In two recent Just Security posts (see here and here), Luis Moreno Ocampo, the first Prosecutor of the International Criminal Court (ICC or the Court), argues that proposals to create a Special Tribunal for the Crime of Aggression (STCoA) “[p]romote nothing short of selective justice.” In order to avoid selectivity, Ocampo suggests amending Art 15bis(5) of the ICC’s Rome Statute to delete 5 words “by that State’s nationals or” – a supposedly “simple revision” that could occur within “a few months” and which would enable the ICC to exercise its jurisdiction over the crime of aggression committed against Ukraine.
Let me at the outset share my conviction that if amending the Rome Statute were an available option, many of those scholars arguing for the establishment of a STCoA would do nothing short of embrace it. However, the reason why many supporters of a STCoA argue for a two-tiered approach (in order to channel the political momentum for accountability for Ukraine and counterweigh the perceived situational selectivity of a STCoA) is their understanding that a quick fix of the Court’s jurisdictional regime is unrealistic.
1. A Preliminary Word on “Selectivity”
The history of the enforcement of international criminal law is tainted with selectivity, partly in the form of victors’ justice, partly in the form of international(ized) enforcement mechanisms created for some situations but not for others. The quest for a permanent and universal international justice mechanism that led to the creation of the ICC in 1998 was guided by the aim of justice and “no impunity” for all victims of crimes under international law. The ICC is a permanent institution with an aspiration of universality that would end the need for ad hoc or post hoc enforcement mechanisms. It has a clear mandate that puts alleged perpetrators on alert to the risk of prosecution from the outset of a conflict. Since its establishment, the Court has rightly become, as former ICC Prosecutor Fatou Bensouda has called it, the center of “a system of global criminal justice.” This central position must be unwaveringly maintained and strengthened.
But the Rome Statute itself is not free from selectivity. The ICC is based on an international treaty, and the Court’s exercise of jurisdiction is linked (and limited) to the territory and nationals of States parties (Art 12(2) ICC Statute); or States accepting the Court’s jurisdiction (Art 12(3) ICC Statute). It is therefore unavoidable that unless the ICC Statuteis universally ratified, the Court’s exercise of jurisdiction remains selective. Article 13(b) of the ICC Statute allows for the referral of situations concerning non-States parties by the United Nations Security Council, an organ that is inherently political and selective. The exercise of prosecutorial discretion in the selection or (de)prioritization of situations and cases affects the perceived legitimacy of the Court (for a longer discussion see this article by Birju Kotecha). This is not an allegation of selective justice at the ICC but a reminder that its very design includes elements of selectivity. Enabling the ICC to exercise its jurisdiction over the crime of aggression in the situation of Ukraine would help to reduce the selective enforcement of international criminal law by the ICC, but it would not end it. More specifically, deleting 5 words in Art 15bis(5) of the ICC Statute would not even end the Court’s selective exercise of jurisdiction over the crime of aggression, but only reduce it (see point 4 below).
2. Establishing Agreement on Amending the Rome Statute Takes Time
The crime of aggression falls within the jurisdiction of the ICC since the adoption of the Rome Statute in 1998 (Art 5(1)(d) ICC Statute). At the time, the crime was included on the basis of a compromise. Although it was listed and recognized as one of “the most serious crimes of concern to the international community as a whole” (Art 5(1) ICC Statute), the Court was unable to exercise its jurisdiction over the crime of aggression until States adopted a provision “defining the crime and setting out the conditions under which the Court shall exercise jurisdiction” over it (Art 5(2) of the 1998 ICC Statute). After twelve years of negotiations, a provision was eventually approved at Kampala, Uganda in 2010, at the first Review Conference of the Rome Statute (RC/Res.6 (the Kampala amendments)). Yet, the Court’s exercise of jurisdiction over the crime of aggression was further delayed “subject to a decision to be taken after 1 January 2017” (Arts 15bis(3), 15ter(3) ICC Statute). After continued intense negotiations, the Assembly of States Parties of the Rome Statute (ASP) decided in December 2017, “to activate the Court’s jurisdiction over the crime of aggression as of 17 July 2018” (ICC-ASP/16/Res.5 (2017 activation decision), para. 1).
In short, the two-decade timeline offers a glimpse into the legal and political complexities involved in these negotiations. Opening this compound of compromises in order to amend the conditions of the Court’s exercise of jurisdiction over the crime of aggression will not prove less challenging. Even if a swift agreement on amending the conditions of the Court’s exercise of jurisdiction over the crime of aggression were possible, it is doubtful that any amendment would enter into force immediately upon adoption. It would arguably be subject to ratification by States (see below 5), which again might involve a lengthy process. It is therefore unrealistic that the ICC Statute could be successfully amended within a few months to offer accountability for the crime of aggression committed against Ukraine. Yet, it is paramount to start this conversation promptly. Not only to enable the Court as soon as possible to enforce the crime of aggression in comparable future situations, but also to start preparations for the mandatory review of “the amendments on the crime of aggression seven years after the beginning of the Court’s exercise of jurisdiction” in 2025 (RC/Res.6).
3. The ICC’s Exercise of Jurisdiction Over the Crime of Aggression
The conditions for the exercise of jurisdiction over the crime of aggression are established in Arts 15bis and 15ter of the ICC Statute. Art 15ter ICC Statute deals with the referral of a situation by the Security Council and corresponds to the ICC’s jurisdictional regime over genocide, crimes against humanity, and war crimes. Art 15bis ICC Statute on the other hand, dealing with the referral of a situation by a State party or a proprio motu investigation by the Prosecutor, deviates significantly from the Statute’s general jurisdictional regime. Recognizing the Security Council’s primary role for international peace and security in accordance with the United Nations Charter, Art 15bis paras. (6)-(9) introduce the additional requirement of authorization of the initiation of an investigation of the crime of aggression, either by the Security Council or, failing that, by the Court’s Pre-Trial Division. At the heart of the Kampala compromise, Art 15bisparas. 4 and 5 ICC Statute limit the Court’s exercise of jurisdiction over the crime of aggression:
The Court may, in accordance with Art 12, exercise jurisdiction over a crime of aggression, arising from an act of aggression committed by a State Party, unless that State Party has previously declared that it does not accept such jurisdiction by lodging a declaration with the Registrar. The withdrawal of such a declaration may be effected at any time and shall be considered by the State Party within three years. (Para. 4).
In respect of a State that is not a party to this Statute, the Court shall not exercise its jurisdiction over the crime of aggression when committed by that State’s nationals or on its territory. (Para. 5
Para. 4 concerns the Court’s exercise of jurisdiction over a crime of aggression arising from an act of aggression committed by a State party. It is based on the understanding that the Statute’s general jurisdictional regime applies, in principle, to the crime of aggression. This understanding recognizes that (a) the crime of aggression has been listed as a crime within the jurisdiction of the Court since 1998 (Art 5 ICC Statute); that (b) in accordance with Art 12(1) ICC Statute “[a] State which becomes a Party to this Statute thereby accepts the jurisdiction of the Court with respect to the crimes referred to in article 5”; and that (c) Art 5(2) ICC Statute provided States with broad discretion in defining the conditions for the Court’s exercise of jurisdiction over the crime of aggression. The Statute’s general jurisdictional regime applies to the crime of aggression in principle, because para (4) provides States parties with the possibility to opt out. In the aftermath of Kampala, a few States contested the compromise enshrined in Art 15bis(4) ICC Statute. The 2017 activation decision ended this discussion on a political level in favour of the minority view, but could not unravel the underlying legal question. The exact contents of Art 15bis(4) ICC Statute thus awaits authoritative judgment by the judges of the ICC (see this article by Dapo Akande and Antonios Tzanakopoulos and this article by Jennifer Trahan).
Para (5) concerns the reach of the Court’s jurisdiction over non-State parties. It discards the Court’s exercise of jurisdiction, which is based alternatively on the principle of territoriality, or the principle of active nationality (Art 12ICC Statute), whenever a non-State party is involved.
4. Deleting 5 Words Would Not End “Selective Justice”
In light of the jurisdictional regime in place, Ocampo’s proposal to delete 5 words in Art 15bis(5) of the ICC Statutecannot end “selective justice for the international crime of aggression.” With a focus on closing the accountability gap in the situation of Ukraine, the proposal deals exclusively with acts of aggression committed by a non-State party. It does not address exceptions to the ICC’s exercise of jurisdiction over acts of aggression committed by a State party.
Furthermore, any possible effect of deleting these 5 words would be limited to acts of aggression by a non-State party against a State party that has accepted the Kampala amendments. The latter point is relevant because, in accordance with Art 12(2) ICC Statute, only an accepting State party may provide the ICC with the necessary jurisdictional link under Arts 15bis(4) and 12 ICC Statute. The conflict between Azerbaijan and Armenia, as well as a U.S. attack in Iraq that killed a senior Iranian military leader would therefore remain beyond the Court’s reach; none of these States are parties to the ICC Statute. Likewise, Ukraine, despite having approved the required constitutional changes to allow ratification in 2016 (with effect from 30 June 2019), has not yet become a State party.
It is also not abundantly clear whether deleting the proposed 5 words in Art 15bis(5) would permit the exercise of jurisdiction over a crime of aggression based on an act of aggression by a non-State party against an accepting State party. On a first glance, it may appear that Art 15bis(4) and (5) are clearly separate, with 15bis(4) regulating the conditions for the exercise of jurisdiction in relation to State parties and Art 15bis(5) addressing non-State parties. But the phrase “[t]he Court may, in accordance with Art 12, exercise jurisdiction over a crime of aggression, arising from an act of aggression committed by a State Party, unless” may – argumentum a contrario – also be interpreted as categorically excluding the exercise of jurisdiction over any act of aggression committed by a non-State party. In that case, submitting a non-State party aggressor under the general jurisdictional regime of Art 12(2) may require amendment of Art 15bis paras (4) and (5) (for instance see the Parliamentarians for Global Action’s Proposal to Amend Article 15bis of the Rome Statute on the Crime of Aggression; full proposal on file with the author) or, at least a clear guideline for interpretation.
Finally, the question arises, whether leaving a reference to the territory of non-State parties in Art 15bis(5) would not undermine the desired effect. The crime of aggression is committed in both the aggressor State as well as the aggressed State. Would there be a jurisdictional link under Art 12(2) of the ICC Statute for acts committed by nationals of a non-State party on the territory of a non-State party (preparation, planning, and initiation)? Would the jurisdictional link provided for acts committed on the territory of the aggressed State (initiation and waging) suffice to allow for the exercise of jurisdiction over preparatory acts?
All these questions can be resolved, but they certainly require thorough discussion among States before adopting an amendment to the aggression provision. It is doubtful all this diplomatic activity could take place within a few months.
5. Adopting an Aggression Amendment Would Raise Broader Political and Legal Questions
In addition to the questions raised above, which are linked to the specific proposal Ocampo suggests, preparing the amendment of the Court’s conditions for the exercise of jurisdiction will undoubtedly open a broader debate.
If an amendment were proposed to end the Court’s selective exercise of jurisdiction over the crime of aggression, the question would arise as to how much selectivity should be repealed. Should the amendment overcome the non-State party exception in Art 15bis(5) of the ICC Statute (and arguably in Art 15bis(4))? Should the non-State party exception be aligned with the status of State parties that have not accepted the aggression amendments (assumption of jurisdiction in accordance with Art 12(2) and the possibility to opt-out as foreseen under Art 15bis(4); see e.g., Proposal to Amend Article 15bis of the Rome Statute on the Crime of Aggression)? If so, this would most probably re-open the debate around the contents of Art 15bis(4). It would also re-open the debate on whether Art 12(3) is applicable to the crime of aggression. Or should the amendment overcome all exceptions from the Court’s jurisdiction, those of non-State parties as well as those State parties that choose to opt-out, and align the Court’s jurisdiction over the crime of aggression with the Court’s jurisdiction over the other three core crimes? Finally, conscious of allegations of selectivity, should an amendment also erase the root cause of future selective jurisdiction over “amended most serious crimes” in Art 121(5) of the ICC Statute?
Procedurally, the perpetual question of the proper amendment regime would need to be resolved. Would mere adoption suffice? Would an amendment of the conditions for the Court’s exercise of jurisdiction over the crime of aggression require entry into force in accordance with Art 121(4) or Art 121(5) of the ICC Statute? Or, would amending an amendment constitute an “inter se regime” that is not explicitly regulated in the ICC Statute (see this article by Fiona Abken and Paulina Rob).
In accordance with Art 121(3) of the ICC Statute, an amendment requires adoption by consensus or a two-thirds majority. True, non-State parties will not be able to vote against the adoption of an amendment at the Assembly of States Parties. However, it would be naïve to think that their positions would not be properly endorsed by close allies or States with political or economic interests. Finally, while I am convinced that impunity for perpetrators of crimes under international law is not a right of their state of nationality, the implications of an amendment with regard to the rights of third States must be addressed.
These issues are resolvable, however, believing these legal questions, which are prone to be discussed alongside or as a pretext for sensitive political questions, could be decided in a quick debate until or during the upcoming session of the Assembly of States Parties is unrealistic.
6. An Aggression Amendment Would Take Time to Enter Into Force
It appears that Ocampo considers the adoption of an amendment by a two-thirds majority of the assembly (a conference) of States parties sufficient for its application. But this approach would be inconsistent with the amendment mechanisms foreseen in the ICC Statute, which require ratification or acceptance of an amendment for its entry into force; a process that may equally take time.
Except for exclusively institutional amendments (Art 122 ICC Statute), the Statute distinguishes two entry into force mechanisms. Amendments to Arts 5, 6, 7, 8, and 8bis of the Statute enter into force “for those States Parties which have accepted the amendment one year after the deposit of their instruments of ratification or acceptance” (Art 121(5) ICC Statute). All other amendments enter into force “for all States Parties one year after instruments of ratification or acceptance have been deposited with the Secretary-General of the United Nations by seven-eighths of them.”
As mentioned above, it is not entirely clear which amendment regime would apply to an amendment of the conditions for the exercise of jurisdiction over the crime of aggression. Since Art 15bis is explicitly not listed in Arts 121(5), it seems that Art 121(4) would necessarily apply. From a temporal point of view, this would substantively delay the practical effect of the amendment, as it would enter into force only if accepted by seven-eighths of States parties (that would be 108 of currently 123 States parties). Yet, it would profit from the highest possible support of States parties, and it would be applicable to all States parties (unless they use the option of withdrawing from the Statute with immediate effect, Art 121(6)). In order to avoid this delay, the Proposal to Amend Article 15bis of the Rome Statute on the Crime of Aggression envisions a decision of the Assembly of States Parties in accordance with Art 58(1)(b) of the Vienna Convention of the Law of Treaties (VCLT) suspending the provisions of Art 121(4) for the amendment of Art 15bis of the ICC Statute. Alternatively, a proposal identifying the procedure to amend an amendment as a gap in the Statute, argues for the application of an entry into force quorum of States corresponding to Art 121(4) within the group of those States parties that have ratified or accepted the aggression amendments (see Abken and Rob; that would be 39 of currently 44 States parties).
A literal interpretation of Art 121(5) thus excludes its application to amendments of Art. 15bis. On the other hand, Art 15bis was introduced on the basis of Art 121(5) and it might be argued that any future amendment to this provision must follow the same procedure. This may be the case, although the ratio for this decision is obsolete. Art 15bis was introduced together with the definition of aggression in Art 8bis. At the time, it was convincingly argued that the substantive and procedural provisions on the crime of aggression were intrinsically linked and therefore required the application of one uniform amendment regime. Although a new amendment only concerns procedural provisions, it has also been argued that “the telos of the holistic approach” chosen in Kampala could be transferred to the revision of the jurisdictional regime (see Abken and Rob).
Art 121(5) foresees a subjective entry into force for each ratifying/accepting State party, which raises a number of questions. Would one ratification suffice to have an objective effect vis-à-vis a non-State party or a non-accepting State party? With regard to the situation in Ukraine, must this one ratification be by (a future State-party) Ukraine? Or would the ratification of Ukraine, be required in any case, independent of the number of ratifications? The Kampala amendments introduced a minimum requirement of 30 ratifications. However, this provision did not alter the entry into force regime of the first sentence of Art 121(5). It served as a precondition for the exercise of jurisdiction, which meanwhile has been fulfilled.
Currently, two versions of the ICC Statute are in force with a view to the crime of aggression. First, the original ICC Statute among States parties who have not accepted the Kampala amendments and in the relation between States parties that have accepted the Kampala amendments and those that have not (see Art 40(3) VCLT on amendments of multilateral treaties). Second, a version of the (amended) Rome Statute among States parties that have accepted the amendments. If 121(5) applied, three versions of the ICC Statute would then exist in parallel until all States parties that have ratified the Kampala amendments ratify the new amendment. In order to avoid the pitfall of creating further sub-regimes, Abken and Rob argue for the suspension of the individual entry into force of the amendment for States parties that have accepted the Kampala amendments until all “Kampala states” will have ratified the new amendment. Legal precautions would also need to be taken to assure that a State party that has not yet accepted the Kampala amendments or a new State party would not be able to ratify the Kampala amendments without the new aggression amendment.
Amending the conditions for the ICC’s exercise of jurisdiction over the crime of aggression includes complex legal and sensitive political questions that will require a thorough discussion among all States parties. Although desirable, this will not be achievable within a few months or within one session of the Assembly of States Parties. Still, it is paramount to start this conversation, particularly with a view to the mandatory review of the provision of aggression in 2025. Undoubtedly, the amendment process would benefit tremendously from the participation of Ukraine as the latest State party to the (amended) ICC Statute (see also here).
In the meantime, accountability for the crime of aggression against Ukraine should not be delayed. An ad hoc or post hoc enforcement mechanisms, such as a STCoA, could fill the enforcement gap of the ICC Statute and thereby reduce the selective enforcement of international criminal law in the broader sense. The answer to past inaction in the face of aggression cannot be continued inaction. It must be judicial relief for the ongoing aggression and a concerted effort to limit or possibly even eliminate elements of selectivity in the enforcement of the law against future aggression.