The actions of what kind of entity can be qualified as ‘armed attack’ under Article 51 (armed attack ratione personae)? It is perhaps best to deal with this extra- ordinarily complex topic in two stages. First, it is to be decided whether only states can commit an armed attack. Second, assuming the first question is answered in the affirmative, the issue becomes what connection individuals ought to have to a given state to call their actions an armed attack (Section 2.4.2).
This is the second argumentative strategy employed by recent proponents of a wide reading of self-defence (next to a widening of the meaning of ‘armed attack’, see Sections 2.2.2 to 2.2.4). They submit that in this traditionally state- centred field of law, self-defence is not conditioned upon an armed attack com- mitted by a state. In effect, ‘armed attack’ is interpreted to mean that it is not dependent upon the actor’s status in international law: ‘armed attack’ does not mean ‘armed attack by a state’. This is held by a minority of international lawyers,156 but their legal arguments are brief.157
This interpretation can be seen as reasonable from a textual viewpoint, for clearly Article 51 does not make explicit who ought to be the actor involved in an armed attack. Accordingly, the argument is developed that ‘armed attacks’ as used in Article 51 need not necessarily be armed attacks committed by or on behalf of a state, because the text is open to this interpretation.158 Article 51 does not use the words ‘armed attack by a state’. One could argue that the wording: ‘if . . . occurs’
does suggest an occurrence independent of state action rather than an act of state. Article 51’s ‘armed attack’ can be approximated to a natural disaster, an occurrence not necessarily dependent on wilful action by a state.
The argument that private acts cannot be excluded from Article 51 is also based on various passages in Security Council Resolutions. Resolution 241, for example, expresses the Council’s concern at ‘the serious situation created in the Democratic Republic of the Congo following the armed attacks committed against that country by foreign forces of mercenaries’.159 However, the text of that reso- lution is less clear than it seems. The Council does not mention a link between the connection these ‘foreign forces of mercenaries’ might have with certain states and the classification of the acts as ‘armed attack’ under Article 51. In fact, the resolution does not mention self-defence at all. Because the French version of Resolution 241 translates what the English version calls ‘armed attack’ as
‘attaque armée’ rather than ‘agression armée’ (as used in Article 51), the drafters might not have intended to use ‘armed attack’ in its narrow technical sense. The
156 This minority has grown considerably since 11 September 2001 and the attendant changes in US policy.
157 Though cf. Kreò (1995) supra note 41 at 149–153, 206–249.
158 Dinstein (2005) supra note 9 at 244–245; Franck (2001) supra note 79 at 840; Kreò (1995) supra note 41 at 207; Miller (1985) supra note 29 at 57.
159 S/RES/241 (1967) Preamble (para 1) (emphasis added). Kreò (1995) supra note 41 at 207–208.
Uncertainty in International Law 36
same argument can be used for Resolution 405, whose French version mentions an act of ‘agression armée’ against Benin, but speaks of ‘armed aggression’ in English.160
In the aftermath of the events of 11 September 2001 the Security Council expressly referred to the inherent right of self-defence in Resolution 1368,161 which has led to the contention that this amounted to a recognition of that right within the specific context of the Al-Qaeda terrorist attacks against targets in the United States irrespective of attributability of these acts to Afghanistan.162 However, the Council did not acknowledge that the right to self-defence applied vis-à-vis Afghanistan (or even vis-à-vis the non-state group ‘Al-Qaeda’), but only
‘recognised’ or ‘reaffirmed’163generally the right to self-defence. It did not even specify that this right applies in the fight against terrorist acts or as against terrorist groups. Furthermore, it may be doubted whether the Charter has empowered the Security Council to make authoritative decisions on the existence vel non of armed attacks or to pronounce on the legality of armed force as self-defence.164
The most interesting arguments are teleo-systematic interpretations of the Charter (mostly by contrahents) or appeals to necessity (mostly by proponents).
In order to unravel the tangled strands of this argument it is best to present the argument of the majority165 (denying that armed attacks can be private in nature)
160 S/RES/405 (1977) Paragraph 2.
161 S/RES/1368 (2001) Preamble (para 3).
162 Franck (2001) supra note 79 at 840. Carsten Stahn takes that resolution, together with NATO’s North Atlantic Council statement (NATO Press Release (2001) 124, 12 September 2001, at:
www.nato.int/docu/pr/ 2001/p01–124e.htm) as ‘a limited clarification of the law’ (Stahn (2004) supra note 28 at 836–838) which significantly loosened the required connection to allow for an attack with merely an ‘external link’ – and thus independent of state action – to be counted as an armed attack (Stahn (2004) supra note 28 at 849–852). See also Randelzhofer (2002b) supra note 28 at 802 (MN 35). Cf. Armed Activities (2005) supra note 26, Separate Opinion Kooijmans at 313–314 (para 28), Separate Opinion Simma at 337 (para 11). Judge Kooijmans has recently criticised the present author’s opinion on this point (as formulated in Kammerhofer (2007) supra note 79 at 99–100): Pieter H. Kooijmans, The legality of the use of force in the recent case law of the International Court of Justice, in: Sienho Yee, Jacques-Yvan Morin (eds), Multiculturalism and international law. Essays in honour of Edward McWhinney (2009) 455–466 at 463–465.
163 S/RES/1368 (2001) Preamble (para 3); S/RES/1373 (2001) Preamble (para 4), respectively. The present author is indebted to André de Hoogh for his arguments against using the resolutions in this way.
164 The Council is authorised to decide when it has ‘taken measures necessary to maintain inter- national peace and security’ (Article 51, first sentence, second part), but that is a categorically different determination; one based on its Chapter VII competences and one utterly divorced from the legality of the action purportedly in self-defence. The Council cannot in any sense
‘authorise’ self-defence, contrary to Thomas Franck’s theory (Thomas M. Franck, Recourse to force. State actions against threats and armed attacks (2002)).
165 As Claus Kreò notes, there is a marked absence of argument by the majority, perhaps because the notion discussed here has been, until recently, a radical and marginal opinion. Kreò writes that ‘often, [this opinion] is implicit in [the authors’] words on Art. 51’ ‘họufig liegt diese [Meinung]
den Ausfỹhrungen zu Art. 51 unausgesprochen zugrunde’. Kreò (1995) supra note 41 at 206 (FN 879).
2.4.1 Self-defence 37
as a coherent argument, whereas the counter-arguments are presented as interjec- tions. While this puts the minority at a disadvantage, it is preferable to proceed in this way, because the majority has a more coherent argument, irrespective of its merits.
The majority starts with the argument that the ‘use of force’ against private individuals is not a threat or use of force and thus not prohibited by Article 2(4) UN Charter. Of course, harm to individuals may be a factual element of a breach of the prohibition, but only when military force is used between states.166 There- fore, the employment of military means only against individuals – since it does not constitute a violation of the ius contra bellum – does not need to be justified as an exercise of the right to self-defence. This would be the case when combating a domestic insurgency by military means or in the unlikely cases that an individual were swimming on the high seas or were to stay on territory considered res nullius.
However, if the ‘defender’ uses military means against a private individual or group staying on the territory of another state, the use of force impinges on the other state – even if the defender seeks to do harm only to the private entity. As Article 2(4)’s content is defined by the majority, the employment of military means by a state against another state is categorically prohibited. Therefore, the
‘defender’ violates Article 2(4) only vis-à-vis the territorial state, not vis-à-vis the individual or group. This use of force needs to be justified vis-à-vis the state in order to be lawful, not vis-à-vis the individual or group.
The acts of private individuals or groups, on the other hand, cannot be a use of force in the sense employed by Article 2(4). The majority contends that the pro- hibition of the threat or use of force is specifically inter-state in character: ‘All Members shall refrain in their international relations from the threat or use of force against . . . any state. . . .’167 Private persons violate international law by commit- ting acts of terrorism, but they are not bound by Article 2(4). Purely private actions cannot violate the prohibition of the threat or use of force. On a more general plane it can be argued that individual humans cannot violate a state’s sovereignty, because for a state’s sovereignty to be violated there needs to be an act of state. There needs to be a violation of the sovereign equality of states, for example through unlawful intervention or an act of aggression. Private acts can- not violate the maxim ‘par in parem non habet imperium’.
The use of military means by a state against individuals staying on another state’s territory constitutes a violation of the prohibition of Article 2(4) exclusively
166 While usually a trans-frontier use of force is required to classify as violation of Article 2(4), there are certain cases in which no crossing of frontiers takes place, e.g. the attack upon military forces stationed abroad or upon warships on the high seas. The necessary condition is the employment of military force by one state against the other.
167 Article 2(4) UN Charter (emphasis added). See Randelzhofer (2002a) supra note 18 at 121 (MN 28); Randelzhofer (2002b) supra note 28 at 802 (MN 34) as examples of the consensus opinion.
The content of the obligation in Article 2(4) is not contested by opponents, e.g. Kreò (1995) supra note 41 at 212: ‘requirement of state action in Article 2(4) UN Charter’ ‘Staatlichkeit- serfordernis des Art. 2 Ziff. 4 SVN’; Dinstein (2005) supra note 9 at 85–91.
Uncertainty in International Law 38
vis-à-vis the second state. That violation must be justified vis-à-vis that very state.
The armed attack as necessary condition for the legality of self-defence under Article 51 must be committed by the state against which force is directed. The defender’s prima facie unlawful action is after all taken against that state (the use of force is always directed against a state) and the defender’s acts are in need of justification. In other words, the territorial state (the guerrillas’ base state) needs to have committed an armed attack for the defender’s actions to be justified under Article 51. Thus, an ‘armed attack’ as used in Article 51 can only be an act attributable to a state.
This conclusion rests on two crucial – but assailable – premises on the nature of self-defence. This is the point where the minority can employ particularly effective counter-arguments, made effective by the fact that they show that this is not the only way the law on self-defence could be conceived, which is implied in the majority argument. However, showing that a solution is not the only conceivable version does not mean that this solution is excluded. Indeed, that might still be how the law is shaped.
The first premise is the identity of acts: self-defence justifies otherwise illegal acts of the defender only because the attacker has itself committed an unlawful act.
Self-defence presupposes illegal conduct by the attacker. In the Charter this means that the event ‘armed attack’ – the main necessary condition for self- defence168 – must be an international wrong. Furthermore, the majority believes that the offending act must be the same sort of infraction as the one to be justified,169 because the trigger for the justification must be a violation of the prohibition violated. Thus, ‘armed attack’ must contain ‘threat or use of force’.
Self-defence is conditional upon a threat or use of force170 amounting to an armed attack. Since the general prohibition may only be violated by states, only states can commit an armed attack.
The interjections against this premise proceed from the assumption that armed attacks as such are not prohibited; they merely form the trigger for lawful self- defence. Force is prohibited and the Charter makes it abundantly clear that only states can threaten or use force vis-à-vis other states. The trigger is not necessarily a violation of international treaty law, because the violation can only be commit- ted by a state. The negative connotation of the French term ‘agression armée’
compared with ‘attaque’171 ought not to be read as containing a specific reference to an international wrong in the technical sense.172
168 Section 2.2.1.
169 Ago (1982) supra note 9 at 54 (para 89): ‘[T]he only international wrong which, exceptionally, makes it permissible for the State to react . . . by recourse to force . . . is an offence which itself constitutes a violation of the ban.’
170 On this narrow point Bowett agrees: ‘. . . the right of self-defence is only available against a use or threat of force which is delictual as being contrary to international law.’ Bowett (1958) supra note 11 at 11.
171 Klein (1964) supra note 28 at 182.
172 Kreò (1995) supra note 41 at 208.
2.4.1 Self-defence 39
Why, continues the minority, ought the violation of international law to be the only trigger for self-defence as justification? The only hindrance is that the attacker’s acts might themselves be justified either as action authorised under Article 42 or as self-defence and thus themselves be justified, which, in turn, would violate the maxim: ‘No self-defence against self-defence.’ This causes a logical problem: if action against self-defence can be self-defence, then the first defender can itself exercise self-defence against the self-defence of the second defender and so cause a vicious circle. Claus Kreò argues that we avoid this happening by supposing that Article 51 requires an attack not itself justified by international law as self-defence or military sanctions, which would mean that private attacks could never be justified.173
Even if one were to suppose that self-defence needs unlawful conduct as a trigger, the minority doubts whether the unlawful conduct specifically needs to be a use of force.
Rather, one could say that self-defence ‘if an armed attack occurs’ is exercised, when an unlawful act under international law is connected to an armed attack in such a way that the existence of the armed attack creates a self-defence situation.174
This means that the connection established by the majority is severed and the event ‘armed attack’ does not have to be a wrong. In the case of a private attack a second necessary condition is added and the first condition (‘armed attack’) is modified. The majority requires that the unlawful threat or use of force is the armed attack (force amounting to an armed attack – Section 2.2.2) in order to trigger self-defence. Kreò, on the other hand, needs, first, an ‘armed attack’
(though using an expanded understanding of that term) committed by any person or group and, second, unlawful conduct by some state in connection with that
‘attack’. The problem with this argument is, however, that if the connection between the unlawful act and the armed attack is severed, the attacker and the entity defended against are different (which will be discussed à propos the second premise).
Minority scholars argue that the mechanics of prohibition and exception are not such as to require the exact contraposition of the trigger and the justified reaction as the same prima facie unlawful acts. The duality of rule and exception does not mean that positive regulation need specify that a wrong constitute the trigger, only that the prohibition obliges states and that the exception may only be exercised by states. This is fulfilled in the case of Articles 2(4) and 51.175 States
173 Kreò (1995) supra note 41 at 208 (FN 885).
174 ‘Vielmehr kann auch dann von einer Selbstverteidigungssituation “im Falle eines bewaffneten Angriffs” gesprochen werden, wenn vửlkerrechtliches Unrecht derart mit der Begehung eines bewaffneten Angriffs im Zusammenhang steht, daò mit der Vornahme des bewaffneten Angriffs der Eintritt einer Selbstverteidigungsituation verbunden ist.’ Kreò (1995) supra note 41 at 209.
175 Kreò (1995) supra note 41 at 211–212.
Uncertainty in International Law 40
would still remain the addressees of rule and exception, even if one were to argue that ‘armed attacks’ may also be committed by private individuals. This argument merely brings to light the problem already mentioned in the preceding paragraph.
States do not need to justify their forcible acts against private individuals. Only if they act forcibly against another state do they need to justify their acts – vis-à-vis that state, not the individuals concerned. If a state attacked by the defender cannot be said to have itself committed the armed attack, then the attacker and the entity ‘defended’ against are different.
The second premise is the identity of entities: self-defence needs (a) an illegal act by the attacker and (b) that the entity that attacks is the only valid target for the defender.
Should the defender violate a third entity’s rights (not the attacker) in the course of its defence, the acts against that third party cannot conceivably be justified as
‘self-defence’.176 The only possible justification would be ‘necessity’,177 as codified in Article 25 of the 2001 Articles on State Responsibility.178 The UN Charter, however, does not recognise such an exception to the prohibition of force. There- fore, the majority concludes that the state which is the target of defensive action has to have committed the armed attack itself. It would be disjointed to allow state A to act against state B because person C not under B’s control had harmed A.
States cannot be made responsible, and much less attacked, for persons they cannot control. It would be a mere fiction if states were thought to control every- thing happening on their territory.179
The opposite position is put to us by Dinstein:
The armed bands or terrorists in Arcadia are not cloaked with a mantle of protection from Utopia. . . . Just as Utopia is entitled to exercise self-defence against an armed attack by Arcadia, it is equally empowered to defend itself against armed bands or terrorists operating from within the Arcadian territory. . . . Utopia may, therefore, dis- patch military units into Arcadian territory, in order to destroy the bases of the hostile
176 Even Bowett admits this: Bowett (1958) supra note 11 at 56.
177 Ago (1982) supra note 9 at 61–62 (para 106); Kreò (1995) supra note 41 at 208. Interestingly, Albrecht Randelzhofer argues that ‘[f]or the purpose of responding to an “armed attack”, the State acting in self-defence is allowed to trespass on foreign territory even when the attack cannot be attributed to the State from whose territory it is proceeding’ (Randelzhofer (2002b) supra note 28 at 799 (MN 29)). For him, actions against non-attacking states may also be justified by self- defence, although the remark above is made with respect to Article 3(f) Definition of Aggression (Placing territory at another state’s disposal) as applied to self-defence.
178 General Assembly, Articles on responsibility of states for internationally wrongful acts, A/RES/
56/83, Annex, 12 December 2001 (ARS 2001). See also two earlier versions: (a) International Law Commission, Draft articles on responsibility of states for internationally wrongful acts 2001 (DARS 2001), in: International Law Commission, Report of the International Law Commission on the work of its fifty-third session 23 April to 1 June and 2 July to 10 August 2001, A/56/10 (2001) at 29–365. (b) International Law Commission, Draft articles state responsibility 1996 (DASR 1996) Article 33, in: International Law Commission, Report of the International Law Commission on the work of its forty-eighth session 6 May–26 July 1996, A/51/10 (1996).
179 Corfu Channel (1949) supra note 22 at 18.
2.4.1 Self-defence 41