The Crime of Aggression and Modes of Liability – Is There Room Only for Principals?

Defendants at Nuremberg


By Volker Nerlich*

The crime of aggression is a “leadership crime.” Not anyone who participates in a war of aggression—for instance, as a member of an aggressor’s army—is to be held criminally responsible. Rather, in keeping with the precedents of Nuremberg and Tokyo, liability attaches only to those high up in the chain of command.[1] Nevertheless, the exact reach of the criminalization has remained largely unclear in post-World War II jurisprudence. The International Military Tribunal in Nuremberg (IMT) convicted twelve of the twenty-two high-level Nazi officials accused of crimes against peace or conspiracy to commit crimes against peace.[2] In the subsequent proceedings before American military tribunals in Nuremberg, crimes against peace and conspiracy to commit crimes against peace were charged in four cases,[3] resulting, however, in only five convictions, all in the Ministries Case.[4] According to Henry T. King, who, like Benjamin Ferencz, was one of the prosecutors in the Nuremberg follow-up trials, “the IMT judgment left open the question of how involved in the policy of aggression an individual would have to be in order to be convicted.”[5]

The definition of the crime of aggression in article 8 bis of the Rome Statute seeks to resolve this question by requiring that the perpetrator be “in a position effectively to exercise control over or to direct the political or military action of a State.”[6] This is a high threshold: The person planning, preparing, initiating, or executing the acts made criminal under article 8 bis must presumably hold a high-ranking position in the aggressor state. What is required is “control” or the ability to “direct” the state’s action. This suggests that only a small number of government leaders, perhaps only the head of state or government or the ministers of defense or foreign affairs, could ever be held guilty of the crime of aggression.[7]

Article 25(3) of the Rome Statute establishes a detailed and differentiated system of modes of liability, which distinguishes between principal perpetrators—who, either alone, together with, or through others commit a crime,[8]—and accomplices—who are involved in instigating, ordering, aiding and abetting, or otherwise contributing to the commission of a crime by one or more principal perpetrators.[9] It is noteworthy that the definition of principal liability in the International Criminal Court’s jurisprudence uses language that resembles the characterization, in article 8 bis(1), of the potential persons criminally responsible for a crime of aggression: According to the case law,[10] a principal is someone who has control over the crime, in the sense of possessing the ability to frustrate its commission. In contrast, all forms of accomplice liability require a lesser form of control over the crime.

During the negotiations on the crime of aggression, there was a debate as to whether article 25’s differentiated participation regime should be made applicable to the crime of aggression or whether the incriminated conduct should be set out comprehensively and conclusively in article 8 bis.[11] The former solution was eventually adopted. Article 25(3) bis provides the link between the crime of aggression and the modes of liability in article 25(3) of the Rome Statute.[12] The provision clarifies that article 25(3) applies to the crime of aggression as well, albeit, “only to persons in a position effectively to exercise control over or to direct the political or military action of a State,”[13] thus copying the language of article 8 bis(1). In light of this formulation, it has been argued that the effect of article 25(3) bis and the adoption of a differentiated approach is “virtually nil”—essentially all those participating in a crime of aggression would be principal perpetrators.[14] At first sight, this would appear to be a reasonable expectation, given the definition of principal perpetration in the case law of the ICC and the limitations that have been adopted on the persons potentially responsible for the crime of aggression.

However, such an understanding would lead to a surprising and arguably unreasonable result. While article 25(3) bis makes the whole range of modes of responsibility in article 25(3) applicable to the crime of aggression, the formulation of the provision would, in effect, negate the applicability of large parts of article 25(3). On this reading, rather than allowing for the application of all sub-sections of article 25(3) to the crime of aggression, article 25(3) bis would actually limit applicability to article 25(3)(a)—perpetration as a principal.

Perhaps a closer look at the interplay between article 25(3) bis and article 25(3) is needed. First, the level of control that is required has to be assessed. Is control only “effective” if it is complete—unified in one man or woman at the helm of a state? Such an understanding of “effective control” would essentially limit the crime of aggression to dictators holding absolute power in a state. This cannot have been intended. Indeed, arguably none of the accused in Nuremberg held such a high level of control over the war-making of Nazi Germany. In addition, political systems based on separation of powers and “checks and balances” would automatically be excluded, as none of the political leaders would, in fact, hold effective control. Thus, control in terms of article 8 bis and article 25(3) bis must be considered to be effective even if it is not complete, as long as the person in question has the power to shape political and military decision-making. Such an understanding would also align with the post-World War II case law, particularly the Nuremberg follow-up trials, in which the tribunals adopted a “shape or influence” standard.[15]

Further, it is noteworthy that, according to the plain language of article 8 bis(1) and article 25(3) bis, the object of control or direction is not the act of aggression itself, but the “political or military action” of a state. Arguably, this must be determined independently of the question of who was at the center of the decision-making with respect to the specific act of aggression giving rise to criminal responsibility. While there must certainly be some link to the exercise of military force—for example, it would be difficult to justify control over the cultural policy of a state as sufficient to make the person a potential perpetrator of or accomplice to a crime of aggression—the group of people controlling or directing the “political or military action” of a state may be larger than the group that actually made the decision to go to war.

If such an approach is accepted, it is conceivable that control over the “political or military action” and the control over the commission of the crime of aggression do not fully align. In a given situation, a government minister may have control over the political action of a state—including in matters of war and peace—but may be involved only indirectly in planning, preparing, initiating, or executing the act of aggression. She may thus be found to have been an accomplice rather than a principal perpetrator of the crime of aggression. The same may apply to high-ranking officials within the military.

Based on such an understanding of article 8 bis and article 25(3) bis, not only those who were at the center of a decision to wage aggressive war could be held accountable, but also those who, while not at the sidelines, were somewhat removed from the decision-making process, as long as they were sufficiently high up in the hierarchy to be able to shape the policy of the aggressor state in that regard. In turn, and depending on the facts of the case, this would allow for the application of different modes of liability to all those who were in a position of control; some might be considered principal perpetrators, while others might be mere accomplices, who, for example, aided and abetted the execution of an act of aggression. Thus, while the crime of aggression is a leadership crime, it is not necessarily a crime only of principals.

* Legal Adviser, International Criminal Court; honorary professor, Faculty of Law, Humboldt-University of Berlin. The views expressed are those of the author and cannot be attributed to the International Criminal Court.

[1] On the post-World War II jurisprudence in this regard, see Carrie McDougall, The Crime of Aggression under the Rome Statute of the International Criminal Court 168–83 (2013).

[2] Namely, Göring, Hess, von Ribbentrop, Keitel, Rosenberg, Frick (no conviction for conspiracy), Funk (no conviction for conspiracy), Dönitz (no conviction for conspiracy), Raeder, Jodl, Seyss-Inquart, and von Neurath. See Judgment, 1 Trial of the Major War Criminals Before the International Military Tribunal 279–341 (1947).

[3] The Farben Case, the Krupp Case, the High Command Case, and the Ministries Case.

[4] Namely, von Weizsaecker, Keppler, Woermann, Lammers, and Koerner. See Judgment, 14 Trials of War Criminals Before the Nuremberg Military Tribunals Under Control Council Law No. 10 323–435 (1952). The convictions against von Weizsaecker and Woermann for aggression, however, were later set aside by the Tribunal on motions of error.

[5] Henry T. King Jr., Nuremberg and Crimes against Peace, 41 Case W. Res. J. Int’l L. 273, 276 (2009).

[6] Rome Statute of the International Criminal Court art. 8 bis, July 17, 1998, 2187 U.N.T.S. 90, rev. 2010 [hereinafter Rome Statute].

[7] Note, however, that the Elements of Crimes for the crime of aggression clarify in a footnote that “[w]ith respect to an act of aggression, more than one person may be in a position that meets these criteria.” Amendments to the Rome Statute of the International Criminal Court, Annex II, 21 n.1, June 11, 2010, A-38544 U.N.T.S.

[8] Rome Statute, supra note 6, art. 25(3)(a).

[9] Id., art. 25(3)(b)–(d).

[10] See, e.g., Prosecutor v. Lubanga, Case No. ICC-01/04-01/06 A 5, Judgment on the Appeal of Mr. Thomas Lubanga Dyilo against his conviction, ¶¶ 458–69 (Dec. 1, 2014),

[11] See Stefan Barriga, Negotiating the Amendments on the Crime of Aggression, in The Travaux Préperatoires of the Crime of Aggression 3, 19–24 (Stefan Barriga & Claus Kreβ eds., 2012); Roger S. Clark, Individual Conduct, in The Crime of Aggression: A Commentary 565, 566 (Claus Kreβ & Stefan Barriga eds., 2017).

[12] Rome Statute, supra note 6, art. 25(3) bis.

[13] Id.

[14] See Kai Ambos, 2 Treatise on International Criminal Law: The Crimes and Sentencing 206–208 (2014); see also Clark, supra note 11, at 582 (“[O]ne suspects that most of the real-life cases can be fitted within article 25(3)(a).”).

[15] See Kevin Jon Heller, Retreat from Nuremberg: The Leadership Requirement in the Crime of Aggression, 18 Eur. J. Int’l L. 477, 480–88 (2007). On the other hand, there were concerns during the negotiations that the “shape or influence” formula “would open the doors too far, especially in relation to democracies where a very large circle of persons could be said to ‘shape or influence’ the State’s action.” See Barriga, supra note 11, at 22.