of Non-International Armed Conflicts.Footnote 66 That proposal was nonetheless rejected by the experts who studied the ICRC's draft. It was suggested that it would tend to make these conflicts worse, as the non-governmental groups would try to attract third States in order to benefit from application of the law of international armed conflict.Footnote 67
The intervention of multinational forces in a non-international armed conflict
We need to begin by recalling that the presence of multinational forces in this context does not necessarily transform them into parties to the conflict. Usually, these troops are not there to take part in the fighting, but are deployed with the aim of conventional peace-keeping. Their mandate does not authorize them in that case to provide support for one or other of the adversaries, but is limited to interposition or observation. Moreover, they may only resort to using armed force in the case of self-defence. Multinational forces must, however, be considered parties to the conflict in two hypotheses. First, it may so happen that they take part directly in the ongoing hostilities by supporting one of the warring entities. The United Nations Organization Mission in the Democratic Republic of Congo (MONUC), for example, provided military support for the government of the Democratic Republic of the Congo in order to repel the offensives launched by the armed opposition.Footnote 68 Secondly, when international troops are deployed without supporting one of the warring camps, their status will be determined in accordance with the criteria normally used to evaluate the existence of a non-international armed conflict. Those troops must be considered as a party to the conflict if their level of involvement reaches the required degree of intensity.Footnote 69 This is not the case if recourse to force is limited to the context of self-defence.
The nature of the armed conflicts considered here is controversial. For most authors,Footnote 70 these situations are to be equated with international armed conflicts. To the extent that the operations concerned are decided, defined and carried out by international organizations, they are by nature included in that category. It is of little relevance in that case whether the opposing party is a State or a non-governmental group. In accordance with that doctrinal position, the hypothesis envisaged constitutes an international armed conflict sui generis, the particular status of the organization involved being enough to qualify it as such. Nonetheless, the application of the law of international armed conflict in this case does pose certain problems. In the hypothesis in which the conflict is between multinational forces and unstructured armed groups, it seems, in particular, difficult to require the parties to comply with the Third Geneva Convention of 1949. Taking another approach, which is that followed by the ICRC, a differentiated application of international humanitarian law must be followed in that case, depending on the parties to the conflict in each individual case in the field. The law of international armed conflict must therefore be applicable when international troops clash with government forces. By contrast, if fighting is between those troops and non-governmental groups, it is the law of non-international armed conflict which must prevail. The legal regime applicable in the same conflict thus varies depending on the adversaries present in each situation.Footnote 71
Non-international armed conflicts taking place on the territory of several States
Both Article 3 common to the Geneva Conventions and Additional Protocol II specify their respective fields of application by clearly stating that the conflict in question takes place on the territory of a State party to those instruments. However, many conflicts between a government and an armed group are in practice carried out on the territory of two or even of several States. Some authors consider that this is a new type of conflict that is not taken into account in the texts currently in force. They refer to such conflicts as ‘transnational armed conflicts’ or ‘extra-State conflicts’ and consider that a specific type of international humanitarian law must apply to them.Footnote 72 It is useful in this respect to differentiate between various scenarios.
‘Exported’ non-international armed conflicts
The parties to a classic non-international armed conflict (within the meaning of common Article 3 or of Additional Protocol II) may well continue their fighting on the territory of one or more third States with the explicit or tacit consent of the government(s) concerned (These are known as ‘exported’ or ‘delocalized’ conflicts, or ‘extraterritorial’ non-international armed conflicts.) In principle, the government forces involved are pursuing the armed group seeking refuge in the territory of a neighbouring State. In that kind of situation, an international armed conflict does not exist since there is no conflict between two or more States (as required by Article 2 common to the Geneva Conventions), given that the intervening State acts with the consent of the territorial sovereign.
The law applicable in this case gives rise to controversy. Some authors consider that what is being dealt with here is a different form of conflict and recommend working out a new form of international humanitarian law, which would constitute a third legal system alongside the law of international armed conflict and the law of non-international armed conflict. They consider that, from the perspective of the parties involved, these armed conflicts are very similar to non-international armed conflicts, as they involve government forces in conflict with armed groups. However, from the territorial point of view, these conflicts are characterized by ‘internationalization’, as they are not confined within the borders of a single State but concern two or more States.Footnote 73 A new legal regime specially adapted to that third category might lead, for example, as Schöndorf suggests, to a combination of, on the one hand, the ‘law of non-combatants of inter-State armed conflicts’ (treatment of civilians in enemy hands, principle of distinction) with, on the other, the ‘law of combatants of intra-State armed conflicts’ (protection and treatment of the wounded, sick and shipwrecked, no status for adversaries taken captive, etc.).Footnote 74 The author thus considers that there is nothing to justify a different kind of protection for non-combatants (in the sense of civilians not taking part directly in hostilities) depending on whether the conflict is inter-State or intra-State. However, this kind of distinction would be acceptable for combatants as, in the case of an intra-State conflict, the members of armed groups do not benefit from the privilege granted to soldiers taking part in international armed conflicts. This solution would allow account to be taken of both the internal (nature of the parties to the conflict) and international (extraterritoriality) aspects of those armed conflicts.
It is nonetheless not certain whether the territorial aspect is indeed a constitutive factor of non-international armed conflict. It may actually be maintained that the reference to the territory of a High Contracting Party in common Article 3 and in Additional Protocol II was simply intended to ensure that the application of the relevant rules is linked to the jurisdiction of a State that has ratified the treaties in question. The aim of this reference would therefore not be to remove ‘exported conflicts’ from the field of application of international humanitarian law. Rather, it can be seen as a reminder – in the context of the law of non-international armed conflict – of the principle according to which the international conventions are only binding upon those States that have submitted to them.Footnote 75 There is then nothing to stop this legal regime being applied, even if hostilities extend beyond the borders of a single State. Moreover, given that the four Geneva Conventions have now been ratified universally, the ICRC adds that, in practice, the territorial criterion in Article 3 has lost its importance. Indeed, as that organization points out, ‘any armed conflict between governmental armed forces and armed groups or between such groups cannot but take place on the territory of one of the Parties to the Convention’.Footnote 76
Cross-border non-international armed conflicts
Another possibility is that State forces enter into conflict with a non-governmental armed group located in the territory of a neighbouring State. In that case, there is thus no spillover or exportation of a pre-existing conflict. The hostilities take place on a cross-border basis. If the armed group acts under the control of its State of residence, the fighting falls within the definition of an international armed conflict between the two States concerned. If, however, this group acts on its own initiative, without being at the service of a government party, it becomes more difficult to categorize the situation. Does an international armed conflict necessarily exist because a State uses armed force on the territory of another State? If that is so, should members of the armed group be considered to be civilians taking part directly in the hostilities? Is it more appropriate to consider that situation to be a cross-border non-international armed conflict even if a parallel international armed conflict between the two States may also be taking place? By contrast, must a non-international armed conflict be deemed to exist solely in the hypothesis according to which the State of residence of the armed group accepts or tolerates intervention by its neighbouring State, the absence of consent leading it to be classified as an international armed conflict?Footnote 77 Lastly, should this be considered a new type of conflict, requiring a specific legal regime that has yet to be defined?Footnote 78
One recent example is that of Lebanon in the summer of 2006. It may be recalled that a high-intensity armed conflict had begun on 12 July following various attacks by Hezbollah's military component on positions and villages in Israeli territory. For instance, eight Israeli soldiers had been killed in the course of those operations and two others taken captive. The Israeli authorities retaliated by launching a ground, air and sea offensive on Lebanon. The hostilities continued until 14 August, when a ceasefire that had been agreed by the two governments concerned entered into effect.Footnote 79
The Commission of Inquiry set up by the United Nations Human Rights Council considered that an international armed conflict had taken place, although, in its view, the Lebanese armed forces had never taken part in the fighting. In its report dated November 2006, it considered that Hezbollah should be considered a militia ‘belonging to a Party to the conflict’, within the meaning of Article 4A(2) of the Third Geneva Convention of 1949. In support of that position, it stressed that Hezbollah, as a legally established political party, was represented in parliament and in the Lebanese government. In addition, for several years Hezbollah had assumed the role of an anti-Israeli resistance movement in southern Lebanon, a fact acknowledged by the President of Lebanon himself, who had called the armed branches of that group ‘national resistance fighters’.Footnote 80 According to the Commission, the war in 2006 thus assumed an international character by virtue of the organic link existing between Hezbollah and the State of Lebanon at that time.
There is nonetheless some doubt about whether the arguments put forward by the Commission really do allow the conclusion to be reached that the hypothesis of Article 4A(2) of the Third Geneva Convention had been realized in the case in point. Actually, those arguments are not enough to show a sufficiently narrow link between the Hezbollah combatants and the Lebanese government. For that link to exist, those combatants need to have been acting ‘on behalf of’ the latter.Footnote 81 Expressed differently, ‘[i]n order for irregulars to qualify as lawful combatants, it appears that international rules and State practice […] require control over them by a Party to an international armed conflict and, by the same token, a relationship of dependence and allegiance of these irregulars vis-à-vis that Party to the conflict’.Footnote 82 This is how the expression ‘belonging to a Party to the conflict’ must be understood in Article 4A(2) of the Third Geneva Convention.Footnote 83 In the case in question, it seems that the required degree of control was not achieved. On the contrary, the Lebanese authorities stated on several occasions that they had not been aware of the attacks that were at the origin of the conflict and that they did not approve of them. They made this statement officially in a letter addressed to the Secretary General and the Security Council of the United Nations.Footnote 84 Consequently, in this case a double legal classification probably needs to be retained. Alongside an international armed conflict between Israel and Lebanon, the war in 2006 constituted a non-international armed conflict between Israel and Hezbollah, whose distinctive feature was that it took place across a border.Footnote 85 It is therefore the nature of the belligerents rather than the transborder character of the situation which in this case constitutes the decisive criterion for classifying the conflict.Footnote 86
That position nonetheless raises certain questions about the application of the law, particularly in connection with combatants who are taken captive. It implies that the Fourth Geneva Convention would have to apply to detained members of Hezbollah to the extent that they were nationals of Lebanon or of a State not entertaining diplomatic relations with Israel.Footnote 87 However, the Israeli soldiers detained by the armed group would benefit only from the protection granted by the law of non-international armed conflicts. That position therefore raises a problem with regard to the equality of the belligerents. If, however, one considers that the law of non-international armed conflict also applies to those detained in Israeli hands, equality has been upheld, but at the cost of a weakening of the applicable standards.
Other observers propose further different readings of the Lebanese conflict. Some of them consider, for instance, that this example is illustrative of a new type of armed conflict, which cannot be classified as an international armed conflict or as a non-international armed conflict, and which, hence, implies the application of a specific international humanitarian law.Footnote 88
The question of the international fight against terrorism
The debate about the nature of cross-border armed conflicts prompts questions about the current clash between some States and Al Qaeda. In the case in point, this conflict takes the form of a series of terrorist attacks and anti-terrorist operations in several countries. Can the sum total of these events then be considered as a (cross-border global) armed conflict to which international humanitarian law would apply?Footnote 89 Does it constitute a new type of armed conflict giving rise to the application of a legal regime that has yet to be established?Footnote 90 Or is it a phenomenon that is not related to armed conflict? The same question could also be asked with regard to transnational criminal groups. Some observers do not hesitate to refer, for example, to the existence of a ‘global war on drugs’.Footnote 91
The key issues in the matter have already been discussed at length.Footnote 92 It is enough to recall at this juncture that the answer needs to be flexible enough to take account of the different types of armed conflict provided for under international humanitarian law. Basically, Al Qaeda's way of operating probably excludes it from being defined as an armed group that could be classified as a party to a global non-international armed conflict. In accordance with the current state of intelligence, it appears, rather, to be a loosely connected, clandestine network of cells.Footnote 93 These cells do not meet the organization criterion for the existence of a non-international armed conflict within the meaning of humanitarian law. Some experts nonetheless think that it is not impossible for a conflict between one or more States and a transnational armed group to reach that level one day.Footnote 94 In the Hamdan v. Rumsfeld case, the United States Supreme Court seems to take that view, considering that Article 3 common to the 1949 Geneva Conventions is applicable to the members of Al Qaeda, and to the persons associated with that organization, who were taken captive during the fight against terrorism.Footnote 95
Apart from that particular problem, in certain contexts the fight against terrorism may also take the form of an armed conflict. That is the case when it results in a clash between States, as was the case when the United States of America attacked Afghanistan in October 2001. That fight may also be the equivalent of a classic (internationalized) non-international armed conflict, as was the case in Afghanistan from 19 June 2002 onwards, on which date a transition government was established. With the support of the international coalition, the newly established authorities were to deal with high-intensity fighting against organized non-government troops, i.e. those of the Taliban.
Conclusion
This review of the different forms of armed conflict in international humanitarian law has shown just how difficult it can be to classify situations of violence and hence to determine the rules that apply. These difficulties are partly related to the legal categories themselves, whose content is often imprecise in the treaty texts establishing them. In that respect, the development of international practice is essential as it enables those categories to be gradually expressed in concrete terms by assessing them in the light of real situations. The most outstanding contribution in this regard is probably that of the ICTY with regard to the concept of non-international armed conflict. The ICTY's case law has not only identified the two constitutive elements of that concept, but has also put forward a wide range of indicative criteria making it possible to verify, on a case-by-case basis, whether each of these components has been achieved. Other elements deriving from the typology of armed conflicts would, however, deserve additional clarification. The case of the Gaza Strip is just one example of the difficulty of taking account of all the dimensions of the concept of occupation. Other uncertainties persist, in particular, about the criteria enabling the beginning or the end of an occupation to be determined.
The classification of situations of armed violence is also often linked to political considerations, as the parties involved endeavour to interpret the facts in accordance with their interests. On the basis of the margin of discretion allowed by the general terms of the legal categories, it is not unusual, for instance, for States to refuse to admit that they are involved in an armed conflict. They prefer to play down the intensity of the situation by claiming to carry out an operation to maintain public order. In so doing, they deny the applicability of humanitarian law. This tendency is encouraged by the fact that there is no independent international body authorized to decide systematically on cases that are likely to relate to one or other form of armed conflict. It is true that the ICRC, whose work is based primarily on international humanitarian law, informs the parties concerned of its assessment of situations, unless it would not be in the interest of the victims to do so. However, those who receive that assessment are not bound by the ICRC's view. Under those conditions, it seems even more important to clarify the relevant concepts, with a view to reducing the scope for interpretation, thus reinforcing the predictability of international humanitarian law.
