The criminal character of aggression, long endured in world history but so rarely prosecuted in any court of law, no longer is assigned only to the legacy of the Nuremberg and Tokyo International Military Tribunals. Although present in the Rome Statute of the International Criminal Court (ICC) from that treaty's inception in July 1998, the crime of aggression was stillborn. There could be no prosecution of aggression before the new court until the crime was activated with a definition and a jurisdictional procedure by which investigations and cases could be initiated. Years of international negotiations led to the Kampala Review Conference from 31 May to 11 June 2010, where the primary set of amendments to the Rome Statute, adopted by consensus, activated the crime of aggression.Footnote ¹ Prior to Kampala, aggression was a theoretical possibility awaiting an operational ignition by the states parties to the Rome Statute. After Kampala, the crime of aggression is on a slow burn until 2017, when it will either fizzle or boldly light a pathway towards accountability.

In this essay I want to examine four discrete issues that will bear upon how the ICC ultimately addresses the crime of aggression. They are emblematic of the complexities that await the ICC, states parties, the Security Council, and non-party states as the crime of aggression takes hold under the Rome Statute.

1. The magnitude test for aggression

The new Article 8 bis (Crime of aggression) distinguishes between a ‘crime of aggression’ in Article 8 bis(1) and an ‘act of aggression’ in Article 8 bis(2). While there is a gravity, or magnitude, threshold of undefined character that must be factored into any determination of the crime of aggression, no such calculation is strictly required for identifying an act of aggression. The most generous reading of Article 8 bis is to regard the crime of aggression as the major transgression that must fall within a gravity context, and the act of aggression as simply identifying the component parts of the crime regardless of the gravity of any particular act of aggression. Under that reading, the gravity test must be met for the big event – the crime of aggression – but not for the smaller sub-events (acts of aggression) that constitute the actions under investigation.

Such an interpretation, however, ignores the reality of how matters are referred to the ICC as well as how one first determines the existence of aggression. Matters, or situations, referred to the ICC by states parties or the UN Security Council are not classified as crimes per se. Alleged genocide, crimes against humanity, war crimes, and now aggression (together, atrocity crimes) are referred as situations that merit investigation for the presence of specific atrocity crimes and of individuals who can be charged with committing such crimes. Therefore, in the beginning, either of these two referring entities do not determine the criminality of any particular act, but reach a decision to refer an overall situation where atrocities appear to be committed and the ICC is being asked to investigate the atrocities for criminal conduct by individuals (not states).

Under Article 13(c), the Prosecutor may seek to initiate the investigation of an atrocity crime, but in practical terms this occurs in the context of a situation of atrocities that the Prosecutor seeks to investigate. The most recent initiation of an investigation, into the Kenyan electoral violence, approved by the Pre-Trial Chamber, strengthens that interpretation of the Article 13(c) authority.Footnote ²

A more logical reading of Article 8 bis would have the Article 13 referring body (Security Council, state party, or, in practical terms, the Prosecutor) refer a situation of alleged aggression to the ICC for investigation, first, as to whether the situation encompasses any of the acts of aggression defined in Article 8 bis(2) and then, if such a determination is positive, whether any such acts evidence the crime of aggression by an individual pursuant to Article 8 bis(1). If the Security Council refers an act of aggression, or better yet a situation of aggression, to the ICC pursuant to Article 13(b) and subject to the conditions of Article 15 ter, there is no magnitude test to be met. Whatever the Security Council refers automatically qualifies for consideration by the Prosecutor and judges, who need only determine whether the Council's referral satisfies Article 8 bis(2) as a particular form of aggression described thereunder and thus eligible for investigation and prosecution by the ICC.

If the Security Council's referral is so categorized by the ICC as an Article 8 bis(2) act of aggression, the next step would be discovering a crime of aggression relating thereto pursuant to Article 8 bis(1) and, of course, an individual suspect responsible for such criminal behaviour. Interestingly, is it possible to satisfy the gravity requirement for a crime of aggression as it would apply to one suspect under Article 8 bis(1) and yet afford no consideration to a gravity requirement under the larger situation of an act of aggression under Article 8 bis(2)? In my view, the answer, unfortunately, is ‘yes’, because Article 8 bis(2) has no magnitude standard.

This dilemma presents an awkward framework for analysis, as one would logically seek a gravity standard for the mega-event of an act of aggression and then supplement that evaluation with an additional gravity requirement for the suspect's participation in the crime of aggression (falling within the larger framework of an act of aggression). That task probably will be easier if the Security Council refers a situation of aggression under Article 13(b), because it presumably would not trouble itself with such a referral unless it met a common-sense gravity standard as a threat to international peace and security.

However, if the referral arises under either Article 13(a) (state party) or for all practical purposes under Article 13(c) (Prosecutor) and subject to Article 15 bis, the matter could become more precarious. The ICC could ignore magnitude issues in evaluating the referral of the situation of aggression and focus strictly on Article 8 bis(2) events that constitute acts of aggression. Having cleared that hurdle, the dilemma arises under Article 8 bis(1) of what constitutes a crime of aggression of sufficient gravity associated with an act of aggression. The referring state party or the Prosecutor may not be as careful or even as prudent as the Security Council would be when the latter uses its Article 13(b) referral power. One might see a state party or the Prosecutor refer a relatively minor cross-border incursion that technically satisfies Article 8 bis(2) for an act of aggression but leaves to a final judicial determination the more complicated issue of identifying a crime of aggression of sufficient gravity.

One can even imagine the ICC's time being wasted with a frivolous referral under Article 13(a) of a relatively de minimis act of aggression that satisfies a state party's narrow political agenda against another state but with no rigorous forethought of whether a crime of aggression, one that would attract the ICC's jurisdiction over any particular individual, has been committed. In other words, a state party could use the ICC essentially for state responsibility allegations without any objective of finding anyone criminally responsible. The manipulation of the ICC for such an aim, namely to shame or politically corner an opposing nation-state, may be possible if the initial state party referral is of a situation of aggression that arguably constitutes an act of aggression under Article 8 bis(2) but may have no prospect of clearing the ‘character, gravity, and scale’ requirements for a ‘manifest violation of the Charter of the United Nations’ under Article 8 bis(1). And the referring state party may be perfectly aware of that probability but still wish to achieve its aim of libelling another country with state responsibility for aggression.

Prosecutorial and judicial discretion will have to suffice as the guardians at the gate of the International Criminal Court for any such politically motivated referrals which are devoid of expectations that the requirements for a crime of aggression can be met. If the Prosecutor were to seek to manipulate the jurisdictional hurdles in similar manner, if only to make some political point against a government, the Pre-Trial Chamber judges presumably would deny the matter for investigation swiftly and with a sharp rebuke.

Surely in good faith, the US delegation to the Kampala talks pushed through an ‘Understanding’ that seeks to inject a gravity standard into determinations of acts of aggression. Understanding No. 7 reads,

However, the wording of Understanding No. 7 is peculiar, as nowhere in Article 8 bis(2) is there any reference to a ‘manifest violation’ or the need for a ‘manifest’ determination. The ‘manifest violation’ language resides exclusively in the definition of a crime of aggression under Article 8 bis(1), where there already exists the triple-hitter standard of ‘character, gravity and scale’ (emphasis added) for manifest violations of the Charter of the United Nations. Understanding No. 7 errs in that there is no comparable requirement under Article 8 bis(2) for there to be any determination of a ‘manifest violation of the Charter of the United Nations’.

Further, a strictly grammatical reading of Understanding No. 7 points to the need to establish only two of the three magnitude requirements in Article 8 bis(1) for a crime of aggression in order to achieve the supposed magnitude standard for an act of aggression. (‘No one component can be significant enough to satisfy the manifest standard by itself.’) In any event, I question whether judges will create any such magnitude standard for determining acts of aggression based on the erroneous formulation of Understanding No. 7 and the simple fact that the Rome Statute, as amended, requires no such determination for acts of aggression that doubtless will frame the initial assessment by the Security Council, states parties, or the Prosecutor for any Article 13 referral to the ICC.

Finally, one should recognize that the distinction between an act of aggression and a crime of aggression arises in Article15 bis, and not always in a logical manner. For example, Article 15 bis(4) describes a crime of aggression ‘arising from an act of aggression committed by a state party’, thus identifying the crime as a subset of the act of aggression. That would suggest the referral of acts of aggression from which the crime of aggression then is investigated, just as situations of other atrocity crimes lead, through Article 13 referrals, to investigations of crimes committed within those situations. A similar rationale applies to Article 15 bis(6), where the determination of an act of aggression precedes the investigation in respect of a crime of aggression. Article 15 bis(7) speaks to a prior determination by the Security Council of an act of aggression before the Prosecutor may proceed with the investigation of a crime of aggression.

In contrast, Article 15 bis(8) permits the Prosecutor to investigate the crime of aggression if the Security Council does not render a determination about acts of aggression. The Prosecutor thus investigates criminal conduct before any determination about acts of aggression. The Pre-Trial Chamber need only first authorize the commencement of the Prosecutor's investigation of a crime of aggression. Whither acts of aggression when the Security Council is silent? The Pre-Trial Chamber should be required to determine that acts of aggression, ideally of some identifiable magnitude, have occurred before authorizing the Prosecutor to investigate possible crimes of aggression arising from such acts of aggression.

2. The nature of Security Council action on aggression

The new Article 15 bis (‘Exercise of jurisdiction over the crime of aggression (State referral, proprio motu)’) omits one significant action by the Security Council that could create a conundrum for states parties, the Prosecutor, and the judges. Article 15 bis(6) requires the Prosecutor first to ascertain, when he or she thinks there is a reasonable basis to proceed with an investigation in respect of a crime of aggression, ‘whether the Security Council has made a determination of an act of aggression committed by the State concerned’. Article 15 bis(7) continues: ‘Where the Security Council has made such a determination, the Prosecutor may proceed with the investigation in respect of a crime of aggression.’ Article 15 bis(8) states that ‘Where no such determination is made within six months after the date of notification, the Prosecutor may proceed with the investigation’.

The question arises, what does ‘determination’ mean with respect to the Security Council? The purpose of Article 15 bis in part is to provide a means to initiate an investigation if the Security Council remains silent on the issue. But the Security Council could make a determination, whether or not it refers to the ICC in the relevant resolution, that no act of aggression has occurred. This is entirely plausible if the Security Council wants to provide guidance to the ICC not to tread any further into the allegations of aggression, but to accomplish this short of a Chapter VII resolution that satisfies the requirements of Article 16 of the Rome Statute. Or the Security Council may wish to make such a determination for strictly political reasons within the context of managing a threat to international peace and security. If there were such a negative determination by the Security Council, can the ICC nonetheless proceed given the lack of a positive determination by the Security Council on an act of aggression? Nothing in Article 15 bis explicitly prohibits the ICC from forging ahead even if the Security Council renders a negative determination. As noted, the only backstop provided for in Article 15 bis(8) is if the Security Council renders a Chapter VII enforcement resolution consistent with Article 16 of the Rome Statute deferring the ICC's investigation and prosecution of the specific matter for at least 12 months. The Security Council members could believe that in fact aggression is occurring between two states but still want the ICC to back down for at least 12 months in order to give negotiators a (perhaps) better chance to stop the fighting.

Article 15 bis(7), as agreed to in Kampala, only makes sense if the meaning of ‘determination’ as used therein means a positive determination on acts of aggression by the Security Council. I had hoped in Kampala that the negotiators would draft Article 15 bis(7) to reflect the reality that the Security Council may (i) make a positive determination on aggression, thus permitting the Prosecutor to proceed with the investigation in respect of a crime of aggression, or (ii) make a negative determination on aggression, thus prohibiting the Prosecutor from proceeding with the investigation in respect of a crime of aggression. The further prospect of a Security Council resolution under Chapter VII consistent with Article 16 of the Rome Statute, as provided in Article 15 bis(8), would arise where the Security Council remains silent on the issue for at least six months. Such silence was the touchstone of negotiators’ speculation for years in the working group discussions and during the negotiations leading to the Rome Statute. Of course, even if the Security Council renders a positive or negative determination on aggression, it still could exercise its Article 16 authority to defer the ICC's investigation and prosecution for one year, or more with additional extensions under Article 16. But the issue here is whether the initial negative determination on aggression by the Security Council should be sufficient to block ICC action.

In the absence of relevant language contemplating a negative determination on aggression, Article 15 bis leaves a yawning gap in the scenarios confronting the ICC on aggression.

3. Temporal jurisdiction for the crime of aggression

As I originally examined in an earlier publication,Footnote ³ delegations in Kampala laboriously agreed to regard the amendments on aggression as entering into force in accordance with Article 121(5) of the Rome Statute rather than Article 121(4). This was a contentious debate, with Japan, for example, strongly advocating adherence to Article 121(4) for any amendment that was outside the scope of Articles 5, 6, 7, or 8, which pertain to the subject-matter jurisdiction of the ICC and are amended pursuant to Article 121(5). Any nod towards Article 121(4) greatly complicates entry into force procedures for the crime of aggression, because the jurisdictional filter for aggression would be grounded in new Articles 15 bis and 15 ter. If the jurisdictional filter requires a seven-eighths majority for ratification or acceptance by all states parties, then it might take a very long time for jurisdiction over the crime of aggression to be fully activated.

The final agreement in Kampala, to which Japan acquiesced while warning that the ‘dubious legal foundation’ of the amendments warranted future action by the Assembly of States Parties, allocated all the aggression amendments to the procedures of Article 121(5), which normally would have meant that they would come into force for a state party one year following the ratification or acceptance of the amendments by that state party. However, the amendments for new Articles 15 bis and 15 ter modify the Article 121(5) procedures with two critical and unusual conditions: (i) the ICC may exercise jurisdiction only over crimes of aggression committed one year after the ratification or acceptance of the amendments by 30 states parties; and (ii) the ICC may exercise jurisdiction only following at least a two-thirds vote of the Assembly of States Parties after 1 January 2017, reconfirming the agreed procedures in the amendment to activate jurisdiction over the crime of aggression.

The first caveat essentially raises the bar for entry into force under Article 121(5) and lowers it under the abandoned Article 121(4) – an artful albeit fragile compromise. The second caveat requires the Assembly of States Parties to revisit the issue in 2017 and determine (by consensus or a two-thirds vote) whether to proceed with the agreed procedures. Such radical tinkering with amendment procedures arguably merits an Article 121(4) amendment of the Rome Statute's amendment procedures – which is what Japan's concerns revealed – but the alternative course described above ultimately prevailed.

As consensus was reached in Kampala, Japan promised to resurrect its concerns at future meetings of the Assembly of States Parties. This may prove a disruptive feature of such meetings and it may encourage the ICC's judges in the future to look back at Japan's arguments and ponder the legality, under the Rome Statute, of what was forged in Kampala. States parties and the ICC itself will need to come to grips with this issue long before 2017 and, ideally, resolve it to everyone's firm assent so that the judges, when challenged by defence counsel in live cases regarding the crime of aggression, at least can rely on a more transparent, reaffirmed, and united interpretation of the Kampala amendments among states parties and leading scholars of the ICC.

4. The patchy terrain of aggression

The pragmatic reality of Kampala is that the geography of liability for the crime of aggression will be very patchy, at least for many years to come. At a minimum, Articles 15 bis(2) and 15 ter(2) inform us that 30 states parties will have accepted the jurisdiction of the ICC over the crime of aggression before the crime can first be referred, investigated, and prosecuted.Footnote ⁴ A number of states parties may not ratify or accept the aggression amendments, in which case they and their nationals will not be subject to the jurisdiction of the ICC for the crime of aggression. Other states parties that ratify the aggression amendments may elect to file a declaration of non-acceptance of the crime within one year with the Registrar pursuant to Articles 121(5) and 15 bis(4), and thus not be exposed to the ICC's jurisdiction for the crime of aggression. Certain of the states parties that choose this path may at some point in the future change their policies and withdraw such declarations on aggression and thus become subject to the ICC's jurisdiction.Footnote ⁵ Finally, the Security Council can determine at any time to extend the ICC's jurisdiction for the crime of aggression over any non-party state and its nationals and even any non-ratifying or non-accepting state party and its nationals by adoption of a UN Charter Chapter VII resolution referring a specific situation of aggression, probably occurring within a limited period of time and perhaps concerning only a limited category of suspects, to the ICC for investigation and prosecution under Rome Statute Articles 13(b) and 15 ter(1) referral authority.

The Prosecutor will need to maintain a map of the world in the office with an updated set of coloured pins stuck on countries’ territories indicating the current coverage and non-coverage for the crime of aggression globally. There will be differently coloured pins on different categories of states parties (those volunteering to be covered by the crime of aggression, those covered by a Security Council resolution, those ratifying the aggression amendments but declaring non-acceptance, those that never ratified or accepted the aggression amendments, and those that withdrew declarations of non-acceptance) and non-party states (those that are covered by a Security Council resolution or have filed a Rome Statute Article 12(3) declaration that invites investigation of atrocity crimes, including the crime of aggression, on its territory). The patchy coverage for the crime of aggression reflects the complexity of the issue, which bears directly on national political and military policies. But at least after Kampala the crime of aggression has taken hold within international criminal law, including its enforcement, and that is no mean feat. The result is a slap at the equality of states, or at least the theory of equality, but most major shifts in the international system begin that way.

No one claimed that aggression would be an easy crime for the International Criminal Court to investigate and prosecute, and the Kampala amendments demonstrated that truism in spades. But I suspect that, twenty years hence, the ICC will have adjudicated several cases of aggression and demonstrated that impunity for policies of aggression will not stand, at least in some parts of the world. Negotiators could have done far worse in Kampala.