By: Anika Winn


The Assembly of State Parties to the Rome Statute activated the jurisdiction of the International Criminal Court (ICC) over the crime of aggression at its 16th Assembly on December 14, 2017 effective on the 17th of July 2018, exactly 20 years to the date on which the Rome Statute was adopted. This was a historic moment in the progress of international criminal law, breathing life into the crime that was described in Nuremberg as the ‘supreme international crime’ (Duff, ‘Defining the Crime of Aggression’, 1 Transnational Legal Theory (2010) at 303). However, the road to activation of the crime of aggression within the jurisdiction of the ICC was a long and tumultuous process, marked by stark disagreement between states and far-reaching compromises. Now, just shy of one year after the effective date, only 35 state parties have accepted the amendments to the Rome Statute on aggression and the potential future of prosecuting the crime of aggression remains unknown (here at 942). Many questions arise as to the ability of the Court to actually prosecute aggressor states. Additionally, many of the outstanding issues and foreseeable challenges are the result of the numerous compromises reached between State Parties in order to achieve the consensus to activate this crime.

Origins of the Crime of Aggression

The 1945 Nuremberg Charter was the first to hold criminally liable those responsible for ‘crimes against peace’, which was defined as the ‘planning, preparation, initiation or waging of a war of aggression’ (Von Braun, ‘Judicial independence at risk: critical issues regarding the Crime of Aggression raised by selected human rights organizations’, 10 Journal of International Criminal Justice (2012) at 113).  The Nuremberg and Tokyo trials set the stage for the inclusion of the ‘crime against peace’ within the development of the new international system, but unfortunately, the experiences of prosecuting aggression was a legacy resigned to these trials for decades to come (here at 897).

Following the Second World War, the United Nations (UN) Charter was adopted and it did include the prohibition on the use of force under Article 2(4), subject to Security Council enforcement measures in Chapter VII. Then in 1974, the UN General Assembly adopted Resolution 3314, where it recommended a definition of the crime of aggression to be used by the UN Security Council. However, it did not seek to define a crime of aggression for which individuals could be held responsible, but only examined the responsibility of states (Von Braun at 113). Finally, after more than 40 years of relative silence on the crime of aggression, the crime seemed to begin to develop momentum again with the birth of the ICC through adoption of the Rome Statute on July 17, 1998. State Parties to the Rome Statute agreed that the crime of aggression would be one of the four core crimes under the jurisdiction of the ICC. However, they could not agree on a definition, and therefore, prosecution for the crime of aggression was deferred to such time that a definition was agreed upon and activated (here at 897).

The Kampala Review Conference

Fast forward to the Kampala Review Conference of 2010. The impression going into the conference was that defining the crime of aggression in the Rome Statute was a done deal, based on the work already done by a special working group (here at 1117). Unfortunately, however, this turned out not to be the case. The most pressing issue for many delegates was the role of the Security Council (Von Braun at 115). The USA in particular, which had a commanding presence at Kampala despite only having observer status, was an outspoken supporter of the position that the Security Council should be the only body that could refer incidents of aggression to the ICC (Lavers, ‘The new Crime of Aggression: a triumph for powerful states’, 18 Journal of Conflict and Security Law (2013) at 514). The United Kingdom and France supported the position of the USA, and this proved to be the most challenging aspect during the conference as they are permanent members of the Security Council (here at 890).

The majority of other states advocated for additional mechanisms for referral of a situation of aggression to the ICC. Many wished for the Court to be as independent as possible from long-established political institutions, particularly the Security Council (Von Braun at 126). Many feared that granting the Security Council sole referral power would further enhance the abilities of the Security Council to shield not only themselves but their allies from prosecution, and instead ‘cherry-pick’ which states would face prosecution (here at 1125). Further, the level of involvement of the Security Council being proposed by the USA, UK and France would be far greater than the involvement they are granted for the prosecution of the other core crimes, which states argued would unduly prejudice and differentiate the crime of aggression (here at 891).

From the outset of Kampala, almost every State party indicated a strong preference for consensus decision-making about the crime of aggression and some even said that they did not want to vote if consensus was not going to be achieved as it would be divisive for the ICC (here at 891). Eventually, State Parties unanimously adopted the text of Article 8 bis as proposed by the special working group, which defined the crime of aggression as a two-step approach. First, Article 8 bis (1) specifies the leadership requirements in order for an individual to be charged with a crime of aggression, as well as a magnitude threshold limiting what acts of aggression would be of a “character, gravity and scale” to constitute a crime of aggression. Then, Article 8 bis (2) provides the ‘component element’ of the crime, which is the actual act of aggression committed by the state. State Parties also unanimously agreed to the text of Articles 15 bis and ter, which enabled both States and the Security Council to refer situations to the ICC as well as for the Prosecutor to investigate proprio motu.

However, the compromise required to reach this consensus resulted in the inclusion of a never-before seen ‘opt out’ provision, increased the number of mechanisms to block a situation of aggression from coming before the Court, and resulted in certain provisions being left intentionally ambiguous and several questions left unanswered (Politi, ‘The ICC and the Crime of Aggression: a dream that came through and the reality ahead’, 10 J. Int. Crim. Justice (2012) at 277-8). Therefore, it was agreed upon through resolution that the ICC’s jurisdiction over the crime of aggression would be activated on a date not before 1 January 2017 and be subject to a minimum of 30 ratifications of the amendments, to allow time to resolve these issues (here at 891).

The New York Meeting

The next ASP to deal with the outstanding issues from Kampala and agree to activate the jurisdiction of the ICC over the crime of aggression was in New York in December 2017 (here at 890). The main dispute in New York was about which states would come under the jurisdiction of the ICC  when a proceeding was triggered either by state referral or by the Prosecutor taking up the matter proprio motu. Two main camps of thought emerged, with one advocating for a model of consent and the other for more of protection model and are explained further in an article by James Hendry. Eventually, after ‘Camp Consent’ refused to budge, the State Parties decided that in the case of state referrals and prosecutions proprio motu, the ICC will only have jurisdiction to prosecute nationals of and on the territory of States Parties  that have ratified the Kampala amendments one year after their ratification or acceptance (here at 942). The ASP also unanimously agreed to activate the jurisdiction of the ICC over the crime of aggression effective 17 July 2018.

Kampala and subsequently New York, were successful in achieving consensus on including and activating the crime of aggression in the Rome Statute. In the end, the advantage of reaching a consensus at both Kampala and New York was that it put a collective strength behind the crime of aggression and the jurisdiction that the ICC has over it. However, the disadvantage was, that to reach consensus at both meetings, significant compromise was required, which created several challenges and left many questions unanswered that the ICC will now have to deal with should a case of aggression come before the Court.

Academic Discussion: Outstanding Issues of Jurisdiction

States at the New York Meeting attempted to clarify which states fell under the jurisdiction of the ICC for the crime of aggression. To start with, nationals of non-States Parties to the Rome Statute are excluded from the Court’s jurisdiction over the crime of aggression when exercised by way of state referral or proprio motu powers (here at 953). Under Article 15 bis the ICC only has jurisdiction over states that are party to the Rome Statute. However, a referral under Article 15 ter is the exception to this as the Security Council can enable the court to exercise its jurisdiction over crimes of aggression committed on the territories or by nationals of State Parties and non-State Parties alike.

The ICC’s jurisdiction over state parties becomes complicated where situations are referred to the Court under Article 15 bis, for two reasons. First, there was (and still is) the question of whether ratification of the Kampala amendments is required for the ICC to have jurisdiction over that state (here at 955-8). The second reason is because of the inclusion of an ‘opt out’ provision which allows for a state party to declare that they do not accept the jurisdiction of the ICC over the crime of aggression, a feature unavailable for any of the other core crimes (here at 290). The opt-out provision will allow aggressor states to avoid prosecution, so long as the declaration is filed with the registrar before any acts of aggression occur. As a result, four different categories of states have essentially emerged under Article 15 bis (4), with each category having its own legal implications: (i) states that have ratified the amendment and have not opted out; (ii) states that have not ratified the amendment and have not opted out; (iii) states that have ratified the amendments and have opted out; (iv) states have not ratified the amendment and have opted out (here).

The first, third and fourth category of states have relatively clear legal implications. The ICC will have jurisdiction under Article 15 bis in the future if the aggressor state has ratified the amendments and has not ‘opted out’ under Article 15 bis (4) (Lavers at 501). The ICC will not have jurisdiction when the aggressor state has opted out, regardless of whether they have ratified the amendment (here at 955-8). When it comes to the first, third and fourth categories, the ratification status of the victim state seems to be irrelevant.

The category with divergent interpretations on whether the ICC will retain jurisdiction is the second category of states which have not ratified the amendment and have not opted out. The main arguments in favour of ICC jurisdiction is that Article 15 bis (4) seems to suggest that the ICC will have jurisdiction over aggression committed by State Parties regardless of whether they have ratified the amendments (here at 955-8). Also, if the normal rules of ICC jurisdiction laid out in Article 12 apply, then aggression committed by an ICC State Party that has not ratified the amendments, on the territory of a State Party that has ratified them will be subject to ICC jurisdiction. Further, according to Article 12(1), state parties to the Rome Statute accept the jurisdiction of the Court with respect to the crimes laid out in Article 5, including the crime of aggression, and the statute remains silent on ratification of the amendments. This supports an interpretation that the Court will retain jurisdiction in the case that an aggressor state has not ratified the amendments and not opted out (here at 955-8).

Alternatively, the main argument in favour of the ICC not having jurisdiction over this second category of states is that it would be in accordance with the language of Article 121(5) (here at 202). On its face, Article 121(5) seems to allow a State Party to avoid the application of ICC jurisdiction by simply not ratifying the amendments. Naturally then, it would follow that state parties would not be required to affirmatively opt of the Court’s jurisdiction over aggression unless they have already ratified the amendments, and if they fail to ratify them, then jurisdiction would not apply to their nationals in the first place (here at 203). However, Article 121(5) does not exist in isolation, and such an interpretation is either directly at odds with other provisions of the Statute or incongruous with them (here at 201-4). Overall, it remains to be determined what will happen in the future if there are situations of aggression committed by states that fall into this second category. What is certain however, is that not clarifying the status of this category could be consequential since currently, 89 of the 123 state parties to the Rome Statute have neither ratified the Kampala amendments, nor opted out (here).

Moving Forward

With the crime of aggression having come into force less than one year ago, it has yet to be determined how it will develop at the Court and several questions may need to be addressed. First, what happens if the if the aggressor state has ratified the amendments and not opted out, but the victim state is not a party to the Rome Statute? It is not clear whether the Court will be required to seek the consent of the victim state to proceed with an investigation and prosecution, or if having jurisdiction over the perpetrator state is sufficient (Politi at 277). Second, what is the position with respect to states that become parties to the Rome Statute after the adoption of the Kampala Amendments? Are they to be regarded as having ratified the amended Rome Statute of 2010 or the original Rome Statute of 1998, which would give them the option of whether they want to ratify the amendments on aggression (here at 942-5)? Third, if at a later date, a state party ratifies the amendments, can they be prosecuted retroactively for crimes of aggression committed in the period between the activation of the crime in 2018 and when they ratified the amendments (here at 291)? Or, can they only be prosecuted for crimes committed one year after the date in which they ratified the amendments? These are just some examples of questions that scholars have raised.

Overall, the crime of aggression seems to be the most recent example of a trend by especially powerful members of the international community, to limit or restrict the ICC (Lavers at 515). As a result of political compromise to activate the crime, the independence of the Prosecutor and ability for the Court to obtain jurisdiction over states was restricted. Ironically, for all the debate and concern over political persecutions, the current regime grants the most politically skewed branch of the UN, the Security Council, the most say in whether an incident of aggression can be investigated by the ICC (Lavers at 504). Overall, as it stands currently, the crime of aggression will rarely, if ever, be prosecuted, thus giving states an undeserved clean bill of health.

Please cite this article as Anika Winn, “The Crime of Aggression: A Political Compromise Resulting in an Ambiguous and Complex State of the Law”, (2019) 3 PKI Global Just J 59.

About the Author

Anika Winn is a third-year law student at the University of Ottawa. She has interned under the General Counsel at The Ottawa Hospital and is currently working as a summer student at Gowling WLG in Calgary, Alberta. She will be clerking at the Court of Queen’s Bench of Alberta beginning in 2020. Additionally, she has published in the field of maternal and refugee health and has conducted public health research in rural Tanzania.


Image: Flickr, Public Domain: German Troops March Through Warsaw by Jger Hugo, 1939 (NARA)