Opt-in vs. Opt-out = opt-in-opt-out?
On the activation of the ICC’s jurisdiction over the crime of aggression
In December 2017, the States Parties to the International Criminal Court (ICC) had the chance to realize in New York what Robert H. Jackson called for in his opening statement before the International Military Tribunal in Nuremberg. While international criminal law was only applied against German aggressors at that time, the U.S. Chief Prosecutor emphasized that the condemnation of aggressive war should be the benchmark for any other nation in the future. When states laid down the foundations of the ICC in 1998, they listed the crime of aggression as one of the crimes under the Court’s jurisdiction. Although States Parties agreed on a definition of the crime of aggression and the conditions for the Court’s exercise of jurisdiction at the 2010 Kampala Conference, it took another seven years before they activated the Court’s jurisdiction. This article sheds light on the most controversial issue prior to activation and explains the content as well as the effects of the resolution as adopted by the Assembly of States Parties (ASP).
The remaining controversy with regard to non-ratifying States Parties
States were divided on the following, very technical question: In case of state referrals or proprio motu investigations, does the ICC have jurisdiction over nationals of States Parties that have not ratified or accepted the aggression amendments and that did not make an opt-out declaration? Intuitively, international lawyers would say “no”. Treaty amendments cannot bind a state without its consent. But this does not answer the question whether active consent (by accepting/ratifying the amendments) or passive consent (by not making an opt-out declaration) is required. Several states, the so-called opt-in camp, argued that active consent by the aggressor and the victim state was necessary. Why? According to the plain meaning of the second sentence of Article 121(5) ICC Statute, the ICC shall not exercise jurisdiction regarding a crime covered by an amendment unless the state of nationality and the state of territoriality have opted in by accepting/ratifying the amendments. This sentence seems to establish an opt-in system in case a new crime is added to the Court’s jurisdiction. The opt-out camp tried to rebut these allegations. Whereas any new crime (e.g. terrorism) needs to fulfill all requirements of Article 121(5), this cannot hold true for the crime of aggression which is already listed in Article 5(1) as a crime under the Court’s jurisdiction. In addition, the application of Article 121(5) second sentence would conflict with the genesis of the Kampala compromise: To bridge the gap between those that did not require the consent of the aggressor state (Article 12(2)(a)) and others that emphasized its importance, delegates established an opt-out system (Article 15bis(4)). Accordingly, the Court can exercise jurisdiction over a crime of aggression committed by nationals of a State Party, unless the aggressor state has previously revoked its consent by opting out. Thus, States agreed on a consent-based regime, but they decided that passive consent sufficiently respects the interests of aggressor states. The opt-out regime would lose its character as a middle ground if it was combined with the opt-in clause of Article 121(5). States could shield their aggressive leaders twice, by not ratifying the amendments and by opting out after ratification.
Establishment of an opt-in-opt-out system but reaffirmation of judicial independence
In the end, it was not the power of legal arguments but the risk of an indefinite postponement of the activation decision that led to the adoption of the resolution, which is mainly based on the opt-in regime. Paragraph 2 reflects almost verbatim Article 121(5) second sentence by stating that the Court shall not exercise its jurisdiction regarding a crime of aggression when committed by a national or on the territory of a State Party that has not ratified or accepted these amendments. If one perceives the resolution as subsequent agreement or practice (Article 31(3) Vienna Convention on the Law of Treaties), the combined reading with Article 15bis(4) leads to an opt-in-opt-out system. Good news for all states that change their mind after ratification. There would be an emergency exit in case they decide to attack other countries after ratification. Bad news for the 35 states that expected judicial protection after ratification. Except for Security Council referrals, their aggression case against non-ratifying States Parties could not be brought before the ICC. End of story? Well, the resolution still leaves some leeway by reaffirming judicial independence (paragraph 3). In case the resolution remains ambiguous as to the scope of jurisdiction, it is up to the Court to decide, see Article 119(1). Indeed, it is quite contradictory to establish an opt-in regime while “noting with appreciation the Report on the facilitation” which summarizes the views of States Parties, including those supporting the opt-out camp. Nonetheless, it will be hard to justify that the pure opt-out system of Article 15bis(4) still applies despite the clear wording of paragraph 2 of the resolution. On the other hand, there is a slim chance that subsequent state conduct might reveal a persisting controversy that needs to be settled by the Court. The subtle objections States Parties expressed after adoption might serve as a first indication. States could further strengthen the impression of disagreement by declaring a temporary opt-out until they ratify the amendments. This seemingly schizophrenic conduct would help to reaffirm the pure opt-out regime as the prevailing one despite the formal adoption of the opt-in clause. Their opt-out declaration would suggest that the default rule is acceptance (Article 15(4)). If states fail to engage in common conduct, paragraph 2 of the resolution cannot be seen as establishing subsequent state practice. Be that as it may, time will tell whether the paragraph on judicial independence will ever become a powerful tool for the Court. At the ASP, however, its inclusion was rather a means to placate the opt-out camp and to reach consensus.
The potential impact of the resolution
In case paragraph 2 is perceived as a subsequent agreement, the ICC’s jurisdiction will be further restricted in case of state referrals and proprio motu investigations, namely, both the aggressor and the victim state need to ratify before the ICC can exercise jurisdiction. If the number of ratifications does not increase, these two trigger mechanisms will only play a minor role in combatting aggression. Instead, referrals by the Security Council will gain in importance. Contrary to States Parties and the Prosecutor, the Security Council can trigger the Court’s jurisdiction even if Non-States Parties or non-ratifying States Parties are involved (Article 15ter). The adopted opt-in clause does not only reduce the number of potential aggression cases from conflicts between 122 States Parties (Kenya opted out) to 35 ratifying States Parties. Indirectly, it also reduces the Court’s autonomy from the Security Council as its exercise of jurisdiction over non-ratifying States Parties would fully depend on a Security Council referral. Ironically, the permanent members already called for the Council’s exclusivity to trigger the Court’s jurisdiction in Kampala. Seven years later, States Parties did not establish exclusivity, but they adopted a resolution that potentially produces the same effect if the number of ratifications does not increase. The condemnation of aggressive war will be a benchmark for any other non-ratifying nation, but only if the Security Council so decides.
Annegret L. Hartig works as a research assistant at the University of Hamburg and writes her PhD thesis on the domestic implementations of the crime of aggression. At the 16th session of the Assembly of States Parties, she had the chance to follow the negotiations on the activation of the International Criminal Court’s jurisdiction over the crime of aggression.
Cite as: Annegret L. Hartig, “Opt-in vs. Opt-out = opt-in-opt-out?”, Völkerrechtsblog, 7 February 2018, doi: 10.17176/20180207-093321.