Guest Post: Self-Defence and Non-State Actors in the Cold War Era – A Response to Marty Lederman

[Tom Ruys is professor of international law at Ghent University. He is the author of ‘Armed Attack’ and Article 51 of the UN Charter (CUP: 2010) and co-editor-in-chief of the Journal on the Use of Force and International Law.]

In a previous post on OJ, Kevin Jon Heller talked about the Israeli intervention in Lebanon in 2006 and its relevance from a jus ad bellum perspective. The post gave rise to a discussion between Kevin and Marty Lederman revolving essentially around the legality of self-defence against attacks by non-State actors in the Charter era. Kevin takes the position that throughout the Cold War attacks by non-State actors were generally only regarded as ‘armed attacks’ in the sense of Article 51 UN Charter inasmuch as they could be imputed to a State. In this context, Kevin among others things quotes a section of my book from 2010. Marty, on the other hand, claims that ever since the 1837 Caroline affair, States have always regarded attacks by non-State actors as ‘armed attacks’ (irrespective of any State involvement), which may trigger the right of self-defence if the necessity and proportionality requirements are met. He adds that there is no clear evidence from the Cold war era to suggest that States have come to embrace a prohibition on the exercise of self-defence against non-State attacks since 1945. Marty notes en passant that the excerpt from my book contains ‘no references to any state practice or opinio iuris’ that would support Kevin’s position. Yet, if this is indeed the case, it is because the excerpt is merely the conclusion of a much more extensive chapter, which does contain ample illustrations in terms of State practice and opinio (similar and other references can moreover be found in the excellent analyses by Olivier Corten (Law Against War) and Christine Gray (International Law and the Use of Force)).

Upon a closer reading of Marty’s comments, I doubt we can find much common ground, since our points of departure seem diametrically opposed. Marty’s view rests on the presumption that pre-Charter precedents, such as the Caroline case, are ‘incorporated’ in Article 51 UN Charter. I essentially agree that the Caroline case has largely constituted the source of inspiration for the customary requirements of necessity and proportionality in the context of the right to self-defence. That is not to say, however, that what was considered permissible back in 1837, was still deemed permissible in 1945. Quite the contrary, even if we leave aside the fact that recourse to war was still regarded as a lawful means of settling inter-State disputes in the 19^th century, the 19^th-century understanding of self-preservation was much broader than the modern concept of self-defence, which developed only gradually in the early 20^th century. The modern jus contra bellum is a product of the late 19^th and mostly early 20^th century. Up until 1912, Oppenheim stated without hesitation that intervention ‘in the interest of the balance of power’ must ‘obviously’ be excused, since ‘an equilibrium between the members of the Family of Nations is an indispensable condition of the very existence of International Law’ (International Law, 2^nd ed.). In other words: pre-Charter precedents must be put into perspective. Instead of relying on 19^th-century precedents, the proper course in dealing with this conundrum is to look at the Charter provisions and at evolutions in State practice in opinio juris since 1945.

Below are some cursory comments that tend to support Kevin’s position:

The adoption of the UN Charter and its aftermath

Even if some official statements and some draft versions of Article 51 refer to ‘attacks by one State against another’, the problem of non-State attacks was widely overlooked at the San Francisco Conference. Still, there are various indications that self-defence in the early Charter era was essentially construed as applying to attacks by one State against another. Kunz, for instance, in 1947 wrote that an armed attack ‘must not only be directed against a State, it must also be made by a State or with the approval of a State’ ((1947) 41 AJIL, 878). And the 1949 Report of the US Senate Committee on Foreign Relations on the North Atlantic Treaty contains the following observation: ‘[T]he words ‘armed attack’ clearly do not mean an incident created by irresponsible groups or individuals but rather an attack by one State upon another… However, if a revolution were aided and abetted by an outside power such assistance might possibly be considered an armed attack.’

Article 3(g) of the Definition of Aggression

Marty Lederman dismisses the idea that the Article 3(g) of the Definition of Aggression renders any support to Kevin’s position, since it ‘doesn’t say a thing about whether and in what circumstances IL prohibits State A from attacking the armed group in the territory of State B.’ That is perhaps textually correct, but if one takes a closer look at the debates within the Fourth Special Committee on the Definition of Aggression, it becomes obvious that the precise wording of this provision had everything to do with the underlying effort to find a compromise on the scope of Article 51 UN Charter (e.g., Italy: ‘The main concern was whether the right of self-defence should apply in cases of indirect aggression…’ UN Doc. A/AC.134/SR.52-66, 100). It all started with the following clause in the draft ‘Thirteen-Power proposal’: ‘When a State is victim in its own territory of subversive and/or terrorist acts by irregular, volunteer or armed bands organized or supported by another State, it may take all reasonable and adequate steps to safeguard its existence and its institutions, without having recourse to the right of individual or collective self-defence against the other State under Article 51 of the Charter’ (Article 7, UN Doc. A.AC.134/L.16). The draft clause fuelled an intense debate over the circumstances in which ‘indirect’ aggression might or might not trigger the right of self-defence. It is thus not too surprising to see that the ICJ used Article 3(g) of the Definition of Aggression in its Nicaragua case as the relevant threshold to determine whether attacks by non-State armed groups could justify the exercise of self-defence. What was surprising, however, about the Court’s approach, was that the Court gave a restrictive interpretation to Article 3(g) by excluding the provision of ‘logistical or other support’ (thus by and largely confining the applicability of Article 51 UN Charter to attacks imputable to a State, leaving little or no room for other forms of ‘substantial involvement’).

State practice in the Cold War period

If we look at State practice prior to 1980, in particular to the interventions by Israel, Portugal and South Africa in neighbouring countries, it is quite clear that State invoking the right of self-defence to justify cross-border military action against non-State actors generally claimed that the other State had somehow ‘sent’ the perpetrators of the initial attacks. By contrast, self-defence claims relying on active or passive support to armed bands or ‘terrorists’ operating on a more autonomous basis, did not meet with legal acceptance from third States. There is no evidence from the early Charter years to suggest that attacks by non-State armed groups could as such qualify as ‘armed attacks’ (irrespective of any State involvement).

It is only in the 1980s and the 1990s that this situation began to change as Israel, and subsequently the United States, increasingly began to claim a broader right to exercise self-defence against attacks by non-State actors (absent State imputability). This evolution must be seen against the background of an evolving political climate, viz. the (quasi-)completion of the decolonization process, and the growing recognition of terrorism as a threat to international peace and security. Cases such as the Israeli raid against the PLO headquarter in Tunis nonetheless illustrate that these claims initially met with a lukewarm, if not outright negative, reaction from third State (e.g., UN Doc. S/PV.2613, § 115 (Greece: ‘Acts of terrorism cannot in any way serve as an excuse for a Government to launch an armed attack on a third country’.).

State practice after the Cold War era and the 9/11 attacks

Marty stresses in his comments that the excerpt from my book contains ‘no references to any state practice or opinio iuris’ from the Cold war era affirming the need for State imputability (or at least State involvement) for attacks by non-State actors to trigger the right of self-defence. By contrast, he suggests the same excerpt refers to a ‘considerable number of interventions’ as well as ‘numerous security doctrines and official statements’ supporting the opposite view that non-State attacks can ipso facto amount to ‘armed attacks’ in the sense of Article 51 UN Charter. These references, however, relate to the post 9/11-era, not to the early Charter years. I readily acknowledge – and here I must depart to some extent from the position of Kevin – that there have been significant evolutions in State practice and opinio juris (cf. the Israeli intervention in Lebanon in 2006 or the strikes against IS in Syria). Yet, I fully agree with Kevin that this evolution is of a relatively recent nature and does not alter the fact that for most of the Charter era attacks by non-State armed groups were not, of themselves, regarded as ‘armed attacks’ triggering the right of self-defence. I hesitate to confirm whether this broader interpretation of the concept of ‘armed attack’ (as encompassing attacks by non-State actors absent any form of State involvement) has become lex lata. Suffice it to say that State practice seems to be heading in that direction (if the ICJ had taken the opportunity to confirm this trend in DRC v. Uganda or Palestinian Wall, we would not be having this discussion), and that the US-led military operations against IS have given rise to quite a number of interesting expressions of opinio juris of various States (for a comprehensive overview, readers may wish to keep an eye on the forthcoming Digest of State Practice in the Journal on the Use of Force and International Law).

(One final thing: like Marty and others I believe an ‘unable and unwilling’ test must be regarded as an aspect of the necessity assessment. By contrast, the question whether there is need for State imputability, or some other form of ‘substantial involvement’ by a State, for cross-border attacks to trigger the right of self-defence, has to do with the interpretation of the concept of ‘armed attack’.)