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

by Tom Ruys

[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 19th century, the 19th-century understanding of self-preservation was much broader than the modern concept of self-defence, which developed only gradually in the early 20th century. The modern jus contra bellum is a product of the late 19th and mostly early 20th 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, 2nd ed.). In other words: pre-Charter precedents must be put into perspective. Instead of relying on 19th-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’.)

8 Responses

  1. Thanks to Tom for the typically learned and careful post. Just two points.

    First, I am in basic agreement with him concerning the post-9/11 era — there is no question state practice is moving towards the unwilling and unable test. I suspect our views concerning custom-formation simply differ somewhat, as views on that issue often do. I am particularly concerned about the tendency in the global north (and I am not including Tom in this criticism) to simply ignore states in the global south that protest against various evolutions in the jus ad bellum. (The 2000 statement by the Non-Aligned Movement being a perfect example.) So I simply think we are a bit further from U/U becoming custom than Tom does.

    Second, regarding necessity: there is no question that, were U/U to become lex lata, a state’s willingness and ability to prevent an NSA from using its territory to launch armed attacks would be properly considered in terms of necessity. My point — and I believe Tom agrees with this — is simply that insofar as armed attacks by NSA’s still require state attribution, however defined, non-attributable attacks by NSAs — such as those where the state is nothing more than unwilling or unable to prevent the attacks — are not armed attacks, do not give rise to the right of self-defence, and thus make consideration of necessity irrelevant.

  2. Regarding Israel and Tunis, the use of force in “reprisal” was condemned in 1985, the U.S. explaining its vote in the S.C.: “We, however, recognize and strongly support the principle that a state subject to continuing terrorist attacks may respond with appropriate use of force to defend against further attacks. This is an aspect of the inherent right of self-defense recognized in the United Nations Charter….” The U.S. has consistently recognized the propriety of self-defense against NSA armed attacks since 1814, 1816, 1817, 1818 and the Caroline incident in 1837. See, e.g., Self-Defense Targetings of Non-State Actors…, 19 J. Transnat’l L. & Pol’y 237, 241-49 (2010) (available at SSRN).
    The 1986 ICJ opinion did not rule on self-defense against a NSA as such. Greece addressed an “attack on a … country,” not “on” a NSA. Gray wrote that “the response of the world in 1998 with respect to the U.S. use of force in self-defense against al Qaeda that “the response of the world was generally muted,” whereas others wrote that most states had “acquiesced” and generally accepted “Article 51’s application. See id. At 247-49 n.29, and Mary Ellen at 97 AJIL 446, 450 (2003) (“use of force in Afghanistan in 2001 was lawful self-defense…. September 11 attacks were part of a series of terrorist actions” that began in 1993 “and would include future attacks,” and Security Council resolutions “reveal the Council’s consensus that armed force in self-defense following terrorist attacks is lawful”). For an extensive list of more than 53 writers in agreement, see 19 J. Transnat’l L. & Pol’y at 239-40 n.3.

  3. Jordan,
    Thank you for your reaction. Two comments from my side:
    – A series of precedents from the early 19th century is not helpful in the present context. Again, these precedents date from an era where there was no general prohibition against recourse to force in international law. Relying on those precedents to establish the (continued) legality of self-defence against attacks by non-State actors is a bit like invoking the interventions in the Ottoman Empire to defend the legality of humanitarian interventions under contemporary international law.
    – Apart from the 19th-century precedents you refer inter alia to the 9/11 attacks and the US strikes following the 1998 embassy bombings. I readily concede these are relevant and important precedents in this context, but again: they relate to the post-Cold War era. I would invite you to find similar cases from the period 1945-1980 that confirm that non-State attacks can ipso facto qualify as ‘armed attacks’. Fact is that throughout the Cold War era, self-defence was conceived either as a reaction to ‘direct aggression’ (an attack by one State against another) or a reaction to ‘indirect aggression’ (which required State imputability of the attacks, or at least substantial State involvement). There was no third option for reactions agains ‘non-State aggression’. State practice from this period (cf. interventions by Portugal, South Africa and Israel and the reactions thereto) indicates that State imputability or substantial State involvement was considered of key importance, even if the action in self-defence did not target State infrastructure but only the non-State presence.

  4. @Jordan,
    Here is what Judge Kooijmans had to say in his Separate Opinion in the Palestinian Wall case. Criticizing the Court for recognizing the existence of an inherent right of self-defence in the case of an armed attack by one State against another State, while ignoring the impact of resolutions 1368(2001) and 1373(2001), he stated the following: ‘This [i.e., the fact that resolution 1373(2001) recognized the right of self-defence without ascribing the 9/11 attacks to a particular State] is the completely new element in these resolutions. This new element is not excluded by the terms of Article 51 since this conditions the exercise of the inherent right of self-defence on a previous armed attack without saying that this armed attack must come from another State even if this has been the generally accepted interpretation for more than 50 years. The Court has regrettably by-passed this new element, the legal implications of which cannot as yet be assessed but which marks undeniably a new approach to the concept of self-defence.’ My point is simply that we need not, and should not, rewrite the history of the jus ad bellum, to assess the impact of novel trends in State practice and opinio juris.

  5. Tom: I appreciate that you don’t wish to even consider “the 19th Century” precedents, but (1) they were part of practice and opinio regarding claims to engage in measures of self-defense, and (2) the UN Charter expressly recognized the existence of an inherent right of self-defense (presumably with some consideration of the 19th Century and early 20th Century precedents — my article also addressed some U.S. practice and opinio re: 20th Century practice), although Article 51 expressly limited such with the requirement that an armed attack occur.
    Without commenting on the remarks of any particular Judge or Justice (e.g., of the present U.S. S.Ct.), one can understand that some have been particularly inattentive to history, especially history demonstrating the formal participatory roles of several non-state actors — even with respect to what we term jus ad bellum. International law is too important to be left to judges.

  6. p.s. I cited the Wall Opinion in my footnote 3 (as a “but see” cite) and Judges Kooijmans’ and Simma’s later opinions in the Congo Case.

  7. p.s.p.s. — a recent posting at ASIL:
    Professor Mark Weller’s ASIL Insight: Striking ISIL (available through ) declares that “[a] non-state entity can mount an ‘armed attack’ … notwithstanding the confused reading often given to the Advisory Opinion in the Wall case….” [footnote: “The ICJ did not in fact hold that activities of non-state actors cannot give rise to self-defense, but that Israel could not invoke self-defense against a non-state actor operating in territory it occupied.”].

  8. “I would invite you to find similar cases from the period 1945-1980 that confirm that non-State attacks can ipso facto qualify as ‘armed attacks’.”

    From Security Council Resolution 84 “The Security Council,
    Having determined that the armed attack upon the Republic of Korea by forces from North Korea constitutes a breach of the peace”

    At the start of the Korean War, only the Russians recognized the government of North Korea. The UN General Assembly explicitly rejected the idea that North Korea had any official government. Generally speaking, the official view is that there was one country called Korea with two occupation zones, a Russian zone in the North and an American zone in the South. The Republic of Korea was recognized as a democratically elected government in the South, but nobody claimed it had any authority over the North or the entire country. Therefore, although the invasion by the North had the appearance of an IAC, legally it was the start of a civil war in which a non-state party in the northern half of a recognized country under occupation with no recognized national government attacked the government of the southern section of the same country. Still if the Security Council can call it an “armed attack” this means that a non-state entity can mount an ‘armed attack’ based on the highest available authority (the Security Council).

Trackbacks and Pingbacks

  1. There are no trackbacks or pingbacks associated with this post at this time.