Lieber Studies Non-State Actor Attribution Volume Series – The Customary Law Status and Relevance of the Attribution Rule in Article 11 ARSIWA

Editors’ note: This post appears as an entry in a series dedicated to Dr Jennifer Maddocks’ monograph, Attribution of Non-State Actor Conduct during Armed Conflict, appearing in the Lieber Studies book series with Oxford University Press.

In her recent book, Jenny Maddocks argues that Article 11 of the Articles on Responsibility of States for Internationally Wrongful Acts (ARSIWA) does not reflect customary international law, although she accepts that it may acquire such status in the future. As is well known, Article 11 ARSIWA attributes to a State private conduct that would not otherwise be attributable to it where the State subsequently acknowledges and adopts that conduct as its own.

Her argument is, however, not entirely clear as to whether it is Article 11, or the concepts of “acknowledgement” and “adoption” that lack a customary basis due to divergent interpretations, although the former appears to be the principal thrust of her thesis. This post argues, by contrast, that Article 11 reflects customary international law and explains both its historical pedigree and its continuing significance.

Article 11 ARSIWA as Customary International Law

The situation addressed by Article 11 has deep roots in the law governing State responsibility for the conduct of individuals. Historically, State responsibility was not based on the concept of attribution but on whether a State was involved in some way in private conduct. Only later was this connection reconceptualised in terms of attribution, while in other contexts it evolved into distinct forms of responsibility, such as complicity.

Hugo Grotius, a foundational figure in the development of international law, recognised two circumstances in which State responsibility could arise in relation to private conduct: patientia; and receptus. The former resembles the modern concept of due diligence, as it concerns a State’s failure to prevent wrongful conduct. The latter imposed responsibility where a State failed to punish or extradite individuals who had committed wrongful acts. Indeed, receptus conveys the idea that the State accepted or recognised the private wrongful act through its conduct (Book II, Chapter XVII, XX & Chapter XXI, III).

A similar idea appears in the writings of Emer de Vattel who in general maintained that the acts of individuals are not imputed to a State unless they are approved or ratified. As he wrote,

If the nation, or its ruler, approve or ratify the act of the citizen, it takes upon itself the act, and may then be regarded by the injured party as the real author of the affront of which the citizen was perhaps only the instrument (para. 74).

This understanding was subsequently reflected in international arbitral practice. In Cotesworth & Powell, the tribunal stated,

One nation is not responsible to another for the acts of its individual citizens, except when it approves or ratifies them. It then becomes a public concern, and the injured party may consider the nation itself the real author of the injury (p. 2082).

Edwin Borchard similarly explained that State responsibility could arise from,

the actual or implied complicity of the government through either direct ratification or approval of the private act, or through implied, tacit, or constructive approval manifested by the state’s negligent failure to prevent the injury, investigate the incident, punish the perpetrators, or enable the victim to pursue available civil remedies (p. 217-18).

He further observed that, “although express ratification of private wrongful conduct was uncommon, several arbitral awards had nevertheless imposed responsibility on that basis.” His analysis was based on State practice, in particular the practice of the United States.

Later writers described this phenomenon as the State’s “condonation” of private conduct (here, paras. 84, 136-40, and here).

Against this historical background, Article 11, ARSIWA should be understood not as an innovation but as the modern expression of a long-established principle. It translates earlier notions of approval, ratification, condonation, and acceptance of private conduct into the contemporary language of attribution. The Tehran Hostages case should therefore be seen not as the origin of the rule embodied in Article 11 but as its modern judicial articulation. That is so even if the International Law Commission (ILC) cites relatively little State practice, and even though some of the practice and opinions it cited support a lower standard.

Since its adoption, Article 11 has been invoked and applied in investment arbitrations (mentioned in Maddocks’s book), by the International Tribunal for the Law of the Sea (ITLOS) (para. 182) and in Makuchyan and Minasyan v. Azerbaijan and Hungary where the European Court of Human Rights (ECtHR) expressly recognised that Article 11 forms part of the existing rules of international law (para. 113). The Court undertook a detailed analysis of this article, endorsed the interpretations of the International Court of Justice (ICJ) and the ILC, which treat acknowledgement and adoption cumulatively, and acknowledged that it sets out a very high threshold of attribution (para. 112).

On this basis the Court did not attribute the impugned act to Azerbaijan although its application of the law to the facts may be debated. As Judge Pinto de Albuquerque opined, granting salary arrears, promoting and pardoning the accused goes beyond endorsement and approval (here, para. 4). Borchard also argued that pardons or amnesties can establish State responsibility as demonstrated by State practice (p. 218). Maddocks is therefore correct to observe that some uncertainty remains regarding the precise attribution threshold of the rule. However, such uncertainty does not call into question the existence of the rule itself or its status as customary international law.

The Continuing Relevance of Article 11 in the Cyber Context

In the cyber context, Austria, Costa Rica, Denmark, the EU, Finland, Ireland, Japan, the Netherlands, New Zealand, Poland, Sweden, Republic of Korea, Thailand and the United States have, in one form or another, accepted Article 11 as the basis for attributing conduct to a State. While certain of the aforementioned States and organisations refer to Article 11 itself or reproduce its language, other States or organisations refer to the customary rules of attribution (the African Union, Australia, Colombia) or the Articles on State Responsibility (see here).

The importance of Article 11 as an attribution criterion lies in the fact that acknowledgement and adoption may be inferred from overall State conduct after the cyber operation has taken place, thereby decreasing thus the scope of plausible deniability (commentary to art. 11, para. 9). According to the Tallinn Manual 2.0, if a State deploys its cyber capabilities to protect a non-State actor in order to continue its actions, the non-State acts can be attributed to that State on the basis of Article 11 (p. 99, para. 16). In the same vein, failure to punish offenders or bring them to justice, amnesties or pardons had in the past given rise to State responsibility for the impugned act (p. 217-18).

For example, suppose patriotic hackers launch a distributed denial-of-service (DDoS) attack against State B, with which their State of nationality, State A, has strained relations. If officials of State A publicly characterise the attack as an act in defence of the nation or praise the hackers for performing their patriotic duty, while at the same time protecting the group’s infrastructure and deliberately refraining from taking action against them, such conduct, taken cumulatively, may amount to an acknowledgement and adoption of the hackers’ conduct as the State’s own within the meaning of Article 11 ARSIWA.

Similarly, suppose a patriotic hacker group exfiltrates confidential diplomatic communications from State B. If State A publicly praises the operation, subsequently relies on the stolen information to support its diplomatic position, rejects requests for the extradition of the perpetrators, and grants them immunity from domestic legal proceedings, the cumulative effect of these actions may likewise amount to an acknowledgement and adoption of the conduct under Article 11.

Conclusion and Interpretative Recommendations

It is true that Article 11 ARSIWA has received little attention in cyberspace or in the law of State responsibility in general. On the one hand, it adopts a broader perspective on the relationships between non-State actors and States and addresses situations where the involvement of States with non-State actors falls below the threshold of direction, instruction, or control. In this respect it complements the other attribution criteria by not allowing States to benefit from non-State acts.

On the other hand, Article 11 preserves the distinction between public and private conduct that lies at the core of the law of State responsibility and maintains the distinction between political and legal attribution by rejecting the urge to treat political rhetoric or praise, and expressions of endorsement, approval or support as attribution criteria. This function of Article 11 may also explain its high threshold compared to its earlier formulations and the fact that the ILC opted for “acknowledgement” and “adoption” instead of endorsement or approval which are the terms used in the Lighthouses arbitration (p. 155) and the Tehran Hostages (para. 74), which the ILC used as evidence of the rule. As noted, although the ECtHR applied the existing high threshold in Makuchyan and Minasyan v. Azerbaijan and Hungary, it also pointed out that further developments may be expected in this area (para. 114).

Taking this as a starting point, and bearing in mind the scarcity of modern case law and scholarly commentary on Article 11 and the fact that the sharp distinction between States and non-State actors on which the ILC’s approach was premised no longer reflects contemporary realities, I argue that there may be merit in adopting a lower threshold of attribution that reflects the origins of the rule and its rationale of catching State complicity in non-State acts.

Moreover, in practice, it is not easy to differentiate between adoption and approval or endorsement. Consequently, if a State shields a person or a group from punishment, continues to support their activities, or colludes with them, this may amount to acknowledgment and adoption of their acts. In Nikolić, for example, the International Criminal Tribunal for the former Yugoslavia interpreted acknowledgement and adoption as “collusion or official involvement” (paras. 64-66).

A further recommendation is for Article 11’s acknowledgement and adoption to be continuous, encompassing all the activities of a non-State actor instead of requiring the separate acknowledgement and adoption of particular non-State conduct. Neither its text nor its underlying rationale necessarily confines its operation to isolated acts. Consequently, where a State adopts a particular non-State actor as an instrument of its State policy, establishes a continuing relationship with that actor, and supports and coordinates its activities militarily, financially and politically, there is a plausible basis for attributing the operations of the non-State actor to that State under Article 11 without the need to establish acknowledgement and adoption anew for each individual act. This will be the case with Iran, Hezbollah, Hamas, and the Houthis. For example, in addition to the political, financial and military support Iran provided to Hezbollah, Iran acted as Hezbollah’s patron insisting that Israel ends its strikes against Hezbollah, securing the inclusion of relevant conditions in the memorandum of understanding and declaring the defence of Hezbollah as its strategic mandate (here & here).

Finally, in my view acknowledgement and adoption may take place ex ante. Although Article 11 is generally understood as addressing subsequent State conduct, its wording does not expressly require that acknowledgement and adoption occur only after the relevant acts have been committed. The example of the Ukrainian IT Army before its integration into the armed forces may be a case in point. An IT entrepreneur founded it, but the Ukrainian Minister of Digital Transformation later announced, “We are creating an IT army. All operational tasks will be presented in the Telegram channel.” To the extent that operations were subsequently carried out in accordance with tasks published through this official channel, a plausible argument can be made that Ukraine acknowledged and adopted those operations without the need to prove instructions, control or direction, or separate acknowledgement and adoption.

In conclusion, I submit that Article 11 remains highly relevant and has considerably greater potential than is commonly acknowledged, particularly as a means of complementing other attribution criteria and capturing forms of State involvement with non-State actors that these attribution criteria do not adequately address.

***

Nicholas Tsagourias is Professor of International Law at the University of Sheffield, UK.

The views expressed are those of the author, and do not necessarily reflect the official position of the United States Military Academy, Department of the Army, or Department of Defense.

Articles of War is a forum for professionals to share opinions and cultivate ideas. Articles of War does not screen articles to fit a particular editorial agenda, nor endorse or advocate material that is published. Authorship does not indicate affiliation with Articles of War, the Lieber Institute, or the United States Military Academy West Point.

Photo credit: Tasnim News Agency via Wikimedia Commons