The Yale Law Journal

January 2016

Reconciling the Crime of Aggression and Complementarity: Unaddressed Tensions and a Way Forward

International Law

abstract. In June 2010, after more than a decade of negotiation, the Assembly of States Parties to the Rome Statute of the International Criminal Court (ICC) agreed on a definition of the crime of aggression. But the Assembly failed to address a critical issue: whether and how prosecutions for aggression fit within the ICC’s complementarity regime. That regime positions the ICC as a court of last resort; domestic prosecutions are preferred except in a very narrow set of circumstances. Thus, an aggressed state could conceivably prosecute, in its own courts, the nationals of an aggressor state. This Note argues that the crime of aggression is a poor fit with the general complementarity regime, creating major tensions. Policymakers have so far overlooked the problem, despite its potential implications for international peace and security, as well as the ICC’s legitimacy. On the one hand, domestic prosecutions of the crime of aggression pose serious risks to international political stability. At the same time, bringing aggression prosecutions only at the ICC would create its own difficulties, such as incomplete concurrence between proceedings at the domestic level and proceedings at the ICC. Because the aggression amendments to the Rome Statute will take effect in 2017, policymakers must address the complementarity question, and soon. After diagnosing the dilemma, this Note offers several recommendations for a way forward.

author.Yale Law School, J.D. 2015. I am deeply grateful to Harold Koh, an inspiring teacher who introduced me to this topic, encouraged me to develop the project into a Note, and provided invaluable comments and enthusiastic support along the way. Many thanks are also due to Beth Van Schaack, who gave generous feedback at critical stages of the writing process. Finally, I thank Nick Crown, Rebecca Lee, and Michael Clemente for their thoughtful and careful editorial support.


In 2010, the Assembly of States Parties (ASP)1 to the Rome Statute of the International Criminal Court (ICC) finally agreed on a definition for the crime of aggression after years of deliberations.2 The ASP adopted the Rome Statute, the treaty creating the ICC, in 1998, but the crime of aggression proved too contentious. The ASP failed to reach a consensus on its definition or jurisdictional regime. Rather than allow that division to derail the entire treaty, the ASP punted on the issue of aggression by writing a promise into the Rome Statute that it would eventually return to it.3 Until that time, the ICC could exercise jurisdiction only over war crimes, crimes against humanity, and genocide.4 More than a decade passed before the ASP, meeting in Kampala, Uganda, fulfilled its promise and amended the Rome Statute to define the crime of aggression and the conditions for the ICC’s jurisdiction over it.

The Kampala Amendments to the Rome Statute defined the crime of aggression as “the planning, preparation, initiation or execution” by a high-level official “of an act of aggression which, by its character, gravity and scale, constitutes a manifest violation of the Charter of the United Nations.”5 When committed by the armed forces of one state against another state, examples of acts that would qualify as “aggression” include invasion or attack, bombardment, blockade of ports or coasts, and sending mercenaries to carry out acts of armed force.6 The ASP also agreed to jurisdictional rules governing the crime and adopted a set of Understandings7 touching on Security Council referrals, temporal jurisdiction, domestic jurisdiction, and the nature of an aggression determination.8

There were many difficult issues on the table in Kampala, as one might expect from a gathering that followed more than twelve years of debate. Yet the ASP devoted scant attention to a critically important issue: whether and how aggression fits within the ICC’s complementarity regime. Complementarity positions the ICC as a court of last resort. Domestic prosecutions are strongly preferred over ICC prosecutions in all but a very narrow set of circumstances.

The crime of aggression could give rise to three types of domestic prosecutions. First, a state could prosecute its own nationals, such as the principals of a former regime. This type of prosecution is unlikely to provoke the concerns that animate this Note. Second, a state with no real connection to an act of aggression could prosecute under extraordinary bases of jurisdiction, such as universal jurisdiction. This type of prosecution has the potential to cause acute political problems, but is unlikely to occur given evidentiary and procedural barriers. Finally, an aggressed state could prosecute the nationals of an aggressor state. This last type of prosecution—in which one state accuses another state of aggression—will likely prompt the most international debate and warrants concern. Such prosecutions could destabilize conflict management and resolution efforts, deter states from undertaking certain military forms of humanitarian intervention, and even undermine the ICC’s legitimacy and stability.9

At the same time, deviating from the standard complementarity framework for the crime of aggression by discouraging domestic prosecutions would create its own set of problems, such as incomplete concurrence between domestic and ICC proceedings. Moreover, the victim state’s ability to prosecute the nationals of an aggressor state may be critical to deterring aggression. Relying primarily or solely on the ICC to prosecute aggression would likely result in fewer prosecutions than would take place if states could also prosecute. By extension, the deterrent power created by the crime’s inclusion in the Rome Statute might be weakened and the uptake of an international antiaggression norm slowed.

In sum, the ASP faces a difficult choice laden with tradeoffs: if it situates the crime of aggression within the ICC’s longstanding complementarity regime, it risks destabilization and threats to the ICC’s legitimacy. But if the ASP rejects complementarity for the crime of aggression and limits or bans domestic prosecutions in favor of ICC prosecutions, it might spark complicated procedural issues and weaken the crime’s deterrent effect and normative power.

The other Rome Statute crimes—genocide, war crimes, and crimes against humanity—do not pose similar challenges vis-à-vis complementarity. Individuals may be held liable for these atrocity crimes without formally implicating the state. In contrast, the crime of aggression inextricably intertwines individual liability with state action.10 First, by definition, only high-level leaders can commit aggression.11 Second, the crime necessarily involves at least two states.12 Finally, liability for aggression hinges on a finding that a statecommitted an act of aggression. These conditions dramatically raise the political stakes of a domestic prosecution for aggression, but need not be present for atrocity prosecutions.13

The unique tension between aggression and complementarity demands the international community’s attention. In 2004, the Special Working Group on the Crime of Aggression14 advised the ASP that the question of fit between aggression and complementarity “merited being revisited” once the ASP agreed on a definition of aggression and the conditions for the ICC’s exercise of jurisdiction.15 The ASP, however, has failed to heed this call. Even after the Kampala Review Conference, the important and complex issues surrounding complementarity and the crime of aggression have remained undertheorized and underdiscussed.

The Kampala Amendments are scheduled to come into force in 2017.16 The tensions between the crime of aggression and complementarity are therefore not front of mind for government lawyers and policymakers. But the States Parties should not delay. Without more sustained attention to the crime of aggression’s fit with complementarity, we risk deeply concerning prosecutions by domestic courts and bewildering procedural questions at the ICC. These prosecutions could significantly affect international peace and security, the prevention of mass atrocities, conflict resolution, and the legitimacy of the ICC. The future of the United States’ relationship with the ICC is also at stake. The relationship is sufficiently fragile that even one badly managed aggression prosecution could undo the slow and steady improvements to the ICC-U.S. relationship made under the Obama Administration.17 The tensions between complementarity and aggression are not easily reconciled and require attention now.

This Note’s objectives are twofold. First, it aims to assist U.S. policymakers by articulating a sensible system of complementarity vis-à-vis the crime of aggression. Second, it seeks to spark dialogue on this relatively unacknowledged problem before 2017.

The Note proceeds in four Parts. Part I describes the history and values underpinning complementarity to frame the subsequent analysis of how complementarity should apply to the crime of aggression. Part II draws on the official records on the adoption of the crime of aggression—the travaux préparatoires—to illustrate how the International Law Commission (ILC), the Special Working Group on the Crime of Aggression, and delegations at Kampala paid little attention to the complementarity issue.

Part III discusses the stakes involved. It argues that the States Parties face a genuine and acute dilemma in navigating complementarity’s application to the crime of aggression. If the States Parties do domestically incorporate and prosecute for aggression, they could jeopardize international political stability. But if the States Parties do not domestically incorporate the crime of aggression, they will be unable to prosecute for aggression, and the ICC may have to navigate incomplete concurrence between proceedings in domestic courts and proceedings at the ICC.

Part IV offers a way forward. It proposes four possible interventions and assesses their political viability and potential impact. The States Parties should (1) establish exclusive ICC jurisdiction or primacy over the crime of aggression; (2) urge the ICC’s Chief Prosecutor to issue an official statement discouraging domestic incorporation; (3) encourage domestic prosecutions for ordinary crimes instead of aggression; and (4) generate a multifactor list to guide domestic prosecutions to avoid the most problematic prosecutions and mitigate the associated harms. I recommend that a coalition of States Parties debate these four interventions at an intersessional meeting, similar to the meetings held during the Princeton Process.18

Before turning to Part I, however, I want to make clear that this Note does not argue that the ICC and the international community would be better off without the crime of aggression in the Rome Statute, nor do I suggest that aggression prosecutions should be avoided. Aggression is a serious crime with significant harms. We should endeavor to deter it and to punish those who perpetrate it. But we must do so with thoughtful attention to the technical details. Failing to do so could destabilize the ICC, undermine promising diplomatic negotiations aimed at deescalating or ending future conflicts, or otherwise jeopardize the pursuit of international peace, justice, and political stability. This Note seeks to shine a light on the tough practicalities inherent in efforts to end impunity for international crimes; it does not promote abandoning the fight altogether.

I. the history and function of complementarity

Although the term “complementarity” does not appear in the Rome Statute, it is routinely used to describe the relationship between the ICC and domestic courts.19 That relationship cements a preference for domestic prosecutions, and positions the ICC as a court of last resort.

The concept of complementarity finds expression in the Rome Statute’s Preamble and Article 1. The Preamble states that “the International Criminal Court established under this Statute shall be complementary to national criminal jurisdictions.”20 Article 1 provides, “[The ICC] . . . shall be complementary to national criminal jurisdictions.”21 In turn, Article 17 sets out the mechanics of complementarity, providing that a case is presumptively inadmissible before the ICC if it “is being investigated or prosecuted by a State which has jurisdiction over it, unless the State is unwilling or unable genuinely to carry out the investigation or prosecution.”22 The complementarity framework does not allow for double jeopardy: a person tried by another court for conduct proscribed by the Rome Statute may not also be tried by the ICC for the same conduct.23

The idea of complementarity arose early on during the preparatory negotiation phase for the ICC­,24 but was not based on past practice.25 Instead, the framework emerged from States Parties’ desire to balance two interests: (1) respecting state sovereignty and, in particular, states’ rights and duties to exercise criminal jurisdiction over international crimes; and (2) ensuring that serious international crimes do not go unpunished.26

Practical considerations also counseled in favor of positioning the ICC as a court of last resort. First, the ICC has limited resources and cannot feasibly prosecute all instances of serious international crimes.27 National authorities, by virtue of their proximity to the scene of the crime, are often better equipped to interview witnesses, collect evidence, and prepare strong cases.28 Second, complementarity comports with the customary international-law principle favoring the exhaustion of “available and effective domestic remedies” before international organs may exercise jurisdiction.29

The decision to adopt a complementarity framework for the ICC was uncontroversial. The debate over complementarity during the Rome negotiations focused “not on its merits or appropriateness, but on perfecting the most agreeable textual approach that would gain state consensus.”30 Commentators have since lauded complementarity as a critically important feature of the ICC’s design and acceptance.31

Under the complementarity framework, the ICC plays two roles: passive and proactive. The first assumes that complementarity serves a gatekeeping function such that the ICC acts only where states are unable or unwilling to do so. The second suggests that the ICC can play an active role in inspiring and encouraging states to investigate and prosecute serious international crimes.32 For instance, many commentators believe that complementarity strongly incentivizes states to incorporate the substantive provisions of the atrocity crimes into their domestic penal codes to ensure the ability to exercise jurisdiction.33 Indeed, since the Rome Statute was adopted in 1998, national laws establishing universal jurisdiction over the atrocity crimes have proliferated.34

In contrast to the atrocity crimes, far fewer countries have incorporated the crime of aggression. Prior to the 2010 Review Conference in Kampala, the vast majority of states did not include the crime of aggression in their domestic penal codes, and most domestic courts lacked jurisdiction over the crime under any basis.35 A 2009 survey of ninety national criminal codes revealed that only twenty-four countries, mostly in Eastern Europe and Central Asia, had statutes containing some provisions relating to the crime of aggression.36 These statutes were largely “rudimentary,”37 “broadly worded, and lack[ing] precision as to the elements of the crime.”38

If a state does not incorporate the crime of aggression into its domestic penal code, it cannot prosecute for aggression, unless it is one of the very few states allowed to exercise universal jurisdiction over the crime.39 Except for the period immediately following World War II, when the Allies held a handful of trials for acts of aggression,40 there have been virtually no domestic criminal prosecutions for the crime of aggression or crimes against the peace.41 Under the proactive theory of complementarity, the ICC’s complementarity regime may inspire more states over time to incorporate the crime of aggression into their domestic penal codes.

This Note is concerned with the challenges that could emerge if many states do in fact domestically incorporate the crime of aggression. Before discussing those challenges, however, it first traces the minimal attention devoted to aggression’s fit with complementarity during the negotiations leading up to the 2010 Kampala Review Conference and at Kampala itself.

II. complementarity and the crime of aggression: the lead up to kampala

The negotiations that preceded the 2010 Kampala Review Conference reveal a persistent trend: States Parties have repeatedly recognized the potential tension between the crime of aggression and complementarity, but have failed to meaningfully engage with that tension and forge a workable resolution. States Parties did not deviate from this trend during the Review Conference. Most delegations paid little attention to the thorny issues inherent in applying a complementarity framework to the crime of aggression.

According to Beth Van Schaack, a public member of the U.S. interagency delegation to the 2010 Kampala Review Conference, issue inundation and apathy explain the dearth of discussion about complementarity. First, the States Parties needed to resolve many open questions going into Kampala. States were still dissatisfied with the crime’s definition and significant jurisdictional issues demanded attention.42 Delegates also had to devote attention to a stocktaking exercise wholly separate from the crime of aggression.43 Many delegations therefore felt unable to add the complementarity problem to their plate. Second, some delegations were simply uninterested in delving into the logistics of how an aggression prosecution would work in practice. To them, including the crime of aggression in the Rome Statute was an important step in realizing the legacy of Nuremberg. They therefore focused their attention on the crime’s symbolic value, not ground-level implementation.44

Delegations’ inadequate time and insufficient interest in the mechanics help explain why they failed to explore tough questions surrounding complementarity at Kampala. This Part details this failure and its origins. It first outlines the pre-Kampala negotiations and describes the Kampala Review Conference itself. It highlights how the States Parties at Kampala adopted Understanding 5, which subtly discourages domestic incorporation of the crime of aggression. The Note then offers a current snapshot of trends in domestic incorporation of the crime of aggression. These trends provide the necessary context for Parts III and IV, which identify present tensions and propose ways forward.

A. Pre-Kampala Aggression Negotiations

1. Deliberations on the Draft International Law Commission Statute (1995)

Concerns about the fit between complementarity and the crime of aggression extend back at least to 1995. That year, delegates to the newly created United Nations Ad Hoc Committee on the Establishment of an International Criminal Court assembled to review and report on a draft United Nations ILC statute.45 The draft provided that the soon-to-be-created ICC should not prosecute aggression unless the Security Council determined that a state had committed an act of aggression.46 Some delegates were skeptical about extending the ICC’s jurisdiction to aggression because few domestic penal codes encompassed the crime.47 Other delegates, however, argued that there was limited domestic legislation on aggression because no international definition existed, and including the crime of aggression within the ICC’s jurisdiction would incentivize states to incorporate and prosecute the crime domestically.48 Despite acknowledging the complementarity problem, the delegates did not meaningfully grapple with or resolve their disagreement.

2. Draft International Law Commission Code of Crimes (1996)

The issue reemerged when the 1996 ILC Draft Code of Crimes offered the strongest anticomplementarity position taken in the negotiations leading up to Rome and Kampala. The draft assigned the ICC exclusive jurisdiction over the crime of aggression and provided for a system of concurrent jurisdiction between States Parties and the ICC over genocide, war crimes, and crimes against humanity.49

The draft recommended exclusive jurisdiction for the crime of aggression because of the crime’s “unique character . . . in the sense that . . . 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.”50 Asking the national court of one state to sit in judgment of another state “would be contrary to the fundamental principle of international law par in pare[m] imperium non habet.”51 This Latin phrase means that an equal has no authority over an equal. It expresses the public international law principle that a sovereign power is equal to and independent of another sovereign power. It is the basis for the act of state doctrine and sovereign immunity,52 and has been called the most “fundamental” and “universally accepted” international law principle.53 Inviting the national court of one state to conclude whether another state has committed aggression would violate this principle. Such violations, in turn, “would have serious implications for international relations and international peace and security.”54 Alain Pellet, an ILC member, explained, “If . . . national courts could try a person for the crime of aggression without a prior filtration process, the Court of Assize of Benghazi, say, could decide that Luxembourg had committed an act of aggression against Mali: [a]n inconceivable and surrealistic situation.”55

The 1996 ILC Draft Code of Crimes did make one exception to the exclusive jurisdiction regime: a state could try its own nationals for the crime of aggression. The logic behind this exception was twofold. First, prosecuting one’s own nationals does not require assessing the actions of another state and thus does not carry the same risks to international political stability.56 Second, such a prosecution might actually be “essential to a process of national reconciliation.”57

3. Initial Deliberations on the Rome Statute (1998)

The Rome Statute drafters ultimately rejected the ILC’s framework. Instead of exclusive jurisdiction for aggression and concurrent jurisdiction for the atrocity crimes, the Rome Statute provides only for a complementarity regime.58 Because the ASP did not immediately agree on the crime of aggression’s definition, however, the States Parties could still explore whether and how the complementarity regime should apply to aggression.59

When the ASP adopted the Rome Statute in 1998, it passed a resolution to create the Preparatory Commission for the Establishment of an International Criminal Court. The Commission was charged with “prepar[ing] proposals for practical arrangements for the establishment and coming into operation of the Court,” as well as “prepar[ing] proposals for a provision on aggression, including the definition and Elements of Crimes of aggression and the conditions under which the International Criminal Court shall exercise its jurisdiction with regard to this crime.”60 The Commission was to present its proposals to the ASP at a Review Conference, “with a view to arriving at an acceptable provision on the crime of aggression for inclusion in [the Rome] Statute.”61

Despite its express mandate, the Commission devoted scant attention to complementarity during its ten sessions, spread out over roughly twenty weeks.62 The Commission did not even mention complementarity as it related to aggression until its fourth session in March 2000. The Preparatory Commission issued a “Preliminary List of Possible Issues Relating to the Crime of Aggression” that included complementarity. The list asked, “How would the provisions of the Statute on complementarity (admissibility, challenges to the Court’s jurisdiction) be applicable to the crime of aggression[?]”63 It then noted that the Rome Statute’s provisions regarding international cooperation and judicial assistance “may require further consideration depending upon the applicability of the principle of complementarity to the crime of aggression.”64 The Italy delegation urged the other states to tackle the issue, questioning whether and how complementarity would operate if national legislation defined aggression differently from the Rome Statute.65

No delegation actually addressed Italy’s questions until the Sixth Session. Certain delegations asked how complementarity would work in practice, and whether the Security Council would play a role.66 Delegations voiced similar concerns during the Ninth Session.67 Unfortunately, the discussion never moved beyond simple acknowledgments that a problem might exist.

4. Meeting of the Special Working Group on the Crime of Aggression (2004)

The complementarity problem received a similarly cursory treatment during the first intersessional meeting of the Special Working Group on the Crime of Aggression, an arm of the ASP. The Group held its meeting at Princeton University in June 2004 and discussed whether the Rome Statute’s provisions on complementarity should apply to the crime of aggression. According to the meeting report, “[t]here was general agreement that no problems seemed to arise from the current provisions [of complementarity] being applicable to the crime of aggression,” but that the issue “merited being revisited” once a definition of aggression and conditions for the ICC’s exercise of jurisdiction were agreed upon.68

B. The 2010 Kampala Review Conference

Against this backdrop of repeated failures to engage with complementarity’s application to aggression, the United States increased its involvement with the ASP as the 2010 Kampala Review Conference approached. The United States had not participated in the negotiations around the crime of aggression during the Bush Administration, but the Obama Administration worked to reengage with the ICC.69 As part of this effort, the United States identified several issues it thought important to tackle before Kampala, including complementarity. In March 2010, U.S. State Department Legal Adviser Harold Koh addressed the topic in his statement at the Resumed Eighth Session of the ASP in New York. Koh posed a series of questions about the mechanics and political viability of situating the crime of aggression within the complementarity framework, suggesting that domestic aggression prosecutions might undermine “state cooperation with the [ICC]” and the ICC’s acceptance in the international community.70

Despite the United States’ increased involvement, the Kampala Review Conference continued the long trend of acknowledging the difficulties that complementarity poses but avoiding any deeper consideration of how to address those difficulties. That the delegates discussed the issue at all was largely due to the questions that the United States raised at the ASP meetings.71 The Conference Chair, Christian Wenaweser of Liechtenstein, responded to those questions in a non-paper that sought to address whether the Kampala Amendments required—or only encouraged—States Parties to domestically incorporate and exercise jurisdiction over the crime of aggression.72

Legal Adviser Koh, speaking on behalf of the U.S. delegation, lauded the non-paper as “a valuable contribution” and expressed support for using understandings to minimize the risk of “unjustified domestic prosecutions.”73 Koh then rearticulated many of the complementarity-related concerns he had expressed in New York less than three months earlier. He observed that domestic aggression prosecutions “would ask the domestic courts of one country to sit in judgment upon the state acts of other countries in a manner highly unlikely to promote peace and security.”74

With these concerns in mind, the United States proposed small but significant changes to the Chair’s draft. The non-paper, with the United States’ additions represented with underlines and deletions represented with strikethroughs, stood as follows:

It is understood that the amendments address the definition of the crime of aggression and the conditions under which the Court shall exercise jurisdiction with respect to this crime 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, and shall not be interpreted as constituting a statement of the definition of “crime of aggression” or “act of aggression” under customary international law.

It is understood that tThe amendments shall therefore not be interpreted as creating the right or obligation to exercise domestic jurisdiction with respect to an act of aggression committed by another State.75

The changes to the first paragraph reflected the disagreement among the States Parties whether the definitions of “crime of aggression” and “act of aggression” accurately expressed customary international law.76 The proposed change was intended to stress that the amendments pertained only to prosecutions under the Rome Statute.77

The second paragraph had been the final sentence of the first and only paragraph in the Chair’s draft. The United States reframed it as a separate understanding to distinguish more clearly “issues of definition versus jurisdictional competency” and to emphasize “that ratification or acceptance of the amendments will not empower or oblige states to incorporate the crime of aggression into their domestic codes or to launch domestic prosecutions for the crime.”78

Over the next two days, Claus Kreß of the German delegation facilitated bilateral, regional, and open informal consultations on the U.S. draft to assess whether it could achieve a consensus.79 The United States’ proposed changes “met with little public discussion or resistance, except insofar as the words ‘customary international law’ were ultimately not included in the final text.”80 The final text comprised two Understandings:

4. 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.

5. It is understood 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.81

On its face, Understanding 5 appears obvious, since the Rome Statute “is generally not designed to create rights or obligations of States Parties with respect to domestic legislation and adjudication.”82 But reading Understanding 5 in light of the United States’ articulated concerns reveals what Van Schaack has called a “subtle preference” against domestic incorporation and prosecution.83 Taken in context, Understanding 5 captures and attempts to respond to the reticence that some delegations at Kampala felt about domestic aggression prosecutions.84

C. Post-Kampala Developments

Since the 2010 Review Conference, commentators have encouraged states to incorporate aggression into their domestic criminal codes.85 Thus far, only six countries—Croatia, the Czech Republic, Ecuador, Luxembourg, Slovenia, and Samoa86—have passed domestic legislation implementing the Kampala Amendments. But others may soon follow. The Dominican Republic approved the text, and the law will enter into force after the president enacts it. Macedonia, New Zealand, Peru, and Venezuela are considering draft criminal code bills containing the crime of aggression as defined at Kampala.87 South Africa also appears likely to incorporate.88 The Global Campaign for Ratification and Implementation of the Kampala Amendments on the Crime of Aggression reports that it expects fifteen other countries to consider incorporating the crime into domestic law.89

Whether a meaningful number of States Parties will actually incorporate the crime into their domestic law remains unclear. Certainly, there is evidence of momentum in that direction. Such momentum, fueled by advocacy from civil society groups, may only increase after the Amendments come into force in 2017.

III. tensions between complementarity and the crime of aggression

The remainder of this Note aims to articulate what remained unsaid at Kampala. First, I identify the core tensions, problems, and pitfalls associated with applying the existing complementarity framework to the crime of aggression. In Section III.A, I analyze the potential risks if states heed the implied message of Understanding 5 and do not domestically incorporate the crime of aggression. Conversely, Section III.B explores the problems that might occur if states do domestically incorporate and prosecute the crime of aggression. Finally, in Part IV, I identify and analyze various interventions that could address these problems.

A. Problems if States Do Not Incorporate

If States Parties choose not to incorporate the crime of aggression into their domestic criminal codes, the ICC would be the primary forum for aggression prosecutions. This approach could produce two notable problems: (1) incomplete concurrence between proceedings at the domestic level and proceedings at the ICC; and (2) diminished uptake of an antiaggression norm.

1. Incomplete Concurrence

First, if States Parties do not incorporate the crime, they risk incomplete concurrence between proceedings at the domestic level and proceedings at the ICC.90 To understand this problem, imagine a series of events that could give rise to charges for both the crime of aggression and an atrocity crime, such as genocide. Most domestic courts, having incorporated the atrocity crimes into their penal codes but not the crime of aggression, could prosecute only the atrocity crime and not aggression. Imagine, then, that the domestic court decided to prosecute the atrocity crime. Because the domestic court would be unavailable to prosecute for aggression, the ICC could then assert jurisdiction pursuant to Article 17(3) of the Rome Statute, and indict and prosecute for the crime of aggression.

To date, the Pre-Trial Chamber of the ICC has used a “same person/same conduct test” to interpret the complementarity principle. In other words, “the Court will stay its hand if a domestic court is prosecuting the same suspect for the same impugned conduct, regardless of how the crime is legally characterized.”91 As noted above, the atrocity crimes and the crime of aggression generally proscribe different conduct.92 Because the conduct giving rise to the domestic court’s atrocity prosecution would differ from the conduct subject to an aggression charge, a prosecution by the ICC for aggression would be fair game, even though both charges stemmed from the same series of events.

Yet even if legally permissible, these parallel prosecutions—a domestic prosecution for an atrocity crime and an ICC prosecution for aggression—would put the domestic court and the ICC in difficult situations. First, tough procedural questions would emerge: Should the domestic and ICC prosecutions move forward at the same time, or would they need to be sequenced? If the prosecutions proceed concurrently, which system would have custody over the defendant? If the ICC ordered the defendant to be transferred to The Hague, it would effectively divest the domestic court of its jurisdiction over the alleged perpetrator. How would “overlapping or potentially contradictory witnesses and evidence” be handled?93 If the prosecutions were sequenced, which jurisdiction would prosecute first? How would Article 20’s commitment to ne bis in idem, its double jeopardy provision, apply?94

Second, institutional considerations would have to be weighed. The ICC’s decision to step in and prosecute for aggression could fuel hostility toward the international tribunal. The prosecution might be understood to undermine the domestic jurisdiction’s sovereignty. It could also discourage the domestic atrocity prosecution, thus subverting the core purpose of complementarity. Alternatively, if the ICC chose not to step in and prosecute for aggression, that decision might undermine its moral role as a bulwark against impunity for serious international crimes.

In sorting through these complicated questions, the ASP will have to consider how best to balance the pursuit of the ICC’s multiple goals: encouraging states to exercise their duty to domestically punish serious international crimes, deterring genuine acts of aggression, ending impunity, and maintaining positive and productive relationships with States Parties.The answers will bear directly on the ICC’s legitimacy and viability as an enduring tribunal.

2. Diminished Uptake of an International Antiaggression Norm

By choosing not to incorporate the crime of aggression, States Parties could stunt the development and broad uptake of an antiaggression norm.95 Resource constraints will prevent the ICC from investigating and prosecuting all acts of aggression.96 Without domestic prosecutions, many acts of aggression could go unpunished, undercutting the deterrent effect that originally motivated the ASP to include aggression in the Rome Statute.97 The Global Campaign for Ratification and Implementation of the Kampala Amendments on the Crime of Aggression has even argued that failing to incorporate is antithetical to the principles underlying the Rome Statute. Indeed, the statute’s preamble provides “that it is the duty of every State to exercise its criminal jurisdiction over those responsible for international crimes.”98

B. Problems if States Do Incorporate

Notwithstanding the difficulties that may arise, Understanding 5 expresses a subtle preference against domestic incorporation of the crime of aggression. Of course, states need not heed this call. Instead, they may choose to follow the path of Croatia, the Czech Republic, Ecuador, Luxembourg, Slovenia, and Samoa, and incorporate the Kampala Amendments into their domestic penal codes. Countries might incorporate the crime to preserve sovereign control over aggression prosecutions—particularly of their own nationals—and preempt the ICC from asserting jurisdiction. They might also seek to promote an antiaggression norm. But serious practical impediments may complicate these good intentions. Domestic prosecutions could exacerbate diplomatic relations and jeopardize international peace and stability.

1. Justiciability, Evidentiary, and Other Jurisprudential Roadblocks to Prosecutions

Domestic prosecutions for aggression could encounter justiciability, evidentiary, immunity, and other legal roadblocks. Theresa McHenry, the Chief of Human Rights and Special Prosecutions at the U.S. Department of Justice and a member of the U.S. Delegation at Kampala, described several such roadblocks shortly after the Review Conference concluded. In many countries, she noted, the crime of aggression will almost certainly implicate questions within the purview of the political branches, such as “the executive’s commander-in-chief authority or congressional power to declare war.”99 Domestic aggression prosecutions might run afoul of various states’ political-question doctrines.100

In addition to justiciability issues, prosecuting states will likely face evidentiary hurdles. To prove that a state action constitutes aggression, the prosecution might need to access information that is classified, governed by executive privilege101 or state secrets, or otherwise controlled by the putative aggressor state. States may also have divergent approaches to the burden of proof on issues such as whether the prosecution must show that an alleged act of aggression was not self-defense.102

Moreover, domestic efforts to prosecute will likely trigger a range of substantive and procedural immunity arguments, including head-of-state immunity and foreign official immunity.103 Yet the act of state doctrine—which provides “that one sovereign should not sit in judgment of another [sovereign]”—is unlikely to bar a domestic court’s prosecution of a nonnational for aggression.104 The doctrine applies only to acts fully executed within the sovereign’s own state. Aggression is, by definition, unlikely to satisfy that requisite. Furthermore, in the United States, the doctrine has generally not served as a bar to proceedings for “international crimes or breaches of jus cogens.”105

Finally, it is highly unlikely that a victim state could, in practice, exercise physical custody over the aggressor state’s leaders.106 Assuming that a victim state’s aggression prosecution depends on having physical custody of the aggressors, this fact alone will doom most prosecutions.

Complementarity might not rescue domestic prosecutions from these practical complications. Should one of these procedural issues impede a domestic prosecution, the ICC may not necessarily have jurisdiction. Article 17 of the Rome Statute provides that the ICC may assume jurisdiction over a case where a state with jurisdiction is unable or unwilling to prosecute.107 But “inability” and “unwillingness,” according to the ICC Chief Prosecutor Bensouda, do not encompass a situation in which a state attempts to prosecute but is stymied by legitimate legal roadblocks. Unwillingness, Bensouda states, “arises if a country lacks the political will to try its own leaders, conducts sham trials in order to let the guilty go free, allows an unjustifiable delay in bringing perpetrators to justice, or does not conduct judicial proceedings independently or impartially.”108 Inability “arises, for example, when the judicial system of a country has collapsed, rendering the state unable to arrest perpetrators or gather evidence.”109 Consequently, if a state initiates an aggression investigation or prosecution that then fails due to justiciability, evidentiary, immunity, or other issues, the ICC may not be able to then step in and assume jurisdiction unless the state self-refers the situation to the ICC. Without a referral, the aggressor would effectively be immune from prosecution.

2. Threats to International Political Stability, Conflict Resolution, and Humanitarian Intervention

Procedural and practical hurdles will likely prevent most domestic aggression prosecutions from reaching a judgment on the merits. Even if domestic prosecutions could surmount these hurdles, they may undermine international political stability. Scholars and practitioners agree on this point. Richard Goldstone, the first chief prosecutor at the International Criminal Tribunal for the former Yugoslavia, asserts that investigating and prosecuting the crime of aggression requires a court to examine what “is inherently a profoundly political decision.”110 Because of the political nature of the crime, Van Schaack warns that a domestic aggression charge “will inevitably generate intense charges of politicization from within and outside the prosecuting state,” “exacerbate relations” between the implicated states, and prompt other states to “inevitably take sides.”111 Similarly, Michael Scharf argues that aggression prosecutions in domestic courts “may be so politically sensitive that they cannot be fairly tried and the attempt to do so would undermine efforts at restoring or maintaining international peace.”112 National court indictments—particularly if employed as a form of “lawfare” to “intimidate and harass another state or its officials”—might interfere with diplomatic negotiations and frustrate efforts at conflict resolution.113

Even the mere specter of an aggression indictment could impede warring parties from negotiating a peace agreement. The risk increases if a state’s prosecutorial bodies are independent from its diplomats, preventing the negotiators from ruling out an aggression prosecution.114 The possibility of domestic prosecution may also chill a state’s willingness to undertake, or cooperate with, humanitarian intervention to stop the commission of atrocities.115 Nicolaos Strapatsas even speculates that the desire and ability to domestically prosecute for aggression could motivate an aggressed state to prolong an armed conflict to ensure the overthrow or removal of the aggressor state’s leader from power.116 This would make it more feasible to obtain physical custody over and prosecute the leader.117

3. Domestic Alterations to the Definition of Aggression

If states incorporate the crime of aggression, they might deviate from the carefully calibrated definition of aggression in the Kampala Amendments. The definition and elements of the crime of aggression, as agreed upon in Kampala, are products of long and challenging consensus negotiations. States could jeopardize the legitimacy of the ASP’s decision to include the crime under the Rome Statute—and the legitimacy of the ICC itself—by creating and prosecuting under a fragmented and incoherent patchwork of aggression variants.

States might adopt narrower or broader versions of the aggression amendments, define aggression differently, or ignore the carefully negotiated protections in the Understandings. For example, a state could fail to incorporate the crime’s leadership requirement. This would expand the range of potential defendants far beyond the scope contemplated by the ASP. A state might limit or exclude “the threshold qualifier ‘manifest,’ which is meant to prevent the ICC from pursuing cases that are not sufficiently grave or that involve state conduct whose illegality under the UN Charter framework is uncertain.”118 A domestic statute might lessen the prosecution’s burden of persuasion from beyond a reasonable doubt. Or a state’s statute could permit “prosecutions for attempted aggression” or could encompass non-state actors.119

To avoid such situations, the ASP should strongly encourage states that do incorporate to use the verbatim text of the Kampala Amendments, including the stipulations contained within the Understandings.120


States face a genuine dilemma with high stakes and strong arguments on both sides. On balance, however, the subtle preference against incorporation, as expressed in Understanding 5, is the prudent approach. Incorporation risks flawed domestic prosecutions that foreclose ICC jurisdiction, international political destabilization and disruptions to conflict resolution efforts, and definitional fragmentation that could permit prosecutions far beyond what the ASP envisioned and endorsed. In contrast, the problems if states do not incorporate and prosecute—limited to incomplete concurrence and a slower antiaggression normative uptake—pose relatively less cause for concern. The drawbacks to nonincorporation are also more amenable to productive countermeasures than are the serious potential harms from domestic incorporation and prosecutions.

IV. a way forward

The Special Working Group on the Crime of Aggression agreed nearly a decade ago that the issues surrounding aggression and complementarity “merited being revisited once agreement had been reached on the definition of aggression and the conditions for exercise of the Court’s jurisdiction.”121 Nonetheless, the uneasy fit between complementarity and the crime of aggression was inadequately addressed in the lead up to the 2010 Kampala Review Conference. Difficult questions persist, and the ASP must address them now, before the Kampala Amendments enter into force in 2017. This Part offers a roadmap for the ASP: it identifies four potential interventions to resolve the tensions between aggression and complementarity, and suggests the means by which the States Parties could engage with those interventions.

A. The Substance: Four Possible Interventions

The ASP must seek solutions to the challenges that will result if states do or do not incorporate and prosecute the crime of aggression. I recommend that the States Parties consider four possible interventions: (1) establish exclusive ICC jurisdiction or primacy over the crime of aggression; (2) supplement Understanding 5 with an official statement from the Chief Prosecutor discouraging domestic incorporation; (3) encourage prosecutions for ordinary crimes instead of aggression; and (4) generate a multifactor list to guide domestic prosecutions.

The first and second interventions limit or discourage domestic incorporation and prosecution. The third intervention provides an alternative, less politicized way for states to punish certain acts of aggression. The fourth intervention seeks to mitigate the harms of domestic aggression prosecutions. Although the first intervention is the cleanest, its adoption is unlikely. I therefore recommend that the States Parties focus instead on mobilizing support for the second, third, and fourth interventions, which are mutually reinforcing and will work best in concert. I further argue that the States Parties should devote the greatest attention to the fourth intervention, the most nuanced and potentially effective intervention of the set.

1. Establish Exclusive ICC Jurisdiction or Primacy over the Crime of Aggression

First, the ASP should consider amending the Rome Statute to establish exclusive jurisdiction for the ICC over the crime of aggression or, alternatively, to establish a primacy framework. Neither approach would affect the jurisdictional regime applied to genocide, crimes against humanity, or war crimes, which would continue to be situated within a complementarity framework.

As describedin Part II, the ILC contemplated an exclusive jurisdiction regime for the crime of aggression in its 1996 Draft Code of Crimes.122 Under such a regime, only the ICC could prosecute the crime of aggression. This approach would be consistent with the principle of par in parem imperium non habet, as one sovereign state would not sit in judgment of another. A possible exception to exclusive jurisdiction, also suggested in the 1996 Draft Code of Crimes, would be to permit a state to try its own nationals for the crime of aggression.

Primacy is similar to, but not as absolute as, exclusive jurisdiction. Under a primacy framework, an international tribunal need not defer to domestic prosecutions; rather, it has the right to trump domestic proceedings.123 The ICTY, ICTR, the Special Court for Sierra Leone, and the Special Tribunal for Lebanon have all adopted a primacy regime.124 For the ICTY, primacy was instituted in response to concerns about the “effectiveness and impartiality” of domestic courts.125 For the ICTR, the concern was less that Rwanda’s courts would try to shield perpetrators from justice, and more that they were too weakened by the genocide to capably prosecute the genocidaires.126 The Security Council established each ad hoc tribunal’s primacy regime under its Chapter VII authority to determine the measures required to “restore ‘international peace and security.’”127 The ICC, of course, does not enjoy Chapter VII provenance. It was created via a multilateral treaty, and is intended to be an enduring, forward-looking institution, rather than responsive to a specific set of atrocities. As such, its design offers more deference to national prerogatives.128

Enabling the ICC to assert exclusive jurisdiction over the crime of aggression, or giving it the opportunity to trump national prosecutions, would avoid the political, legal, and procedural problems associated with domestic aggression prosecutions and described in Section III.B. However, many states would likely view such power as an illicit incursion on their sovereignty. States might also perceive these frameworks as disrupting the delicate balance between preserving national sovereignty and combating impunity for international crimes. This balance, which complementarity aims to strike, has been key to the ICC’s acceptance and legitimacy in the international community.129 Exclusive jurisdiction or primacy might also force the ICC to choose between two undesirable options: overburdening itself with cases or declining to pursue most potential prosecutions. Either choice could hinder the development of a strong antiaggression norm, as discussed in Section III.A.2.

It would also be politically difficult to establish exclusive jurisdiction or primacy. Doing so would require amending Article 17 of the Rome Statute, as well as Article 15 bis and Article 15 ter. To amend the Rome Statute, a State Party must propose the amendment, a majority of those present at the next meeting of the ASP must vote to consider the proposal, and a two-thirds majority of States Parties must then vote to adopt it.130 It took twelve years for the ASP to agree to the aggression amendments, and there is already widespread fatigue following Kampala. It is thus unlikely that there will be much appetite for another amendment in the near future. So, although this proposal would provide the cleanest solution to the challenges posed by domestic aggression prosecutions, it is almost certainly politically infeasible.

2. Supplement Understanding 5 with a Diplomatic Overture from the Chief Prosecutor To Discourage Domestic Incorporation

A softer, more feasible approach to discouraging domestic incorporation would be to petition ICC Chief Prosecutor Bensouda to make a diplomatic overture toward States Parties. Bensouda could urge States Parties to give more space to the ICC to take the lead on aggression prosecutions and generally refrain from bringing such prosecutions domestically. In so doing, she could promote the notion that a state can respect its international obligations regarding serious crimes like aggression by not acting when ICC prosecution would be more appropriate and effective.131

This approach shares the same goal as the first proposal—limiting domestic incorporation and prosecution—but does not require the ASP to surmount any procedural hurdles or formally reconfigure the sovereignty-impunity balance that complementarity aims to strike.

That strength, however, is also a weakness. A diplomatic overture from the chief prosecutor is merely that—an overture. Whether a State Party pays it any heed will depend on how much respect the state has for the chief prosecutor, how much faith the state has in the ICC’s ability to prosecute the crime of aggression, and how important it is to the state—for symbolic, dignitary, political, or other reasons—to conduct its own prosecution. A diplomatic overture is an important part of the shifting norms and expectations on how the crime of aggression is best prosecuted, but is not a sufficient solution in itself. States Parties should concurrently pursue the third and fourth proposals.132

Before discussing those proposals, however, it is important to acknowledge the major drawback to my first and second proposals. If the ASP adopts a system of ICC exclusive jurisdiction or primacy,133 or if States Parties heed the chief prosecutor’s recommendation not to domestically incorporate the crime of aggression, the ICC may have to contend with the problems of incomplete concurrence between ICC and domestic prosecutions.

Van Schaack has given this potential problem significant attention and offers one possible solution: before the aggression amendments enter into force in 2017, the Chief Prosecutor should “announce . . . the intention to stay . . . her hand in the event that genuine domestic prosecutions are going forward on the basis of charges of genocide, crimes against humanity, or war crimes, even if domestic aggression charges are not available, are legally barred, or are not forthcoming.”134 The sole exception would arise in cases “in which the crime of aggression is the primary or central charge to arise out of a particular situation such that atrocity crimes are non-existent or largely peripheral.”135

Van Schaack’s proposal would allow the ICC to maintain “de facto exclusivity over the crime of aggression,” thus obviating some of the problems associated with domestic incorporation and prosecution.136 Yet it would also empower national courts “to take the lead on prosecuting atrocity crimes” and therefore avoid encumbering the ICC with a massive caseload.137 This is a smart and practical approach to promoting antiatrocity norms, encouraging states to fulfill their duty to prosecute international crimes, combating impunity, and conserving the ICC’s limited resources. It avoids the practical coordination challenges of concurrently prosecuting the same individual for different crimes arising out of the same events. Furthermore, it establishes a default position of avoiding politically volatile aggression prosecutions while leaving space for the ICC to intervene where aggression truly warrants prosecution.138

3. Prosecute for Violations of Ordinary Criminal Law, Not the Rome Statute

The ASP can further guard against worrisome domestic prosecutions by encouraging national courts to prosecute an aggressor under ordinary criminal law. Though the defendant’s conduct could trigger an aggression indictment, the state would apply its ordinary domestic criminal code instead. This would avoid the par in parem imperium non habet problem of states judging the conduct of other states. Ordinary criminal prosecutions would not require a finding that a state had committed an act of aggression, and would instead target individual acts.139

But, ordinary criminal law prosecutions will not always be possible. Whether a domestic analogue to aggression exists will require a case-by-case determination. For example, cross-border terrorism or using weapons of mass destruction would constitute aggression under the Kampala Amendments, and would violate U.S. law.140 Yet these examples are rare. Because Article 8 bis includes a leadership element and requires that an act of aggression have the “character, gravity and scale” to constitute “a manifest violation of the Charter of the United Nations,” the likelihood that a domestic analogue to the crime of aggression will be available is slim.141

Assuming that a state could find an analogous ordinary crime, the benefit to this approach is clear. The state could convict and punish the perpetrators for their harmful actions. In turn, the state would avoid charges of politicization and threats to political stability and conflict resolution that would flow from an aggression prosecution.142 Because of the Rome Statute’s double-jeopardy provision, prosecuting a defendant for ordinary crimes instead of aggression in a national court would preclude a subsequent aggression prosecution by the ICC.143 This approach therefore captures the very values that motivated the adoption of the complementarity framework: respecting state sovereignty and ensuring that serious international crimes not go unpunished.

The drawback of this approach is that prosecuting an act of aggression as an ordinary crime instead of as a crime of aggression will carry less expressive value.144 As a result, prosecuting ordinary crimes may do less to contribute to the spread of an antiaggression norm.145 This limitation, however, is outweighed by the benefit of avoiding the potentially destabilizing ramifications of a domestic aggression prosecution.

4. Generate a Multifactor List To Guide Domestic Prosecutions

Realistically, a diplomatic overture from the ICC Chief Prosecutor will not deter all States Parties from domestically incorporating the crime of aggression. Moreover, ordinary criminal-law prosecutions in lieu of aggression prosecutions will not always be feasible or desirable. At least some States Parties will proceed with domestic incorporation, particularly after the Kampala Amendments come into force in 2017. The ASP should therefore focus primarily on developing guidance to help domestic jurisdictions decide whether and when to indict and prosecute for aggression.

In particular, the ASP should create a multifactor list to target and prevent the riskiest domestic prosecutions. Its audience should be domestic decision makers who choose when to investigate, indict, and try criminal cases, such as Ministries of Justice and high-level public prosecutors. The ASP should aim to reach those prosecutors who are especially likely to pursue, or have a demonstrated history of pursuing, politically sensitive cases with little regard for the repercussions. These bellicose prosecutors, however, are also most likely to ignore the need for caution. Accordingly, the ASP should model the tone of the list on Understanding 5 and its “subtle preference” against domestic incorporation and prosecution.146 The ASP should tactfully but clearly convey that the international community will disfavor prosecutions undertaken with relative insensitivity to political reverberations—particularly those prosecutions that might undermine conflict resolution efforts in response to mass atrocities and human rights violations—and that prosecutors who undertake such actions risk reputational harm.

The list would not bind the States Parties, as it would not formally amend the Rome Statute. The ASP, however, could champion the list via diplomatic outreach and strongly recommend it as a tool to Ministries of Justice and public prosecutors’ offices. To enhance the list’s credibility, States Parties should publicly announce their commitment to it. Through widespread expressions of support and consistent application, the list could gradually become part of customary international law. G. John Ikenberry wrote of the Princeton Principles on Universal Jurisdiction that “[a]t the very least,” the list would serve as “a gold standard” and make egregious conduct “harder to get away with.”147

The first factor on the list should be whether the United Nations Security Council has determined that a state committed an act of aggression.148 The factor should be understood as dispositive—a threshold precondition for a domestic aggression prosecution. Determining whether an act of aggression has occurred “is one of the Council’s core Charter functions,”149 and relying on a United Nations Security Council determination will lend greater legitimacy to the state’s indictment. A Security Council determination would also relieve the prosecuting state from a highly politicized assessment of whether the putative aggressor state in fact committed an act of aggression. For these same reasons, a Security Council determination was central to the ILC’s approach in its draft statute for an international criminal court.150

In addition to the first, dispositive factor, the list should contain several discretionary factors, including: (1) whether the state(s) with territorial-based or national-based jurisdiction are unable or unwilling to prosecute (for cases in which the prosecuting state seeks to exercise nonterritorial jurisdiction); (2) whether the prosecution is against a country’s own nationals;151 and (3) whether prosecutions for ordinary crimes are infeasible.152

Of the four proposals, the multifactor list to guide domestic action has the most nuance and the greatest potential for impact. The list responds to the problems identified in both Sections III.A and III.B. It does not seek to foreclose all domestic prosecutions and vest exclusive or primary jurisdiction in the ICC, and thus does not raise the risk of incomplete concurrence or impede the uptake of an international antiaggression norm.153 The list instead recognizes that many states likely will incorporate the crime of aggression into their domestic penal codes and guides those states to refrain from launching prosecutions that could jeopardize political stability and conflict resolution efforts.154 The multifactor list is the most promising of the four proposals, but is not exclusive of the second and third proposals. Rather, the three proposals are mutually reinforcing and should therefore be pursued simultaneously.

B. The Form: An Intersessional Gathering

It would be unwise for the United States to wait until the 2017 Review Conference to raise these issues with sympathetic partners. There is too much fatigue regarding the crime of aggression, coupled with a sense of fait accompli post-Kampala. Some members of the ASP even resist the idea that a 2017 Review Conference is needed at all.155

I therefore recommend that the United States organize a Chatham-House-style gathering for States Parties to engage in a thoughtful, deliberative process about the practicalities left unexamined in Kampala.156 The States Parties would wrestle with the implications of situating the crime of aggression inside a complementarity framework and aim to generate a set of best practices. A best practice guide, in turn, could influence concrete action at a 2017 Review Conference or, at a minimum, guide States Parties in their decisions about incorporation and subsequent prosecutions.157

If the effort to convene a pre-2017 intersessional gathering should fail, interested States Parties could instead focus on the final decision that the ASP must take to activate the Kampala Amendments. Article 15(3) bis states, “The Court shall exercise jurisdiction over the crime of aggression in accordance with this article, 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.”158 Concerned States Parties could invoke this requirement to raise (and attempt to resolve) some of the difficult questions around complementarity and to put a finer point on the subtle preference against domestic incorporation of the crime of aggression. Still, states should not wait until this decision juncture to engage States Parties in a dialogue about possible interventions. If progress is to be made, a longer-term engagement strategy such as the one proposed here is required.159

In organizing the proposed gathering, the conveners should aim to replicate the Princeton Process’s informal, intersessional meetings. These meetings generated fruitful, productive discussions, according to Stefan Barriga, who from 2003 to 2010 served as principal legal advisor to the primary negotiators on the crime of aggression.160 He attributed the success to the following considerations: a “serene environment” that helped inspire a “friendly and constructive atmosphere”; discussions structured around “individual contributions . . . generally understood not to be binding on the government that nominated the respective participant”; the “active involvement of NGO representatives” who offered “great expertise . . . thereby elevating the quality of the exchange”; a “pro-active approach” taken by the Chairman, as demonstrated by his submission of informal non-papers and use of questionnaires to shape the agendas; intermeeting communication over an e-mail list; and a general “sense of camaraderie” that generated “interactive, focused, open and frank” discussions.161

Of course, not everyone agrees with Barriga’s view that the Princeton Process was a resounding success, for it skirted many difficult political issues. Kreß observed that it “focused on legal and technical questions. . . . [R]elatively little time was wasted with rehearsing the divergent and well-entrenched views” on the highly political question of the Security Council’s role.162 Instead, Kreß explained, “[M]ost intellectual energy was applied to reducing the differences of opinions on a host of other issues of a less politically sensitive nature.”163 Koh noted that the Princeton Process was “unable to bridge very significant differences of views among states” even though the participants addressed “at length” the jurisdictional prerequisites for an ICC prosecution for aggression.164

Ideally, these drawbacks to the Princeton Process will be less salient at the gathering proposed here, given its singular, explicit focus on the highly political issues surrounding complementarity. Delegations will attend because of their interest in and commitment to resolving these tough questions, creating a rigorous, results-oriented environment.


The ASP took twelve years to fulfill the promise it wrote into the Rome Statute: to set out the conditions under which the ICC would exercise jurisdiction over the crime of aggression. During those twelve years, the international community repeatedly confessed that a complicated problem loomed on the horizon, but declined to grapple with its implications. Even at the 2010 Kampala Review Conference, the ASP failed to meaningfully wrestle with the tensions between the Rome Statute’s complementarity framework and the crime of aggression.

This Note diagnosed the problems that could arise if states do or do not incorporate the crime of aggression into their domestic penal codes and charted an agenda around which the United States and concerned States Parties might convene. Admittedly, there is strong inertia against reviving this unfinished conversation post-Kampala. But it would be a shame if States Parties allow that inertia to undermine the global community’s hard-fought effort to define, punish, and deter a serious international crime. Without action to resolve the complementarity question, the aggression amendments could give way to highly politicized or badly managed prosecutions that endanger conflict resolution prospects; flawed domestic prosecutions that bar subsequent ICC intervention; or an incoherent patchwork of domestic aggression variants. In short, the amendments could give rise to prosecutions that endanger, rather than promote, peace and justice. In turn, these risks threaten to undermine the ICC itself and the United States’ relationship with it, both of which are fragile.

Though these issues may not yet feel urgent, they must be addressed before policymakers’ neglect manifests in an acute crisis. Once ratified, the aggression amendments and their concomitant norms will be much harder to change. The United States and other concerned states must enlist States Parties to address these questions and devise an approach before 2017 arrives.