1. Introduction
This article grew out of a workshop held by the International Committee of the Red Cross (ICRC) in Geneva in January 2012. The subject of the workshop was the use of force in armed conflicts and the interplay between the law enforcement and conduct of hostilities paradigms.Footnote 1 As a background to the workshop the organisers presented participants with a list of hypothetical situations that were to be discussed. The first situation contemplated a state (X) in which there is an ongoing armed conflict between its armed forces and the Rebels (an organised armed group). X is a large state and the hostilities are confined to an area in the south of the country. The authorities have sound intelligence information that the leader of the Rebels is visiting his wife, who lives in a house in the north of the country. They have decided to ‘neutralise’ him. Which legal regime applies to their actions? May the state authorities use lethal force against him as he is a fighter belonging to a party to the non-international armed conflict (NIAC), even though he poses no imminent threat and it would be feasible to arrest him? Alternatively are they confined to the rules of law enforcement as he is away from the arena of hostilities?
In the discussion between the experts who participated in the meeting there was a wide division of opinion on the applicable legal regime. Many argued that, as the individual involved was a fighter in an armed conflict, he could be targeted according to the rules on the conduct of hostilities in armed conflicts. They conceded that, from a policy point of view, the state authorities might not choose this path of action, but opined that legally they were perfectly entitled to do so. Other experts had serious reservations about this approach. Their view was that the law enforcement regime has precedence in an internal armed conflict, and that the mere fact that there is an ongoing armed conflict in the state concerned does not imply that the state may use lethal force against a fighter outside the context of the hostilities themselves, when law enforcement measures are perfectly feasible.
This difference of opinion goes to the core of the debate over the conditions for legitimate use of lethal force against those whom a state regards as ‘internal enemies’. This debate has intensified since the so-called ‘War on Terror’ began. Whether use of lethal force is legitimate depends first and foremost on the legal regime that is applicable in the particular circumstances. Does a law enforcement, criminal justice, policing regime apply? If so, state agents may use lethal force only when it becomes absolutely necessary to do so in defined circumstances, the most important of which is the protection of others against unlawful violence. This is generally taken to mean that such force may be used only in the face of an imminent danger that cannot be thwarted by other, less violent, means. Certainly a state may not plan in advance to use lethal force against a person regarded as a criminal or an enemy of the state; it may use such force only if it becomes absolutely necessary in the particular circumstances. If, on the other hand, the law relating to the conduct of hostilities in an armed conflict regime applies, the legitimacy of force will initially be determined by the status of the particular individual. If he or she is a combatant (privileged or non-privileged) it will be legitimate to use lethal force, unless the individual is hors de combat.Footnote 2
Given the wide disparity between the two paradigms, it is essential to clarify which of them applies in defined circumstances. One of the contributors of this article has addressed this question elsewhere and it will not be discussed here.Footnote 3 The question we address in this article is somewhat narrower than the general question. It starts from the premise that a non-international armed conflict is indeed taking place in the territory of the state in question. It would seem, therefore, that in the actual conduct of hostilities the armed forces of the state may use lethal force against fighters of the non-state party which is involved in the conflict, even if at the given moment they do not immediately threaten the lives of others. However, is this licence to use lethal force restricted to the battlefield, or does it extend to the whole area of the state in question, or even to the territory of another state that agrees to the use of such force? Put more bluntly, does the law relating to the conduct of hostilities in an armed conflict regime apply away from the battlefield when it is perfectly feasible for the state authorities to use the law enforcement paradigm against active members of the non-state actor group which is a party to the conflict?
It is widely considered that in an international armed conflict (IAC) combatants of one party may target combatants of the enemy at any time and in any place, as long as they are not hors de combat. In recent years there have been some attempts to qualify this principle. Below we mention the view of Nils Melzer that use of force against enemy combatants is always subject to the notion of military necessity, and that this may mean that in certain circumstances a combatant should be captured rather than killed.Footnote 4 More recently, Ryan Goodman has cast doubt on the historical accuracy of the prevailing view that allows use of force even when it would be perfectly feasible to capture an enemy combatant.Footnote 5 We have no intention of taking a stand on Goodman's view here.Footnote 6 For the purposes of our discussion we shall simply assume that in international armed conflicts the prevailing view reflects lex lata. Our argument is that even if this is the situation in IACs there are cogent reasons for taking a different position in NIACs.
Conducting a discussion of the scenario presented above requires returning to fundamental principles relating to the scope of application of the norms of armed conflict in non-international armed conflicts. Is it justified to draw a parallel between such conflicts and international armed conflicts, in which parties ostensibly have a licence to use force against the enemy's combatants as long as they are not hors de combat? Alternatively do we have to make a principled distinction between these two types of armed conflict, based on the assumptions that, in its relations with persons who are subject to its jurisdiction, the primary duties of the state are its human rights duties, that these duties include the duty to respect and protect the right to life, and that nobody may be arbitrarily deprived of their right to life. True, according to the view taken by the International Court of Justice in the Nuclear Weapons case, in an armed conflict the lex specialis of armed conflict law will determine when deprivation of life is arbitrary,Footnote 7 but is there indeed a lex specialis in international law which grants states licence to use lethal force against ‘fighters’ in a NIAC outside the context of hostilities?
Our thesis in this article is that in non-international armed conflicts there is no lex specialis in international law that sanctions the use of lethal force outside the context of hostilities. In other words, to the extent that customary international law allows states to resort to the conduct of hostilities paradigm of force in NIACs, this licence is restricted to exactly that: namely the actual conduct of hostilities. We concede that it is not always clear whether the context is one of hostilities or not, and that the distinction between the arena of hostilities and other arenas may be flexible and unclear.Footnote 8 We therefore do not argue for a rigid geographical line between the area of hostilities and other areas, but refer rather to the ‘context of hostilities’. The further away we are from the hostilities themselves, the less justification there is for use of force under the conduct of hostilities paradigm.
The first treaty norms relating to non-international armed conflicts appear in Common Article 3 of the 1949 Geneva Conventions. These norms merely place humanitarian restrictions on parties to a NIAC. They cannot, and should not, be read as having imported the wide ‘licence to kill’ implicit in the law relating to international armed conflicts into the realm of NIACs. Given the asymmetrical nature of non-international armed conflicts, and the fundamental human rights obligations of states towards all individuals subject to their jurisdiction, use of lethal force under the armed conflict paradigm should be restricted to those situations in which normal law enforcement mechanisms are not feasible.
Our argument proceeds along the following lines. The next section is devoted to discussion of Common Article 3 of the Geneva Conventions. We shall examine the assumptions behind this pivotal provision and pose the following question: given that Article 3 imposes fundamental norms of international humanitarian law on parties to a non-international armed conflict, must we necessarily conclude that all the norms relating to the law of armed conflict also apply? In other words, if we revert to the once prevalent distinction between Hague and Geneva law, must we conclude that when norms of Geneva Law apply to an armed conflict, the basic norms of Hague law must also apply? Based on a textual and historical analysis of Common Article 3, we shall argue that the answer to these questions is negative. Common Article 3 was introduced into the 1949 Geneva Conventions on the assumption that, unless such a provision appeared in an international convention, what the parties in a NIAC were permitted to do was regulated solely by the domestic law of the state in whose territory the armed conflict was taking place. It was only at a later stage that attempts were made to draw a wider parallel between international and non-international armed conflicts, and to import into the latter norms which were part of ‘the laws and customs of war’.
We then (in Section 3) consider the implications of the preceding discussion. Our argument here is that, although there are obviously parallels between international and non-international armed conflicts, they are limited by the different nature of these two types of conflict. In the context of IACs the question of whether the law of hostilities will apply to the use of lethal force or not will be dependent on the concrete situation and not solely on the mere existence of an ongoing armed conflict. The state involved may resort to the norms relating to the conduct of hostilities only in those concrete situations in which the scope and level of organised armed violence are such that a policing, law-enforcement model of law is clearly inappropriate. When a policing model of law may reasonably be employed, there is no justification for employing the rules relating to the conduct of hostilities merely on the strength of the argument that an internal armed conflict is taking place in the state's territory.
There has been a growing tendency in recent years to narrow the gap between the law applicable in IACs and NIACs. Some have argued for doing away with the distinction altogether.Footnote 9 In Section 4 we discuss this tendency as reflected in the jurisprudence of international tribunals, especially the International Criminal Tribunals for the former Yugoslavia (ICTY) and Rwanda (ICTR). We show that the application of humanitarian norms applicable in IACs to NIACs so as to encompass conduct that is far removed from the battlefield was motivated by the desire both to extend the jurisdiction of international criminal tribunals and to grant greater protection to victims of NIACs. Reverting to the distinction between humanitarian norms and the norms relating to the conduct of hostilities, we argue that extension of the former does not necessarily drag extension of the latter in its wake.
Following the scenario presented at the ICRC workshop, the analysis in this article rests on the existence of the type of non-international armed conflict contemplated both in Common Article 3 and Additional Protocol IIFootnote 10 – namely one occurring in the territory of a state party. In referring to NIACs we are referring solely to internal armed conflicts between a state's armed forces and organised armed groups.Footnote 11 It is now widely (but not universally) accepted that a transnational conflict between a state and an organised armed group that meets the substantive criteria of a non-international armed conflict – namely, the intensity and scope of armed force and the degree of organisation – must also be considered to be a non-international armed conflict.Footnote 12 The extent to which our analysis may be applied to such conflicts requires further research and will not be discussed here.
2. The Laws and Customs of War and Humanitarian Law
2.1 The Emergence of Humanitarian Law
The laws and customs of war have existed since ancient times, but it was only in the nineteenth century that a new branch of law, ‘international humanitarian law’, was developed. The first steps in the creation of this branch of law were the establishment of the International Committee of the Red Cross in 1863 and the adoption of the first international convention in this field – the Geneva Convention on the Amelioration of the Condition of the Wounded in Armies in the Field – in 1864. Soon after the adoption of the 1864 Convention, the 1868 Saint Petersburg Declaration Renouncing the Use of Explosive Projectiles under 400 Grammes Weight in Times of War proclaimed in its Preamble that ‘the progress of civilization should have the effect of alleviating as much as possible the calamities of war’.Footnote 13 The Declaration stated that the only legitimate object of war which states should endeavour to achieve is to weaken the military forces of the enemy, and that for this purpose it is sufficient to disable the greatest possible number of men. This object would be exceeded by the use of weapons which needlessly aggravate the suffering of disabled men or render their death inevitable. Subsequent humanitarian conventions include the Geneva Protocol for the Prohibition of the Use of Asphyxiating, Poisonous or Other Gases, and of Bacteriological Methods of Warfare (1925),Footnote 14 the two Geneva Conventions of 1929,Footnote 15 and the four Geneva Conventions of 1949.Footnote 16
In parallel with these purely humanitarian instruments, attempts were made to codify the laws and customs of war in the instruments adopted at the two Peace Conferences which took place in The Hague in 1899 and 1907. Six conventions and declarations were adopted in 1899 and 14 were adopted in 1907. While some of these conventions include provisions of a humanitarian nature, they relate first and foremost to the norms relating to the conduct of hostilities in armed conflicts. Thus it was that the laws of war developed along two tracks: the Hague track, relating mainly to conduct of hostilities, and the Geneva track, relating solely to humanitarian concerns.
‘Hague law’ was never devoid of humanitarian aspects.Footnote 17 However, ‘Geneva law’ consisted entirely of humanitarian norms, and did not address issues relating to the conduct of hostilities. It aimed at placing restrictions and some positive obligations on parties involved in international armed conflicts, so as to grant protection to the wounded and sick and to prisoners of war and to allow for the work of humanitarian organisations, foremost among which is the ICRC.Footnote 18 A prime example of this restricted nature of Geneva law is Geneva Convention IV (GC IV) of 1949. Despite its name – Convention Relative to Protection of Civilian Persons in Time of War – the Convention does not deal with norms that aim to protect civilians during conduct of hostilities, such as those that prohibit making civilians the object of an attack or launching indiscriminate attacks. It deals solely with the protection of civilians who find themselves in the hands of the enemy, in its territory or in occupied territory.
Since the adoption of Additional Protocol I (AP I) to the Geneva Conventions in 1977, the distinction between Hague and Geneva law has become blurred.Footnote 19 While purporting to be a Protocol which supplements the Geneva Conventions of 1949, AP I does much more than that: it attempts to codify the laws and customs of war relating to the conduct of hostilities. The result is an amalgamation of norms relating both to the conduct of hostilities and to humanitarian protection in one international ‘Geneva convention’ which relates to international armed conflicts. This incorporation of ‘Hague law’ into a ‘Geneva Convention’ led, it would seem, to some confusion that is reflected in the change of terminology. As ‘Geneva law’ was traditionally associated with humanitarian norms, and as all the norms of law were now incorporated in ‘Geneva law’, these norms gradually became known as ‘international humanitarian law’ (IHL).Footnote 20 The term ‘humanitarian law’ thereby lost its original meaning and is now used as a term that is synonymous with the ‘laws and customs of war’ or the law of armed conflict (LOAC), leading Professor Michael Reisman to refer to it as ‘Orwellian’.Footnote 21 We regard this use of the term IHL as an unfortunate development. There is nothing humanitarian in laws which allow a party to a conflict to kill combatants of the enemy even when they do not impose an immediate threat, and even to kill civilians during an attack, provided that the expected damage to them is proportionate to the anticipated military advantage.Footnote 22
Use of the term ‘international humanitarian law’ as a catch-all name for the laws and customs of war (or jus in bello) has led to confusion in that it creates the false impression that the application of humanitarian norms in a given situation necessarily implies that all the laws and customs of war also apply. Acts of violence which are criminal under the legal systems of all countries, and are certainly unlawful under the laws and customs of war, may easily be seen as violations of fundamental humanitarian norms, which should apply in all armed conflicts, international and non-international. In order to reach the conclusion that mistreatment of prisoners, intentional killing of persons who are not taking an active part in hostilities, persecution of ethnic groups, and aiding and abetting in murder and persecution on racial grounds are all severe violations of universal humanitarian norms, one does not need to consider when it is lawful to kill a person who has actively been participating in hostilities. However, we question the very assumption that the application of humanitarian norms in a given context necessarily implies that all the laws and customs of war also apply. We argue that historically there was no necessary connection between the application of international humanitarian norms in a given situation – namely armed conflicts not of an international character – and importation into that situation of all the laws and customs that apply in armed conflicts. Furthermore, we argue below that there are persuasive considerations of principle as to why there should not be an assumption that the application of humanitarian norms implies that all the laws and customs of war apply too.
2.2 The Internationalisation of Internal Armed Conflicts: Common Article 3
The breakthrough in the application of some international humanitarian norms to non-international armed conflicts occurring in the territory of a state came with the adoption of Common Article 3 of the 1949 Geneva Conventions.Footnote 23 Common Article 3 was a radical innovation in international law in two respects. First, it regulates the relationship between the authorities of a state and persons in that state's territory and subject to its jurisdiction, rather than the relationship between states – the traditional sphere of application of international law.Footnote 24 Secondly, it purports to impose international obligations on non-state actors, who are not, nor could be, parties to the Conventions, and are therefore not directly bound by their provisions.
From the text of Common Article 3, it is abundantly clear that its sole concern is with humanitarian protection of persons not taking an active part in hostilities, care for the wounded and sick and opening the door for the ICRC to offer its services to the parties involved in an internal armed conflict, without this being regarded as interference in the internal affairs of the state. There is nothing in the text of Common Article 3 itself to suggest that the laws and customs of war relating to the conduct of hostilities that apply in international armed conflicts will also apply in non-international armed conflicts. But was this not the implied assumption behind the adoption of Common Article 3? Was it not based on the idea that the application of international humanitarian norms was necessary because the laws and customs of war relating to the conduct of hostilities apply in such conflicts? A positive answer to this question would appear to rest on the argument that there is no room, and essentially no need, for humanitarian norms in an international convention dealing with times of war, unless the laws and customs of war apply. In other words, the argument is that humanitarian law (once called Geneva law) must go hand in hand with laws and customs relating to hostilities (once called Hague law).Footnote 25 We do not accept this argument. On the contrary, we shall argue that such an assumption neither lay behind the adoption of Common Article 3 nor is it substantively justified.
There have, of course, been major developments in the law of internal armed conflict since the adoption of Common Article 3 in 1949. We discuss some of these developments below.
2.2.1 The Adoption of Common Article 3 – Historical Background
The idea that the laws and customs of war should in principle be applied to civil wars is not new.Footnote 26 However, with the rise of the modern state in the post-Westphalian era, international law has been concerned solely with the relations between states, and the laws and customs of war were no exception. The relationship between the authorities of the state and insurgents or rebels was regarded as a matter that was generally of no concern to the international community.Footnote 27 It was therefore not subjected to regulation under international law but was regulated solely under the domestic political and legal system.Footnote 28 In internal armed conflicts before the adoption of the 1949 Geneva Conventions, to the extent that states observed the laws and customs of war, this was done tacitly, generally on the basis of reciprocity. During the Spanish Civil War, for example, the ICRC obtained the agreement of both sides to the conflict to respect, to some extent, the principles of the 1864 and 1906 Geneva Conventions.Footnote 29 The total discretion of a state involved in an internal armed conflict whether or not to recognise a situation of belligerency provided no certainty as to if and when the laws and customs of war would be applied in such conflicts.Footnote 30
Proposals to apply international humanitarian norms to non-international armed conflicts between states and rebel groups were raised by the ICRC before the adoption of the 1949 Geneva Conventions.Footnote 31 However, they were strongly resisted by states.
The road to the adoption of Common Article 3 was paved with controversy.Footnote 32 The initial draft article, which had been submitted by the ICRC to the seventeenth International Red Cross Conference held in Stockholm in August 1948, stated that in cases of non-international conflicts of any kind, ‘the implementing of the provisions of the present Geneva Convention shall be obligatory on each of the adversaries’.Footnote 33 It immediately became apparent that the very idea of applying the Conventions in NIACs was going to meet with heavy opposition.Footnote 34 States considered that such a provision would violate their sovereignty and amount to an unacceptable interference in their legitimate efforts to repress criminal insurgency.Footnote 35 While some states argued for the widest possible application of humanitarian standards in internal armed conflicts, the most that states could agree upon was Common Article 3, which provides only for the application of minimal humanitarian norms in an armed conflict not of an international character occurring in the territory of one of the High Contracting Parties.Footnote 36
Once it had been agreed that some humanitarian standards should be applied in internal armed conflicts, there was controversy over whether to make a distinction between different types of conflict, on the basis of the threshold of armed force. This idea was shelved and Common Article 3 contains no such distinction. The ICRC, faced with criticism from all sides of the committee, acknowledged that the text did not reflect in full its complete ideal of humanitarian concern, but at the same time stated that it ensured minimal protection to victims of NIACs.Footnote 37
As the Diplomatic Conference was convened solely to deal with ‘Geneva law’, there was no discussion of the conduct of hostilities norms that should apply in such conflicts. The discussion related solely to the application of humanitarian norms. The principal concern of many states was that application of these norms might imply that members of rebel or insurgent groups have the privilege to fight that is enjoyed by combatants in international armed conflicts. In drafting Common Article 3, great care was taken to ensure that no such implication could be drawn. Thus the final paragraph of this article expressly states that application of the humanitarian norms ‘shall not affect the legal status of the Parties to the conflict’.
2.2.2 The ICRC Commentary
Common Article 3 refers to an ‘armed conflict not of an international character’, but does not define the term. Since the article allows the ICRC to offer its services in the case of such a conflict, it was only natural that the ICRC argued for a wide definition, which in fact implied a rather low scope and level of armed violence.Footnote 38 This is clearly reflected in the ICRC Commentaries on the 1949 Conventions that were published in the 1950s under the general editorship of Jean Pictet. For our purposes, the most important element in the discussion of the definition of a non-international armed conflict is the manner in which the Commentary attempts to alleviate fears of states that a wide definition of NIAC would involve restrictions on the way in which they could fight their internal enemies. Consider the following excerpt from the Commentary relating to Common Article 3:Footnote 39
The Article should be applied as widely as possible. There can be no reason against this. For, contrary to what may have been thought, the Article in its reduced form does not in any way limit the right of a State to put down rebellion. Nor does it increase in the slightest the authority of the rebel party.
The Commentary goes on to explain that states could not really object to the widest possible application of Common Article 3, since that article includes ‘a few essential rules’ which all governments observe daily under their own rules of domestic law, even when dealing with common criminals.Footnote 40 Clearly, then, Pictet's view was that the import of Common Article 3 was purely humanitarian and had no implications regarding the application of the general laws and customs of war in armed conflicts to which Common Article 3 applies.Footnote 41
The Commentary also explains why it was necessary to add the provision according to which the application of humanitarian norms does not affect the legal status of the Parties to the conflict:Footnote 42
This clause is essential. Without it, neither Article 3, nor any other Article in its place, would ever have been adopted. It meets the fear – always the same one – that the application of the Convention, even to a very limited extent, in cases of civil war may interfere with the de jure Government's lawful suppression of the revolt, or that it may confer belligerent status, and consequently increased authority and power, upon the adverse Party.
In other words, while applying some of the most fundamental humanitarian norms in non-international armed conflicts, the states which adopted the 1949 Geneva Conventions rejected the application in NIACs of the fundamental principle of the laws and customs of war, namely that combatants have the right to fight.Footnote 43 Non-state actors who participate in non-international armed conflicts could be subject to criminal liability under domestic law, even if their acts were consistent with the laws and customs of war that apply in international armed conflicts. If apprehended by the state, such ‘fighters’ would not enjoy prisoner-of-war status and could be prosecuted as ordinary criminals.
But is there any sense in applying the humanitarian norms of war in a context in which the general laws and customs of war do not apply? Originally, Geneva law was developed in order to ‘humanise’ situations of interstate wars in which the traditional laws and customs of war applied. Could the same not be said for non-international armed conflicts? The adoption of Common Article 3 must be seen in its immediate historical context.Footnote 44
The Stockholm conference – in which the original drafts of the 1949 Geneva Conventions to be submitted to the Diplomatic Conference were drawn up – was held in August 1948, before the adoption of the Universal Declaration of Human Rights by the UN General Assembly. The Diplomatic Conference took place after adoption of the Universal Declaration, but the implications of that Declaration were not considered during the Conference. There was no awareness of the potential development of international human rights law, the very object of which would be to place constraints on what a state was permitted to do in its relationship with persons subject to its jurisdiction. On the contrary, as long as no aliens were involved, international law placed no restrictions on a state's freedom of action. In discussing the exclusion of nationals of a state from the definition of protected persons under GC IV, the Pictet Commentary explains that the Convention ‘remains faithful to a recognized principle of international law: it does not interfere in a state's relations with its own nationals’.Footnote 45
Given this assumption, it is clear why it was thought necessary to adopt Common Article 3 and to impose humanitarian restrictions on parties in a non-international armed conflict. The assumption was not that in the absence of such a provision the parties would be free to act under Hague law without the humanitarian restrictions imposed by Geneva law, but that in the absence of such a provision the parties (and particularly the state party involved) would be free to act as they liked and would be subject to no restrictions under international law. It was therefore essential to adopt a provision that would subject them to the most fundamental humanitarian demands that should apply in any situation of armed conflict, whatever the basic rules applicable to fighting may be.
In conclusion, the original impetus for extending the application of humanitarian norms relating to the conduct of hostilities in internal armed conflicts was clear: the arguments over the text of Common Article 3 did not relate to the conduct of hostilities in NIACs, but only to the extent and scope of the humanitarian norms that would be imposed on the parties. States were certainly not prepared to apply the provisions of the Third Geneva Convention relative to the Treatment of Prisoners of War in non-international armed conflicts.Footnote 46 Agreement could only be reached on applying the most fundamental humanitarian norms that were to be respected in relation to all persons who were taking no active part in hostilities.
3. IACs, NIACs and the Conduct of Hostilities
3.1 Customary International Law and NIACs
Our fundamental argument is that the application of international humanitarian norms in armed conflicts not of an international character was never meant to imply that all the norms and customs of war which apply in international armed conflicts apply in such conflicts too. As the discussion of the background to the adoption of Common Article 3 of the Geneva Conventions reveals, historically there is little support for the idea that such norms and customs of war were applicable in internal armed conflicts unless a situation of belligerency had been recognised by the state involved. As mentioned above, such recognition was extremely rare.
While it seems clear that there was never any intention, in adopting Common Article 3, to import all of the norms and customs of war that apply in IACs into the law relating to NIACs, one must ask whether over time some of these norms and customs of war have become part of the customary international law that applies to such conflicts. We stress that the question cannot relate to all the norms that apply in international armed conflicts, as it is abundantly clear to all that many of these norms do not apply. Thus, for example, there is no recognised combatant status in non-international armed conflicts.
Once we concede that the most that can be argued is that some of the norms that apply in IACs may have become part of the customary law applicable in NIACs, we are faced with a question to which there is no simple answer. For, even if one assumes, for example, that state practice and opinio juris allow for the conduct of hostilities in internal armed conflicts according to the norms that apply in international armed conflicts, one may not conclude that outside the context of hostilities the rules of international armed conflict have gained the same status.
There are significant differences between situations of international and non-international armed conflicts which favour the view that the IAC norms do not apply outside the specific context of hostilities themselves. One would need strong proof of state practice and opinio juris to show that, notwithstanding these differences, under customary law the norms regarding lethal force against persons who have the privileges of combatants in international armed conflicts apply in relation to ‘fighters’ in non-international armed conflicts who have no such privileges. As we discuss below, to this very day there is a division of opinion as to whether there is a specific category of ‘fighters’ (albeit unprivileged) in NIACs, or simply some civilians who take an active part in hostilities and others who do not take part. Given this division of opinion, it is hard to see how state practice and opinio juris could have crystallised around the power to use lethal force against a category of persons whose very existence is questionable.
We have found no evidence on state practice and opinio juris relating to the application of norms relating to the conduct of hostilities in NIACs outside the context of hostilities themselves. To the extent that an argument is made (as it was during the ICRC workshop mentioned in the Introduction) that lethal force may be used against fighters in NIACs at any time and place, it would seem to rely on the simplistic assumption that all norms that apply in IACs apply in NIACs too. This is a weak basis for finding a norm of customary international law. We shall now show why there is no justification for the assumption that all conduct of hostilities rules of IACs also apply in NIACs.
3.2 Between IACs and NIACs
Given their humanitarian nature, restrictions on the means and methods of fighting that apply in international armed conflicts, such as the prohibition on the use of certain weapons, should obviously apply in non-international armed conflicts too. This is reflected in recent arms restriction conventions that expressly apply to NIACs.Footnote 47 Nevertheless, like Sandesh Sivakumaran, we maintain that there remain fundamental distinctions between the two types of armed conflict and that these differences are highly relevant when discussing amalgamation of the two regimes of law.Footnote 48 Application in NIACs of the humanitarian restrictions that are imposed on parties in IACs does not, and cannot, wave away these distinctions. As part of our argument rests on these distinctions, it is important to lay them out.
3.2.1 Nature of the Regulated Relationship
The first fundamental and principled difference between IACs and NIACs lies in the very nature of the relationship which is the concern of the two LOAC regimes. The laws of international armed conflicts are part and parcel of classical international law. They relate to the relationship between states that are involved in armed conflict and rest on notions of common interest, reciprocity and symmetry. Until the development of international human rights law, which today has some place in IACs, there was no other system of legal norms to regulate that relationship. Conceptually that relationship could not be regulated by the domestic legal system of one of the states involved in the armed conflict. Like all other issues concerning the relationship between states, it was therefore only natural that, if it were to be subject to legal regulation, the only system of law under which it could be regulated was the law of nations, now known as international law. The international norms of armed conflict, jus in bello or the laws and customs of war, were not developed to modify or mitigate other applicable norms, but as the primary source of norms for the particular relationship involved – that between warring states (with some implications for neutral states too).
In the typical non-international armed conflict contemplated in both Common Article 3 of the Geneva Conventions and Additional Protocol II – namely an armed conflict not of an international character occurring in the territory of a state – LOAC norms concern the relationship between the authorities of the state and organised armed groups of persons who are subject to that state's jurisdiction.Footnote 49 There is no conceptual reason why this relationship could not be regulated solely under the domestic laws of the state concerned. The introduction of international law into this essentially domestic arena, once regarded as within the sole prerogative of sovereign states, was not a function of the relationship between the fighting parties; its motivation was to mitigate the prevailing domestic political and legal regime so as to force states and other parties involved to comply with humanitarian norms, even if those parties are hostile or indifferent to those norms.
This difference between the two regimes is not one of purely historical interest. To this day the primary norms that regulate the relationship between states involved in an international armed conflict are the laws and customs of war, or LOAC, while the primary legal norms that apply to the parties in an internal armed conflict are those that regulate the relationship between the governors and the governed – namely the norms of the domestic constitutional and legal regime and the international human rights obligations of the state involved. It is by no means self-evident that when an internal armed conflict occurs, the state may employ laws and customs of war which are incompatible with those obligations. On the contrary, the statement of the International Court of Justice that a state's departure from those norms must be compatible with its power to derogate from its human rights obligations in a state of emergency would seem to be most pertinent in the case of such a conflict,Footnote 50 for there is no room for the argument that, in such a case, human rights conventions do not apply to the relationship involved – namely the relationship between a state and persons subject to its jurisdiction. It is not surprising, therefore, that the cases dealing with the application of human rights conventions to a state involved in an armed conflict have related to the extraterritorial application of human rights conventions. In the one major situation in which cases relating to an internal armed conflict reached the European Court of Human Rights, the state involved did not attempt to challenge the applicability of the European Convention on Human Rights.Footnote 51
This essential difference between international and non-international armed conflicts is reflected in the very definition of those conflicts. In the absence of any other regime to regulate the relationship between states, the international law of armed conflict applies ‘whenever there is a resort to armed force between states’;Footnote 52 its application is not dependent on the level or scope of that armed force. Conversely, as the primary legal regime that applies to NIAC is the internal domestic regime (subject always to the state's international human rights obligations), the law of armed conflict does not apply to conflicts in a state which involve the use of armed force unless they meet demands relating to the scope and level of violence and to the level of organisation behind the non-state actors involved. Article 1(2) of Additional Protocol II expressly stipulates that the Protocol ‘shall 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, as not being armed conflicts’.Footnote 53
This provision has some significance for the issue under discussion here. Whatever the state interests may have been in introducing this express provision in AP II, implicit in the provision is recognition that such situations must be dealt with under regular, law-enforcement procedures. European states were involved in drafting the Additional Protocol and today most of them are parties to it. One must assume that they were aware of their obligations under the European Convention on Human Rights, under which lethal force may not be used unless absolutely necessary ‘in action lawfully taken for the purpose of quelling a riot or insurrection’. Under the Convention there can be no doubt whatsoever that when it is perfectly feasible to detain a person, it cannot be absolutely necessary to use lethal force against him even if the purpose is to quell a riot or insurrection. It is difficult to accept that if the situation moves from one of riot or insurrection to one of internal armed conflict, the notion of ‘absolute necessity’ undergoes such a radical transformation that lethal force may be used against persons merely because of their status, even though they could easily have been detained.
3.2.2 The Principle of Distinction
The principle of distinction is regarded as the most fundamental of the norms that apply in cases of armed conflict.Footnote 54 The principle first appears in Article 22 of the Lieber Code (1863), which states:Footnote 55
Nevertheless, as civilization has advanced during the last centuries, so has likewise steadily advanced, especially in war on land, the distinction between the private individual belonging to a hostile country and the hostile country itself, with its men in arms. The principle has been more and more acknowledged that the unarmed citizen is to be spared in person, property, and honor as much as the exigencies of war will admit.
Shortly after the Lieber Code was adopted by US President Lincoln as ‘Instructions for the Armies of the United States in the Field’, the principle of distinction was strengthened by the 1868 Saint Petersburg Declaration, which stated that ‘the only legitimate object which States should endeavor to accomplish during war is to weaken the military forces of the enemy’.Footnote 56 The underlying assumption of this Declaration was that only acts of violence against military forces and military targets are legitimate acts of war.
The first express mention of the principle of distinction in an international convention was in Article 48 of Additional Protocol I, 1977.Footnote 57 This article lays out the ‘Basic Rule’ that ‘the Parties to the conflict shall at all times distinguish between the civilian population and combatants and between civilian objects and military objects …’ The principle of distinction was described as one of the cardinal principles of IHLFootnote 58 and as the ‘grandfather of all principles’ of the law of armed conflict.Footnote 59 As a fundamental principle of armed conflict law, the distinction principle seemingly applies in all types of armed conflict, international and non-international. There are, however, subtle differences in the way in which the principle is presented in the treaty provisions relating to each type of conflict. In applying the principle in non-international conflicts there are also difficulties that do not arise in international conflicts.
In international armed conflicts the principle of distinction has two sides to it: one restrictive, the other permissive. On the one hand, the principle serves to protect civilians and civilian objects from attack.Footnote 60 On the other hand, it recognises the implicit licence of international law to use armed force against combatants and military objects. This dual nature of the principle is reflected in the very way the ‘Basic Rule’ is presented in Additional Protocol I, which relates, of course, to international armed conflicts. After requiring that the parties to an IAC distinguish between the civilian population and combatants, and between civilian objects and military objectives, Article 48 of the Protocol states that the Parties ‘shall direct their operations against military objectives’.Footnote 61 Article 51(2) adds that ‘[t]he civilian population as such, as well as individual civilians, shall not be the object of an attack’.
We see then that Additional Protocol I recognises both sides of the coin: the protective side is reflected in the prohibition on making civilians the object of an attack. The permissive side is reflected in the express provision that the parties shall direct their operations against military objectives.Footnote 62 When we turn to Additional Protocol II, which deals with non-international armed conflicts, we find only the protective rule.Footnote 63 This lends support to our argument that the sole impetus of extending the application of the principle of distinction to NIACs was to guarantee the protection of persons not taking an active part in hostilities.Footnote 64 It contains no implication that licence was being recognised under international law to use force against persons who do not meet this definition.Footnote 65
One would assume that, when the principle of distinction applies, we would have clear criteria for distinguishing between combatants and civilians. However, while the conventions relating to international armed conflicts contain a definition of the term ‘combatant’ and provide that a civilian is every person who is not a combatant,Footnote 66 neither Common Article 3 nor AP II make any mention of the term ‘combatant’. This was not an oversight. In IACs combatants have the right to participate in hostilities and therefore enjoy two related privileges: (i) immunity from prosecution for acts that are normally criminal, but are lawful under the laws and customs of war, such as killing an enemy combatant who was not hors de combat; and (ii) prisoner of war status if apprehended by the enemy. The states that drew up both the Geneva Conventions and its Additional Protocols were not prepared to grant these privileges to domestic rebels or insurgents. As we saw above, one of their main concerns in extending the application of international humanitarian norms to non-international armed conflicts was that this should have no implications as regards the status of those who take up arms against the authorities of the state. Thus, both Common Article 3 and AP II lack any mention of the term ‘combatants’ or any similar term, such as ‘fighters’.
The absence of any reference to combatants or fighters in non-international armed conflicts creates problems on a number of levels. First, there is an ongoing debate as to whether in such conflicts members of the armed groups of the non-state party should be considered as combatants, as belonging to a special category of fighters, or merely as civilians who take an active part in hostilities.Footnote 67 In the 2005 ICRC study on customary law, the ICRC concedes that in the context of a non-international armed conflict ‘practice is ambiguous as to whether members of armed opposition groups are considered members of armed forces or civilians’.Footnote 68 The debate on this distinction between fighters and civilians who take an active part in hostilities is not purely a matter of semantics. If fighters are the equivalent of combatants in international armed conflicts they may be attacked at any time (provided they are not hors de combat); civilians taking an active or a direct part in hostilities may be attacked only for such time as they take such part.Footnote 69 In the absence of provisions relating to combatants or fighters, as opposed to the situation in IACs,Footnote 70 in NIACs there are no norms that require fighters to wear uniforms or to have a fixed distinctive sign recognisable at a distance. Thus, even if one takes the view that there is a special category of combatant or fighter in NIACs, there is an inherent difficulty of identification, especially in situations in which the persons are not actively engaged in fighting. Finally, if we accept that members of the armed forces of the state party involved in a NIAC are to be regarded as combatants, the absence of a recognised and identifiable group of combatants on the side of the non-state actors weakens the symmetry between the parties.
The potential lack of symmetry between members of the armed forces of states and members of armed groups was raised during preparatory discussions for the ICRC ‘Interpretive Guidance on Direct Participation in Hostilities’.Footnote 71 Some scholars contended that, given the core principle of symmetry in LOAC, it is unacceptable that a member of the state's armed forces commits an international crime if he kills a member of an armed group who could easily have been captured, whereas if a member of that armed group kills a member of the state's armed forces who was not at the time engaged in hostilities, he merely commits a crime under the state's domestic law.Footnote 72
The ICRC attempted to resolve this problem of symmetry by making a distinction between civilians and members of organised armed groups who have a ‘continuous combat function’. For the purposes of the principle of distinction, the latter are analogous with combatants in international armed conflicts, and are thus legitimate targets even when not taking direct part in hostilities. This approach does not solve the problem of identification, which is an inherent problem in an armed conflict situation in which there is no external way of distinguishing between fighters and civilians. Removed from the conduct of hostilities area, identification will necessarily involve some degree of uncertainty.
3.2.3 Comparing Additional Protocols I and II
Additional Protocol II (AP II) aims to extend the norms of IHL which are applicable in certain types of non-international armed conflict, namely those which meet the conditions laid down in Article 1. According to this article, the Protocol applies only in armed conflicts between a government and dissident armed forces or other organised armed groups under responsible command, which have gained control over part of the territory and have the ability to implement the Protocol.Footnote 73 It is now widely accepted that following the adoption of AP II there are two types of non-international armed conflict: those which meet the conditions of the above article so that they are covered both by Common Article 3 and AP II, and those which do not meet those conditions and are therefore covered only by Common Article 3 (and subsequent treaties on means and methods of warfare which apply to all NIACs).Footnote 74
During the sessions of the Diplomatic Conference at which the two additional protocols to the Geneva Conventions were prepared, there were some voices in favour of unification of all the humanitarian norms in one treaty.Footnote 75 However, they were voices in the wilderness that received little support, and two separate additional protocols were adopted: the first dealing with IACs; the second dealing with NIACs. The difference between the two protocols is striking. While Additional Protocol I (AP I) purports to be an international treaty that supplements the 1949 Geneva Conventions (Article 1.3), it goes much further than those Conventions in that it relates not only to humanitarian norms but also contains detailed provisions on the conduct of hostilities, a topic ignored in the Geneva Conventions themselves. By incorporating both humanitarian norms and norms on the conduct of hostilities in one instrument, AP I is generally regarded as a treaty which abolished the distinction between Hague law and Geneva law.Footnote 76 AP II is far less extensive and the few norms it contains relating to the conduct of hostilities are all protective in nature.Footnote 77
AP I contains 102 articles and an annex relating to identification. AP II, on the other hand, contains only 28 articles, which are almost entirely devoted to (i) humanitarian norms relating to humane treatment of persons not taking a direct part in hostilities and persons deprived of their liberty; (ii) fair trials for persons prosecuted for criminal offences related to the armed conflict; (iii) treatment of the wounded, sick and shipwrecked; and (iv) protection of the civilian population, cultural objects and places of worship. AP I includes many provisions regulating the methods and means of warfare, which are conspicuous by their absence in AP II.
Besides the lack of norms relating to the conduct of hostilities that form the backbone of AP I, there are two other major differences between the two Additional Protocols that are highly relevant for the present discussion. The first is that AP I expressly states that members of the armed forces of the parties to a conflict ‘are combatants, that is to say, they have the right to participate directly in hostilities’.Footnote 78 While there has been some discussion in the literature over whether this right implies a right under international law to kill combatants of the adversary, or merely immunity from prosecution for doing so, the result is the same – a licence under international law to kill combatants, and even to kill civilians if they are not the object of the attack and the attack is not indiscriminate (that is to say, it meets the demands of proportionality).Footnote 79
As we saw above, the term ‘combatants’ is not mentioned in AP II, and there is no indication that anyone has the ‘right’ to participate in hostilities. The reasons for this ‘omission’ are well known and have been widely discussed in the literature – states were never prepared to accept that insurgents, rebels or members of dissident forces or organised armed groups should enjoy the privileges enjoyed by members of the armed forces of states involved in international armed conflicts.Footnote 80 On the contrary, one of the main objections to any application of humanitarian norms in non-international armed conflicts was that it would grant an element of legitimacy to those who challenge state authority by force. Only assurances that the application of such norms would not affect the legal status of the parties to a conflict, so that states could continue to regard members of rebel forces as criminals, paved the way for the adoption of Common Article 3. In this respect not much changed between 1949 and 1977.Footnote 81
Another, somewhat more subtle, difference between the two Additional Protocols relates to the presentation of the Martens Clause in each of them. This well-known clause was first incorporated in both the 1899 and 1907 Hague Conventions. The text adopted in the later Convention is regarded as the accepted formula of the idea that when specific norms do not exist in treaties dealing with the laws of war, ‘the inhabitants and belligerents remain under the protection and the rule of the principles of the law of nations, as they result from the usages established among civilized peoples, from the laws of humanity, and the dictates of the public conscience’.Footnote 82 In Article 1.2 of AP I the Martens Clause appears, with minor variations, in the formula of the Hague ConventionFootnote 83 whereas in the Preamble to AP II a modified version of the clause appears. Following the line of AP II itself, in which no mention is made of combatants, rather than referring to ‘civilians and combatants’, as AP I does, the text refers to ‘the human person’. More significantly, no mention is made of the principles of the law of nations or international law derived from established custom, but solely of ‘the principles of humanity and the dictates of the public conscience’. Apparently the states did not think there were any principles of international law that had not been incorporated in Common Article 3 or in AP II itself that were relevant to the protection of ‘the human person’ in non-international armed conflicts. So much for the idea that, even at the time when AP II was adopted (1977), there were principles of customary international law that applied in NIACs. The authors of the ICRC Commentary on AP II admit as much. In commenting on the AP II version of the Martens Clause they state: ‘It was apparently felt that the regulation of non-international armed conflicts was too recent a matter for State practice to have sufficiently developed in this field.’Footnote 84
AP II does not refer to combatants or fighters. It does, however, refer to civilians – seemingly a reference which makes sense only if there are also some persons who are non-civilians or combatants (albeit non-privileged ones). Article 13.3 also adopts the principle relating to direct participation of civilians in hostilities that appears in Article 51.3 of AP I, namely that civilians shall not be the object of an attack ‘unless and for such time as they take a direct part in hostilities’. The authors of the Commentary make the assumption that members of organised armed groups are not civilians under AP II, and that they may be attacked at any time.Footnote 85 In the absence of customary law relating to non-international armed conflicts, it is not at all clear on what they base this assumption. An alternative interpretation, which is more in line with the absence of any reference to non-civilians, is that in NIACs the legal vulnerability to attack is conduct based, rather than status based. In other words, the distinction is between those who take a direct part in hostilities and those who do not. Only the former may be attacked, and only during such time as they are taking part in such hostilities.
If one accepts this interpretation of AP II, one still has to ask why reference is made to civilians, rather than to persons (as in its version of the Martens Clause). One possible answer is that members of the armed forces of a state who take part in a non-international armed conflict are certainly not civilians. True, this would create a disparity between the parties to the conflict. On the one side, that of the state's armed forces, there would be non-civilians or combatants; on the other side, that of the non-state actors, there would only be civilians taking part in hostilities. Ostensibly this would be inconsistent with the very foundations of LOAC, which rest on symmetry between the parties.
The problem, however, is that non-international armed conflicts are, by their very nature, asymmetrical. The only way to change that is to apply principles that emanate from the recognition of belligerency in such conflicts, something that states have, of course, vigorously rejected. As long as there is no recognition of belligerency, under which there will be privileged combatants on both sides of the conflict, such conflicts will remain asymmetrical: non-actors who attack members of the armed forces of the state will be subject to criminal liability under domestic law, while members of the armed forces who attack non-state actors will not be subject to such liability. This inherent imbalance makes the parallel between international and non-international armed conflicts incomplete.
What conclusions may be drawn as to the effect of the Additional Protocols on the application of laws relating to the conduct of hostilities in non-international armed conflicts? Even if we limit ourselves to those NIACs covered by AP II, there seems to be no justification for arguing that the adoption of that protocol represented a radical departure from the legal situation under Common Article 3 of the Geneva Conventions, described above. Fundamentally, AP II simply represents an extension of the humanitarian norms imposed on parties in non-international armed conflicts. As we have tried to show, the lack of any mention of the right to participate in hostilities, the assumption that there was no customary law relating to such conflicts, and the absence of detailed norms regulating the conduct of hostilities belie the idea that AP II rests on notions of a wholesale importation of the laws and customs of war relating to IACs into the international norms that relate to NIACs.
3.3 Conclusions
There is no basis for arguing that the customary rules that apply in international armed conflicts – according to which combatants may be attacked at any time, as long as they are not hors de combat – apply in non-international armed conflicts too. It seems clear that in conducting hostilities in NIACs, the parties (including, of course, the state concerned) are not expected to use a policing paradigm of law and may resort to the rules on conduct of hostilities that apply in IACs. Our argument, however, is that, given the primary duty of the state towards persons subject to its jurisdiction to respect and ensure their right to life, and the fact that there is no recognised status of combatant in non-international armed conflict, this licence for the state to resort to the conduct of hostilities paradigm is restricted to the actual context of hostilities, where resort to a policing paradigm is clearly unrealistic. Where the state can reasonably employ the policing model of law, it is required to do so.
4. Narrowing the Gap between IACs and NIACs
There are indeed those who call for amalgamation of the norms of international and non-international armed conflict, or at least for narrowing the gap between them.Footnote 86 However, those who favour this approach are generally motivated by the desire to extend all of the means of protection offered in IACs to NIACs.Footnote 87 The main impetus for the amalgamation of IHL that applies in IACs and NIACs has come from the jurisprudence of the International Criminal Tribunals for the former Yugoslavia (ICTY) and Rwanda (ICTR). Close examination of this jurisprudence reveals, however, that the only questions that have arisen before those tribunals relate to the extension of humanitarian norms. The wide use of the term IHL as a term that is synonymous with the laws and customs of war may have led to the false impression that the extension of humanitarian norms brings in its wake the extension of norms relating to the conduct of hostilities.
4.1 International Humanitarian Law in the ICTY and ICTR
The establishment of the ICTY was a watershed in the development of IHL, and especially of international criminal law. For the first time since the establishment of the Nuremburg and Tokyo tribunals by the victorious allies after the Second World War an international criminal tribunal had competence to try persons charged with war crimes, crimes against humanity and genocide. The jurisprudence of that court has become a major source of authority for the norms of international humanitarian law.
The international community was horrified by the atrocities that were being committed during the wars that broke out in the territory of Yugoslavia after the collapse of the USSR and the fall of the communist regimes in the countries that had been in the USSR sphere. The desire to prevent impunity for atrocities committed in the conflict led to a series of resolutions by the UN Security Council (UNSC) which expressed grave alarm over reports of widespread violations of IHL in the territory of Yugoslavia,Footnote 88 and eventually to the establishment of a special criminal tribunal in order to try persons charged with war crimes and crimes against humanity.Footnote 89 Prior to the establishment of the ICTY the UNSC had requested the Secretary-General to establish a Commission of Experts to examine and analyse information submitted in accordance with previous resolutions on the situation in Yugoslavia ‘with a view to providing the Secretary-General with its conclusions on the evidence of grave breaches of the Geneva Conventions and other violations of international humanitarian law committed in the territory of the former Yugoslavia’.Footnote 90
The Report of this Commission is highly instructive. In discussing the legal basis for criminal liability of persons charged with participation in atrocities committed in Yugoslavia, the Commission quite rightly saw fit to distinguish between the law applicable in international armed conflicts and in non-international armed conflicts. In discussing the latter, the Commission stated:Footnote 91
The treaty-based law applicable to internal armed conflicts is relatively recent and is contained in Common Article 3 of the Geneva Conventions, Additional Protocol II, and Article 19 of the 1954 Hague Convention on Cultural Property. It is unlikely that there is any body of customary international law applicable to internal armed conflict which does not find its root in these treaty provisions. It is probable that Common Article 3 would be viewed as a statement of customary international law, but unlikely that the other instruments would be so viewed. In particular, there does not appear to be a customary international law applicable to internal armed conflicts which includes the concept of war crimes.
The body of customary international law applicable to international armed conflicts includes the concept of war crimes, and a wide range of provisions also stated in Hague Convention IV of 1907, the Geneva Conventions of 1949 and, to some extent, the provisions of Additional Protocol I.
It must be observed that the violations of the laws or customs of war referred to in Article 3 of the statute of the International Tribunal are offences when committed in international, but not in internal armed conflicts.
The UNSC Resolution, which implemented the principled decision to establish the Tribunal, adopted the Statute for the Tribunal which was attached to the Report by the Secretary-General. In that Report the Secretary-General states thatFootnote 92
the principle of nullum crimen sine lege requires that the international tribunal should apply rules of international humanitarian law which are beyond any doubt part of customary law so that the problem of adherence by some but not all states to specific conventions does not arise.
The Report makes no mention of the distinction between international and non-international armed conflicts and ignores the issue later mentioned in the Commission of Experts Report that ‘there does not appear to be a customary international law applicable to internal armed conflicts which includes the concept of war crimes’.
The ICTY Statute states that the Tribunal will have the power to prosecute persons charged with grave breaches of the Geneva Conventions, violations of the laws or customs of war, genocide and crimes against humanity. In defining the violations of the laws and customs of war, the grave breaches of the Geneva Convention and the crime of genocide, the ICTY Statute does not expressly mention non-international armed conflicts.Footnote 93 Only in the context of crimes against humanity does the Statute mention internal armed conflicts.Footnote 94
In the first ICTY case, the famous Tadić case, the defendant initially challenged the very existence of criminal liability in international law for acts carried out in the context of a non-international armed conflict and on this basis he claimed that the Tribunal lacked jurisdiction ratione materiae. Footnote 95 While the Appeals Tribunal accepted that the term ‘grave breaches’ of the Geneva Conventions applies only in international armed conflicts, it held that the term ‘violations of the laws or customs of war’ (set out in Article 3 of the Statute) includes, inter alia, serious violations of Common Article 3 of the Geneva Conventions. The Tribunal determined that such violations resulted in international criminal liability and that it therefore had jurisdiction to deal with them.Footnote 96 The Tadić decision was, in fact, the first judicial ruling that rejected the view of the Commission of Experts cited above, by holding that there is an extensive body of customary international law applicable to internal armed conflicts, and that violation of these rules may involve individual criminal responsibility.Footnote 97
The Appeals Tribunal in the Tadić case asserted that over time the traditional dichotomy between international and internal armed conflicts has become weakened as the ‘state-sovereignty-oriented approach has been gradually supplanted by a human-being-oriented approach’.Footnote 98 The Tribunal stated that since the Spanish Civil War, rules of international armed conflict protecting civilians from attack have been applied to non-international armed conflicts.Footnote 99 It proceeded to examine which humanitarian norms were applied in NIACs, and held that these related to two fields: (i) ‘general rules or principles designed to protect civilians or civilian objects from the hostilities or, more generally, to protect those who do not (or no longer) take active part in hostilities’;Footnote 100 and (ii) restrictions on means and methods of warfare, especially the use of certain types of weapon.Footnote 101
The ICTY is a criminal tribunal. As such it is concerned with the protective rules of armed conflict. The questions raised in the Tadić case related solely to the extension to NIACs of the forms of protection that exist in relation to IACs. The fundamental idea was that humanitarian norms should apply in all situations of armed conflict, irrespective of their character. As the Appeals Chamber stated:Footnote 102
Elementary considerations of humanity and common sense make it preposterous that the use by states of weapons prohibited in armed conflicts between themselves be allowed when states try to put down rebellion by their own nationals on their own territory. What is inhumane, and consequently proscribed, in international wars, cannot but be inhumane and inadmissible in civil strife.
Other scholars have written about the expansionist phase of international criminal law.Footnote 103 The ICTY's very jurisdiction is dependent on the application of IHL and it is therefore hardly surprising that it took an expansionist view of the application of IHL norms. As Ohlin states: ‘The more conduct that is governed by IHL, the more conduct can be described as a war crime and prosecuted in a court of law.’Footnote 104 It cannot be assumed, therefore, that extending the application of the restrictive norms of IHL to all the territory of the state in whose territory the NIAC is taking place implies that the permissive norms relating to the conduct of hostilities apply throughout the territory too.
Even in dealing with the specific criminal context, the Tribunal saw fit to mention that blurring the distinction between IACs and NIACs was not without limitations. The Appeals Chamber noted that internal armed conflict is not regulated in all its aspects by general international law. In particular, it noted two limitations:Footnote 105
(i) only a number of rules and principles governing international armed conflicts have gradually been extended to apply to NIACs; and (ii) this extension has not taken place in the form of a full and mechanical transplant of those rules to NIACs; rather, the general essence of those rules, and not the detailed regulation they may contain, has become applicable to NIACs.
Another argument raised by the defendant in the Tadić case ostensibly has important implications for the present discussion. The defendant, who was charged with serious crimes relating to his treatment of prisoners in the Omarska camp in the Prijedor area of Bosnia, claimed that there was no armed conflict of any kind in the Prijedor region at the relevant time. Since under its Statute, the ICTY has competence to ‘prosecute persons for serious violations of international humanitarian law committed in the territory of the former Yugoslavia since 1991’, in the absence of a nexus between the crime with which he was charged and an armed conflict, the defendant claimed that the Tribunal lacked jurisdiction to try him.Footnote 106 In rejecting this argument the Tribunal held that there was no basis for claiming that international humanitarian law applies only in areas in which fighting was taking place. On the contrary, it applies in all the territory of the state involved. As the Court stated, ‘until a peaceful settlement is reached international humanitarian law continues to apply in the whole territory of the warring states or, in the case of NIACs, the whole territory under the control of a party, whether or not actual combat takes place there’.Footnote 107
Looking at the other decisions of the ICTY and ICTR, in which the Tribunals consistently rejected the claim according to which IHL is restricted to the geographical and temporal spheres of hostilities, it is quite apparent that the impetus for widening the scope of application of IHL in non-international armed conflicts was mainly jurisdictional: in order to convict perpetrators of war crimes, a nexus to an armed conflict had to be proved. The broad territorial interpretation must be understood as an attempt to extend the application of humanitarian norms and international criminal liability for their violation.Footnote 108 In the Blaškić case the ICTY referred to Tadić and noted that ‘international humanitarian law continues to apply … in the case of NIACs, [in] the whole territory under the control of a party’,Footnote 109 and that ‘a given municipality need not be prey to armed confrontation for the standards of international humanitarian law to apply there’.Footnote 110 Similarly, the Tribunal continued to refer to IHL in the Kordić and Čerkez ruling, in which it stated that:Footnote 111
in order for norms of international humanitarian law to apply in relation to a particular location, there need not be actual combat activities in that location. All that is required is a showing that a state of armed conflict existed in the larger territory of which a given location forms a part.
In the Kunarac case, the Trial Chamber held that a state of armed conflict existed in the entire region relevant to the indictment and emphasised that IHL applies continuously to the whole of the territory under the control of one of the parties, whether or not actual combat takes place there;Footnote 112 the Appeals Chamber reaffirmed this position.Footnote 113 The jurisprudence of the International Criminal Tribunal for Rwanda (ICTR) offers similar interpretation of the geographical scope of IHL in NIACs, affirming that Common Article 3 ‘must be applied in the whole territory of the state engaged in the conflict’Footnote 114 and that IHL ‘extends throughout the territory of the state where the hostilities are occurring’.Footnote 115
There can be little argument with these statements made by both of the Tribunals, provided, of course, that we restrict them to what they actually say, and to the type of case in which they were actually applied. The Tribunals did not refer to the means and methods of war, nor to the conduct of hostilities, but to international humanitarian law. More specifically, the ICTY referred in the Tadić case to those humanitarian norms which apply to the treatment of prisoners. It seems almost self-evident that whichever norms apply to the conduct of hostilities, and wherever those norms may apply, persons with a nexus to an armed conflict who are held in detention by a party to the conflict must be treated according to the humanitarian norms that apply to the treatment of prisoners or detainees. Any other conclusion would lead to ridiculous results as, by removing prisoners from the battle arena, a party to an armed conflict would be able to free itself of the norms of humanitarian law which apply to those prisoners. There is no substantive reason why all the humanitarian norms that apply to prisoners should not apply, wherever they are held. Thus, in addition to the jurisdictional motive behind the expansion of the territorial scope of IHL during NIACs, such extension is essential in order to ensure that parties to an armed conflict cannot relieve themselves of humanitarian restraints by simply relocating individuals and directing operations away from the battlefield. One of the arguments against the US attempt to free itself from the constraints of IHL by holding persons captured on the battlefield in Afghanistan in Guantanamo Bay was precisely that such persons could not be denied the protection of IHL by their removal from the theatre of hostilities.Footnote 116
While there are ample examples of cases in which the application of IHL was extended in order to ensure the enforcement of humanitarian norms, we have found no case in which the argument was made that there should be a wider extension of the laws and customs of war relating to the conduct of hostilities so as to allow more leeway for the parties to kill, wound or cause damage with impunity.Footnote 117 It is here where use of the term ‘international humanitarian law’ causes confusion. While the ICTY seems to have used this term in its original sense – namely as referring first and foremost to those norms relating to armed conflict the aim of which is to protect those who are not, or are no longer, taking an active part in hostilities – given the use of this term as a general term for LOAC, the assumption seems to have become that the extension applies to all the rules of LOAC, including those relating to the conduct of hostilities. We argue below, however, that even if the rules relating to the conduct of hostilities apply on the battlefield during NIACs, the fact that humanitarian norms apply in the whole territory of the state involved does not imply that those rules apply there too; rather, if applicable at all in cases of NIAC, those rules are applicable in the theatre of hostilities alone.
4.2 Conduct of Hostilities in NIACs
Hostilities in non-international armed conflicts are often concentrated in comparatively limited and identifiable territories within the states in which they occur. For example, the armed conflict between the Sri Lankan armed forces and Tamil Tigers was largely confined to the northern and eastern regions of Sri Lanka.Footnote 118 The conflict between the armed forces of Uganda and the Lords Resistance Army is limited to the northern and north-eastern regions of Uganda;Footnote 119 the conflict between the ‘Huthis’ rebels and the armed forces of Yemen was mainly concentrated in the most northern region of Yemen;Footnote 120 and the various armed conflicts that have plagued the Democratic Republic of the Congo over the last two decades have all occurred within its east and north-east regions.Footnote 121
In the cases described there would appear to be a geographical divide between areas of hostilities and other areas. There may also be cases in which, while the geographical line between areas of hostilities and other areas is hazy, there is still a clear differentiation between the context of hostilities and contexts in which normal law enforcement continues. Hence we argue for a distinction between the contexts of hostilities and law enforcement, rather than a distinction that rests on the geographical area in which those hostilities are taking place. We are not arguing for a rigid line between geographical areas in which the conduct of hostilities norms apply and areas in which only humanitarian aspects of armed conflict apply. Rather, we stress that the issue is one of context: even if a ‘fighter’ is in the geographical area in which hostilities are taking place, he may not be targeted when it is perfectly reasonable to arrest him. On the other hand, if the ‘fighter’ is outside the area of hostilities but is surrounded by other fighters and is likely to engage any forces that come to arrest him in combat, it will be acceptable to employ the hostilities model against him.
In international armed conflicts the relationship between a party to the conflict and combatants and civilians of the enemy party is regulated first and foremost by LOAC. In the course of the armed conflict persons who come under the control of the hostile party may be regarded as subject to its jurisdiction for the purposes of the application of human rights conventions. Thus, for example, when territory becomes occupied in the course of an armed conflict, the occupying power's obligations under human rights treaties to which it is a party will apply to its relationship with persons in the occupied territory.Footnote 122 In Al Skeini, the European Court of Human Rights went further and held that these obligations would apply extraterritorially whenever a state involved in an armed conflict exercises ‘the public powers normally to be exercised’ by the government of a territory, with that government's ‘consent, invitation or acquiescence’.Footnote 123 Application of the human rights treaty obligations in such situations is a function of the control resulting from the armed conflict and not of the ‘normal’ relationship between those persons and that state.
In internal armed conflicts, the position is the opposite. All persons involved in such armed conflicts are already within the jurisdiction of the state when the armed conflict begins and are therefore protected by human rights conventions to which that state is a party. The state therefore has a dual role in relation to fighters and other persons with a nexus with the armed conflict. On the one hand, it must respect and ensure their rights; on the other hand, it may have certain powers to conduct hostilities against non-state actors who are a party to a non-international armed conflict according to the rules relating to the conduct of hostilities in armed conflicts. Given this dual function of the state, it may not extend the laws relating to the conduct of hostilities beyond the context of hostilities.Footnote 124
Even if we take the position that international human rights norms apply in international armed conflicts, their application vis-à-vis enemy forces is far more limited than it is in relation to members of armed groups involved in an internal armed conflict. In the former situation the state involved may have the duty to respect the right to life of all, in the sense that use of lethal force against the enemy forces must be in accordance with the laws and customs of war.Footnote 125 Towards persons who are not subject to its jurisdiction in the wide sense, it cannot, and does not, have the duty to ensure their right to life. In internal armed conflicts, on the other hand, the state has the duty both to respect and ensure the right to life of all persons who are subject to its jurisdiction. This must surely have implications in situations where it can frustrate the danger posed by such persons without resort to lethal force.
During the drafting of Additional Protocol II, concerns over the unjustified extension of LOAC to areas removed from the battlefield during internal armed conflicts were raised, as government experts ‘considered it inconceivable that, in the case of a disturbance in one specific part of a territory (in a town, for instance) the whole territory of the state should be subjected to the application of the Protocol’.Footnote 126 It is contended here that in situations of NIACs, LOAC should not apply in the whole territory of the state, but rather only in the specific context of hostilities;Footnote 127 in other contexts, in which territory is under the effective control of the state, and where the ability of the state to apply policing law-enforcement measures is evident, there is no justification for applying the conduct of hostilities paradigm and the law enforcement regime should apply. Hence the state will have justification to use lethal force only when it is absolutely necessary to do so.Footnote 128 This approach is similar to the notion presented in the 2009 ICRC ‘Interpretive Guidance on the notion of Direct Participation in Hostilities’. In Chapter IX, entitled ‘Restraints on the Use of Force in Direct Attack’, it is stated:Footnote 129
The practical importance of the restraining function [of the principles of military necessity and humanity] will increase with the ability of a party to the conflict to control the circumstances and area in which its military operations are conducted, and may become decisive where armed forces operate against selected individuals in situations comparable to peacetime policing. In practice, such considerations are likely to become particularly relevant where a party to the conflict exercises effective control, most notably in occupied territories and non-international armed conflicts.
When it comes to non-international armed conflicts, we consider that this position is correct not because of the restraining function of the principles of military necessity and humanity (which are again an amalgam of Geneva law and Hague law), but because the applicable paradigm is one of law enforcement rather than of conduct of hostilities.
It would seem that the ICRC adopted our thinking in its approach to the conflict in Syria. As early as May 2012, the ICRC maintained that the situation in Homs met the requirements of intensity, duration and organisation, and was therefore a non-international armed conflict.Footnote 130 At the same time it emphasised that this ‘does not mean that all areas throughout the country are affected by hostilities’.Footnote 131 Initially, the ICRC identified a restricted number of areas that were affected by hostilities and were therefore subject to the application of LOAC, but as the conflict intensified and hostilities spread it acknowledged that other areas had become affected by hostilities and, as a result, LOAC applied ‘wherever hostilities between government forces and opposition groups are taking place across the country’.Footnote 132 Of exceptional relevance for our discussion is the ICRC's statement according to which measures taken outside the areas affected by hostilities, including those against demonstrations with the purpose of restoring law and order, ‘must respect international law and standards governing the use of force in law enforcement operations’.Footnote 133 This indicated that, at least with regard to the situation in Syria, the ICRC adopted the approach that we advance here: whereas LOAC applies in the context of hostilities, the law enforcement regime governs other contexts where the state party is clearly in control.
5. Summary
We started with a simple question. Assuming a non-international armed conflict is taking place in the territory of a single state, may the forces of that state use lethal force against a member of the armed group which is involved in the armed conflict, even in a context in which using regular law enforcement measures, such as arrest and detention, would clearly be feasible? We are now in a position to summarise our arguments for a negative reply to this seemingly simple question:
1. International regulation of non-international armed conflicts began with the adoption of Common Article 3 to the 1949 Geneva Conventions. The aim and function of this provision were solely to subject the parties involved in a non-international armed conflict to fundamental humanitarian norms that should apply in all situations, and there was no implied assumption that in NIACs the norms relating to the conduct of hostilities in international armed conflicts apply. The 1977 Additional Protocols to the Geneva Conventions also maintained a clear distinction between IACs and NIACs. While Additional Protocol I contains detailed norms relating to the conduct of hostilities, and a specific norm acknowledging the right of combatants to participate in hostilities, such norms are conspicuous by their absence in Additional Protocol II.
2. The Commission of Experts established in 1993 by the UN Secretary-General, at the request of the UN Security Council, to examine the legal issues involved in establishing an international criminal tribunal for Yugoslavia, was of the view that no customary law exists relating to the commission of war crimes in non-international armed conflicts. We are of the view that there is also little, if any, evidence on state practice and opinio juris that would allow us to conclude that under customary international law states may use lethal force against members of armed groups except in the context of hostilities in which the application of the law enforcement paradigm is clearly not feasible.
3. To the extent that an argument is made that ‘fighters’ in non-international armed conflicts may be targeted at any time and in any place, it would seem to be based on a simplistic assumption that the norms applied in IACs also apply in NIACs. This assumption is a weak basis for that argument. There are significant differences between IACs and NIACs which explain why there cannot be a mechanistic importation of all IAC norms into NIAC situations.
4. In its jurisprudence, the ICTY has extended the application of humanitarian norms that apply in IACs to NIACs and has held that such norms apply throughout the territory of the state in which the NIAC is taking place. This jurisprudence has not related to extending to NIACs the IAC norms relating to the conduct of hostilities. Given the differences between IACs and NIACs, such an extension cannot be deduced from the extension of humanitarian norms.
5. In internal armed conflicts, states have a dual function: respect and the protection of the human rights of all persons subject to their jurisdiction and involvement as a party in an armed conflict with some of those persons. The latter role has the upper hand only in the context of hostilities themselves; outside this context the state's human rights obligations take precedence. Given the precedence of its human rights obligations, a state may not employ lethal force against a member of an armed group involved in an internal armed conflict in a situation in which it is perfectly feasible to employ law enforcement mechanisms. It follows that when it is perfectly feasible for the authorities of the state to use normal law enforcement mechanisms (arrest, detention and prosecution) against persons subject to its jurisdiction, it must employ those mechanisms even if such persons belong to armed groups who are involved in an internal armed conflict. As in all other cases of law enforcement, lethal force may be employed only when it is absolutely necessary to do so. This principle does not fall aside simply because an armed conflict is taking place in the state.
