The Activation of the Crime of Aggression in Perspective

In its final hours soon after midnight of 14 December 2017, the 16^th Assembly of States Parties to the Rome Statute of the International Criminal Court decided to activate the Court‘s jurisdiction over the crime of aggression. This is the effect of operational paragraph 1 of resolution ICC-ASP/16/Res.5. But in the same breath, the Assembly in operative paragraph 2 confirmed “that in the case of a State referral or propio motu investigation 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.”

As is well known, whether or not the Court can exercise its jurisdiction over a crime of aggression committed by a national or on the territory of a State Party to the Rome Statute that has not ratified the crime of aggression amendments was subject to intense controversy and negotiations in the run-up to the activation decision. In fact, the Assembly recognized this in preambular paragraph 4 of the resolution, where it made approving reference to the report of the facilitation process led by the Austrian delegation summarizing the diverging legal views held by States Parties on this issue. (In the following, I assume some familiarity with the controversy between what could be called the adherents of the “restrictive” and “extensive” positions. For more explanations see the posts prior to the activation decision by Dapo Akande, Stefan Barriga and Astrid Reisinger Coracini).

So how did the Assembly arrive at operative paragraph 2? What is the Court to make of a resolution that, on the one hand, confirms one legal view while, on the other hand, notes with appreciation the summary of the diverging views of States Parties, and finally, in operative paragraph 3, reaffirms the independence of the judges of the Court? Dapo Akande, Kevin Jon Heller and Jennifer Trahan have already commented on this outcome. The following is an account from the viewpoint of the Swiss delegation witnessing and engaging in the negotiations.

On Wednesday 13 December, the Assembly had before it a revised discussion paper prepared by the Austrian facilitation (available here) that presented three options for an activation decision: firstly, a simple activation whereby the Assembly would not say anything on the Kampala compromise beyond recognizing that different views existed. Secondly, an endorsement of the restrictive group’s position confirming exemption from the Court’s exercise of jurisdiction for all non-ratifying States Parties, in effect exempting them just like non-States Parties to the Rome Statute. Thirdly, a middle ground whereby non-ratifying States Parties that put or have put their restrictive positions on record would be assured exemption, i.e. that the Court would not exercise its jurisdiction over crimes of aggression committed by their nationals or on their territory.

It is important to recognize that the third option went well beyond a “pragmatic opt-out” on the basis of paragraph 4 of article 15bis of the Rome Statute (as amended), an idea that had been introduced at an earlier stage. An opt-out declaration provides exemption for a crime of aggression “arising from an act of aggression committed by a State Party”, that is, only for the acts of the nationals of a non-ratifying State Party. In contrast, the facilitator’s third option foresaw exemption for jurisdiction based on nationality and territory. Similar to a “pragmatic opt-out”, however, States Parties that wanted to be out needed to have stated their position in one way or another. The compromise foresaw that Canada, Colombia, France, Japan, Norway and UK would have been “out” without further ado due to their position already being reflected in the report of the facilitation. Others could have resorted to an explanation of position at the time of adoption of the resolution. Overall, the third option significantly built upon a joint effort on the part of Brazil, New Zealand and Portugal to come up with a “third way” to achieve consensus, with the difference that the three delegations had proposed the compilation of an annex to the resolution to identify exempt States Parties.

From the vantage point of the extensive group arguing that the Court would by default have jurisdiction over the nationals of non-ratifying States Parties (which included Switzerland), this was an extremely forthcoming proposal. In contrast to the opt-out declaration foreseen in Kampala, States Parties wishing to exempt themselves could do so in an as painless form imaginable, by simply restating their legal convictions, respectful of the sovereign view of each State Party. Many delegations believed that this solution would go a long way in achieving consensus.

Confirming this, the discussion of the three options on Wednesday 13 December revealed a stark picture: a large majority of States Parties that took the floor (I counted 37 out of 46) spoke out in favor of option 1, dismissed option 2 as not constituting a compromise given the existence of contending legal views, and/or signaled readiness to explore option 3. Seven expressed their continued preference for option 2 or voiced some unease with option 3. Only two States Parties, the most prominent proponents of the restrictive group, openly rejected option 3.

On the basis of this exchange, the Austrian delegation leading the facilitation process prepared a draft resolution (ICC-ASP/16/L.9, available here) that sought to build on option 3. For example, it added time for States Parties to put their legal position on record. This version of the draft resolution formed the basis for final discussions on 14 December and represented, along with a revised version (available here), a concluding effort on the part of the very laudable Austrian facilitation to achieve consensus.

And yet, the draft resolution did not pass. Quite the contrary: at Thursday 5 PM, the same two most prominent proponents of the restrictive group boldly circulated their own draft resolution reintroducing option 2 practically unchanged, and it became evident that no consensus could be obtained on the draft resolution submitted by the facilitator. From 9 PM onwards, it was up to the Presidency of the Assembly, Vice-Presidents Sergio Ugalde (Costa Rica) and Sebastiano Cardi (Italy), to attempt a final deal.

To understand the remaining hours of the conference, it is important to recall two procedural realities. Firstly, the activation decision required a two thirds majority of all States Parties (the same majority for approving amendments to the Statute) on the basis of paragraph 3 of article 15bis of the Rome Statute, as agreed in Kampala. In other words, no less than 82 positive votes were necessary to achieve activation. This was a tall order, given that barely 100 delegations were represented in the room and given that since its establishment in 2002, the Assembly has never voted on a resolution. Presumably, only few delegations had instructions for a possible vote.

The second procedural reality was that postponement would have buried activation. Although the Assembly would have, technically speaking, been free to activate the Court’s jurisdiction at any subsequent session, given the realities of multilateral negotiations, the outcome would very likely have been eternal postponement. After all these years, the crime of aggression would have failed to overcome a last final hurdle.

As a result, any States Parties willing to take the issue to a vote were in a very strong position to hold the activation decision hostage to an agreement on their view of jurisdiction. This was widely understood by States Parties.

Facing this reality, the two Vice-Presidents, after a series of further consultations and additional commotion over elements of the resolution, at around 11 PM on Thursday submitted to the Assembly what became ICC-ASP/16/Res.5 as a final take-it-or-leave-it text, taking on board option 2 confirming the restrictive position rather than compromise option 3. But the resolution also contained a new operative paragraph 3 reaffirming the judicial independence of the judges of the Court.

This additional reference was responsible for almost an extra two hours of suspense before adoption of the resolution. Pushing the envelope, the two most prominent proponents of the restrictive group filed objections to it. However, in the absence of a consensus to delete or relocate the paragraph to the preambular part, the draft resolution was brought to action unchanged – and passed, to the great relief of delegations and thundering applause, at around Friday 0:40 AM.

How the activation decision was arrived at is relevant in at least two ways. For one thing, it is difficult to label the resolution as a case of subsequent agreement or practice under article 31(3) of the Vienna Convention of the Law of Treaties. Granted, the Assembly formally confirmed one legal position on the Court’s reach over the crime of aggression. And granted, accepting a disagreeable position in order to make gains elsewhere or in order to arrive at an overarching deal is the daily bread of multilateral diplomacy. But it is also very evident that this confirmation does not reflect the “actual” view of most, let alone all, States Parties to the Rome Statute, and was arrived at due to the procedural hurdles imposed on activation. Tellingly, a number States Parties chose to reiterate their legal views in their statements following adoption of the resolution. As a result, a paradox presents itself: both the formal resolution adopted by the Assembly and the “actual” opinio iuris of States Parties underlying the resolution would seem to matter.

Another important consideration is that the Assembly did not agree upon a particular legal basis for operative paragraph 2. During the course of negotiations, different legal avenues for the Assembly’s authority for option 3 and more broadly for any determination of the Assembly on jurisdiction outside the formal amendment procedures of Rome Statute were discussed. Following a proposal by Liechtenstein, the 13 December facilitator’s discussion paper based itself on paragraphs 3 of articles 15bis and 15ter of the Rome Statute, which arguably leaves the door open for a qualified, more than binary “on or off” activation decision by the Assembly. The adopted resolution makes passing reference to this in preambular paragraph 3. But in contrast to the discussion paper, the resolution adopted does not formally link its determination in operative paragraph 2 to the proposed legal basis or in fact any other discussed basis, including subsequent practice. The two most prominent proponents of the restrictive group themselves had cast doubt over their validity during negotiations, arguing for example that articles 15bis and 15ter could formally not be binding for States Parties that had not ratified the crime of aggression amendments and thus could not form the basis for a decision by the Assembly. This adds a layer of uncertainty over the legal effect of operative paragraph 2 of the resolution. The uncertainty only disappears if one assumes the paragraph simply reflects what Kampala implied all along, but, after all the controversy, how can that be readily assumed?

What remains is an operative paragraph 2 that, like the second sentence of paragraph 5 of article 121 of the Rome Statute which it seeks to leverage, stands in contradiction to paragraph 4 of article 15bis of the Rome Statute. Paragraph 4 of article 15bis literally asserts that the Court has jurisdiction over a crime of aggression “arising from an act of aggression 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.” In this sense, it is somewhat difficult to argue that operative paragraph 2 is simply a case of interpreting or clarifying the crime of aggression amendments. If the intended point of operative paragraph 2 is to revise paragraph 4 of article 15bis, the problem is that it was not passed pursuant to the Statute’s amendment provisions. One could argue that the Assembly did no more than clarify that the second sentence of paragraph 5 of article 121 fully applies, thus conditioning the application of article 15bis on a State Party’s ratification or acceptance, but this nonetheless would seem to imply a revision of the plain reading of article 15bis.

Be it as it may, with the Assembly having achieved activation, it is now up to the Court to determine the extent of its jurisdiction over acts of aggression committed by nationals or on the territory of non-ratifying States Parties. It is for this reason that the reference to the independence of the judges in operative paragraph 3 is so important. Pending a clear pronouncement by the Court, the only way for any State Party to legally ensure that its nationals fall outside the remit of the ICC’s jurisdiction continues to be to file an opt-out declaration in accordance with paragraph 4 of article 15bis of the Rome Statute.

In all of this, let us not forget that the activation of the crime of aggression is meant to be a contribution to the preservation of peace and the prevention of the most serious crimes of concern to the international community as a whole. More than 70 years after the Nuremberg and Tokyo trials, the ICC has received the historic opportunity to strengthen the prohibition of the use of force as enshrined in the UN Charter and completed the Rome Statute as originally drafted. This is the perspective we should preserve.