Politics and Justice at the International Criminal Court

Abstract

The International Criminal Court (ICC) is a legal institution embedded in international politics. Politics shaped the Rome Statute of the ICC, which is rooted in norms and rules of European lineage and security interests of party states. Politics constrains and influences the operation of the Court, which has adapted in response to oversight and governance of the Assembly of States Parties, and to political actions extrinsic to institutional rules. The ICC also has political effects in situation states. A brief history shows that application of Rome Statute triggers across state parties with different social conditions skewed geographic distribution of its investigations and prosecutions towards Africa, a structural bias that catalysed a legitimation crisis for the ICC. Subsequent exercises of expansive jurisdiction aimed at nationals of non-African, non-party states – including Israel and some of the world's great powers – have dampened African complaints and advanced the ICC agenda, but intensified non-legitimacy claims by powerful non-party states. To survive, Court organs must follow legal mandates, yet be responsive to pressing international political demands, continuously risking the legitimacy of the ICC as a legal institution and adverse political reactions by antagonised governments. Careful management of the tension between law and politics at the ICC may modestly reduce antagonism towards the Court, but that tension cannot be resolved, and confrontations over the ICC's legitimacy are certain to recur.

1. Introduction

When the Rome Statute entered into force, constituting the International Criminal Court (ICC), many were hopeful that the Court might dampen violence and atrocities in the world. The ICC offered a promise of enhanced accountability for perpetrators of atrocities, justice for survivors, deterrence of atrocities, and removal of perpetrators from situations where they could commit more atrocities. Many expected that the ICC might achieve all this as a pure legal institution, applying law to facts, treating similar cases similarly, and speaking law to power – behaviour that creates a sense that justice is being done and lends law legitimacy in domestic and international legal systems.¹

So, it should not be surprising that the Court's mantra has been exactly that: take decisions for legal reasons exclusively. The first Prosecutor, Luis Moreno Ocampo, repeatedly asserted that politics would not enter into his decisions, stating at the outset: ‘I shall not be involved in political considerations. I have to respect scrupulously my legal limits’.² The second Prosecutor, Fatou Bensouda, stated her policy: ‘We will do so with unyielding commitment to end impunity for mass crimes and in total independence, but we can only do so in strict conformity with the Rome Statute legal framework’.³ The third and current Prosecutor, Karim Khan, declared early in his tenure: ‘As I proceed to discharge my responsibilities, I will ensure that investigations by my office are conducted objectively and independently’.⁴ Moreover, the ICC has all the hallmarks of a legal institution: a statute elaborately defining the crimes addressed;⁵ strict rules of procedure and evidence;⁶ specified rights of the accused;⁷ a hierarchy of Chambers (Pre-trial, Trial, and Appeals);⁸ the decorum and pomp of a courtroom; refined modes of argumentation; and expansive case law.⁹

However, like all international organisations, the ICC is nested in global politics. International criminal law has been deeply affected by politics from the beginning. The trial of Conradin Von Hohenstaufen in 1268 is said to be the first post-classical European trial for war crimes.¹⁰ Conradin, who carried the title of King of Jerusalem, was tried purportedly for murder and pillaging in Tagliacozzo, a town outside Naples, found guilty of a crime ‘against the laws of God and man’, and was executed. However, historians who have studied the events argue that this was actually a political trial and execution, directed by the Pope, who preferred that a French noble control the Naples region.¹¹

Centuries later, the parties to the Treaty of Versailles indicted Kaiser Wilhelm II for ‘a supreme offence of crimes against international morality and the sanctity of treaties’ and specified that a tribunal was to be constituted to try the accused;¹² others were also to be tried for various war crimes.¹³ However, the special tribunal for Wilhelm was never constituted and he was never tried; the Kaiser was exiled to the Netherlands; the British did not consistently pressure the Dutch to extradite him; and the United States government feared German unrest if he were extradited and tried.¹⁴ Politics again interfered with justice.

After the Second World War, Emperor Hirohito of Japan was not charged by the International Military Tribunal for the Far East,¹⁵ despite his having done nothing to stop the Rape of Nanking where 200,000 people were killed,¹⁶ and nothing to stop the sexual slavery of 300,000 ‘comfort women’.¹⁷ The political reason for not trying the Emperor is made clear in a 1946 message to Washington from General Douglas MacArthur, arguing that Hirohito's indictment and execution could trigger a ‘tremendous convulsion among the Japanese people’ and ‘a condition of underground chaos and disorder amounting to guerilla warfare’.¹⁸ MacArthur estimated that if Hirohito were tried, one million occupation troops might be needed and ‘all hope of introducing democratic methods would disappear’.¹⁹

In Europe, the Nuremburg trials²⁰ were also skewed by politics. Soviet forces had perpetrated mass atrocities, including mass rape and murder, as they drove Nazi forces back into Germany.²¹ When the Nuremberg Charter, which established the Tribunal, was being negotiated, representatives of the United States and the United Kingdom pressed for some Soviets to be tried. Not surprisingly, Stalin objected and threatened to end all Allied cooperation in Europe; no Soviets were tried.²²

All this suggests that in the international context, Lady Justice might not be blind. There is no shortage of arguments that international justice is merely ‘victor's justice’,²³ and most of the examples above of earlier international justice efforts may be seen in that light, with powerful states championing and establishing international justice mechanisms that selectively prosecute military and civilian leaders of weaker states whose military has usually lost a conflict. In so far as the ICC has been perceived as a European-led effort which has mainly investigated and prosecuted people from former European colonies in Africa,²⁴ it too may be seen as analogous to victor's justice, to the extent that the powerful constituted a court that is prosecuting the weak.

Another posited tension between law and politics at the ICC suggests that, at least in some cases, justice through international criminal tribunals such as the ICC might be more likely to deter peace than deter atrocities: once a leader has perpetrated atrocities and faces the threat of prosecution, the argument goes, that leader is less likely than otherwise to conclude a peace agreement, which would risk them ending up in the dock. They are therefore more likely to fight to the end, with a view to a victory that would surround them with a guard that can protect against their arrest.²⁵ A contrary position is that the threat of justice, accountability, and prison will deter atrocities.²⁶ A third view, with considerable statistical evidence behind it, is that both phenomena have been taking place during the ICC era: civilian killing by state party governments is being deterred and diminished in civil conflicts, but civil conflicts are lasting longer, presumably after a leader has engaged in atrocities.²⁷

This article focuses on the broader question of how ICC organs and party states have navigated the tension between law and politics. As a legal institution embedded in global politics, the ICC was born with a congenital contradiction. It derives legitimacy by doing what is appropriate under Rome Statute principles and rules, pursuing justice according to the rule of law, applying the law as written to facts to determine the legality of an act or omission. However, legal action pursuant to the Rome Statute often challenges the interests of powerful political actors, and may appear illegitimate to targeted leaders and their followers, especially to those in non-party states subjected to the Court's jurisdiction. To survive, the ICC must often act politically, influenced by pressure from powerful actors to exercise its authority in ways they favour. Wedged between the law of the Rome Statute and global politics, a logic of appropriateness and a logic of consequences,²⁸ the ICC faces challenges from which it cannot completely escape.

This antinomy is illustrated here historically. Ideationally, the establishment of the ICC became a fundamentally European project, driven by norms, reflected in detailed substantive rules, procedures and practices, which had been developed over centuries in Europe and accepted widely there. Those norms and rules were accepted by many elites in western-influenced former colonies in the global south. The world's great powers – the United States, Russia, China, and India – are not state parties, as a result of a combination of normative differences and security interests.

During the ICC's first decade of operation, a legally conservative application of Rome Statute triggers, in which rules and their application were not stretched,²⁹ resulted in investigations and prosecutions of only Africans. That is not surprising, as Rome Statute crimes are more likely to take place in poor, weak or failed party states than in developed European party states with established borders and strong human rights norms.³⁰ Moreover, while the United Nations (UN) Security Council may refer cases to the ICC,³¹ the permanent members have blocked or vetoed Security Council efforts to refer situations that could result in trying their nationals or those of allies. Atrocities at least as grave as those in Africa were being perpetrated in non-party states like Syria, Iraq, Myanmar, and China, where the terms of the Rome Statute and powerful UN Security Council permanent members made it difficult to assert ICC jurisdiction.

While there is no consensus on what is ‘justice’, which is an essentially contested concept,³² many intuitively accept arguments based on the ‘like cases’ maxim – that similar cases should be treated similarly, and dissimilar cases dissimilarly.³³ Hence, many have seen the overall pattern of ICC investigations and prosecutions as unjust.³⁴ An African political reaction – based on arguments that Africa is being treated differently from the global north, and fuelled by prosecution of powerful African leaders – challenged the ICC's legitimacy, generated African claims of European neo-colonialism, and resulted in many African states ceasing to cooperate with the Court; some even threatened to withdraw from the ICC, thus imperilling its survival.

Starting in 2011, the ICC adapted with the election of two successive prosecutors generally favoured by African party states, and expansive assertions of jurisdiction that stretched its legal authority to investigate nationals of non-party states, such as those of the United States in Afghanistan, Myanmar in Bangladesh, and Israel in ‘the State of Palestine’. The Court also launched investigations of Russia in Georgia and Ukraine. Those events modestly dampened African complaints, but rekindled another legitimacy challenge – this one from non-party states that found their leaders in the crosshairs of a treaty regime they have not joined, with definitions of crimes or procedures they do not accept, sometimes facing criminal referrals from ‘states’ they do not recognise, and subject to legal determinations they contest.

Facing legitimacy challenges from within and without, the ICC finds itself in a bind, which will continue to shape and challenge its agenda. Section 2 of this article explains the ICC and its agenda, codified as the Rome Statute, as a political outcome, shaped largely by European norms and security interests of party states. Section 3 shows that politics always hangs over the head of the Court: the Prosecutor and the Court are constrained by and responsive to politics, some of which is intrinsic to the oversight and governance functions established in the Rome Statute, and other that is extrinsic to institutional rules. It also elaborates a dynamic historical account of the operation of the ICC, its legitimation crises, and its efforts to adapt, outlined above. Section 4 briefly describes some important political consequences of ICC actions in situation states, which should be taken into account in decisions of the Prosecutor and the Court, and informs strategic interaction between the Prosecutor and leaders in situation countries.

Section 5 concludes that the Court's most fundamental challenge is that it must be responsive to both its legal mandate and international political demands to survive. In applying the law of the Rome Statute, the Court has triggered claims of bias, assertions of the Court's illegitimacy, and political confrontation with and retaliatory actions by powerful party and non-party states. Yet adapting to these challenges by taking actions attentive to politics risks the ICC's legitimacy as a legal institution. While the tension between law and politics can be dampened by various changes in ICC policies, that tension and associated challenges cannot be fully resolved.

2. Explaining the ICC agenda and membership: European norms and the security interests of states

States that have become parties to the Rome Statute are those that accept the particular principles, rules and procedures set forth in the Statute (which embody largely European norms about laws of war), and that perceive their security is not weakened by joining the regime.

Ideas – beliefs held by people, which may be normative, ontological or epistemological – explain many foreign policies, international institutions and other political outcomes.³⁵ While it is frequently asserted that human rights are ‘universal’, the particulars of human rights and humanitarian law embodied in the Rome Statute are deeply rooted in western norms centred in Europe. These norms, hardened into laws of war and accompanied by western rules of procedure and practice, drove the development, form and content of the Rome Statute.³⁶ For many states and rulers, those norms have shaped their perception of the national interest. Yet those norms, and the particular legal definitions, rules and procedures into which they hardened, are not shared universally. Moreover, they stand in opposition to security interests of states and rulers in much of the world. Understanding the normative roots of the Rome Statute, and the security interests of states and rulers, helps to explain both the Rome Statute agenda and limits on the scope of ICC membership.

2.1. Western lineage

Following the International Military Tribunals after the Second World War, slow and halting negotiations to establish a permanent international criminal court ensued through the International Law Commission, the UN General Assembly and other venues, but it was not until 1998, at a moment of western unipolarity, that the ICC was finally constituted.³⁷ The Soviet Union had fallen; Russia was not very strong; and China possessed a fraction of its current economic and military power.

In that context, it is not surprising that the Rome Statute's agenda was supported largely by the west, reflecting western geopolitical interests in maintaining a territorial status quo. Yet among western and western-influenced states, there were significant political cleavages – transatlantic and north-south – and those cleavages influenced the Rome Statute agenda and membership.

The crimes and procedural rules defined in the Rome Statute are of European origin. Many people think of the laws and institutions addressing war crimes and genocide, and international criminal law more generally, as rooted in the Nuremburg trials at the end of the Second World War. Its foundations are much older than that.³⁸

Probably the most important modern western work in the development of international criminal law is Hugo Grotius, Laws of War and Peace,³⁹ written during the Thirty Years War, which at that time was the most destructive war Europe had experienced. Grotius rested many premises on the work of Ancient Greek and Roman philosophers, as well as Thomas Aquinas⁴⁰ and other scholasticists,⁴¹ and employed natural law reasoning, deducing central principles from those thinkers, and from the ‘Old Law’ (the Torah) and the ‘New Law’ (the New Testament).⁴² Using ‘right reason’, making deductions from those western sources, Grotius developed what are foundational principles of international humanitarian law today: (i) jus ad bellum: law on the right to wage war – namely, the principle that the only legitimate reasons for use of force are self-defence or righting a wrong; and (ii) jus in bello: law in warfare, establishing a civilian-military distinction, according to which military targets in wartime are legitimate but civilian targets are not, and a principle of proportionality, that in the use of force, military action must be proportionate to the wrong being righted.

These fundamental principles, of European pedigree, are echoed in the basic contemporary instruments of international criminal law. Chapter VII of the UN Charter sets the contemporary legal bases for the use of force, providing that the only legitimate legal bases are self-defence (individual or collective) or Security Council authorisation.⁴³ The Geneva Conventions and the Additional Protocols detail contemporary law in warfare. Article 48 of Additional Protocol I requires all parties to distinguish between civilians and combatants, while Article 50 defines a ‘civilian’, by way of reference to the Third Geneva Convention or Article 43, as ‘any person who is not’ a member of an armed force, militia, or openly using arms to resist an invading force.⁴⁴ Protocol I requires parties to ‘do everything feasible to verify that the objectives to be attacked are not civilians or civilian objects’,⁴⁵ and to ‘take all feasible precautions in the choice of means and methods of attack, with a view to avoiding, and in any event to minimizing, incidental loss of civilian life, injury to civilians and damage to civilian objects’.⁴⁶ The International Criminal Tribunals for the former Yugoslavia (ICTY) and Rwanda (ICTR) each defined crimes prohibiting genocide, crimes against humanity, and war crimes, again reflecting Grotian principles and incorporating much of the Geneva Conventions’ elaboration of those principles.⁴⁷

Similarly, the Rome Statute's prohibitions and definitions of genocide,⁴⁸ crimes against humanity⁴⁹ and war crimes⁵⁰ are aimed in large part at upholding the Grotian jus in bello principles of a civilian–military distinction and proportionality, and borrow heavily from the Geneva Conventions and Additional Protocols. Article 8 bis upholds Grotian jus ad bellum principles, contextualised to the United Nations era, defining the crime of aggression as using force for any reason not permitted by the UN Charter.⁵¹

Institutionally the Court is also western. Its rules of procedure and evidence⁵² are based largely on those used in common law and civil law countries. Procedure is a mix of adversarial legalism, as in common law countries, and inquisitorial, as in the civil law tradition.⁵³ This western bias in procedure and evidence, and in which only certain types of domestic criminal justice process satisfy the ICC's principle of complementarity,⁵⁴ have been a focus of critique from Islamic law perspectives,⁵⁵ Asian perspectives,⁵⁶ and from commentators sympathetic to traditional or restorative forms of justice and reconciliation.⁵⁷

Finally, the model for the Rome Statute, from which much of its text is lifted, was the statute of the ICTY, drafted largely by rapporteurs appointed by the Conference on Security and Cooperation in Europe, as well as by commissions of jurists from France and Italy.⁵⁸ It is hard to argue that the Rome Statute is not a fundamentally European instrument.

2.2. Arab states, Islamic Law, and the Rome Statute: An example of ideational differences

The western principles and definitions embedded in the Rome Statute, with deep roots in Judeo-Christian thought, conflict with ideas in many other societies. Most non-western societies maintain some deeply pedigreed approaches to what is permissible in warfare, but they differ in important ways from the Rome Statute. Confucian cultures can point to Sun Tzu's Art of War, which requires humane treatment of captives,⁵⁹ but there are no other meaningful humanitarian prescripts in the work; moreover, it advocates plundering the enemy's territory, which is a war crime under the Rome Statute,⁶⁰ and the focus of the work is on maximising the prospects of military victory, not on humanitarian norms. The Manusmriti from 200 BC may be cited as an authoritative repository of Hindu law that elaborates several just war principles, stating that war should be avoided by negotiation and reconciliation; that if war becomes necessary, a soldier must never harm unarmed civilians, non-combatants, or someone who has surrendered, and use of force should be proportionate.⁶¹ Neither of these works may be considered anywhere near as comprehensive or detailed as western humanitarian principles and laws. Moreover, while some of these non-western principles or laws concerning warfare bear similarities with those of the west, the rules diverge in various important respects.

By way of example, it is useful to consider how particular ideas, definitions, and prohibitions in the Rome Statute conflict with at least some interpretations of Sharia,⁶² particularly in many Arab states in and around the Persian Gulf.⁶³ Divergent interpretations of Sharia within Arab states, where some interpretations run contrary to the Rome Statute, suggest there is polarisation or lack of consensus within those countries on Rome Statute particulars, affecting whether certain Arab states could commit to the Statute.⁶⁴

The Islamic legal tradition is characterised by a range of views and debates that vary across time and space.⁶⁵ Many Islamic law scholars argue that the Qur'an not only contains ‘justice’ as one of its core principles, but delimits war in several ways that are consistent with western and Rome Statute norms, such as permitting war to be waged only in self-defence,⁶⁶ distinguishing between civilians and combatants, prohibiting execution or improper treatment of prisoners of war,⁶⁷ and prohibiting plunder and the destruction of civilian objects. Hence, all Muslim countries have signed the Geneva Conventions, and most Muslim scholars and states hold that international humanitarian law is consonant with the spirit and ultimate objectives of Islam.⁶⁸

Yet some of the rules developed in the past by classical Muslim jurists are inconsistent with international humanitarian law, and the positions of some contemporary Arab state Islamic law jurists are inconsistent with the Rome Statute's embodiment of western international humanitarian law.⁶⁹ The conflict between some of these interpretations of Sharia and international human rights law is reflected, for example, in the practice of Arab states to make reservations, known as the ‘Sharia reservations’, to human rights treaties, intended to avoid incompatibilities between the treaty provisions and Sharia rules and principles.⁷⁰

A few examples illustrate conflicts between some interpretations of Sharia in Arab countries and the law of the Rome Statute. Most broadly, some Islamic law scholars, jurists, leaders, and Muslim-majority states have expressed a fundamental concern about the sources of law that would be applied in their land, that joining the ICC would usurp the Sharia's exclusive jurisdiction in those states, substituting the law of God with the law of man.⁷¹ This general concern reverberated in particular instances in Rome Statute negotiations, such as in arguments that the detailed procedural rules and standards employed by the ICC conflict with the comparatively abstract Islamic law approach of interpreting Sharia principles,⁷² and in the efforts of several Arab state delegations to exclude various elements of crimes of gender violence, including those related to forced pregnancy.⁷³

Legitimate bases for the use of force are another area of some contention. As suggested above, while there was general consensus among classical Islamic jurists that self-defence is a legitimate basis for initiating warfare, there was disagreement on whether disbelief, such as polytheism, could justify warfare,⁷⁴ and some contemporary Islamic extremists have interpreted Sharia to permit the use of force to expel unbelievers from Muslim lands.⁷⁵ This basis for use of force, of course, is not recognised by Grotian principles, the UN Charter or the Rome Statute, which could treat such action as forcible transfer of a population, a crime against humanity under Article 7(2)(d) of the Rome Statute.

Similarly, as suggested above, the Qur'an distinguishes between civilians and combatants, but ‘civilian’ has been a subject of dispute among Islamic jurists. Most define ‘civilians’ to include women, children, the elderly, clergy and hired workers, but some have also treated all ‘adult’ males past puberty or over 15 years of age, physically able to engage in combat, as legitimate military targets.⁷⁶ Taken together, this definition of ‘civilian’ is narrower than the protected category of ‘civilians’ in the Geneva Conventions and Rome Statute: a widespread or systematic murder of males over 15 years of age, physically able to engage in combat, would constitute a crime against humanity under Article 7 of the Rome Statute.

An extremist Al Qaeda interpretation broadened ‘the definition of active participation [in armed conflict] to include roles that directly assist the enemy’, expanding the class of legitimate targets to include those providing financial and political support to the enemy.⁷⁷ Al Qaeda also argued that the principle of reciprocity in warfare supersedes the principle of distinction and cited Qur'an 2:194 (‘attack your attacker in like manner’) to permit the reciprocal targeting of civilians.⁷⁸ Daish (the Islamic State of Iraq and the Levant, ISIS) legal arguments overlapped with those of Al Qaeda on this point.⁷⁹ These extremist views have been sharply criticised by more mainstream Muslim jurists and public figures,⁸⁰ but they are part of a range of views among Islamic law scholars in the Arab world, which do not accord with western doctrine embodied in the Rome Statute.

Behaviour in accord with other interpretations of Sharia mandates may also be criminal under the Rome Statute. Corporal punishments laid out in Islamic law are contrary to modern, ‘universal’ human rights standards, including the prohibition of torture in the Rome Statute.⁸¹ In the Al-Hassan case, the ICC Pre-Trial Chamber agreed with the Prosecutor's contention that applying Sharia punishments in Timbuktu could have amounted to torture, and that applying Sharia law there could prove to be an organisational policy to commit a widespread or systematic attack on the civilian population. Taken together, that particular application of Sharia law could constitute a crime against humanity under Article 7(2)(e) of the Statute.⁸²

Hence, normative differences help to explain why all European states have joined the Rome Statute, whereas only five of the 22 Arab League states have joined.⁸³ Other explanations have been offered, such as general suspicion of western powers, or some Arab autocratic leaders’ fear of prosecution,⁸⁴ but those explanations could be epiphenomenal of the ideational differences identified above.

2.3. Norms, security interests, and ICC membership

The Rome Statute's European ideational pedigree, combined with the varying security interests of states and rulers, largely explains the composition of ICC membership. Societies in which elites have internalised European human rights norms, such as some former colonies or states that mimic a perceived western script of modernity,⁸⁵ shared a motivation to join the ICC. Conversely, societies with ideas about sovereignty, government, justice or human rights that diverge from those embedded in the Rome Statute have had little internal motivation to join the ICC.

Moreover, many states and rulers have security interests that run in opposition to Rome Statute rules and approaches. Some autocratic rulers need to crush internal or external opposition violently in order to retain or enhance their power or authority.⁸⁶ Some states, particularly great powers, have defensive military entanglements or aggressive ambitions that increase the likelihood of their engagement in armed conflict in which civilians may be harmed, thereby exposing their rulers, commanders or soldiers to the risk of Rome Statute liability greater than that of weaker states’ leaders.⁸⁷ Conversely, as is explored further in Sections 2 and 3 below, for some rulers, ICC membership offers a tool for action against opposition militias that harm civilians and commit Rome Statute crimes.

2.3.1. China

Consider why China did not sign the Rome Statute and has been suspicious of the Court. Ideationally, the approach of Confucian legalism to influencing human behaviour differs markedly from western legalism⁸⁸ and, while contested, some argue that ‘Asian values’ do not prioritise human rights.⁸⁹ Moreover, Marxist thought – modified by Lenin, Mao, and recently Xi Jinping⁹⁰ – not only stands in opposition to democratic governance, at least during the transition to an egalitarian communist society,⁹¹ but rationalises violence against, or mass purges of, those opposed to the socialist state,⁹² as well as international violence against imperialist powers,⁹³ which would likely be criminal under the Rome Statute. Chinese attitudes towards the west are influenced by the 1839–1949 ‘century of humiliation’⁹⁴ in which China was occupied and coerced into concessions, ending with western support of Chiang Kai-Shek, in opposition to Mao Tse-Tung and the Chinese Revolution. Western refusal to recognise the People's Republic of China for another few decades further strained Sino-western relations. China's suspicion of western driven international criminal adjudication, in particular, was furthered by the Tokyo Trials of 1946, with China concluding correctly that the United States made politically motivated decisions which hampered justice, planting ‘seeds of distrust’ of international criminal adjudication, which have lasted to the present day in China.⁹⁵

Chinese government domestic security behaviour is consistent with the rationales of the Chinese Communist Party (CCP) for crushing internal opposition through violent means, action that would contravene the Rome Statute. Several Rome Statute provisions limit state behaviour in non-international armed conflict,⁹⁶ reflecting a normative view that certain restraints on violence apply not only to international armed conflict but also to how a state treats those within its borders.⁹⁷ In the Rome Statute negotiations, China, like many autocracies, opposed allowing any of the defined crimes to apply in non-international armed conflict or to domestic affairs.⁹⁸ This stance is consistent with decades of Chinese government atrocities against its own people, which include mass killings during the Cultural Revolution,⁹⁹ Tibetan repression,¹⁰⁰ and the contemporary genocide in Xinjiang.¹⁰¹

The global revisionist ambitions of the CCP constitute an international security basis for its opposition to the ICC. It is noted above that the Court was founded at a unipolar moment, when the west was dominant. In that context, it is not surprising that the Rome Statute effectively supports the status quo global order: it is a legal-political effort to restrain violence and armed conflict, including violence that would be a means of undermining the extant global political-economic order. Geopolitically revisionist leaders, like Xi, have demonstrated or announced their willingness to use violence to change territorial borders and to revise the global order.¹⁰² Rome Statute provisions aimed at delimiting violence could constrain that behaviour – another reason why contemporary revisionist powers like China are not parties to the Rome Statute, and why they consider it a potential threat. Moreover, China continues to assert legally suspect territorial claims with bordering states or entities – including India, Japan, Taiwan, and Vietnam – and clearly illegal territorial claims over the South and East China Seas.¹⁰³ Use of force to assert those claims could subject Chinese leaders to ICC jurisdiction, especially if it were to become a state party.¹⁰⁴

2.3.2. Russia

Similarly, the Russian government's hostility towards the Court may be explained by its internal and international security objectives, and the ideas that define them. Domestically, setting aside a brief flirtation with democracy under Yeltsin, Russia has only a Tsarist or autocratic tradition, and has long engaged in repressive atrocities perpetrated against dissidents, with millions killed during Stalin's reign, for example. Contemporary Russian fears of being subjected to ICC jurisdiction over non-international armed conflict, particularly in the Caucasian region of Russia,¹⁰⁵ have been a specific source of its hostility towards the Court.

Internationally, Putin's revisionist ambitions, which he sees as an effort to restore Russia's rightful geopolitical position, forms another basis for its opposition to the ICC. While Tsarist Russia was influential in the early development of international humanitarian law, the 1899 Hague Convention in particular, and the USSR played a key role in establishing the Nuremberg and Tokyo trials,¹⁰⁶ contemporary Russia soured on international criminal tribunals in the 1990s, when it became clear that most ICTY prosecutions would focus on long-time Slavic allies: Serbian leaders and military personnel, Bosnian Serbs, and nationals from the Republika Srpska.¹⁰⁷ Russian scholars further questioned the legitimacy and legality of the ICTY after NATO's bombing of Kosovo in 1999.¹⁰⁸ While Yeltsin's Russia signed the Rome Statute in 1998, Putin's Russia never ratified it and became openly hostile towards the Court, incensed initially by the ICC's engagement in the situation in Georgia (South Ossetia).¹⁰⁹ This was aggravated by the ICC opening a preliminary examination of the situation in Ukraine after Russia's 2014 incursion in the East,¹¹⁰ and finally by the ICC Prosecutor's investigation of the situation in Ukraine after Russia's 2022 aggressive invasion;¹¹¹ Russia responded to this, in part, by charging 92 members of Ukraine's military with crimes against humanity. One scholar has suggested that engagement in armed conflicts is now a permanent facet of Russian foreign policy, driving Russia's efforts to weaken the ICC;¹¹² another has argued that Russia was particularly concerned about ICC determinations of crimes of aggression.¹¹³

2.3.3. United States

The United States’ decision to not join the ICC regime is more complicated than those of the states discussed above. In contrast to those states, the US government shared with Europeans support for the principle of criminal accountability for mass atrocities, but the US approach to human rights, humanitarian law, and particular definitions and procedures elaborated in the Rome Statute differ considerably from the European approach.¹¹⁴ Moreover, the United States has national security interests that many argue demand more limited ICC jurisdiction than that favoured by European states. The United States built a global order that champions principles of democracy, human rights and accountability for mass atrocities,¹¹⁵ but it has been ladened with global military commitments and responsibilities to maintain that order. With more than 750 military bases in 80 countries,¹¹⁶ and military commitments to maintain order on every continent, US military personnel would face, and have faced, substantial risks of liability under the Rome Statute, even if illegal behaviour was inadvertent or unauthorised by superiors, whereas most ICC state parties have comparatively few or no military entanglements or risks. Hence, the biggest US objection to the Rome Statute was the extent of prosecutorial discretion. The US government wanted the ICC to take only cases referred to it by the UN Security Council, decisions over which the United States has a veto. There were other objections to Rome Statute provisions that generated substantial political concern in the United States: for example, many made the contested argument that the ICC denies fundamental US constitutional due process guarantees, such as the right to trial by jury, which may not be discarded by treaty.¹¹⁷

The US government also has strongly objected to a Rome Statute provision that targets leaders of Israel,¹¹⁸ the most important US military ally in the Middle East. An overtly political provision, added at Rome, could treat certain Israeli facilitators of West Bank settlements policy as war criminals, despite the wording of that provision having no pedigree as a war crime, and voluntary settlement of occupied territory by non-state actors never before having been treated as a war crime.¹¹⁹ Rome Statute Article 8(2)(b)(viii) elaborates as a war crime ‘[t]he transfer, directly or indirectly, by the Occupying Powers of parts of its own civilian population into the territory it occupies’. The Fourth Geneva Convention has similar language, but with one crucial difference: Article 49(6) states that ‘[t]he Occupying Power shall not deport or transfer parts of its own civilian population into the territory it occupies’, but does not include ‘directly or indirectly’, which substantially expands the prohibited behaviour. According to the official commentary on the Fourth Geneva Convention, the sixth paragraph of Article 49 was intended to prevent a practice adopted during the Second World War, by which the Nazi German government identified people of German blood and forced them to move to annexed territory.¹²⁰ Lebanon and Syria proposed the ‘directly or indirectly’ language in order to target the Israeli government's policies of allowing settlements to be established or inhabited voluntarily. Israel was obviously displeased with the Lebanese/Syrian proposal, as was the United States, but only eight states voted against the language.¹²¹ While Israeli government support of settlements in the Occupied Territories may be morally reprehensible and self-defeating policy, adding the modified language to Rome Statute Article 8(2) was overtly political, a deviation from the main project of Article 8, which generally codifies well-established war crimes. It is the only crime in the Rome Statute aimed at only one state, and it is a provision that Israel's ally, the United States, could not support. Primarily for the foregoing reasons, the United States purported to ‘unsign’ the Rome Statute and is not a state party.

2.3.4. North-south politics

Finally, north-south politics also shaped the Rome Statute, albeit in limited ways. Two examples are illustrative. First, in the early 1990s, during the International Law Commission's drafting process to create the Draft Statute for an International Criminal Court, some countries in the global south favoured including crimes beyond those that would find their way into the Rome Statute, but those crimes were excluded: for example, illicit narcotics trafficking; recruitment, use, financing, and training of mercenaries; and wilful and severe damage to the environment. Of course, these crimes have not been considered part of international humanitarian law, which was the impetus for the Rome Statute, but it is worth noting that the excluded criminal behaviour is often perpetrated by actors from the global north, from states that opposed including them in the Statute. As discussed below, the Rome Statute has been criticised for targeting African countries; some critical analysts suggest that bias is explained partly by the exclusion of the crimes above: if they had been included in the Statute, then the ICC would be pursuing more cases against perpetrators from the north.¹²²

Second, many find it surprising that the use of chemical and biological weapons is not illegal per se under the Rome Statute.¹²³ There was a proposal to make it illegal, but the global south was opposed. Several developing countries took the position that because possession or use of nuclear weapons would not be illegal under the Rome Statute, then they would not agree to prohibiting the possession or use of chemical or biological weapons, regarded by some as the ‘poor man's’ nuclear weapons.¹²⁴ Hence, despite overwhelming international support for criminalising the use of such weapons, as a concession to the global south their criminalisation is excluded.

2.3.5. Western-dominated membership

Building an institution with European ideational lineage, and European-influenced conceptions and definitions of law, procedure and practice, was a political decision. Excluding crimes favoured by much of the global south was a political decision. Including crimes associated with non-international armed conflict; establishing broad prosecutorial discretion; not limiting jurisdiction to referrals by the UN Security Council; adding a new war crime that targets Israeli settlements – all were political decisions.

The resulting agenda has had ramifications not only for the operation of the ICC, discussed below, but for its membership. While the Statute's European lineage and influence suggest neither that all western countries have joined the regime, nor that all non-western countries reject all Rome Statute norms, it should not be surprising that the ICC's western ideational bias and European influence is mirrored in the composition of states party to the Rome Statute: almost all of Europe, and most of its former colonies in North America, South America, and Africa are state parties. In contrast, only a handful of countries east of the 30^th east meridian are ICC state parties (that is, east of a vertical line on a world map, roughly running east of Finland and the Democratic Republic of the Congo (DRC)); few Arab states are state parties; and a much smaller proportion of Muslim-dominated states are parties to the Rome Statute than Christian-dominated states.

3. Law, politics, and adaptation

Law and politics drive ICC operations. Demands and constraints on the operation of the primary organs of the Court – the Prosecutor, Chambers, and the Registry – may be conceptualised into three nested categories. The operational law of the Rome Statute includes mandates for and constraints on the organs, but on many crucial matters the Statute offers them broad discretion.¹²⁵ The application and interpretation of that law are nested in the ICC's constitutional governance structure: the breadth of the organs’ discretion and interpretation of discretionary provisions may be limited or shaped by the Assembly of States Parties (ASP), pursuant to the Statute's constitutional governance provisions.¹²⁶ Both the operational law of the ICC and its constitutional governance structure are nested in raw politics: Court organs and the ASP face pressures extrinsic to the Statute by way of state behaviour that is independent of the Statute or its formal governance structures.¹²⁷ These extrinsic political pressures, including threats to exit, have had the most substantial effects on the direction of the Court.

These three drivers operate dynamically and interact in ways that tell a legal and political history of the Court. During the Court's first decade, the UN Security Council and some African leaders referred several situations to the Prosecutor, who is mandated to investigate party state self-referrals, resulting in an exclusive focus on African states in its early years. This led to claims, elaborated below, of an anti-Africa bias and assertions that the ICC is a neocolonial project, challenging the Court's legitimacy, and subjecting it to a chorus of populist demands in Africa to withdraw from the ICC.¹²⁸ Out of necessity, the ICC adapted. The ASP took various actions to appease African party states.¹²⁹ Seemingly, so did Chambers and the Prosecutor, targeting several non-party states, including Israel, initially for actions not traditionally addressed by international criminal law.

3.1. Legal mandates as constraints?

The agenda set by negotiators at the Rome Conference, reflecting the interests and norms of party states, directs and constrains ICC organs in many important ways. For example, the Statute confers limited subject-matter jurisdiction over four core crimes: genocide, crimes against humanity, war crimes, and the crime of aggression. No other crime may be charged. Moreover, there are relatively unambiguous terms in the Statute: ‘bullets’, ‘bombs’, and ‘hospital’ are nouns that lend themselves to relatively straightforward application, for example.

Yet the Statute offers the Prosecutor and judges considerable flexibility. Law has ambiguities intrinsic to language,¹³⁰ and international law is notoriously flexible, as diplomatic negotiators often resolve drafting disagreements by adopting ambiguous language or leaving gaps that subsequently afford international actors a range of interpretations and associated behaviour.¹³¹ Moreover, the law of the Rome Statute is particularly flexible, as a result of diplomatic negotiators’ decisions to invest considerable discretion in the Prosecutor and Chambers. For example, based solely on the terms of the Statute, the Prosecutor has unfettered discretion about how many resources to commit to one situation under investigation versus another. When investigating a situation, the Prosecutor is directed to pursue those ‘criminally responsible’,¹³² but the Prosecutor may decide precisely whom to investigate, against whom to seek an arrest warrant, and how many arrest warrants to seek. Whether a particular case is admissible is based partly on whether the alleged crimes have sufficient ‘gravity’.¹³³ What does that mean? What must be the scale of atrocities to deem a situation of sufficient ‘gravity’? One hundred people murdered in a particular situation? A thousand killed? A million? The Office of the Prosecutor has adopted a policy document which includes several factors that it considers¹³⁴ – which include the scale of the crimes, the sadism with which crimes are committed, and so on – but an unweighted list of factors is merely a framework for decision making that effectively replicates the ambiguity inherent in ‘gravity’.

Subject to the Pre-Trial Chamber's oversight, the Prosecutor may also decide whether to refrain from investigating if doing so ‘would not serve the interests of justice’.¹³⁵ For millennia, philosophers and jurists have debated what is ‘justice’.¹³⁶ Similarly, in applying Article 8(2)(b)(iv), the Court must decide whether an attack was launched with knowledge that it would cause ‘incidental loss of life … which would be clearly excessive in relation to the concrete and direct overall military advantage anticipated’ – a radically ambiguous balance.

Given the ambiguity of many of the Statute's mandates, limits and standards, what then is the next line of constraints on the Court and the Prosecutor? What stops the ICC from having a run-away Prosecutor, or the opposite: a Prosecutor who ignores horrific crimes? Or a Prosecutor who is biased in investigating rebel groups to the exclusion of government forces? Or Chambers that makes law in ways that deviate from what the Court's founders intended, or from what is fair and even-handed, or from what is corrupt or politically driven?

3.2. Adaptation governance politics: The Assembly of States Parties

Under the Rome Statute, the Assembly of States Parties (ASP) has several authorities that could constrain, correct or direct the organs of the Court, but those authorities have been exercised modestly. The primary mechanisms are (i) an oversight authority,¹³⁷ which is intended to enhance the ‘efficiency and economy’ of the Court, and may bring to light unjust behaviour, but which does not formally intervene in judicial matters; (ii) legislative authority, which is limited by the difficulty of meeting majority or supermajority voting thresholds; (iii) budgetary control, which has been used to enhance efficiency, but never to influence particular cases or situations;¹³⁸ and (iv) election of the Prosecutor, Registrar and the judges.¹³⁹

3.2.1. Oversight

The oversight authority is limited by the principles of non-interference in prosecutorial activities and judicial independence.¹⁴⁰ Article 112(2)(b) of the Rome Statute provides that the ASP shall provide management oversight of the Presidency, the Prosecutor, and the Registrar regarding the administration of the Court; Article 112(4) of the Statute provides that the Assembly may establish such subsidiary bodies as may be necessary, including an independent oversight mechanism for inspection, evaluation, and investigation of the Court, in order to enhance its efficiency and economy. However, Article 42(1) provides that the Office of the Prosecutor shall act independently as a separate organ of the Court, and Article 42(2) provides that the Office shall be headed by the Prosecutor, who shall have full authority over the management and administration of the Office, including the staff, facilities, and other resources.

Oversight usually has originated with the actions of one or a few states that build a broader consensus within the ASP. As in most international organisations, the policy process is typically initiated by a powerful state or coalition of like-minded states, which slowly expands the circle of states involved in developing a proposal, until it is presented as an action item to the plenary body as a whole.¹⁴¹ The United Kingdom, for example, launched a high-profile effort to reform the ICC at the 17th ASP meeting in 2018, suggesting that the Court had spent too much money for too few convictions, made insufficient use of performance indicators to increase its efficiency, failed to investigate allegations that had been made about the first Prosecutor, and not adequately respected the principle of complementarity.¹⁴² The Foreign and Commonwealth Office followed the ASP statement by convening a subgroup of state parties at Lancaster House (London) in May 2020 to consider Court reforms, an action resented by excluded states. Court President Chile Eboe-Osuji then reinvigorated the reform effort, sending a letter to the ASP President, requesting the establishment of an independent expert assessment of Court practice, which led to the establishment of an Independent Expert Review (IER) at the ASP in December 2019. That review generated 384 recommendations for reform, covering all aspects of the Court's operations, including governance, human resources, ethics, budget, and use of performance indicators;¹⁴³ Chambers’ working methods, efficiency, and processes and procedures; prosecutorial criteria for case selection, prioritisation, hibernation, and closure; and Registry policies on victim representation.¹⁴⁴ An ASP-approved process has been under way since then to implement many of those reforms,¹⁴⁵ and while IER-recommended reforms may subtly shift the direction of the Court, perhaps increasing efficiency and performance, the initiative has not interfered in particular matters before the Court.

Hence, formal ASP oversight has not been used to influence judicial or prosecutorial behaviour substantially on particular matters. Effective oversight may help the Court to accomplish its objectives by bringing a systematic, disciplined approach to evaluate and improve ‘the efficiency and economy’ of the Court, assisting the Assembly and the various organs of the Court in the effective discharge of their responsibilities. At the same time, the independence of the Prosecutor and Chambers is critical for the credibility of the Court, so the ASP and the Independent Oversight Mechanism have engaged with the organs in only limited ways, and they have never interfered in decisions to investigate particular situations or to prosecute particular individuals.

3.2.2. Legislative authority

The ASP is further endowed with legislative authority. Article 112(7) of the Rome Statute provides that the ASP should strive to take decisions by consensus, but that if a consensus cannot be reached, the ASP may take procedural decisions by a simple majority of state parties present, or substantive decisions by a two-thirds majority of those present (provided there is a quorum). Rome Statute amendments require adoption by two-thirds of state parties.¹⁴⁶ In practice, those majorities and supermajorities have not formed easily. The highest-profile legislative actions have included amendments to fully establish the crime of aggression and activation of the amendments. Combined with the norm of non-interference in judicial or prosecutorial decisions, the ASP has never taken a substantive or procedural decision that has interfered with a specific investigation, prosecution or judgment.

3.2.3. Budgetary authority and contributions

Budgets are used commonly to control or direct organisational behaviour.¹⁴⁷ ICC operations are funded by state parties and the Court's annual budget is approved by the ASP. The Court has spent over 2 billion euros since its inception, and the ASP approved an annual budget of 155 million euros in 2022.¹⁴⁸ Various factions of state parties, often complaining of inefficiency, have limited their contributions, supported a reduced budget, or adopted a policy of zero nominal growth of the budget. However, neither the level nor the allocation of the budget has ever been used by the ASP in conjunction with a mandate to any Court organ to pursue or refrain from a particular investigation or prosecution, or to pressure Chambers to take any particular position on the law.

3.2.4. Election of the Prosecutor, judges, and the Registrar

Elections are inherently political, and the ASP elects the Prosecutor, the Registrar, and the judges – the most important figures at the ICC.

The composition of the bench reflects the political nature of the elections, yet there is no evidence that elections have directly skewed judicial opinions. Nearly a third of ICC judges have held some form of diplomatic position for their country, leading some to question the relevant legal knowledge and legal analytic skills, independence, and impartiality of those former diplomats.¹⁴⁹ Lengthy and expensive campaigns are often required for judicial nominees and poorer party states are less likely than wealthier states to successfully back their nominees. Voting blocs of the state parties, and vote trading, operate to drive results that are highly politicised. In 2011, the ASP established an Advisory Committee on Nominations of Judges,¹⁵⁰ but many suggest that the Committee's assessments have lacked rigour and have had minimal impact on the voting process or on limiting political interference in the election of unqualified judges. Taken together, these practices have yielded a bench of judges who are not uniformly skilled in legal analysis, not geographically representative of the state parties, and allegations that some are not impartial, although few – other than non-prevailing parties in judicial decisions – have claimed that any particular judicial decision has been politicised.

The ASP elections of the second and third Prosecutors were transparently political, but again there is little evidence of a direct effect on any particular investigation or prosecution. There have been three ICC Prosecutors. Election of the first Prosecutor, Luis Moreno Ocampo, was the least politicised. As a young lawyer, Ocampo had been deputy prosecutor in the trial of the Argentine military junta; immediately prior to his election he was the Robert F. Kennedy Visiting Professor at Harvard Law School and a board member of Transparency International. He was elected by consensus.

By the end of Ocampo's tenure, all of the Office of the Prosecutor investigations involved Africa; many of the continent's leaders were claiming that Africa was being targeted unfairly; and an African Union summit had resolved not to enforce arrest warrants issued against African leaders. There was strong pressure in some quarters for Ocampo's replacement to be African.¹⁵¹ Many hoped that the election of Ocampo's successor would be free from political considerations and centred exclusively on electing the most qualified nominee. The ASP elected Fatou Bensouda of Gambia, who had been Ocampo's deputy and was unquestionably qualified, but while others on the short list were also qualified, Bensouda was the only short-listed African. Throughout her tenure, Bensouda argued that the ICC was not biased against Africa; she declined to open any new investigations in Africa via her proprio motu authority – except in Burundi, a tiny, weak country, where crimes had been committed too overtly to ignore – and instead opened investigations outside Africa, in Afghanistan, Bangladesh/Myanmar, Georgia, Palestine, the Philippines, Ukraine, and Venezuela.

When Bensouda's tenure ended in 2020, many African states were still complaining of an anti-Africa bias. Kenyan leaders, who had been the subject of arrest warrants for their role in 2007 post-election violence, were particularly active and vocal in their attacks on the Court. In a contested election, the ASP chose as the third Prosecutor Karim Khan. Khan is a well-respected British barrister, with substantial international criminal law knowledge and experience, who was well known at the ICC for having served as defence counsel to William Ruto, one of the Kenyan leaders who had been charged with crimes against humanity in connection with the 2007 post-election violence.

3.2.5. Conclusion regarding ASP governance politics

In summary, while the ASP is a political body and has influenced the organs of the Court and its direction, none of the governance mechanisms have been used to surgically direct the Prosecutor or judges with regard to any particular investigation, prosecution or decision. Some mechanisms, such as the oversight authority, have been used to modestly improve the functioning of the Court; others, such as election of the Prosecutors, have helped to assuage the political concerns of African states, without sacrificing the quality of the Prosecutors.

3.3. Adaptation and raw politics extrinsic or adjacent to the Rome Statute governance system

The most substantial political influences on the ICC have been state actions extrinsic or adjacent to the formal governance structures of the ICC. Some actions taken by party states and non-party states, respectively, are described below, followed by illustrations of ways in which those actions appear to have directly influenced the organs of the Court, triggered action or yielded inaction by the Prosecutor, or catalysed action by the ASP.

3.3.1. Adapting in response to state parties’ actions

State parties have pursued a range of strategies and tactics, wholly or partly extrinsic to Rome Statute procedures and institutions, to influence the operation of the ICC. For example, in response to concern about Russia's attack on Ukraine, several states provided the Office of the Prosecutor with additional funds, independent of those states’ budgetary obligations, or seconded staff to the Office, to support its investigation.

Several actions by African states since 2011 illustrate strategies, outside the Rome Statute framework, that appear to have influenced the ICC. By 2011, many African heads of state and others had become concerned about the Court's direction, based on a combination of self-interest and principle. In 2009, the Pre-Trial Chamber issued an arrest warrant for Omar Al-Bashir,¹⁵² a sitting head of state, despite the doctrine of immunity for sitting heads of state,¹⁵³ based on seemingly sound reasoning on which the Appeals Chamber would later elaborate: Rome Statute Article 27(2) provides that ‘immunities or special procedural rules which may attach to the official capacity of a person, whether under national or international law, shall not bar the Court from exercising its jurisdiction over such a person’.¹⁵⁴ Later, in 2012, ICC charges were confirmed against two popular Kenyan politicians, Uhuru Kenyatta and William Ruto,¹⁵⁵ who would soon be elected President and Vice-President, respectively.

More significantly, by 2011 all of the situations under investigation by the Prosecutor were in Africa, while several glaringly criminal situations outside Africa, many outside the Court's jurisdiction (such as Palestine and Syria), were not being investigated. To some, this appeared to be an anti-African bias, but of the seven African situations under investigation at that time, six had been referred to the Prosecutor in ways that left no discretion to refuse investigation. Specifically, four of the situations had been referred to the Prosecutor by the governments of the situation states themselves (Central African Republic, DRC, Kenya, Uganda),¹⁵⁶ referrals that required an investigation.¹⁵⁷ Many commentators have argued that they were referred by those governments’ leaders in a self-interested effort to enlist an outside agent's actions against rebel groups operating in those countries.¹⁵⁸ Two of the seven situations had been referred to the Prosecutor by the UN Security Council (Libya and Sudan), also requiring an investigation.¹⁵⁹ Only one of the seven investigations (Côte d'Ivoire) had been undertaken pursuant to the Prosecutor's proprio motu authority, an investigation that required and received consideration and pre-approval by the Pre-Trial Chamber.¹⁶⁰

The African bias claim was first asserted most prominently by leaders of three states: Kenya, Sudan, and South Africa. While the Kenyan government had referred the 2007 post-election violence situation to the Prosecutor for investigation, that government had fallen out of power, and the newly elected government was to be led by Uhuru Kenyatta and William Ruto, two targets of the ICC investigation. Kenyatta and Ruto turned the situation into a populist electoral strategy, arguing that the ICC is a neocolonial expression of European power aimed at Africa. That message was embraced by the Sudanese government, led by Omar Al-Bashir, who was the subject of an ICC arrest warrant for genocide in Darfur. The position was then adopted as a populist claim by politicians in South Africa, accepted by a majority of African governments and much of the African populace,¹⁶¹ and expressed provocatively by the African Union Chair, Ethiopia's Prime Minister: ‘The ICC is hunting Africans’.¹⁶²

Beginning in 2009, the African Union, which includes both party and non-party states, began taking decisions that challenged the ICC. The Union first adopted a resolution to amend the Rome Statute to give the UN General Assembly authority to defer the proceedings of the Court and to grant heads of state immunity from prosecution during their time in office.¹⁶³ This was followed by another resolution directing its members to cease cooperating with the Court in executing the Bashir arrest warrant,¹⁶⁴ despite the obligation on state parties to arrest an accused for whom a warrant has been issued.¹⁶⁵ Several African state parties welcomed visits by Omar Al-Bashir, despite an arrest warrant having been issued, based on claims that the Court was biased against African states and that it was impermissible to arrest a sitting head of a non-party state, even when the situation had been referred to the Court by the UN Security Council. That legal argument was rejected by the Appeals Chamber,¹⁶⁶ and the African Union later adopted a resolution in 2018 to take its argument to the International Court of Justice.¹⁶⁷

Even more challenging, in 2017 the African Union passed a resolution suggesting that its member states should withdraw from the Court en masse,¹⁶⁸ and issued a proposal for an African Criminal Court.¹⁶⁹ Gambia, South Africa, and Burundi threatened to withdraw from the ICC. Burundi did so; the Gambian government decided to remain; and South Africa reversed course only after its Supreme Court ruled that the process used by the government was unconstitutional. Exit, or the threat of exit, may be potent,¹⁷⁰ and the possibility that the campaign could gather steam, with as many as 34 African states withdrawing from the ICC, was deeply concerning.

So, how did the ICC respond? As described above, the ASP elected the second Prosecutor, Fatou Bensouda of the Gambia, an African jurist, and when her tenure ended, it elected as its third Prosecutor Karim Khan, who had been defence counsel for William Ruto in the Kenya case. The election of persons with credibility in Africa might have helped to dampen African opposition to the Court.¹⁷¹

What did the new Prosecutor do? Fatou Bensouda began investigations into several situations outside Africa, some in which nationals of non-state parties were the likely targets, including investigating actions of Americans in Afghanistan,¹⁷² Russians in Georgia,¹⁷³ Russians again in Ukraine,¹⁷⁴ Burmese for crimes against the Rohingya,¹⁷⁵ and Israelis in the situation in the State of Palestine.¹⁷⁶ Bensouda also launched an investigation into the situation in the Philippines,¹⁷⁷ a state party that responded by withdrawing from the ICC (though Philippines nationals remain subject to the jurisdiction of the Court for crimes committed before withdrawal). While she used her proprio motu authority to pursue one new African case in a small country – the situation in Burundi,¹⁷⁸ where government figures had perpetrated flagrant and well-publicised war crimes – she did not use that authority to investigate several other situations in Africa that seemed equally grave; these included the Banyamulengue-Bembe violence in the DRC,¹⁷⁹ renewed Hema-Lendu violence there,¹⁸⁰ the killing of 500 people and displacement of 275,000 in Burkina Faso,¹⁸¹ and Boko Haram's killing of 35,000 people in Nigeria since 2009.¹⁸²

This turn away from Africa, towards investigations of situations involving non-state parties, required some creative jurisdictional theories. For example, Myanmar is not a state party, and the Rohingya victims are not nationals of a state party, so jurisdiction over the situation was not easily established. However, many Rohingya were forcibly displaced to Bangladesh, a state party, so the Prosecutor's investigation focuses on forced displacement, a crime that was completed in Bangladesh, conferring jurisdiction, a theory endorsed by the Pre-Trial Chamber.¹⁸³ The investigation of Russian actions in Ukraine is based on Ukraine, a non-party state, having conferred jurisdiction on the Court,¹⁸⁴ followed by state party referrals of the situation, triggering the investigation. Jurisdiction over the situation in Palestine entails reasoning, described below, that is creative and contested.

Not even Chambers seems to have escaped concerns about Africa. Judge Chile Eboe-Osuji, a widely respected former President of the Court, has said that many at the ICC wanted the Kenya case to ‘go away’.¹⁸⁵ He has argued that ‘[t]he organs of the Court must take policy considerations into account’ as ‘failure to do so risks their legitimacy’.¹⁸⁶ Hence, on grounds of ‘policy considerations’, the ASP adopted Rule 134 quater and the Court permitted William Ruto, an accused who at the time was Kenya's Vice President, to remain in Kenya during trial hearings,¹⁸⁷ despite the clear mandate of Rome Statute Article 63(1): ‘The accused shall be present at trial’.¹⁸⁸

In 2021, the Pre-Trial Chamber affirmed Prosecutor Bensouda's request for authorisation of an investigation into the situation in the State of Palestine, another territory outside Africa, with reasoning that is controversial. In April 2012, on grounds that Palestine was not a state, Prosecutor Ocampo had rejected the Palestinian Authority's attempt to confer jurisdiction.¹⁸⁹ Subsequently, in November 2012, the Palestinian Authority went to the UN General Assembly, which voted, 138 to 9 (with 41 abstentions), to grant Palestine observer state status.¹⁹⁰ In 2015, the Palestinian Authority attempted to deposit an instrument of ratification of the Rome Statute with the UN Secretary General, who accepted it. Article 125(3) of the Rome Statute provides for this means of acceding to the Statute, but it also provides that only a state may become a state party. The President of the ICC then welcomed the ‘State of Palestine’ as the 123rd state party. In 2018, the Palestinian Authority referred the ‘Situation in the State of Palestine’ to Prosecutor Bensouda, a self-interested move in that the clearest targets of any ensuing investigation would be the Palestinian Authority's chief nemeses – Israel and Hamas. The Prosecutor responded by launching a Preliminary Examination on ‘policy grounds’,¹⁹¹ leaving open the legal question of whether the Court had jurisdiction in the matter. In 2020, Prosecutor Bensouda decided to initiate an investigation and sought a ruling from the Pre-Trial Chamber on the scope of the Court's territorial jurisdiction. That question forced the Court to consider whether Palestine is a state, because only a state could have deposited an instrument of ratification of the Statute. Forty-four amici briefs and eight government briefs¹⁹² focused on three alternative possibilities.

Two of the possibilities revolved around the four ‘objective’ elements for determining the existence of a state, specified in the Montevideo Convention: (i) permanent population; (ii) a defined territory; (iii) a government that exercises effective control over a territory; and (iv) capacity to enter into relations with other states.¹⁹³ The first position was that Palestine does not have a government which controls a territory that constitutes ‘Palestine’. Which government exercises control? The Palestinian Authority? Hamas? Which territory does this government control? Gaza? The West Bank? Both? ‘Area C’, comprising 60 per cent of West Bank territory, which is under Israeli control exclusively, under the terms of the Oslo Accords?¹⁹⁴ ‘Area B’, where – also under Oslo – the Palestinian Authority exercises administrative control, but shares security control with Israeli authorities? Or only ‘Area A’, for which the Palestinian Authority exercises exclusive administrative and police authority under Oslo? In any event, under the Oslo Accords, final borders are the subject of negotiation. If one concludes ‘objectively’ that Palestine is not a state, then is very hard to conclude it is a state party.

A second stance is that Palestine objectively is a state under the Montevideo Convention, and is widely recognised, so it is a state party. There are plenty of precedents for interpreting the ‘objective’ Montevideo elements liberally. In 1960, Congo was widely recognised as a state, despite its engagement in a civil war with no effective governance over much of its territory. Similarly, even though borders were in dispute, several states have been recognised: Bosnia in 1992; Croatia in 1992; Israel in 1948. Hence, the argument goes, one is free to assert that Palestine is also a state – and so is a state party – despite having parallel competing governments in Gaza and the West Bank, and disputed borders. This liberal interpretation of the ‘objective’ legal criteria transforms a decision purportedly based on the Montevideo Convention into one based largely on the politics of statehood recognition, effectively consistent with a constitutive theory of statehood.¹⁹⁵

A third stance is somewhat more nuanced and rooted in one of the most important treaties in the region, the Interim Agreement of 1995, which continues to serve as a basis for rights and responsibilities in the Occupied Territories. This treaty was at the core of the comprehensive, clear-headed, 164-page partially dissenting opinion: Judge Kovács argued that Palestine is a state, is widely recognised as a state, and is a state party, but the Court has no jurisdiction over Israelis because the Interim Agreement denies the Palestinians criminal jurisdiction over Israelis,¹⁹⁶ and a state party may not confer on the ICC criminal jurisdiction it does not have.

What did the Court do? The majority concluded it did not need to decide whether Palestine is a state! The majority laid the decision at the feet of the United Nations, arguing that Palestine is a state party by virtue of the Palestinian Authority having deposited a ratification instrument with the Secretary-General, who accepted it – and the State of Palestine encompasses the pre-1967 territory of Gaza and the West Bank, including East Jerusalem.¹⁹⁷ This argument belies the main question, as only a state could have deposited an instrument of ratification. It implies that the General Assembly's grant of observer state status decided Palestinian statehood for the Secretary General (which therefore accepted the deposit) and so for the Court. In short, a majority vote of the General Assembly made Palestine a state and a state party to the Rome Statute. Behaviourally, it is widely recognised that General Assembly votes are political. Legally, under the UN Charter, the General Assembly may discuss issues, receive reports, and make certain recommendations,¹⁹⁸ but neither the General Assembly nor the Secretary General has legislative authority,¹⁹⁹ and the question before the Pre-Trial Chamber was a legal question: whether Palestine is a state. No matter. Other than international lawyers, few will see that the majority's legal analysis is wanting; the decision was politically acceptable to an overwhelming proportion of party states that diplomatically support the Palestinian cause; and it was convenient politics for an institution eager to find cases outside Africa.

3.3.2. Influence of non-party states

Powerful states that are not parties to the Rome Statute have also influenced ICC activities. Most importantly, the Security Council may refer (or not refer) situations of apparent criminality to the Prosecutor for investigation,²⁰⁰ establishing ICC jurisdiction that might not otherwise exist. Of course, the permanent members of the Security Council, four out of five of which are not state parties to the Rome Statute, have weighted power in the Security Council by means of their authority to exercise a veto. In practice, such referrals have been inconsistent across situations of grave criminality. The Security Council referred the situations in Libya and Sudan (Darfur) to the Court, but has not referred some other situations of mass atrocities, such as Syria and Ukraine on account of a veto or expected veto by Russia, or the situations in Myanmar, China (Xinjiang), and China (Tibet) on account of a veto or expected veto by China. These inconsistencies seem to violate the ‘like cases’ maxim – treating similar cases similarly – and has ignited complaints by human rights non-governmental organisations²⁰¹ and from the global south²⁰² that this pattern is unfair and illegitimate.

Powerful non-party states also have leverage by offering or refusing to take actions that can assist the Court. The United States, at times, has assisted the Court and, at other moments, has attacked it. In the Court's early years, the US government was wary of the ICC, concerned that US military or civilian leaders might be subject to criminal investigation or prosecution by the Court, and enacted the American Servicemembers Protection Act (ASPA),²⁰³ statutorily limiting ways in which the US government may assist or cooperate with the Court and authorising certain actions that could be taken against the Court.²⁰⁴ Understanding US concerns and preferring not to engage the United States as an active opponent of the Court, while cautiously attempting to establish the Office of the Prosecutor, the first Prosecutor, Ocampo, met with American diplomats, indicated that he trusted the US military justice system, and suggested that Americans would not be a focus of his investigations or prosecutions.²⁰⁵ Consistent with that stance, Ocampo did not open an investigation of US behaviour in Afghanistan or in Iraq; he rejected a declaration that was intended to open an investigation into the situation in Palestine following operation ‘Cast Lead’; and he focused on other cases where there were glaring atrocities in Africa, over which ICC jurisdiction was clear.

By 2008, the United States had shifted its policy towards the Court from firm opposition to a form of constructive engagement: where US interests dovetailed with those of the Court, the US government would cooperate with or support the Court's activities, to the extent consistent with the ASPA.²⁰⁶ At the request of the National Security Council, the Department of Justice prepared a memorandum in 2010, distilling ways in which the US government may assist the Court without running afoul of the law.²⁰⁷ These activities were determined to include furnishing certain informational assistance (including intelligence, law enforcement information, diplomatic reporting, investigative actions, and testimony) to the ICC for particular ICC cases involving foreign nationals; training ICC personnel; detailing US government employees to the ICC where assistance would be limited to particular cases involving foreign nationals; sponsoring or voting for resolutions in international fora that refer matters to the ICC or support the ICC approach to a particular matter; communicating US government views to the ICC; and encouraging foreign governments to materially assist the ICC.²⁰⁸

Hence, when US government policy has favoured ICC actions, the US government has assisted Court activities. For example, it provided satellite imagery to the Office of the Prosecutor, evidencing movements of armed groups in the Democratic Republic of the Congo and in Darfur; when animated, that satellite imagery shows, for example, that an armed group entered a series of villages, and shows how the villages looked before and after, demonstrating that the villages were set afire and burned to the ground.²⁰⁹ The United States has also provided signals intelligence, intercepted cell communications and radio communications, which may be used as evidence.²¹⁰ It also sent 100 special forces to the Central African Republic to help to look for ICC indictee Joseph Kony. While US law prohibits its government from extraditing persons to the Court,²¹¹ it nonetheless creatively assisted the Court by organising ICC indictee Bosco Ntaganda's safe passage from the US Embassy in Kigali, to which he had surrendered, to the Kigali airport to take a direct, one-way flight from Kigali to The Hague, where he was arrested on arrival.²¹²

Conversely, there have been times when the United States ceased cooperation and instead openly attacked the Court. For example, after the second Prosecutor opened an investigation into the situation in Afghanistan, which included clear evidence of torture by US forces, and sought Pre-Trial Chamber approval of an investigation into the situation in the State of Palestine, Trump administration National Security Advisor John Bolton openly criticised the ICC, deeming it a threat to US citizens and national security;²¹³ President Trump issued an Executive Order that imposed sanctions on the Prosecutor and others in her Office, prohibiting their travel to the United States, freezing their dollar-denominated assets, and prohibiting any US person from providing material assistance to the Court under threat of criminal liability.²¹⁴ President Trump's successor, Joe Biden, rescinded the Trump sanctions shortly after taking office in 2021²¹⁵ and withdrew US forces from Afghanistan in August of that year. Almost immediately thereafter, when the Taliban resumed its rule, the third Prosecutor announced his decision to focus his office's ‘investigations in Afghanistan on crimes allegedly committed by the Taliban and the Islamic State … and to deprioritize other aspects of this investigation’.²¹⁶

4. Political effects of the ICC in states

ICC actions and omissions have important political effects in situations under investigation. There is good evidence that, at least in its first decade of operation, ICC action advanced important objectives of those who have championed international justice. For example, statistical analyses of data across countries have shown that, during the ICC's first decade or so, ratification of the Rome Statute was significantly associated with a reduced rate of civilian killings by party state governments, compared with the pre-ratification rate, particularly if the Prosecutor had opened an investigation or where human rights groups were operating.²¹⁷ Another study found that opening an ICC investigation was significantly related to prosecution of those responsible for international crimes by domestic courts in situation countries,²¹⁸ presumably as a result of the Rome Statute's complementarity principle, which provides that the Court cannot proceed with a case if it is being prosecuted genuinely by a state.²¹⁹

At the individual country level, there is also evidence that some ICC actions have dampened violence. Joseph Kony fled from Uganda and went into hiding shortly after the ICC issued a warrant for his arrest. Ocampo's prosecutorial strategy in the DRC culminated in the arrest and removal of five powerful rebel militia and political leaders across northern and eastern theatres of conflict in Congo, and the arrest of one Congolese militia leader pursuant to an international warrant catalysed demobilisation of the militia under the arrestee's control.²²⁰

Within Israel, there is reason to believe that ICC law and activity might have influenced domestic law and practice. While the Israeli Supreme Court has never explicitly referred to the risk that Israeli nationals would be prosecuted before the ICC, decisions in recent years suggest that it is aware of that risk.²²¹ In 2006, for example, in the second targeted killings case, Chief Justice Barak's mention of international criminal tribunals’ jurisprudence led the Court to decide that certain matters were justiciable,²²² and in the Israeli Supreme Court's 2020 Settlement Regularization Law opinion on the appropriation of Palestinian property for settlements on the West Bank, international criminal law terminology was used to deem that law to be unconstitutional.²²³ In 2011, the Turkel Commission prepared significant military justice reforms²²⁴ at precisely the time that ICC Prosecutor Ocampo was considering whether to open an investigation of the situation in the region, pursuant to a declaration of jurisdiction by the Palestinian Authority, and how to proceed on the referral by Turkey concerning the maritime incident of 2010. As the Turkel reforms were being adopted, Ocampo formally rejected the Palestinian declaration;²²⁵ shortly thereafter his successor formally ended the investigation of the maritime incident of 2010.²²⁶

In contrast, some consequences of ICC action have been normatively problematic. For example, whenever a state party government has referred a case to the Prosecutor for investigation, it has served its own self-interest: targeting the behaviour of opposition militias or key rebel leaders, facilitating the Prosecutor's access to evidence against the leaders of opposing militias, and not cooperating with the Prosecutor's efforts to obtain evidence of crimes perpetrated by the referring government or allied militia leaders. In the situations in the DRC, for example, ICC arrest warrants were not sought for President Joseph Kabila or other Congolese leaders, who presided over an army that had engaged in war crimes; and ICC indictee Bosco Ntaganda, ‘The Terminator’, was absorbed into the Congolese army as a general, rather than being arrested by the government, as required by the Rome Statute. Similarly, arrest warrants were never sought for Ugandan President Yoweri Museveni and his cadre, despite their leadership of an army that had perpetrated atrocities. The horizontal inequity of pursuing leaders of opposition militias that perpetrated crimes, but not government leaders of armies that perpetrated crimes, might be seen as contradicting the ‘like cases’ maxim.

5. Conclusion: Squaring the circle of law and politics

The ICC can be understood only as a legal institution embedded in politics. Its agenda, codified as the Rome Statute, is a political outcome, strongly influenced by European ideas, and reflecting the normative and material interests of party states. The institution is responsive to international political pressure, particularly from state parties; and the ICC and its actions have political consequences in situation countries.

As a legal institution, however, the ICC risks losing its legitimacy if it is perceived as a political animal.²²⁷ Hence, the organs of the Court must always act under the colour of law. Statements of successive Prosecutors that they will not let politics influence their decisions, quoted at the beginning of this article, are consistent with maintaining the Court's legitimacy as a legal institution. In so far as outcomes are as expected, party states have an interest in letting the Court do its work, an agenda to which they signed, without interference. Accordingly, most state parties, particularly those in Europe, have shared a norm of not pressuring the Prosecutor or Chambers to handle any particular investigation or prosecution in any particular way, expressed in the Statute as respecting the independence of the Prosecutor and Chambers.

Metaphorically, the state parties are the principal and the ICC organs are its agents for ending impunity for atrocity crimes, but there has been considerable agent slack:²²⁸ politics leads the ICC on a loose leash – a leash that is sometimes yanked back to correct the Court's direction. This enables the ICC to act as a legal institution, investigating and prosecuting crimes, when raw politics might not otherwise catalyse or permit action. Moreover, as an international legal instrument, the Rome Statute may affect domestic law in party states that give direct effect to or transpose elements of it into domestic codes or court decisions. Even in some non-party states, like Israel, some international criminal law has been incorporated into domestic law via court decisions or domestic codes, perhaps because of the threat of ICC investigations or of cascading global norms.²²⁹ The resulting body of domestic law and norms, combined with the ever-present threat of ICC criminal prosecution, seems to have marginally dampened civilian killings by state party governments, at least during the Court's first decade of operation, especially where human rights groups could use that law to challenge the government in court or in the media.²³⁰ Being a legal institution matters.

Nonetheless, as a legal institution embedded in politics, the ICC recurrently faces legitimacy challenges – not only when it responds to politics, but also when it follows the law. It is squeezed at both ends. Legitimacy is fundamentally a subjective and normative concept, existing in the beliefs of individuals about the rightfulness of a ruler or an institution; those beliefs may be enhanced or diminished by either normative or material appeal.²³¹

Many of the African governments, leaders, and rebels who began to challenge the ICC's legitimacy in the 2010s had both material and principled reasons for doing so, even though the Prosecutors were following their legal mandates. The ICC indictments of African leaders such as Bashir, Kenyatta, and Ruto gave them and their followers obvious material reasons for declaring the Court's actions illegitimate. Similarly, ICC-accused rebels in places like the DRC and Uganda had material reasons for claiming ICC illegitimacy as they became prosecutorial targets. However, those rebels also had principled bases for questioning the ICC's legitimacy, arguing that government leaders that referred their countries’ situations to the ICC Prosecutor were not seriously investigated or charged with crimes, despite evidence that they knew their militaries were engaged in a pattern of committing crimes that they did not try to stop. However, the broadest legitimacy challenge from Africa has been the perception and argument by African state leaders of horizontal inequity, rooted in the application of the Rome Statute's jurisdictional triggers to party states with vastly different social conditions, skewing the geographic distribution of investigations among state parties. In parts of Africa, rule of law is weak, law of war norms are weak, and state failure abounds, leaving large swathes of territory virtually anarchic and subject to atrocities perpetrated by competing militias and armies. These conditions do not exist in most other state parties, certainly not in Europe.²³²

At the same time, it is difficult to establish jurisdiction over situations with crimes perpetrated by nationals of non-party states, unless committed on the territory of a state party. A non-party state may confer jurisdiction on the Court for crimes committed on its territory, triggering a preliminary examination, but that can move to the investigation phase only following a state party referral, or the Prosecutor's exercise of their proprio motu authority, with Pre-Trial Chamber approval – a cumbersome process that has been used only rarely. The UN Security Council may also refer a situation to the Prosecutor, conferring jurisdiction and automatically triggering an investigation, but Security Council referrals have been rare because of the veto power of the permanent members, which are engaged in deepening geopolitical competition with each other. Glaringly, the UN Security Council has failed to refer the China (Xinjing), Myanmar (Rohingya), Syria or Ukraine situations to the ICC.

The resulting geographic focus of early ICC cases in Africa set off a legitimation crisis,²³³ rooted partly in a material appeal to victimhood, and partly in principled appeals based on assertions of neocolonialism and arguments rooted in violation of the ‘like cases’ maxim.²³⁴ ICC investigations into situations in African states exclusively, all but one of which were mandated by the Statute, while lacking authority to launch investigations in places like Syria and Palestine, became seen by many Africans as horizontally unjust, triggering a loss of confidence in the institution across much of Africa.

Some politically driven, horizontally inequitable actions that have been taken less transparently bear less salience, or do not appear to challenge the interests of any state parties, have not caused much of a stir. In some cases, those political actions seem to have enhanced the Court's legitimacy, at least for particular audiences. For example, the election of a Prosecutor from Gambia, Fatou Bensouda, seems to have been influenced by the politics of appeasing certain African states that were complaining of bias, with some threatening to exit from the ICC; that political action is likely to have enhanced the institution's legitimacy in Africa, without damaging ICC legitimacy elsewhere, as a result of Bensouda's strong qualification for the job on the merits. Similarly, the Court's decision on ‘policy grounds’ to permit hearings concerning Kenyan indictees in absentia, despite unambiguous Rome Statute language requiring their presence, is likely to have marginally dampened populist Kenyan complaints about the ICC, without engendering complaints.

The African legitimation crisis has been dampened most effectively, but not fully resolved, by launching investigations of several non-African situations, including investigating nationals of non-party states – for example, Afghanistan (partly targeting U.S. nationals), Georgia and Ukraine (Russians), Bangladesh (Myanmar), and Palestine (Israel). These often have been supported by creative jurisdictional theories, almost always endorsed by Chambers.

That geographic enlargement of investigations is simultaneously politically expedient, in so far as it dampens African assertions of horizontal inequity, and is consistent with the ICC mission of ending impunity for crimes defined in the Rome Statute. However, to the extent that those investigations target nationals of states that are not parties to the ICC regime, they exacerbate tension with and further delegitimate the ICC in the eyes of powerful elements in those states. For material reasons, targeted leaders and their supporters are sure to be repelled by such actions. On principle, it should not be surprising that non-party states find it unfair and unjust to have their nationals subject to the jurisdiction of a legal regime their country opted not to join, in some cases because the Rome Statute criminalises behaviour that their country does not consider criminal, follows procedures they consider unfair, or rests on referrals from ‘states’ they do not recognise.²³⁵ These ICC investigations recapitulate ideational and interest-based disagreements in the Rome Statute negotiations, cementing the refusal of non-party states to join the Court, increasing diplomatic tension between those states and the ICC, and risking overt and covert actions against the Court. For those states, ICC investigations and arrest warrants lack legitimacy. We live in a world that is increasingly bifurcated by states that support and those that oppose the Court.

The Court's biggest contemporary political challenges are likely to continue to be from African states and powerful non-party states. Powerful elements in both categories of states – some of the world's most powerful states and some of the weakest – now question the legitimacy of the ICC. African party state governments continue to endure popular legitimacy concerns about Court bias and a perceived European neocolonial agenda – yet crimes continue to be perpetrated in Africa, promising further investigations there, as the ICC follows its legal mandates. New cases are far less likely to emerge in western Europe or developed countries, where rule of law, democracy, and observance of human rights are more likely to prevail. Non-party states – the United States, Russia, and China, for example – can be expected to challenge the Court, if they or their allies are subjected to investigation or prosecution, and to pressure their allies to reduce cooperation with the Court.²³⁶

More broadly, if world order continues to deteriorate – with revisionist powers or their proxies challenging territorial borders, perpetrating atrocities, or triggering existential wars – then the logic of the appropriateness of fighting within the confines of humanitarian law will come under increased pressure, the priorities of influential states may change, the ICC will face greater political challenges, and the antinomy between law and politics at the ICC may become perilously exposed.

ICC organs will no doubt continue trying to manage the contradictory logics of law and politics. Management of that tension might be enhanced by the Court building the capacity of the Office of the Prosecutor to analyse international and domestic politics so that it can operate more deftly in the international political environment. The Court could also help to shift the task of accountability towards national courts and regional organisations, which may enjoy more legitimacy than action in The Hague, by more aggressively pursuing a policy of positive complementarity, offering greater support to states with direct jurisdiction over a situation, states asserting universal jurisdiction, and regional organisations that are pursuing accountability.²³⁷ Or the Prosecutor could raise the bar for investigations and prosecutions, pursuing only the gravest crimes that are being perpetrated on the largest scale;²³⁸ these are the cases that would most glaringly burnish the ideals for which the Court stands, so they should enjoy the greatest legitimacy.²³⁹ Such policy changes may modestly reduce antagonism towards the Court, but nothing can square the circle of law and politics at the ICC.