Proxy Wars and International Criminal Law

After the Russian invasion of Ukraine in February 2022, there has been a resurgence in the discussions around aggression and the homonym crime. However, the debate has mainly centred around the jurisdictional limitations of the ICC and the leadership requirement of the crime (see, for example, here, here, and here). Still, the issue of indirect aggression, this is committing aggression through proxies, has attracted little attention.

Nonetheless, the increasing importance of non-state actors in the international arena and the growing instrumentalization by States to do their bidding and push their political agenda, normally through acts of force (see, for example, Iran’s relationship with the Houthis in Yemen or Hezbollah in Lebanon) begs for clarification on the criminalization of such conduct.

In this post I am going to critically approach Art. 8bis(2)(g) of the Rome Statute, challenging the traditional interpretation of the materialization of the State conduct element of the crime. But, given the complex nature of the crime of aggression, let me briefly present a breakdown of its elements and its traditional understanding.

The Traditional Interpretation of the Crime of Aggression

The crime of aggression in Art. 8bis of the Rome Statute combines a state act element with individual conduct elements: the individual acts of planning, preparation, initiation or execution carried out by a person in a position of leadership must be anchored in a state act of aggression (which manifestly violates the UN Charter). The existence of the latter is a conditio sine qua non for the individual criminal responsibility (see, Kreß, p. 312; Hajdin, p. 14).

In the case of indirect aggression, where a State utilises non-state actors or proxies to do their bidding, the challenge is compounded by the necessity to attribute the conduct of these non-state actors to the State itself. This complexity is heightened by the manner in which the Rome Statute defines the “act of aggression in Art. 8bis(2)(g), which differs from the approach taken both by the International Court of Justice (ICJ) and the International Law Commission (ILC). Specifically, it does not incorporate any developments that have occurred over time in the ILC Articles on the Responsibility of States for International Wrongful Acts (ASR). Instead, it replicates the wording of Resolution 3314(XXIX), adopted in 1974, thus requiring “the sending by or on behalf” of armed bands by the State or the State’s “substantial involvement”.

This formulation has been traditionally understood to reflect the customary rules of attribution of acts of non-state actors to a State, rather than creating a particular act of aggression. This was the position expressed by different States in the Report of the Special Committee that preceded Resolution 3314 (see, for example, the position of Canada and Mexico), and suggested by the ICJ when interpreting the provision in the context of self-defence (See, Nicaragua §195, Armed Activities, §146).

The debate has thus centred around defining which rule of attribution each formulation of letter (g) referred to. Some have proposed that the phrase “sending by or on behalf” is best understood in light of Article 4 of the ASR, particularly in view of the similar wording used by the ICJ in the Nicaragua case (§109) when determining the extent to which the actions of the Contra forces could be attributed to the United States. This interpretation was subsequently endorsed in the Bosnian Genocide case (§ 401).

It can be observed that the concept of “substantial involvement” is aligned with the remaining provisions of the attribution rule set forth in Article 8 of the ASR. This provision involves determining whether a non-State actor is deemed to be acting under the control of the State in question. This led yet again to a debate regarding the appropriate control test to be employed. While the ICJ requires an effective control, some authors have suggested to use the less strict overall control test to overcome the high evidentiary threshold of the latter (see, for example, Kreß, p. 449). In any case, the requisite degree of control would have to be exercised over each and every act constitutive of the act of aggression (Bosnian Genocide, § 401). This element, in isolation, sets a dangerously high threshold for determining the existence of an act of aggression, allowing state leaders to escape international criminal responsibility by using proxies, albeit arming, equipping and preparing them.

A Further Reading of “Substantive Involvement”

It might as well be that accommodating the formulation of Art. 8bis(2)(g) to the existing rules of state responsibility is a futile exercise, as there is no satisfactory answer that can reconcile the wording of such provision with the existing rules of attribution.

Therefore, I suggest that a better avenue to tackle this matter is considering the “substantive involvement” as an inherent and constitutive element of an act of aggression. Thus, attributing to the State the acts constituting the required involvement – normally carried out by its de jure organs – would be enough to determine the state conduct element of the crime of aggression thereof. However, this would be an insufficient criterion for attributing every individual act carried out by the armed bands.

This argumentation is coherent with the object and purpose of the prohibition of aggression, both in customary law and in criminal law. The crime of aggression seeks to safeguard the prohibition on the use of force enshrined in Art. 2(4) of the UN Charter. This is evidenced by the explicit reference in Art. 8bis of the Rome Statute to the UN Charter and Resolution 3314(XXIX), which refers to aggression as the “most serious and dangerous form of the illegal use of force”.

Furthermore, the ICJ held that the mere organization, provision of equipment, training, and military support to armed groups, even if not sufficient to attribute the acts in question to the State, suffice to constitutes a prohibited use of force (Nicaragua, §228; Armed Activities, §161-163). Likewise, given that aggression is defined as a manifest violation of the prohibition of the use of force, it is coherent to argue that an act of aggression can be committed as well just by the substantial involvement of the State, without requiring any further attribution, provided that the acts of the supported non-state actor are of enough gravity. Rejecting such interpretation risks rendering the “substantive involvement” construction without any effet utile.

Therefore, the “substantive involvement” of the State should be understood by its ordinary meaning, which is wide enough to cover a broad range of acts that do not amount to the exercise of control. It is submitted that providing material and logistic support, financing, training, arming, equipping, aiding or organizing (understood in a broad sense), or encouraging would be enough to satisfy the “substantive involvement” requirement, provided that such acts were, in fact, “substantive” as to allow or facilitate the armed group to commit the constitutive acts of aggression.

El Salvador and the US advanced a similar interpretation in the Nicaragua case, which was later endorsed by Judge Schwebel’s dissenting opinion (§155-167). Its rejection could be explained by the ICJ’s reluctance to accept the possibility that self-defense could be exercised against non-State actors, a position that has consistently held.

Additionally, this interpretation reflects the common understanding of the international community when the Definition of Aggression was adopted.

First, the UN GA Resolution 2625(1970) considered that organizing or encouraging the activities of subversive armed groups was enough to constitute a violation of the prohibition on the use of force, despite lacking attribution of the group’s acts.

Second, during the negotiations of Resolution 3314(XXIX), Western States pushed for a broader understanding of indirect aggression in the “Six-power Draft”, considering that “organizing, supporting or directing armed bands” was enough to constitute an act of aggression.

Third, the UN SC in its Resolution 507(1982) considered that an act of “mercenary aggression” had been committed by allowing the mercenaries to prepare and execute attacks from South Africa’s territory, without attributing those acts to the State (see, also, Resolution 419 (1977)).

Finally, this interpretation is supported by contemporary doctrine from that time (Rifaat, p. 273-4) and more recent scholarship (Milanovic, p. 334, Sayapin, p. 260).

Recent legal developments also suggest that at least some States are moving (or returning) to this broad understanding. In fact, the African Union Non-Aggression and Common Defence Pact, which entered into force in December 2009, in Art. 1(c)(viii) includes within the acts of aggression the State-sponsored support to armed groups, mercenaries or other organized transnational criminal groups that may carry out hostile acts against a State.

Furthermore, the presented interpretation is well suited to the accountability structure of the crime of aggression. The qualification of the acts constituting the substantial involvement as the act of aggression necessarily excludes the prosecution of the members of the armed group acting under the substantial involvement of the foreign State for that crime. Only the individual in control of the state would be held accountable for the aggression itself, whereas the commander of the armed group could be prosecuted for the individual violations committed in every specific act constitutive of aggression.

The construction of the inter-state prohibition of aggression follows a virtually identical structure. Only the State itself can be held responsible for aggression, but it would be accountable for the aggression as a whole, rather than for each and every constitutive act. However, the armed group or its members could be held domestically responsible for the violations committed in each and every specific act constitutive of the aggression. Consequently, the inability to attribute the individual acts of the armed group to the State does not result in a lack of accountability, as the State would be held responsible for the entirety of the group’s operations.

Finally, this broad understanding would not automatically turn every State support to an armed group into an act of aggression and risk the prosecution for the crime, since a crime of aggression would still require a manifest violation of the UN Charter by its character, gravity and scale, and the fulfilment of the required mens rea.

Concluding Thoughts

By incorporating Resolution 3314 into the text of the Rome Statute without providing any indication on how to reconcile it with the current developments on the law of state responsibility, the Kampala agreement perpetuated the conceptual controversies and challenges already present in 1974. Moreover, using a definition developed in the context of inter-state relations with sui generis formulations in the sphere of international criminal law requires a delicate balancing exercise with different areas of the law, each plagued with its own legal hurdles.

This blog post suggests a broader understanding of the “substantive involvement” of the State in order to determine the required state conduct element of the crime of aggression. Such an interpretation is coherent with the object and purpose of the crime of aggression and has found support among States and doctrine now and before. Furthermore, this understanding would also facilitate the inclusion within the crime of social and contemporary realities that otherwise would require an extensive amount of evidence, facilitating the fight against impunity.

Be that as it may, it would be desirable for the Assembly of States Parties of the Rome Statute to clarify the constitutive elements of this indirect crime of aggression, although this possibility seems unlikely. Thus, it would be up to the judges and the prosecution at the ICC to clarify the constitutive elements of this particular form of indirect aggression, if a crime of this nature ever falls within its jurisdiction and is brought to its attention.