The Crime of Aggression at the International Criminal Court
Abstract and Keywords
This chapter focuses on the International Criminal Court’s jurisdiction over the crime of aggression. The discussion provides background to the crime of aggression and the resulting criminal accountability of the guilty party, paying particular attention to UN General Assembly’s adoption in 1974 of a resolution addressing aggression by states rather than the crimes of individuals and is designed as guidance for the Security Council when considering whether an act is one of ‘aggression’. The chapter examines the amendments to the ICC Rome Statute defining ‘act of aggression’ and ‘crime of aggression’ adopted at the ICC Review Conference in Kampala, Uganda, in 2010. It also discusses the uncertainties and ambiguities in the process for activating ICC jurisdiction over the crime of aggression. It considers the possible institutional effects of such jurisdiction on the UN Security Council and the ICC itself, as well as its long-term consequences for the jus ad bellum.
Those seeking to uphold the international prohibition on the use of force by one state against the territorial integrity or political independence of another often favour the idea of criminally punishing governmental leaders who initiate such force. Indeed, at least since the prosecution of the major political and military leaders at the Nuremberg and Tokyo war crimes tribunals, many states and individuals have sought to establish a global criminal tribunal for prosecuting government officials who plan and unleash interstate aggression. Throughout most of the 20th century, that aspiration remained unfulfilled, but in 1998 120 states adopted the Rome Statute establishing the International Criminal Court (ICC).1 The Rome Statute, to which 122 states are a party as of September 2014,2 contemplated the activation of ICC (p. 534) jurisdiction over the crime of aggression, but only at a future point when the crime would be defined and the conditions for its operation would be elaborated.
At the ICC Review Conference, held in 2010 at Kampala, Uganda, the states parties reached major decisions towards that end, settling upon definitions for ‘act of aggression’ and ‘crime of aggression’, and making the jurisdiction potentially available even in the absence of a referral from the Security Council. At the same time, the states parties decided that the ICC’s jurisdiction over this crime will not become operative until sometime after 1 January 2017 pursuant to a further decision by the states parties. Even then, the ICC’s jurisdiction will be limited over this crime, since there are exceptions available for states parties that wish to avoid exposure to such jurisdiction and the jurisdiction will not extend to states that are not parties to the Rome Statute.3 Moreover, considerable uncertainties and ambiguities exist concerning the exact process for activating the jurisdiction, the manner in which the jurisdiction operates once it is activated, its institutional effects on the Security Council and the ICC itself, and its long-term implications for the jus ad bellum.
II. Antecedents: From Versailles to Kampala
As noted elsewhere in this volume and discussed in depth in specialized treatises,4 individuals were first held criminally accountable for waging a war of aggression at the International Military Tribunal convened at Nuremberg in the aftermath of the Second World War. The road to Nuremberg was an uneven one. The 1919 Treaty of Versailles after the First World War called for the arrest and trial of German officials,5 notably Kaiser Wilhelm II, but the Kaiser lived out his life comfortably in the Netherlands and the 1921 ‘Leipzig trials’ of other officials before the German Supreme Court were inconsequential.6 Such trials were for violations of the laws (p. 535) and customs of war, not for aggression or any other unlawful use of force, though the unconsummated trial of the Kaiser would have been ‘for a supreme offence against international morality and the sanctity of treaties.’7
In the 1928 Kellogg–Briand Pact (or Pact of Paris), the states parties solemnly declared ‘that they condemn recourse to war for the solution of international controversies and renounce it as an instrument of national policy in their relations with one another.’8 The Pact, however, did not specify criminal liability either for states or for individuals in the event of a violation of the Pact; whether the norm set forth in the Pact reflected a general rule of international law or one binding solely upon those states that had ratified the Pact was uncertain. As such, after the outbreak of the Second World War, many believed that no ‘international agreement criminalising wars of aggression was in force in 1939, and therefore, on the basis of the nullum crimen sine lege principle, the Allies were not legally entitled to prosecute the top Nazi leaders for aggression.’9
Nevertheless, at the San Francisco conference in April 1945, the ‘enforcement arrangements’ committee asserted that:
It being the original intent and meaning of the Kellogg-Briand Pact…that any person in the service of any Party-State who violated its provisions…should be held individually responsible for these acts, it is declared that the aggressions of the Axis states since the signing of the Pact violated its provisions and that the persons in the service of such Axis states are individually responsible for such acts and may be brought to trial and punishment before any United Nations court or other tribunal of competent jurisdiction which may secure custody of such persons or any of them.10
As such, when the Charter establishing the International Military Tribunal to prosecute major war criminals of the European Axis powers was adopted in August 1945, it provided jurisdiction to the Tribunal over:
Crimes Against Peace: namely, planning, preparation, initiation or waging of a war of aggression, or a war in violation of international treaties, agreements or assurances, or preparation in a common plan or conspiracy for the accomplishment of any of the foregoing. …11
In implementation of the Charter, and with reliance on the Kellogg–Briand Pact and other treaties and agreements, the first trial of the major war criminals proceeded in Nuremberg to prosecute and convict eight defendants for crimes against (p. 536) peace. Although invited to do so by the prosecution, the Tribunal did not convict any defendants for ‘a war in violation of international treaties, agreements or assurances.’12 Rather, the Tribunal only convicted defendants for conducting (or conspiring to conduct) a ‘war of aggression’, although it took into account pre-war international agreements in finding that aggression had been outlawed. When parsing the facts, the Tribunal characterized the Anschluss in Austria and the German administration in parts of Czechoslovakia as ‘aggressive actions’, inasmuch as they were shown to be part of a plan for aggressive wars against other countries.13 By contrast, the Tribunal characterized the uses of force against Belgium, Denmark, Greece, Luxembourg, the Netherlands, Norway, Poland, the Soviet Union, the US, and Yugoslavia as ‘aggressive wars’.14 The convicted defendants were found to have been knowingly involved in activities ‘not too far removed from the time of decision and of action’, and all to have ‘contributed to the initiation of the war in an important and “aggressive” role.’15
When judging the culpability of the defendants for such conduct, the Tribunal developed two counts relating to the crime of aggression. Count one concerned the conduct of broadly engaging in a common plan to prepare, initiate, and wage aggression. This count could be shown, for instance, by establishing that the defendant participated in four secret conferences from 1937 to 1939 at which Adolph Hitler revealed his plans for invading other countries.16 Count two concerned the conduct of planning or waging a particular war of aggression, such as the conviction of Admiral Karl Doenitz for the waging of submarine warfare.17 Eight of the defendants at the first trial were convicted of counts one and two, while four were convicted only of count two.
The International Military Tribunal for the Far East (Tokyo Tribunal), established by a special proclamation issued by the Supreme Commander of the Allied Powers in the East (General Douglas MacArthur), also possessed jurisdiction over crimes against peace.18 Twenty-eight Japanese senior political and military leaders were brought before the Tribunal; count one charged them as ‘leaders, organisers, instigators, or accomplices in the formulation or execution of a common plan or conspiracy…to wage wars of aggression, and war or wars in violation of international (p. 537) law’, while other counts concerned waging unprovoked war specifically against the British Commonwealth, China, France (in Indochina), the Netherlands, the Soviet Union, and the US. Though two died and one was declared incompetent, the remaining defendants were all convicted of crimes against peace.19
The UN General Assembly in 1946 affirmed the principles of international law set forth in both the Charter of the Nuremberg Tribunal and its judgment of the major war criminals.20 Moreover, in 1950 the International Law Commission codified a series of principles reflecting that charter and judgment, including with respect to the crime of aggression.21 Some saw these steps as a prelude to the creation of a permanent international criminal court, but international criminal tribunals disappeared after Nuremberg and Tokyo. The political divide of the Cold War, the resistance of the major powers to scrutiny of their uses of force, and the slowly evolving structures in international law for addressing rights and obligations of individuals held those aspirations in check.
Perhaps the most important development in this period was the adoption in 1974 by the UN General Assembly of a resolution, to which was annexed a document entitled ‘Definition of Aggression’.22 Though generally referred to as a ‘definition’, the resolution is probably best understood as a series of factors set forth by the General Assembly as guidance for the Security Council when considering whether an act of ‘aggression’ has occurred; no single factor standing alone was meant to be determinative, but had to be weighed by the Council in the context of all relevant circumstances. Article 1 of the annex stated: ‘Aggression is the use of armed force by a State against the sovereignty, territorial integrity or political independence of another State, or in any other manner inconsistent with the Charter of the United Nations, as set out in this Definition.’23 Article 2 then asserted that a first use of force was prima facie an act of aggression, ‘although the Security Council may, in conformity with the Charter, conclude that a determination that an act of aggression has been committed would not be justified in the light of other relevant circumstances, including the fact that the acts concerned or their consequences are not of sufficient gravity.’24 Article 3 provided a non-exhaustive list of seven types of act that a state (p. 538) might commit that would presumptively constitute aggression, but reiterated that a determination in each case was to be made by the Security Council.25 As such, the resolution’s annex did not directly equate all uses of force that violated Article 2(4) of the UN Charter with ‘aggression’. Rather, it provided in Article 1 that ‘aggression’ might consist of the types of force ‘as set out in this Definition’, identified in Articles 2 and 3 certain illustrative acts, and then left it to the Security Council to decide for any given incident whether ‘aggression’ in fact had occurred.
Importantly for present purposes, the resolution was focused on state responsibility for aggression, not on the criminal responsibility of individuals; it principally ‘deals with aggression by States, not with the crimes of individuals, and is designed as a guide for the Security Council, not as a definition for judicial use.’26 The only reference to ‘crime’ appears in Article 5(1) of the annex, which (echoing Nuremberg) asserted that a ‘war of aggression is a crime against international peace.’27 As such, the ‘drafters of the Definition thereby signaled clearly that not every act of aggression constitutes a crime against peace; only war of aggression does.’28
After the end of the Cold War, some thought was given to the prosecution of Iraqi leaders for war crimes during Iraq’s 1990 invasion of Kuwait,29 but no steps were taken to do so. Nevertheless, having begun in that crisis to exercise more robustly its Chapter VII powers, the Security Council did establish in 1993 the International Criminal Tribunal for the former Yugoslavia (ICTY)30 and in 1994 the International Criminal Tribunal for Rwanda (ICTR)31 for the prosecution of persons who committed war crimes, crimes against humanity, or genocide in those conflicts. Neither tribunal possessed jurisdiction over the crime of aggression, nor did other ad hoc tribunals or special courts that emerged relating to atrocities in Cambodia, East Timor, Kosovo, Lebanon, or Sierra Leone,32 usually because the circumstances of the crisis were largely internal in nature.
The creation of the ICTY and ICTR, however, helped to break through the political log-jam holding back the creation of a permanent International Criminal Court, which was founded with the adoption of the Rome Statute in 1998 and its entry into (p. 539) force in 2002. While sufficient consensus existed in 1998 regarding the operation of the ICC’s jurisdiction over war crimes, crimes against humanity, and genocide, agreement could not be reached on what was meant by the ‘crime of aggression’ or on what the role should be for the Security Council in determining whether an act of aggression had occurred prior to the ICC exercising its criminal jurisdiction.33 Ultimately, a compromise was reached in 1998: while Article 5(1) of the Rome Statute provided that the Court would have jurisdiction over all four types of crime— war crimes, crimes against humanity, genocide, and ‘the crime of aggression’—Article 5(2) stated that the Court would only exercise jurisdiction over the latter crime ‘once a provision is adopted in accordance with articles 121 and 123 defining the crime and setting out the conditions under which the Court shall exercise jurisdiction with respect to this crime.’34
Article 121 of the Rome Statute concerns the amendment process, while Article 123 provides that seven years after entry into force of the Rome Statute a ‘review conference’ shall be convened to consider such amendments, including to ‘the list of crimes contained in Article 5.’35 The Rome Statute entered into force on 1 July 2002, such that by late 2009 the time was ripe for convening the review conference, which was held in Kampala, Uganda from 31 May to 11 June 2010.
III. The Kampala Amendments
In the years preceding Kampala, a Special Working Group on the Crime of Aggression met and prepared documents for adoption at the review conference.36 After difficult negotiations,37 the review conference adopted a resolution on the crime of aggression (Resolution RC/Res 6), to which was annexed four amendments to the Rome Statute (Annex I), as well as certain amendments to the ‘Elements of Crimes’ previously adopted by the Assembly of States Parties (Annex II), and certain interpretive understandings (Annex III).38
(p. 540) Of particular interest are the four amendments. The first amendment simply deletes Article 5(2) from the Rome Statute.39 The second amendment creates a new Article 8bis, which defines both an ‘act of aggression’ and the ‘crime of aggression’, derived verbatim from the definitions proposed by the Special Working Group. Under Article 8bis(2), ‘act of aggression’ means:
the use of armed force by a State against the sovereignty, territorial integrity or political independence of another State, or in any other manner inconsistent with the Charter of the United Nations. Any of the following acts, regardless of a declaration of war, shall in accordance with the United Nations General Assembly resolution 3314 (XXIX) of 14 December 1974, qualify as an act of aggression: [the amendment then lists the seven types of act identified in that resolution].40
Under Article 8bis(1), ‘crime of aggression’ means:
the planning, preparation, initiation or execution, by a person in a position effectively to exercise control over or to direct the political or military action of a State, of an act of aggression which, by its character, gravity and scale, constitutes a manifest violation of the Charter of the United Nations.41
The ICC’s definition of ‘act of aggression’ draws heavily on the General Assembly’s 1974 resolution, but in ways not provided for in that resolution. The first sentence of Article 8bis(2) essentially equates aggression with any violation of Article 2(4) of the UN Charter. The second sentence of Article 8bis(2) appears to assert, without qualification, that all the acts enumerated in the 1974 resolution constitute aggression. As noted previously, this was not the approach taken in the 1974 General Assembly definition, which viewed such acts as presumptively constituting aggression, but subject to a determination of the Security Council, which might conclude that ‘relevant circumstances, including the fact that the acts concerned or their consequences are not of sufficient gravity’, indicated that ‘aggression’ had not occurred.42
Kampala’s ‘crime of aggression’ by its terms is a leadership crime; the defendant must hold a position by which he or she ‘effectively…exercise[s] control over or…direct[s] the political or military action of a State.’43 The language adopted excludes non-governmental actors, such as persons leading a terrorist group (eg Al Qaeda), leaders of an insurgency, or industrialists in a country even if they have substantial (p. 541) involvement in and influence upon governmental conduct.44 Like Nuremberg, the ‘planning’, ‘preparation’, or ‘initiation’ of an act of aggression falls within the scope of the crime; the term ‘execution’ is used in place of Nuremberg’s ‘waging’ of a war. Unlike Nuremberg, there is no requirement that the conduct relate to a ‘war of aggression.’ Rather, all acts falling within the scope of Article 2(4) of the UN Charter, including those set forth in Resolution 3314 (XXIX), may serve as a basis for finding a crime of aggression, as long as they are a ‘manifest’ violation of the Charter. The requirement of a ‘manifest’ violation is not found in the UN Charter or in the General Assembly’s 1974 resolution, so it presents a new standard that will be discussed later in this chapter.
The third amendment creates a new Article 15bis, which addresses the ability of the ICC to exercise jurisdiction over the crime of aggression in a given situation based on either a referral by a state or the prosecutor’s own initiative (proprio motu).45 The fourth amendment provides for exercise of such jurisdiction when the Security Council refers to the ICC a situation in which the crime of aggression appears to have been committed.46
These amendments addressed several outstanding issues concerning how the crime of aggression would operate before the ICC, but leave open many others, both in terms of procedure and substance.
IV. Procedural Issues
A. Rome Statute Article 121(5) Provides the Process for Entry into Force
According to the resolution on aggression that was adopted at Kampala, all the amendments concerning the crime of aggression ‘are subject to ratification or acceptance and shall enter into force in accordance with article 121, paragraph 5’ of the Rome Statute.47 Article 121(5) was apparently regarded as the relevant basis for bringing the amendments into force because, as previously noted, one of the Kampala amendments formally alters Article 5 of the Rome Statute,48 and Article 121(5) applies to ‘any’ amendment to Article 5.49 Article 121(5) provides:
Any amendment to Articles 5, 6, 7, and 8 of this Statute shall enter into force for those States Parties which have accepted the amendment one year after the deposit of (p. 542) their instruments of ratification or acceptance. In respect of a State Party which has not accepted the amendment, the Court shall not exercise its jurisdiction regarding a crime covered by the amendment when committed by that State Party’s nationals or on its territory.
Arguably not all the amendments adopted at Kampala concerning the crime of aggression had to be brought into force pursuant to Article 121(5), since most of the amendments do not amend Articles 5, 6, 7, and 8. The selection of a single process for all the amendments appears to reflect a conclusion by the states parties that it was appropriate to proceed in toto either under Article 121(5) or under an alternative procedure set forth in Article 121(4). That conclusion may have been reached because the amendments, as a whole, activate a new form of ICC criminal jurisdiction (the basic concern of Art 121(5)) and thus it was appropriate for all the amendments to be ‘subject to ratification or acceptance [and to] enter into force in accordance with article [121(5)].’50
B. A Package Deal?
At present, states are deciding whether to ratify or accept the amendments; 15 had done so as of September 2014.51 Whether states parties must ratify these amendments as a package or can pick and choose as among them may be debated. On the one hand, the package consists of four amendments and other decisions that were orchestrated all at once as part of a series of compromises. On the other hand, there is not a single amendment, but four amendments, and there is no prohibition or restriction within the package requiring that these amendments be ratified or accepted by a state in toto.
Article 121(5) provides that amendments to Article 5 of the Statute enter into force for any state party one year after the deposit of the instrument of ratification or acceptance.52 However, embedded within these amendments are two additional requirements that must be met before the ICC can actually exercise jurisdiction over an alleged crime of aggression. First, the ICC can only exercise jurisdiction (p. 543) over an alleged crime that occurs more than one year after 30 states have ratified or accepted the amendments.53
Second, the ICC may only exercise jurisdiction ‘subject to a decision to be taken after 1 January 2017 by the same majority of states parties as is required for the adoption of an amendment to the Statute.’54 While that language is a bit ambiguous in meaning, it is best understood to mean that jurisdiction may not be exercised until after a decision by at least a two-thirds majority of the Assembly of States Parties, occurring sometime in 2017 or thereafter, expressing approval of the ICC exercising such jurisdiction. It is not clear whether the states parties must take a decision with respect to all of these amendments as a package or can take a decision only with respect to some portions of that package, but it seems likely that the states parties will proceed on them together.
C. Temporal Exercise of Jurisdiction by the ICC
One uncertainty that arises concerns the ICC’s temporal jurisdiction: can the ICC only exercise jurisdiction over an alleged crime that is temporally committed after both of the previously mentioned requirements are met? The language of the amendments does not say as much and, indeed, could be read as allowing ICC jurisdiction in, for example, 2017 over a crime committed in 2015, if at that point one year had elapsed since 30 states had ratified or accepted the amendments. The Kampala Review Conference, however, adopted an understanding on this issue.55 The language of the understanding appears to provide that any alleged crime of aggression must occur after both of the two requirements are met. Only crimes occurring after the further decision of the states parties in 2017 (or thereafter) could fall within the scope of the ICC’s jurisdiction, not crimes that occur in earlier years, even if 30 states had ratified or accepted the amendments before 2017.56
D. Jurisdiction of the ICC over States Parties that Do Not Ratify/Accept
Assuming that the Security Council has not acted, an important issue is whether a state party that does not ratify or accept the amendments is exposed to the ICC’s (p. 544) jurisdiction over the crime of aggression. In other words, does the failure to ratify the amendments mean that the state party is not exposed or must the state party affirmatively opt out if it wishes to avoid the ICC’s jurisdiction?
As noted earlier, Article 121(5) of the Rome Statute provides in its second sentence that, for a state party that has not ratified or accepted an amendment, ‘the Court shall not exercise its jurisdiction regarding a crime covered by the amendment when committed by that state party’s nationals or on its territory.’57 The ordinary meaning of that sentence is that if a state party does not ratify or accept the amendments adopted at Kampala, then the ICC has no Article 15bis jurisdiction over that state party’s nationals.58 The language in French, which is equally authentic, seems even clearer: ‘La Court n’exerce pas sa competence á l’égard d’un crime…’59
That interpretation might be bolstered by comparison to an analogous provision of the Rome Statute. Article 124 is a ‘transitional provision’ that allows a state party to declare (in language similar to Art 121(5)) that, for a period of seven years, ‘it does not accept the jurisdiction of the Court with respect to the category of crimes referred to in article 8 when a crime is alleged to have been committed by its nationals or on its territory.’60 To the extent that Article 124 is interpreted as placing the state party entirely outside the scope of the ICC’s jurisdiction during that time, then it would seem that a comparable result should operate for Article 121(5).61
This interpretation comports with background rules on the amendment of treaties, which provide that an ‘amending agreement does not bind any State already a party to the treaty which does not become a party to the amending agreement’,62 and which otherwise accord protections for a treaty party from amendments.63 Arguably, exposure of a state’s nationals (or persons who commit the crime in the territory of the state) to the ICC’s jurisdiction is not the same as the state itself being bound by the amendment, but presumably the state’s obligations under the Rome Statute to cooperate with the ICC would be viewed as applying to the investigation and surrender of such nationals (or persons), and in that sense the amendments are altering the state’s treaty obligations.
An alternative interpretation advanced during the period leading up to and at Kampala maintained that, once the ICC’s ‘aggression’ jurisdiction is activated, it has certain effects for every state party (unless that state party affirmatively opts out of the jurisdiction, as permitted by the amendments64). This interpretation, which (p. 545) became known as the ‘positive understanding’,65 argued that the second sentence of Article 121(5) only precludes ICC jurisdiction arising under a new amendment over the non-ratifying state party when the jurisdiction is predicated solely on the conduct having occurred in the territory of the non-ratifying state party or by a national of that state party. Yet, under this interpretation, the second sentence does not prevent the Court from exercising jurisdiction over a national of the non-ratifying state party when that national’s conduct was taken against or occurred within the territory of a state party that has ratified or accepted the amendment. The lynchpin of this theory is Article 12(2) of the Rome Statute, which allows the Court to exercise its jurisdiction over the nationals of states that are not party to the Statute whenever their conduct occurs on the territory of a state party.
In essence, this interpretation places the state party that fails to ratify the Kampala amendments in a position (with respect to the crime of aggression) analogous to that of a state that fails to ratify the Rome Statute (with respect to war crimes, crimes against humanity, and genocide); neither state’s nationals are exposed to the Court’s jurisdiction by virtue of that state’s consent, but those nationals are exposed when they take action that falls within the ambit of another state’s consent to the Court’s jurisdiction. Thus, even a state party that has not ratified or accepted the Kampala amendments may find its leaders exposed to prosecution for the crime of aggression, as long as the aggression occurs in the territory of a state party that has ratified or accepted the amendments (just as the nationals of a non-state party can be exposed to the Court’s other forms of jurisdiction under Art 12).
Proponents of the ‘positive understanding’ relied in part on the fact that there is an ‘opt-out’ procedure in new Article 15bis(4).66 Arguably no such procedure is needed if a state party can effectively opt out of the ICC’s jurisdiction simply by not ratifying the new amendments. In the first instance, this argument assumes that the amendments were carefully drafted, with a coherent legal theory understood and accepted by all states parties, which may not have been the case.67 In any event, there are reasons why a state party might ratify the amendments, but then opt out of Article 15bis, including: (1) a desire to support the overall scheme, including the definition of aggression and the role of the Security Council, but a desire not to expose itself to the ICC’s jurisdiction under Article 15bis; (2) a potential interest in (p. 546) exposing itself to Article 15bis, but with a short-term preference for waiting to see how the ICC will begin exercising its jurisdiction before doing so; and (3) a desire only to expose itself partially to the ICC’s jurisdiction and thus use the opt-out declaration to carve out some areas where the state party would not be exposed (discussed later in the chapter).68
The ‘positive understanding’ was contentious at Kampala for various reasons.69 Japan, in particular, argued forcefully at Kampala that such an interpretation of the Article 121 amendment procedures effectively amends Article 121, a step not possible except through the amendment procedures set forth in Article 121(4).70
A second objection was that such an interpretation technically appeared to open the door for the Assembly of States Parties, at any time it could secure a two-thirds vote, to adopt amendments modifying Article 5 of the Rome Statute, so as to add new crimes pursuant to whatever amendment procedures it so chose, thereby exposing the nationals of even dissenting states parties to those crimes based on the states parties’ prior acceptance of Article 12. Indeed, if the ‘positive understanding’ theory of Article 121(5) was correct, it would seem to apply equally to the ‘Belgian’ Amendment adopted at Kampala criminalizing the use of certain weapons in a non-international armed conflict (meaning that nationals of states parties that do not ratify or accept that amendment may nevertheless be prosecuted for that crime).71
Due to such concerns, a different interpretation (referred to by some as the ‘softened consent-based regime’) has been advanced as an explanation of how best to understand the outcome from Kampala.72 This interpretation focuses on the purported combined effects of Articles 5(2), 12(1), and 121(3), as well as the ability of non-ratifying states parties to ‘opt out’ of the ICC’s jurisdiction under new Article 15 bis. In essence, this interpretation argues that every state party to the Rome Statute has accepted, in Article 5, that the ICC has jurisdiction over the crime of aggression (as opposed to other possible crimes). Further, every state party has accepted that such jurisdiction may be exercised ‘once a provision is adopted in accordance with articles 121 and 123 defining the crime and setting out the conditions under which (p. 547) the Court shall exercise jurisdiction with respect to this crime.’ That decision now having been made at Kampala, all states parties are exposed to the effects of the amendments relating to aggression and their nationals may be subject to the ICC’s ‘aggression’ jurisdiction if they commit the crime of aggression in the territory of a state party that has ratified or accepted the amendments. At the same time, the delegates at Kampala chose to allow states parties to opt out of the amendments by lodging an affirmative declaration to that effect. By grounding itself in Article 5(2), this interpretation is responsive to the second objection noted earlier to the ‘positive understanding’, by limiting the interpretation solely to the crime of aggression.
Yet the first objection noted previously appears to remain valid; it is hard to see how this variation could be the amendment process for the crime of aggression that the states meeting in Rome had in mind. Why would states in Rome establish an amendment process that strongly protected their interests for any changes relating to crimes other than aggression (allowing them to avoid exposure to those crimes by non-ratification), but create an amendment process for the crime of aggression that leaves them vulnerable to whatever conditions thought desirable by a two-thirds decision of the Assembly? If anything, the unusual nature of the crime of aggression, as compared with other possible ICC crimes, suggests the need for greater deference to state consent concerning exposure to that crime, since the crime of aggression has very important and inescapable implications for the responsibility of the state itself, not just the individual.73 Thus, while the idea of a ‘softened consent-based regime’ may well have been a ‘creative’ political compromise that sought to ‘bridge the gap’ at Kampala,74 it remains to be seen whether the states parties and the ICC itself regard such an interpretation of the Rome Statute as correct.
E. The Ability of States Parties to Opt Out of ICC Jurisdiction
Assuming that a state party is exposed to the effects of the amendment, a further issue arises with respect to the ‘opt-out’ procedure under Article 15bis, which (p. 548) provides that the Court may not exercise jurisdiction over a crime arising from the conduct of a national of that state party if the ‘State Party has previously declared that it does not accept such jurisdiction by lodging a declaration with the Registrar’ of the ICC.75
Such language leads to an obvious question: the declaration must be lodged ‘previous’ to what? How late in the game can the state party file such a declaration? Must the declaration be lodged prior to the date on which the alleged aggression occurred, including the planning and preparation for such aggression? Or can it be lodged at any point up until the date that a state party refers the matter to the ICC or that the Prosecutor initiates an investigation proprio motu? How about up until the issuance of an ICC arrest warrant? Obviously, the answer is rather important in terms of the ability of a state party to shield its leaders from ICC prosecutions.
One answer might be that the Declaration must be filed prior to a state party’s ratification or acceptance of the amendments. The Kampala resolution seems to suggest this approach, when it ‘notes that any State Party may lodge a declaration referred to in Article 15 bis prior to ratification or acceptance.’76 Indeed, that language may confirm that the ‘positive understanding’ or ‘softened consent-based regime’ interpretations are incorrect, since the scheme appears to link the issue of ‘opting out’ to the situation where a state party has ratified or accepted the amendments and does not link it to some other situation, such as opting out in the absence of ratification or acceptance. Seen in this way, the scheme has a certain coherence: a state party is not exposed to the Court’s jurisdiction over the crime of aggression until it ratifies or accepts the amendment and, if it seeks to opt out of Article 15bis jurisdiction in whole or in part, it must so declare before its ratification or acceptance.
A different question is whether a state party faces a binary choice of ‘opt in’ or ‘opt out’, or whether there are intermediate positions where a state party can opt out with respect to certain circumstances. A state party might seek to opt out of the ICC’s jurisdiction solely with respect to crimes of aggression arising from that state’s military operations against a specified country, such as a neighbour. More broadly, a state party might seek to opt out solely with respect to actions against a class of countries, such as any country that is not a state party to the Rome Statute or any country that is a state party but has opted out of the crime of aggression. Perhaps a state party would seek to opt out from the ICC’s jurisdiction for a specified time period or a specified subject matter area, such as uses of force over maritime resources. For example, a state party might seek to opt out of the ICC’s ‘aggression’ jurisdiction solely with respect to alleged crimes arising from that state’s involvement in a military operation of the North Atlantic Treaty Organization (NATO). One need look (p. 549) no further than the practice of states’ acceptance of the compulsory jurisdiction of the International Court of Justice (ICJ) to find ample evidence of temporal, geographic, and subject-matter carve-outs to jurisdiction; nothing in the text of Article 15bis forbids doing the same.
F. The Problem of a Lack of Reciprocity
A different concern arises given that the structure of this ‘opt-out’ does not carry with it a coherent notion of reciprocity. Assume that state X and state Y both ratify the amendments, but state X lodges a declaration by which it opts out of the crime of aggression, while state Y does not file any such declaration. If state X invades state Y, the ICC will not have jurisdiction to investigate and indict state X’s leaders, but if state Y invades state X, state Y’s leaders will be exposed to the ICC’s jurisdiction. If an armed conflict breaks out between the two states with both sides claiming that the other is the aggressor, apparently the ICC can investigate the leaders of state Y but not the leaders of state X.
At first glance, this lack of reciprocity seems inconsistent with jurisdiction as it normally operates before international courts and tribunals. In that context, the standard rule is that any step taken by one state to restrict its exposure to the tribunal’s jurisdiction can be invoked by the other state in the event that a bilateral dispute arises. Yet in this instance, the lack of reciprocity for ICC jurisdiction may be explained on the basis that the ICC is not adjudicating bilateral disputes between states; rather, it is prosecuting individuals for wrongful conduct because the state of which those individuals are leaders accepted the possibility of such prosecutions. According to this argument, there is no problem of reciprocity because this does not concern an interstate relationship.
Yet there may be a political problem, in that the absence of reciprocity when an armed conflict erupts seems inherently unfair and could undermine the ICC’s appearance as an even-handed institution. In any event, this lack of reciprocity may provide a strong incentive for states parties to opt out of the Court’s jurisdiction over this crime at least vis-à-vis other states parties that have opted out. The theory behind reciprocity (eg as part of the compulsory jurisdiction system of the ICJ) is, in part, that a state will be induced to join the system, since otherwise it cannot benefit from that system. By contrast, with the ICC’s jurisdiction over the crime of aggression, states parties do not automatically benefit by accepting the ICC’s jurisdiction over this crime (in this sense, it differs from the ICC’s other heads of jurisdiction, for which there is no opt-out). Time will tell whether ratifying states refrain from opting out vis-à-vis other states parties that have opted out simply because they see it as the right thing to do; yet unconditional acceptance of the ICC’s jurisdiction appears to provide no automatic benefits with respect to foreign invaders.
(p. 550) G. The Position of Non-State Parties
The part of the amendments that probably received the most immediate public attention concerned the treatment of non-state parties. Article 15bis provides that the Court has no jurisdiction over the crime of aggression with respect to a state that is not a party to the Rome Statute ‘when committed by that State’s nationals or on its territory.’77 This provision forecloses exercise of Article 15bis jurisdiction over nationals of states that are not parties to the Rome Statute, including at present China, Russia, and the US.
While many lament this carve-out from the Court’s jurisdiction as drastically scaling back the scope of the regime, the decision to do so ultimately may be in the best interests of the Court. The crime of aggression is more than just a crime associated with a particular individual; as noted previously, it is a crime that relates much more closely to the state and state policy as a whole. When adjudicating the wrongfulness of state conduct before international courts and tribunals, the international legal system is built upon the idea of express state consent, because dispute settlement institutions only operate effectively when they are premised upon such acceptance ab initio by states. The compulsory jurisdiction of the ICJ, although only accepted by some 70 states, has historically operated quite effectively in terms of participation and compliance, precisely because those states have affirmatively chosen to opt in to the system. Had this exemption not been included, the legitimacy of the ICC might have been seriously damaged, especially if states parties were provided the ability to opt out of this new jurisdiction, while non-state parties were not.
H. The Problem of Alleged Aggression by a Coalition
The existence of ICC jurisdiction over some states but not over other states, however, may well raise a dilemma for the ICC if faced with alleged aggression by a group or coalition of states. Assume that a group of four states engages in alleged aggression against a fifth state. State A is a state party to the Rome Statute that has ratified the amendments without reservation, state B is a state party that has ratified the amendments but opted out of the ICC’s jurisdiction, state C is a state party that has not ratified the amendments, and state D is a non-state party. In the absence of a Security Council referral, the ICC presumably only has jurisdiction over action by the leaders of state A (if the ‘positive understanding’ or ‘softened consent-based regime’ interpretations is correct, it would also have jurisdiction over state C). Yet when investigating and prosecuting that alleged aggression, it seems inevitable that the ICC, in essence, would have to assess the culpability of states B, C, and D. The leaders of those states presumably would not be defendants in the ICC’s courtroom (p. 551) but, given the nature of the crime at issue, the ICC in effect would be discussing and passing upon the conduct of those other leaders.
There is precedent in other settings, such as the ICJ, for the international tribunal to decline to pass upon the merits of the case because of a lack of jurisdiction over other parties whose rights and obligations are inextricably woven into the case.78 Given that the ICC will no doubt confront this difficult issue in situations where alleged aggression involves a group of states, the Assembly of States Parties would do well to consider, in advance of the matter reaching the Court, the best approach legally and politically that the Court should take.
I. Security Council Referral of a Situation of Aggression
A further conundrum concerns Article 15ter, which addresses the ICC’s jurisdiction over the crime of aggression based upon Security Council referral. Once the hurdles previously discussed concerning the ratification of 30 states and the further decision of the Assembly after 2016 are met, a question arises as to whether Article 15 ter operates even as against states parties that have not ratified or accepted the amendments.
As noted earlier, Article 121(5) indicates that the Court shall not exercise jurisdiction regarding a crime covered by an amendment when committed by nationals (or on the territory) of a state that has not ratified or accepted the amendment. Consequently, it would appear that the process for Security Council referral established under Article 15ter might not operate with respect to states parties that do not ratify or accept the amendments.
At Kampala, however, the states parties adopted an Understanding that ‘the Court shall exercise jurisdiction over the crime of aggression on the basis of a Security Council referral in accordance with Article 13, paragraph b, of the Statute irrespective of whether the State concerned has accepted the Court’s jurisdiction in this regard.’79 The legal theory here appears to be that Article 15ter is principally addressing the capacity of the ICC, as an institution, to receive Security Council referrals and is consistent with the view that the Council has the power to confer authority upon the ICC to so act. Once the amendment enters into force and this (p. 552) capacity is activated, then the ICC can receive referrals from the Security Council with respect to any situation involving any country that the Council, in the exercise of its Chapter VII authority, deems appropriate for ICC scrutiny.80
V. Substantive and Institutional Issues
A. The Contours of ‘Act’ and ‘Crime’ of Aggression
Turning to more substantive issues, the contours of what constitutes an ‘act’ and ‘crime’ of aggression are not well defined in the Kampala amendments. As noted previously, the ‘act of aggression’ in Article 8bis essentially is equated with a violation of Article 2(4) of the UN Charter, and further illuminated by the types of acts identified in the General Assembly’s 1974 resolution.81 The ‘crime of aggression’ is the ‘planning, preparation, initiation or execution’ of an act of aggression by a senior leader, but only in situations where the act ‘by its character, gravity and scale’ constitutes a ‘manifest violation’ of the UN Charter.82 Since these same definitions had been developed by the Special Working Group in advance of Kampala, many commentators raised concerns even at that time about the uncertain and difficult line-drawing that arises from the vagueness of the definitions. Some of the concerns related to the principle of nullum crimen sine lege, which requires that a criminal law be reasonably clear to a defendant at the time he commits his allegedly wrongful conduct,83 while other concerns focused on the institutional burdens that such definitions may place upon the ICC.84
(p. 553) Two of the Understandings adopted at Kampala are responsive to such concerns. One provides that ‘aggression is the most serious and dangerous form of illegal use of force’, and a determination that aggression has occurred ‘requires consideration of all the circumstances of each particular case, including the gravity of the acts concerned and their consequences, in accordance with the Charter of the United Nations.’85 The other clarifies that determining the existence of a ‘manifest’ violation of the Charter requires findings with respect to each of the three elements identified in the definition of the crime: character, gravity, and scale.86 Specifically, the Understanding provides that ‘No one component can be significant enough to satisfy the manifest standard by itself.’
Such clarifications are very useful, but some very basic questions remain. For example, it is unclear exactly what kinds of action, by their ‘character, gravity, and scale’, rise to the level of a ‘manifest violation’ of the UN Charter. One possibility is that the standard will be interpreted as encompassing only unlawful uses of force undertaken at a very high level of coercion. If so, then the unprovoked and massive invasion by one state of another state would presumably fall within the scope of the crime of aggression, such as Iraq’s August 1990 invasion of Kuwait. By contrast, the sinking by a North Korean submarine of a South Korean warship, the Cheonan, in March 2010, even though resulting in the death of 46 sailors, might not fall under the crime of aggression if the Court concluded that—even though such conduct might violate the UN Charter—the gravity and scale of the conduct were insufficient.87
If the ‘manifest violation’ standard is interpreted in this way, there may be few prosecutions for crimes of aggression before the ICC since aggression of that scale very rarely happens. For that reason, such an interpretation might be attractive to the ICC Prosecutor and judges, at least initially, as it may allow them to avoid the ICC becoming entangled in numerous incidents of alleged aggression. At the same time, the non-prosecution of cases under such a high standard might have the unfortunate effect of sub silentio condoning lesser uses of force. Every time the ICC announces that a particular transnational use of force does not fall within the scope of its jurisdiction over the crime of aggression (eg the sinking of the Cheonan), the state whose conduct is in question may seek to exploit that decision to promote the idea that its conduct was not wrongful.88
(p. 554) Another possibility is that the ‘manifest violation’ standard will be interpreted as emphasizing not the magnitude of the coercion but, rather, whether regarding the conduct as unlawful is clear or obvious to all relevant observers.89 Under such a standard, perhaps both Iraq’s invasion of Kuwait and North Korea’s sinking of the Cheonan would be viewed as ‘manifest’ violations of the UN Charter, since they were generally condemned by states as unlawful, whereas more contested actions would not be viewed as ‘manifest’ violations. Thus, NATO’s bombing campaign against Serbia in 1999 might not fall within the Kampala definition of an act of aggression because ‘reasonable’ people disagreed about whether that intervention was lawful, with some arguing that it was permissible in order to protect the fundamental human rights of Kosovar Albanians. Such an interpretation may also have an attraction for the ICC Prosecutor and Judges, as it would allow for ICC action in a range of situations large and small, but only when, as a political matter, virtually the entire international community views the conduct as unlawful. Such an approach might trigger highly spirited, public debates by states over the legality of the relevant conduct, as a means of signalling to the Court their views on the matter. A problem with this approach, however, is that it casts the Court in the role of bending to the political will of the international community, rather than adhering to more objective standards of law. Further, since there will often be considerable differences of views about the legality of a use of force, here too the Court’s jurisdiction may rarely be triggered.90
A third possibility is that the ‘manifest violation’ standard will be viewed as encompassing less coercive and less blatantly unlawful transnational uses of force. Here, ‘manifest’ might be viewed as excluding small-scale skirmishes or ‘frontier incidents’, but other uses of force could constitute a crime of aggression. If so, then the ICC may be confronted with passing upon some very difficult factual scenarios. In addition to the Kosovo and North Korea incidents, the ICC might need to assess incidents such as the intervention in Liberia of the Economic Community of West African States (ECOWAS) in 1990, the series of wars that has plagued the Great Lakes region of Africa for the past 20 years, Ethiopia’s 2007 and 2011 interventions (p. 555) in Somalia (which highlight the difficulty of assessing an ‘invitation’ for intervention by an authority that no longer governs the relevant territory), or Colombia’s 2008 attack in Ecuador upon guerrillas of the Revolutionary Armed Forces of Colombia (FARC), to name just a few. Even in a situation where the Security Council has authorized a use of force, there may well be calls for the ICC to assess the circumstances, such as whether NATO’s air campaign against the government of Libya in 2011 constituted an act of aggression because it exceeded the authorization issued by the UN Security Council. At the time of all these actions, there were differing views about their permissibility under the UN Charter, with the positions taken by governments and observers often driven more by political concerns than by well-defined legal criteria.91
B. Institutional Integrity of the Court
The typical answer to the dilemma of imprecise definitions seems to be that those definitions will be refined in practice by the ICC Prosecutor and Judges (at least absent Security Council involvement), who by prosecuting and convicting defendants (or by not doing so) will sketch out over time what is meant by an ‘act’ and ‘crime’ of aggression. Considerable uncertainties with respect to the Nuremberg and Tokyo Charters were refined in practice, including with respect to crimes against the peace.
Even so, refinement of these concepts will pose a considerable challenge for the ICC.92 How exactly can prosecutors and judges form a view as to whether a state is acting in self-defence, in the often complicated context of inchoate threats, indirect action, and reactions that span time and space? How does the ICC gauge whether that defence was necessary or proportionate? Whether that defence was properly undertaken in anticipation of an attack? Whether a use of force to protect human rights is consistent with the UN Charter? What legal standards, of the kind necessary for pursuing criminal charges, will the Court rely upon?
Given the lack of bright lines in this area, given the extraordinary publicity attendant to transnational uses of force, and given the inescapable focus on senior government officials with respect to the crime of aggression, it appears possible that the Court will be an enormous lightning rod for intense political scrutiny with (p. 556) respect to any transnational use of force that falls within the scope of its jurisdiction. To the extent that some observers think the current ICC case against Sudan’s President Omar al-Bashir or Kenya’s President Uhuru Kenyatta, and the adverse reaction of many states parties to those cases,93 is damaging the Court, one can imagine the same scenario potentially playing out multiple times in the context of demands for criminal charges against sitting heads of state or government across the globe for aggression.
C. Collateral Effects on the Jus ad Bellum
Adoption of the definitions on ‘act’ and ‘crime’ of aggression may have collateral implications outside the criminal context, especially on rules relating to the jus ad bellum. As previously noted, the definition of ‘act of aggression’ is equated with any use of force in violation of the UN Charter. As such, the notion in the jus ad bellum of ‘aggression’ as a particularly egregious violation of Article 2(4)—worse than a threat to or breach of the peace94—may be diminished, allowing for less variegation in methods of condemnation and response. Cognizant of such possibilities, the states parties at Kampala adopted an understanding which may blunt the crossover effects of these definitions upon general international law:
It is understood that the amendments that address the definition of the act of aggression and the crime of aggression do so for the purpose of this Statute only. The amendments shall, in accordance with article 10 of the Rome Statute, not be interpreted as limiting or prejudicing in any way existing or developing rules of international law for purposes other than this Statute.95
Even so, the long-term practice of the ICC in prosecuting or not prosecuting particular conduct as a ‘crime of aggression’ may affect conventional understandings as to what are permissible and impermissible uses of force. As suggested earlier, adoption of a high standard for what constitutes a crime of aggression within the jurisdiction of the ICC might serve to condone lesser uses of unlawful force. A low standard for what constitutes a crime of aggression could end up deterring low levels of undesirable coercion, but it might also inhibit lawful uses of force that help to keep aggressors in check. Since the line dividing permissible force and impermissible aggression is not clear, any deterrent effect upon wrongful uses of force might affect lawful uses of force, at least on the margin where a state is contemplating using force to help another state or to end atrocities.
(p. 557) For example, assume that NATO’s 1999 intervention was lawful, but Ethiopia’s 2007 intervention in Somalia was not. Both NATO and Ethiopia asserted that they were intervening lawfully and for good purposes. But how are the leaders of the two interventions to know that their use of force is permissible or impermissible under the criminal standards of the ICC? Assuming that there is a deterrent effect, then, in the absence of bright lines, both uses of force may well be deterred, even though only one of them is unlawful. The worst-case scenario would be if unlawful uses of force are not deterred by the ICC’s jurisdiction (on a theory that law is simply not effective against aggressor governments), while lawful uses of force are deterred.
D. Effects upon the Exercise of National Jurisdiction
A different concern arises with respect to the effects of the amendments upon the exercise of national jurisdiction. The Rome Statute is predicated on the notion that the ICC is not the primary avenue for prosecutions. Rather, states parties are expected to prosecute the crimes identified in the Rome Statute and the ICC is only ‘complementary’ to that jurisdiction by acting when a given state party in a given situation is unable or unwilling to act.96
Activation of the crime of aggression may have the effect of encouraging states parties to criminalize aggression in their national laws. While in the abstract that may be regarded as an important means for stigmatizing aggression, making it less likely that a leader will resort to it, there may also be adverse repercussions. If such conduct is criminalized in national law then, when a conflict erupts, each side might have no political choice but to indict and prosecute in absentia the leaders of the other side for aggression. Doing so may serve simply to raise the stakes for both sides, making it harder for them to find a political solution to the crisis. To avoid self-interested prosecutions, more impartial third states might be expected to pursue the prosecutions, but third states may not wish to assume that role.
Perhaps with that in mind, the Kampala Review Conference adopted an Understanding that ‘the amendments shall not be interpreted as creating the right or obligation to exercise domestic jurisdiction with respect to an act of aggression committed by another State.’97 That Understanding might help in discouraging states from adopting national laws that allow for the prosecution of foreign leaders, leaving it more likely that, if such laws are adopted, they will focus only on potential prosecution of domestic leaders. However, the Understanding does not preclude national laws that encompass crimes of aggression by foreign leaders, and a state may be inclined to adopt such a law if it is thought necessary to maintain ICC (p. 558) deference to its courts in situations that involve a mixture of ICC crimes.98 If such laws are enacted, they may have the unfortunate effect of entrenching nationalist passions within relatively inflexible legal proceedings, making it harder to resolve inter-state conflict.
E. Institutional Role of the Security Council
Finally, there are concerns about the effects of this new ICC jurisdiction upon the Security Council. Arguments that the Council should be the exclusive ‘trigger’ for the ICC’s jurisdiction on the crime of aggression failed to garner sufficient support in Kampala. The Council has a role under both Article 15bis and Article 15ter, but not an exclusive role.
One possibility is that the outcome at Kampala will help to promote a partnership between the Council and the ICC. The Council has in two cases regarded the ICC as a useful means for addressing certain conflicts (Darfur in 200599 and Libya in 2011100). In addition to its existing options for responding to aggression, such as imposing economic sanctions or authorizing the use of military force, there will now be an option for the Council to refer a situation of possible aggression to the ICC under Article 15ter. Perhaps the ease with which the Council may refer a situation involving aggression to the ICC will play a role in helping to deter aggression, though it remains unclear whether the Council’s current ability to refer situations involving non-aggression crimes has helped to deter such crimes.
One uncertainty is whether, after the ICC’s ‘aggression’ jurisdiction is activated, the Council can limit a referral (legally or politically) to just crimes other than aggression, or whether the referral must allow the ICC to pursue all possible crimes relating to that situation. If the latter is the case, then perhaps establishment of Article 15ter will have the unfortunate effect of inhibiting Security Council referrals that would have occurred prior to activation of ‘aggression’ jurisdiction, if the Council cannot agree on allowing the ICC to pursue the crime of aggression for that particular situation.
Another unfortunate outcome would be for the Council to refer a situation to the ICC simply to avoid otherwise dealing with the matter. Confronted with a difficult situation of transnational armed conflict, the Council might find it attractive to send the matter to the ICC so as to simply move it off the Council’s agenda—‘burying’ it in a years-long legal process at the ICC. In other words, it is possible that the Council might use the referral process as a means of escaping its responsibility to address aggression.
(p. 559) Further problems may arise if the Council and the ICC are acting not as partners, but at cross-purposes, such as when the Council is actively engaged in trying to resolve an armed conflict and the ICC injects itself uninvited into the situation by means of Article 15bis. The Council might have decided not to declare a certain crisis as involving an act of aggression, perhaps out of a sense that it would aggravate the situation, only to have the ICC bring charges against senior leaders involved in the crisis for the crime of aggression. For example, there is some support for the proposition that when NATO commenced its bombing campaign of Serbia in 1999, NATO leaders contemplated simply extracting a bilateral settlement with President Milošević not unlike that which occurred in 1995 with the Dayton Accords. However, in the midst of the bombing campaign, the ICTY indicted several senior Serbian leaders, including Milošević. Marc Weller has suggested that the issuance of the indictment during the hostilities ‘triggered a strategic shift’, one in which the nature of the hostilities transformed from mere ‘coercive diplomacy’ into outright ‘war’,101 a transformation not anticipated by NATO states. According to Weller, after issuance of the indictment
President Slobodan Milosevic, the most prominent of the indictees, and his close associates were no longer an indispensable element to a resolution of the Kosovo conflict. Instead, they were to be considered as probable war criminals that needed to be defeated militarily. Indeed, this fact was reflected in the fact that at the end of the conflict, no peace settlement as such was concluded with the Belgrade leadership. Instead, the peace terms were established by way of…Security Council resolution…102
If that is correct, then a further worry is that the Security Council’s ability to manage a situation of armed conflict, which may already be compromised by existing ICC jurisdiction, might be even more affected through further expansion of the ICC’s jurisdiction. Though unlikely, it is even possible that, when a war breaks out between states A and B, the Council might declare that state A has committed an act of aggression, while the ICC brings charges against leaders of state B for a crime of aggression. Article 15bis expressly provides that a ‘determination of an act of aggression by an organ outside the Court shall be without prejudice to the Court’s own findings under this Statute.’103 This problem may be ameliorated by the fact that some states will not be exposed to the ICC’s jurisdiction. It might also benefit from the Security Council’s ability to postpone, on a year-by-year basis, action by the ICC with respect to a particular situation.104 But relying on Council deferral forces the Council to extend a benefit to the aggressor (by turning off ICC jurisdiction) right at the point where the Council is attempting to place pressure on that aggressor, which may prove to be a difficult if not impossible balancing act. And, since the (p. 560) leaders of the aggressor can never be certain that the Council will postpone prosecutions indefinitely, they will have to react in a manner that discounts the certainty of such deferrals.
Whether the activation of jurisdiction at the ICC over the crime of aggression ultimately serves to deter aggression or to punish those who initiate it remains to be seen.105 For now, despite its successes, there emerged from Kampala considerable uncertainties concerning procedural and substantive aspects of how the crime of aggression at the ICC will actually operate. Prior to 2017, there remain opportunities for resolving some of these issues. Greater clarity as to the procedural aspects of the Court’s jurisdiction may occur as states parties ratify or accept the amendments, as some opt out of the jurisdiction, and as other states parties that do not ratify or accept the amendments make known their views as to the effects of non-ratification. Further, at the meeting of the Assembly of States Parties after 2016, there will be an opportunity in the course of its decision to clarify matters further. As such, those interested in the effective functioning of the ICC’s jurisdiction over the crime of aggression, and in the efficacy of international norms on the use of force generally, should not view Kampala as the final word on the crime of aggression, but as an opportunity to continue to grapple with the very real and very challenging issues that still remain.
(1) Rome Statute of the International Criminal Court, 17 July 1998, 2187 UNTS 3.
(2) The states parties to the Rome Statute may be found at: <http://www.icc-cpi.int/en_menus/asp/states%20parties/Pages/the%20states%20parties%20to%20the%20rome%20statute.aspx/>.
(3) For the sake of simplicity, this chapter will refer to a state (or state party) being exposed to the ICC’s jurisdiction over the crime of aggression, though what is actually meant is exposure for a national of that state or a person accused of committing the crime on the territory of that state.
(4) See eg Randall Lesaffer, ‘Too Much History: From War as Sanction to the Sanctioning of War’, Chapter 1 in this volume, at 35; Nico Schrijver, ‘The Ban on the Use of Force in the UN Charter’, Chapter 21 in this volume, at 465; Telford Taylor, The Anatomy of the Nuremberg Trials: A Personal Memoir (Boston, MA: Little Brown & Co, 1992).
(5) Treaty of Versailles, Arts 227–30, in Lawrence Martin (ed), I Treaties of Peace, 1919–1923 (New York: Carnegie Endowment for International Peace, 2007), 121–2.
(6) See Claud Mullins, The Leipzig Trials (Livingston: H. F. & G. Witherby, 1921); Kirsten Sellers, ‘Delegitimizing Aggression: First Steps and False Starts after the First World War’ (2012) 10 Journal of International Criminal Justice 7. The Kaiser died in the Netherlands in 1941.
(7) Treaty of Versailles, Art 227 in Treaties of Peace, 121.
(8) General Treaty for the Renunciation of War, 27 Aug 1928, 46 Stat 2343, 94 LNTS 57.
(9) Page Wilson, Aggression, Crime and International Security: Moral, Political and Legal Dimensions of International Relations (Abingdon: Routledge, 2009), 55; see at 50 (‘the UK, Soviet Union and France raised many of the same objections to the criminality of aggression which had been played out earlier within US bureaucratic circles’).
(10) UN Conference on International Organisation 104 (4 Apr 1945). The ‘enforcement arrangements’ committee at San Francisco was a technical committee set up under Commission III, which addressed Security Council measures.
(11) See Agreement for the Prosecution of the Major War Criminals of the European Axis Powers and Charter of the International Military Tribunal, Charter, Art 6(1), 8 Aug 1945, 82 UNTS 279.
(12) Wilson, Aggression, Crime and International Security, 52; see generally George Ginsburgs and Vladimir N. Kudriavtsev (eds), The Nuremberg Trial and International Law (Leiden: Martinus Nijhoff, 1990).
(13) International Military Tribunal Proceedings, vol XXII, 433, 536, 555.
(14) International Military Tribunal Proceedings, vol XXII, 427, 445–58, 562. At the first trial, the tribunal did not address whether the wars against France and the UK were aggression.
(15) Thomas Weigend, ‘“In General a Principle of Justice”: The Debate on the “Crime against Peace” in the Wake of the Nuremberg Judgment’ (2012) 10 Journal of International Criminal Justice 41, 43.
(16) International Military Tribunal Proceedings, vol XXII, 467–8.
(17) International Military Tribunal Proceedings, vol XXII, 554–7.
(18) See Special Proclamation: Establishment of an International Military Tribunal for the Far East, in The Tokyo War Crimes Trial, Vol I: Pre-Trial Documents, Transcript of the Proceedings in Open Session, 1–2.
(19) See Yuma Totani, The Tokyo War Crimes Trial: The Pursuit of Justice in the Wake of World War II (Cambridge, MA: Harvard University Press, 2009); Neil Boister and Robert Cryer, The Tokyo International Tribunal: A Reappraisal (Oxford: Oxford University Press, 2008); Timothy P. Maga, Judgment at Tokyo: The Japanese War Crimes Trials (Lexington, KY: University Press of Kentucky, 2001).
(20) GA Res 95 (I), A/236 (1946), 1144.
(21) ‘Principles of International Law Recognized in the Charter of the Nürnberg Tribunal and in the Judgment of the Tribunal, with Commentaries’, Report of the International Law Commission on the Work of its Second Session, UN GAOR, 5th Sess, Supp No 12, A/1316 (1950), reprinted in Yearbook of the International Law Commission, 1950, vol II (2), 374.
(22) Definition of Aggression, GA Res 3314 (XXIX) (1974); see Bengt Broms, ‘The Definition of Aggression’ (1977-I) 154 Recueil des cours de l’Académie de droit international de la Haye 348; Julius Stone, ‘Holes and Loopholes in the 1974 Definition of Aggression’ (1977) 71 American Journal of International Law 224.
(23) GA Res 3314 (XXIX), Art 1.
(24) GA Res 3314 (XXIX), Art 2.
(25) GA Res 3314 (XXIX), Art 3 (‘Any of the following acts, regardless of a declaration of war, shall, subject to and in accordance with the provisions of article 2, qualify as an act of aggression…’) (emphasis added).
(26) Report of the International Law Commission on the Work of its Forty-Sixth Session, A/49/10 (1994), 38.
(27) ILC, Report 1994, Art 5(1) (emphasis added).
(28) Yoram Dinstein, War, Aggression and Self-Defence (5th edn, Cambridge: Cambridge University Press, 2011), 135.
(29) See SC Res 674 (29 Oct 1990), paras 1–2 (referring to Iraq’s accountability for war crimes and calling upon states to collect information regarding such conduct).
(30) SC Res 827 (1993); see Jean-Paul Bazelaire and Thierry Cretin, La justice internationale, son évolution, son avenir, de Nuremberg à La Haye (Paris: Presses universitaires de France, 2000).
(31) SC Res 955 (1994).
(32) See Guénaël Mettraux, International Crimes and the Ad-Hoc Tribunals (Oxford: Oxford University Press, 2006); Cesare P. R. Romano et al (eds), Internationalized Criminal Courts and Tribunals: Sierra Leone, East Timor, Kosovo, and Cambodia (Oxford: Oxford University Press, 2004).
(33) See Mahnoush H. Arsanjani, ‘The Rome Statute of the International Criminal Court’ (1999) 93 American Journal of International Law 22, 29–30.
(34) Rome Statute, Art 5.
(35) Rome Statute, Art 123.
(36) For a compendium, see Stefan Barriga, Wolfgang Danspeckgruber, and Christian Wenaweser (eds), The Princeton Process on the Crime of Aggression: Materials of the Special Working Group on the Crime of Aggression, 2003-2009 (Boulder, CO: Lynne Rienner, 2009).
(37) For a compendium of documents concerning the negotiations leading up to and at Kampala, see Stefan Barriga and Claus Kreß (eds), The Travaux Préparatoires of the Crime of Aggression (Cambridge: Cambridge University Press, 2012).
(38) International Criminal Court Assembly of States Parties, Review Conference, Resolution RC/Res 6 (11 June 2010) (hereinafter ‘RC/Res 6’).
(39) RC/Resolution 6, Annex I, para 1 (‘Article 5, paragraph 2, of the Statute is deleted’). The final documents of the Review Conference may be found at (2010) 49 ILM 1325 and in Barriga and Kreß, The Travaux Préparatoires of the Crime of Aggression, 99.
(40) RC/Res 6, Annex I, Art 8bis, para 2.
(41) RC/Res 6, Annex I, Art 8bis, para 1.
(42) For a discussion of the alteration at Kampala of the 1974 language, see Stefan Barriga, ‘Negotiating the Amendments on the Crime of Aggression’ in Barraga and Kreß, The Travaux Préparatoires of the Crime of Aggression, 3, 26.
(43) See Roger S. Clark, ‘Amendments to the Rome Statute of the International Criminal Court Considered at the first Review Conference on the Court, Kampala, 31 May–11 June 2010’ (2010) 2 Goettingen Journal of International Law 689, 696–7.
(44) Mauro Politi, ‘The ICC and the Crime of Aggression’ (2012) 10 Journal of International Criminal Justice 267, 285–7.
(45) RC/Res 6, Annex I, Art 15bis.
(46) RC/Res 6, Annex I, Art 15ter.
(47) RC/Res 6, Annex I, para 1.
(48) RC/Res 6, Annex I, para 1.
(49) Rome Statute, Art 121, para 5.
(50) See Beth Van Schaack, ‘Negotiating at the Interface of Power and Law: The Crime of Aggression’ (2011) 49 Columbia Journal of Transnational Law 505, 531 (‘the legislative history of the ICC Statute somewhat favoured considering Article 121(5) as the lex specialis for any amendments to the substantive criminal provisions’). For a cogent discussion of the ambiguities that existed in the Rome Statute prior to Kampala for how to bring into operation the crime of aggression, see Roger S. Clark, ‘Ambiguities in Articles 5(2), 121 and 123 of the Rome Statute’ (2009) 41 Case Western Reserve Journal of International Law 413.
(51) For information on the states that have ratified the amendments, see <http://treaties.un.org/pages/ViewDetails.aspx?src=TREATY&mtdsg_no=XVIII-10-b&chapter=18&lang=en>.
(52) Rome Statute, Art 121, para 5.
(53) RC/Res 6, Annex I, Art 15bis, para 2; Art 15ter, para 2.
(54) RC/Res 6, Annex I, Art 15bis, para 3; Art 15ter, para 3.
(55) RC/Res 6, Annex III, Understanding No 3.
(56) For an analysis of the legal effects of the understandings, finding that they are supplementary means of interpretation, see Kevin Jon Heller, ‘The Uncertain Legal Status of the Aggression Understandings’ (2012) 10 Journal of International Criminal Justice 229.
(57) Rome Statute, Art 121, para 5.
(58) At Kampala, this interpretation was referred to as the ‘negative understanding’.
(59) See Andreas Zimmermann, ‘Amending the Amendment Provisions of the Rome Statute’ (2012) 10 Journal of International Criminal Justice 209, 217.
(60) Rome Statute, Art 124.
(61) Zimmermann, ‘Amending the Amendment Provisions of the Rome Statute’, 217.
(62) Vienna Convention on the Law of Treaties, Art 40(4), 1155 UNTS 331.
(63) Zimmermann, ‘Amending the Amendment Provisions of the Rome Statute’, 210–11.
(64) See RC/Res 6, Annex I, Art 15bis, para 4.
(65) See Claus Kreß and Leonie von Holtzendorff, ‘The Kampala Compromise on the Crime of Aggression’ (2010) 8 Journal of International Criminal Justice 1179, 1197–8; Van Schaack, ‘Negotiating at the Interface of Power and Law’, 528.
(66) Art 15bis provides that the Court may not exercise jurisdiction over a crime arising from the conduct of that state party if the ‘State Party has previously declared that it does not accept such jurisdiction by lodging a declaration with the Registrar’ of the ICC. RC/Res 6, Annex I, Art 15bis, para 4.
(67) Robert Heinsch, ‘The Crime of Aggression After Kampala: Success or Burden for the Future?’ (2010) 2 Goettingen Journal of International Law 713, 739. (‘In the end, one could get the impression that paragraph 4 has been hastily inserted in Article 15bis without bringing it completely in coherence with the articles dealing with the amendment procedure’); see Barriga, ‘Negotiating the Amendments’, 53–7.
(68) See van Schaack, ‘Negotiating at the Interface of Power and Law’, 586.
(69) See eg Kreß and von Holtzendorff, ‘The Kampala Compromise on the Crime of Aggression’, 1198 (‘one can hardly deny that the “positive understanding” of [para 121]’s second sentence does not naturally flow from its wording’); Politi, ‘The ICC and the Crime of Aggression’, 280 (‘The main issue is whether the States Parties [at Kampala] were legally entitled to establish a new regime that would, in substance, modify the provisions in Article 121 without having recourse to the formal procedure for amending the Statute’).
(70) See Politi, ‘The ICC and the Crime of Aggression’, 281; Kreß and von Holtzendorff, ‘The Kampala Compromise on the Crime of Aggression’, 1212.
(71) International Criminal Court Assembly of States Parties, Review Conference, Resolution RC/Res 5 (11 June 2010).
(72) Kreß and von Holtzendorff, ‘The Kampala Compromise on the Crime of Aggression’, 1212–16; Politi, ‘The ICC and the Crime of Aggression’, 278–83.
(73) See Dapo Akande, ‘Prosecuting Aggression: The Consent Problem and the Role of the Security Council’, Oxford Legal Studies Research Paper No 10/2011, 15–17 (Feb 2011), available at <http://ssrn.com/abstract=1762806> (arguing that in the lead-up to Kampala, because a determination that an individual has committed the crime of aggression requires a prior determination that a state has committed an act of aggression and a breach of the UN Charter, the ICC would act in violation of the consent principle in cases where the alleged aggressor state has not provided prior consent to the Court’s jurisdiction); see also Report of the International Law Commission on the Work of its Forty-Eighth Session, A/51/10 (1996), 30 (‘An individual cannot incur responsibility for this crime in the absence of aggression committed by a State. Thus, a court cannot determine the question of individual criminal responsibility for this crime without considering as a preliminary matter the question of aggression by a State’).
(74) Kreß and von Holtzendorff, ‘The Kampala Compromise on the Crime of Aggression’, 1213–14.
(75) RC/Res 6, Annex I, Art 15bis, para 4.
(76) RC/Res 6, para 1.
(77) RC/Res 6, Annex I, Art 15bis, para 5.
(78) See Monetary Gold Removed from Rome in 1943 (Italy v. France, United Kingdom of Great Britain and Northern Ireland, United States of America), Preliminary Question, ICJ (15 June 1954), 19 (absence of Albania precluded a decision by the Court); East Timor (Portugal v. Australia), ICJ (30 June 1995), 90 (absence of Indonesia precluded decision by the Court); see also Advisory Opinion on the Status of Eastern Carelia, PCIJ, Ser B, No 5 (23 July 1923) (absence of Russia precludes provision of advisory opinion); but see Certain Phosphate Lands in Nauru (Nauru v. Australia), Preliminary Objections, ICJ (26 June 1992), 240, 255–62 (no bar to the proceedings unless the Court has to determine the responsibility of the non-consenting state as a prerequisite to determining the claims before it).
(79) RC/Res 6, Annex III, Understanding No 2.
(80) Clark, Amendments to the Rome Statute, 702–3.
(81) RC/Res 6, Annex I, Art 8bis, para 2.
(82) RC/Res 6, Annex I, Art 8bis, para 1.
(83) See Andreas Paulus, ‘Second Thoughts on the Crime of Aggression’ (2010) 20 European Journal of International Law 1117, 1123; Michael Glennon, ‘The Blank-Prose Crime of Aggression’ (2010) 35 Yale Journal of International Law 71; Sean D. Murphy, ‘Aggression, Legitimacy and the International Criminal Court’ (2009) 20 European Journal of International Law 1147; but see Marko Milanovic, ‘Aggression and Legality: Custom in Kampala’ (2012) 10 Journal of International Criminal Justice 165, 170–1, 184 (finding the definition vague but not so vague as to violate the principle of legality); Heinsch, ‘The Crime of Aggression After Kampala’, 724–6 (same).
(84) See eg Sean D. Murphy, ‘Criminalizing Humanitarian Intervention’ (2009) 41 Case Western Reserve Journal of International Law 241; David Scheffer, ‘A Pragmatic Approach to Jurisdictional and Definitional Requirements for the Crime of Aggression in the Rome Statute’ (2009) 41 Case Western Reserve Journal of International Law 397; Claus Kreß, ‘The Crime of Aggression Before the First Review of the ICC Statute’ (2007) 20 Leiden Journal of International Law 851; Mauro Politi and Giuseppe Nesi (eds), The International Criminal Court and the Crime of Aggression (Farnham: Ashgate, 2004); Andreas Paulus, ‘Peace Through Justice? The Future of the Crime of Aggression in a Time of Crisis’ (2004) 50 Wayne Law Review 1.
(85) RC/Res 6, Annex III, Understanding No 6.
(86) RC/Res 6, Annex III, Understanding No 7.
(87) On 20 May, South Korea formally accused North Korea of sinking the Cheonan, based on an investigation that it conducted in conjunction with Australia, Canada, Sweden, and the UK. See Choe Sang-Hun, ‘South Korea Publicly Blames the North for Ship’s Sinking’, New York Times, 19 May 2010, A1; Choe Sang-Hun, ‘North Korea Denies Sinking Navy Ship’, New York Times, 17 Apr 2010, A8.
(88) See Mary Ellen O’Connell and Mirakmal Niyazmatov, ‘What is Aggression?’ (2012) 10 Journal of International Criminal Justice 189, 191 (‘it is imperative that what international law prohibits as aggression not be undermined by the political realities of the Review Conference’); but see Heinsch, ‘The Crime of Aggression After Kampala’, 731 (‘There is no danger that the prohibition of the use of the use of force laid down in Article 2(4) of the UN Charter will be undermined by this construction. Rather, any kind of penalisation of only a certain (manifest) form of aggression will in the long run strengthen the general norm as well’).
(89) The amended ‘elements of crimes’ provide that the ‘term “manifest” is an objective qualification.’ RC/Res 6, Annex II, Introduction, para 3. See James Potter, ‘The Threshold in the Proposed Definition of the Crime of Aggression’ (2008) 6 New Zealand Yearbook of International Law 155; Heinsch, ‘The Crime of Aggression After Kampala’, 730 (‘If one has a look at the travaux préparatoires, it becomes clear that the idea behind this qualifier is to exclude all violations of the prohibition of the use of force which are controversial’).
(90) Paulus, ‘Second Thoughts on the Crime of Aggression’, 1121 (‘What…is obvious for one is completely obscure to the other, in particular in international law’).
(91) According to Kreß, ‘The Crime of Aggression Before the First Review of the ICC Statute’, 858 (‘One essential albeit regrettable aspect of this reality is the existence of a grey area in the international legal framework. Reasonable international lawyers may legitimately disagree in their assessment of the lex lata for this crime, depending inter alia on how recent international practice is seen and weighed’).
(92) Politi, ‘The ICC and the Crime of Aggression’, 284 (‘the task that the Chambers are due to face in dealing with the questions raised by an alleged state aggression appear to be quite complex both in substance and at different procedural junctures, especially when determinations by the Security Council or the [ICC’s Pre-Trial Division] have taken place’).
(93) See eg Neil MacFarquhar and Marlise Simons, ‘Bashir Defies War Crime Arrest Order’, New York Times, 6 Mar 2009, A10 (reporting that ‘China joined the African Union and the Arab League in calling on the United Nations Security Council to use its powers to suspend the [al-Bashir] indictment for a year, lest it stymie peace talks between the government and the rebels in Darfur’).
(94) See eg UN Charter, Art 39.
(95) RC/Res 6, Annex III, Understanding No 4.
(96) Rome Statute, Preamble and Arts 1, 17; see Mohammed El Zeidy, The Principle of Complementarity in International Criminal Law (Leiden: Martinus Nijhoff, 2008).
(97) RC/Res 6, Annex III, Understanding No 5.
(98) See Beth Van Schaack, ‘Par in Parem Imperium Non Habet: Complementarity and the Crime of Aggression’ (2012) 10 Journal of International Criminal Justice 133.
(99) SC Res 1593 (2005).
(100) SC Res 1970 (2011).
(101) Marc Weller, Contested Statehood: Kosovo’s Struggle for Independence (Oxford: Oxford University Press, 2009), 167.
(102) Weller, Contested Statehood, 74.
(103) RC/Res 6, Annex I, Art 15bis, para 9.
(104) RC/Res 6, Annex I, Art 15bis, paras 8 and 16.
(105) Marko Milanovic, ‘Aggression and Legality: Custom in Kampala’ (2012) 10 Journal of International Criminal Justice 165, 166 (‘It may turn out to be nothing more than a diplomatic dud, a waste of everybody’s time. The ICC’s ponderous progress in dealing even with “ordinary” crimes in its jurisdiction does not inspire much confidence’).