Tilting at windmills: Reparations and the International Criminal Court

Abstract

More than 20 years on from the signing of the Rome Statute, delivering victim-centred justice through reparations has been fraught with legal and practical challenges. The Court’s jurisprudence on reparations only began to emerge from 2012 and struggles to find purchase on implementation on the ground. In its first few cases of Lubanga, Katanga, and Al Mahdi the eligibility and forms of reparations have been limited to certain victims, subject to years of litigation, and faced difficulties in delivery due to ongoing insecurity. This is perhaps felt most acutely in the Bemba case, where more than 5,000 victims of murder, rape and pillage were waiting for redress, and the defendant was not indigent, but where he was later acquitted on appeal, thereby extinguishing reparation proceedings. This article critically appraises the jurisprudence and practice of the International Criminal Court (ICC) on reparations. It looks at competing principles and rationales for reparations at the Court in light of comparative practice in international human rights law and transitional justice processes to consider what is needed to ensure that the ICC is able to deliver on its reparations mandate. An underpinning argument is that reparations at the ICC cannot be seen in isolation from other reparation practices in the states where the Court operates. Reparative complementarity for victims of international crimes is essential to maximize the positive impact that the fulfilment of this right can have on victims and not to sacrifice the legitimacy of the Court, nor quixotically strive for the impossible.

1. Introduction

Reparations for international crimes seem quixotic given the harm caused and the multitude of victims, making them difficult to place within the frameworks of international criminal tribunals. ¹ Judges at ad hoc tribunals for the former Yugoslavia and Rwanda were unable to award reparations, ² instead they recommended it should be delivered through an international claims commission that never materialized. ³ In the early drafts of the ICC Statute there were concerns that including reparations would be too complex due to the scale of victimization. ⁴ States were also reluctant to include state responsibility within the statute, as it could detract from individual criminal responsibility. ⁵ Instead delegates recognized that victims of international crimes have a right to reparation against the convicted person, but acknowledged that this cannot be ‘interpreted as prejudicing the rights of victims under national or international law’. ⁶ However, this has caused down-river consequences at the ICC, where reparations can only be ordered against convicted persons, most of whom are indigent.

This article contributes to the literature on how to improve the work of the ICC on reparations, ⁷ reflecting on the past, present, and future role of reparations within the Court bearing in mind the comparative practice, and adopting a more holistic approach to reparations that builds on the concept of reparative complementarity. To this end, the article begins by outlining the evolving jurisprudence of reparations at the ICC, paying particular attention to the Court’s determinations on three areas of contention: criminal responsibility; victimhood; and reparation measures. It then considers the role of the Trust Fund for Victims (TFV, Trust Fund) in operationalizing reparations. The last section of this article considers the future of the Court’s reparation process, examining the issues of security, funding and the work of the Trust Fund, a tripartite breakdown of concerns constraining the work of the Court today. It critically situates these issues in light of the broader theoretical concern of reparative justice at the ICC as a set of responses on reparation including domestic practice where the Court has jurisdiction, to enhance the fulfilment of this right for victims of international crimes. This is a natural consequence of reparative complementarity, but it is also the way forward in concretizing reparations for victims before the ICC.

2. Emerging jurisprudence on reparations at the ICC

More than 20 years after the adoption of the Rome Statute, the jurisprudence of the Court on reparation remains underdeveloped. At the time of writing this article, only four judgments on reparation have been handed down, with limited implementation. ⁸ Expectations about reparation for international crimes nevertheless remain high. This has not been abated by the information and at times rhetoric of Court officials about ensuring ‘justice for victims of international crimes’. ⁹ While the ICC has been slow to respond to its mandate on reparations, other courts, such as the Inter-American Court of Human Rights (IACtHR), have crafted ground-breaking procedural and substantive principles of reparation that have been translated into comprehensive orders on reparations. ¹⁰ Even other tribunals, like the European Court of Human Rights, certainly more restrained on adjudication on reparations than the Inter-American counterpart, have also refined their approach to reparations by ordering, in some cases, ‘general measures’ and forms of restitution beyond compensation. ¹¹

Reparations, alongside provisions for victim participation, are two of the landmark provisions of the Rome Statute, indicating victim-orientated justice, with some going as far as to say that it constitutes a restorative turn in international criminal justice adjudication. ¹² Despite drawing from international human rights law, ¹³ the exclusion of state responsibility and the focus on compensation, restitution and rehabilitation within the language of the Rome Statute, indicates, as Sperfeldt points out, a selective borrowing of the right to reparation developed in human rights law. ¹⁴ The Court’s reparation jurisprudence has been slow to develop, with some of its core principles and essential requirements in awarding reparations only clarified through the 2015 Appeals Chamber decision ¹⁵ and order on minimum elements for reparations in the Lubanga case. ¹⁶ While these principles were not intended to be binding on other trial chambers adjudicating on reparations, but instead they were meant to be adapted or expanded upon, ¹⁷ they have become entrenched as they were followed in the Katanga and Al Mahdi cases. ¹⁸ Only in the Ntaganda case were they expanded. ¹⁹ While there have been some innovations in these subsequent cases, the Lubanga reparation principles have very much framed the development of reparations at the Court.

The Lubanga principles draw from the UN Basic Principles and Guidelines on the Right to a Remedy and Reparation ²⁰ as well as the reparation jurisprudence of the Inter-American and European Courts of Human Rights. The UN Basic Principles provide ‘clarity about the right to a remedy and reparation’, ²¹ but little content on how to successfully achieve ‘adequate, prompt and effective’ reparation in the face of scarce resources. The Lubanga principles set out a more implementation-focused basis for reparations at the Court, including the scope of beneficiaries, dignity and non-discrimination, consultation, modalities of reparations, the rights of the defence, and obligations of other actors. Some of these principles, such as non-stigmatization, are not usually found in other reparation guidelines or judgments. Instead, they reflect, for example, the 2007 Paris Principles on Child Soldiers, ²² exhibiting the particular facts of the Lubanga case. This may bring out interesting jurisprudence in the future regarding the treatment of victims of conflict-related sexual violence in trying to minimize stigma. The ICC has so far established the general foundation to reparations, but much more can be built upon to make it effective.

Importantly, despite the fact that reparation orders at the ICC have tried to borrow from human rights law, they are framed in narrower terms. At the ICC, reparations can be ordered against the convicted person, for the extent of his/her criminal liability, and to those that are recognized as victims by the Court. ²³ This is in contrast to human rights courts where reparation is awarded against the state and is focused on remedying proven victims’ harm, ²⁴ even in the absence of an investigation/prosecution of the individual perpetrator(s) responsible for the violations. Something similar happens in the context of transitional justice practice when domestic reparation programmes are set up by states to deal, in a flexible evidentiary manner and in a prompt way, with the various harms suffered by thousands to even millions of victims that qualify as beneficiaries. ²⁵ The narrowness of the criminal proceedings when it comes to reparations has a direct impact on the consideration of harm and forms of reparation that the ICC could order. Here we would like to focus on three key issues arising in the Court’s emerging reparation jurisprudence: criminal responsibility; victimhood; and reparation measures.

2.1 Criminal responsibility

Under Article 75(2) the Court can make a reparations order ‘directly against a convicted person’. This, according to the Appeals Chamber in the case of Lubanga, means that ‘the reparations order must not go beyond the crimes for which [the perpetrator] was convicted’. ²⁶ The difficulty with this approach is the asymmetry between the individual convicted person and the hundreds and thousands of victims with potential claims for reparation. This asymmetry is augmented when the Court goes after one individual or a few of them, who have allegedly committed the most serious crimes that denote the existence of an armed group or a criminal organization. ²⁷ However, as these groups are not themselves subject to trial, the scope of liability and therefore of reparation is severely limited. This is apparent in the Lubanga and Ntaganda cases, where two members of the Ituri UPC/FPLC militia were before the Court, but a number of charges were excluded during the trial against Lubanga, and Ntaganda was not prosecuted for alleged crimes his organization committed in North Kivu. ²⁸ That said, the Court did hold both of them jointly liable for recruiting and using child soldiers. ²⁹

Reparations are unlikely to contribute to broader goals of reconciliation or transformation, where only one or a handful of perpetrator(s), from one-side or at a particular point in time of a conflict, are found responsible. The ICC is necessarily selective and exemplary, in that it cannot prosecute all perpetrators, but concentrates its efforts on those most responsible, providing a myopic narrative of violence. This leaves little comfort for victims who expect international justice to work on their behalf, given the inability or unwillingness of domestic systems to ensure justice, including reparations for the harm they have suffered. The focus on individual perpetrators also limits the extent of their financial liability to the extent of their criminal responsibility for the crimes of which they are found to be guilty. So far, all convicted persons before the ICC have been indigent. Initially the Trial Chamber in Lubanga believed that such indigence meant that reparations could not be awarded against Mr Lubanga. This has since been changed by the Appeals Chamber in the Lubanga case, and followed in subsequent cases, where the Court held that indigence does not affect a convicted person’s liability. ³⁰

The 2015 Lubanga appeals decision also set down the principle that liability for reparations should be proportionate to harm caused and, inter alia, to the person’s participation in the commission of crimes. ³¹ Other chambers have nevertheless not always followed this position, which upholds joint liability. In Lubanga the judges found that, regardless of the degree of a perpetrator’s participation in a crime, ‘no single co-perpetrator bears all of the liability for the crimes committed’. ³² Whereas the Appeals Chamber in the Katanga case deemed that it is ‘not, per se, inappropriate to hold the person liable for the full amount necessary to repair the harm’. ³³ In a more recent Appeals Chamber decision on Mr Lubanga’s liability, the Chamber considered that liability could be mitigated by the convicted person where they can show their contribution to reduce the harm caused to victims, such as having helped to demobilize child soldiers, ‘especially those assessed as samples by the Trial Chamber’. ³⁴

The liability cost for reparations is rising as the Court is concluding more complex cases with thousands of victims. In December 2017 the Trial Chamber set Mr Lubanga’s liability to $10,000,000 and included not only the 425 identified victims before the Court, but also around 3,000 non-identified victims. ³⁵ In 2021 Mr Ntaganda, Mr Lubanga’s subordinate, who was convicted of 18 crimes, had his financial liability set at $30,000,000 reflecting the cost of repair for the harm caused to at least 3,500 direct victims. ³⁶ In Katanga the defence tried to argue that given his lower liability of aiding and abetting he should have a more reasonable, nominal share of financial liability (set at $1,000,000) with the Trust Fund having a greater share of the $3,752,620 award. ³⁷ This was rejected by the Appeals Chamber. ³⁸

The financial liability of the perpetrator needs to be critically interrogated for the finality of reparation claims, seizing and freezing assets and securing implementation of what is ordered. In the Al Mahdi case the Chamber recognized that the financial circumstances of the perpetrator affect implementation, such as paying compensation in instalments, but it did not want to ‘impose hardships’ on Mr Al Mahdi that would ‘make it impossible for him to reintegrate into society upon his release’. ³⁹ This was in response to the defence’s submission that his liability should end once he had served his nine-year sentence. Yet this heightens tensions around the balance between the convicted person’s rights and his or her obligations to victims, where the reintegration of the defendant is a consideration for sentencing and early release. The issue remains within the competence of the Presidency to monitor once a defendant is released, but it would require state co-operation and some effort to oversee the financial situation of such individuals. Furthermore, there has been little thought on how the Court and states will oversee and enforce reparation orders against convicted persons once they are released. Do they confiscate any earnings they have, disincentivizing them from officially declaring any job they have? Do they take a percentage contribution from their income to be used towards the reparation debt they have at the ICC, allowing them to provide for themselves and their family, or for the matter to be dropped entirely, given the likely costs in monitoring and enforcement? Would remunerated work in detention be able to be transferred to victims? The Court has yet to address the long-term implementation and implications of reparations.

The issue of identifying and seizing assets of the perpetrator is central in ensuring that there are sufficient financial resources available to provide meaningful and adequate redress to victims. ⁴⁰ Unfortunately, the work done by the Court, and the international co-operation provided by states parties, has been grossly inadequate. This was evident in the Bemba case where his shares in a Congolese beer company were not frozen, ⁴¹ and one of his jets was left to rot on a runway in Faro to the extent that the parking ticket on it was over €980,000, the keys were lost and the plane was only good for scrap. ⁴² Following his acquittal, Mr Bemba has faced ongoing difficulties for such assets to be unfrozen. ⁴³

More reflection is needed at the Court in striking a balance between the scope of liability of the perpetrator and victims’ right to reparation. Victims might have received some reparative benefits from other sources since they were harmed including any help or support provided to them by other actors aimed at wiping out damage. In the case of Al Mahdi, where UNESCO repaired the damage caused to the mausoleums in Timbuktu, the Court held that such efforts do not alter the harm caused or the liability of the convicted person. ⁴⁴ In Bemba, the defence asked that ‘victims disclose all cumulative benefits they received from third parties, be it the State or international organizations, since the moment they were harmed’, ⁴⁵ with the aim to reduce such benefits from any potential order against Mr. Bemba.

We are not arguing that the Court has aimed to limit reparation as much as possible, but rather to note the tension that exists in its existential crisis in ensuring fair trials for suspected perpetrators and offer vindication and remedy for victims. ⁴⁶ While some chambers would want to acknowledge the centrality of the right to reparation that victims have, the issue is not settled in the jurisprudence of the Court. Part of the problem is the criminal law mentality that continues to apply in the consideration of reparations at the ICC and the lack of a standard of effectiveness as used in human rights law. ⁴⁷ The practice of the ICC on the scope of liability of the convicted person is narrowly conceived compared to states in international human rights adjudication or transitional justice programmes. Indeed, under international human rights law, a state has an obligation to ensure reparations for all harm caused as a result of human rights violations, whether by private or public actors. ⁴⁸ In transitional justice practice individuals have been held liable along with the state, obliging both to fulfil reparation, given the scale and seriousness of the violations. ⁴⁹ The ability to pay is often not a factor in such decisions.

Human rights law and transitional justice also struggle with reparation implementation, partly due to the scarcity of resources and/or the lack of will to comply with what has been ordered. ⁵⁰ Furthermore, as international practice on reparations continue to develop, a lack of resources and the prospects of implementation, are having an impact on the orders given by regional human rights courts which, for example, are awarding financial compensation against states in light of its financial situation. ⁵¹ This financial feasibility can also have an impact on the scope of victimhood.

2.2 Victimhood

While the ICC Rules elaborate a wide concept of victimhood, ⁵² who is considered to be an eligible victim for reparations and for what harms remains a key source of contention at the ICC. Eligibility for reparations requires that victims must have suffered personal harm as a result of the crime that the defendant has been convicted of and which fall within the defined locations and dates specified by the Court. ⁵³ Generally, reparations have been focused on direct victims as those who suffer harm themselves, and sufficiently close family members, often referred to as indirect victims, taking a contextual approach to understanding ‘family’. ⁵⁴ This distinction reflects that direct victims need a ‘causal link must exist between the harm suffered and the crimes of which an accused is found guilty’, whereas indirect victims suffer loss, injury or damage due to their ‘relationship’ to the direct victim. ⁵⁵ Legal entities can also be recognized as victims, including schools (Lubanga), hospitals (Ntaganda), and companies or international organizations (Al Mahdi). ⁵⁶ The scope of eligible victims can vary between cases due to their circumstances. ⁵⁷

In determining whether someone is a victim in relation to a crime, causation must be shown as a ‘but-for relationship’ between the harm and the crime, which must be a proximate cause of the harm. ⁵⁸ A proximate cause is limited to those acts that ‘are closely connected to the result of that act and that are significant enough to justify a finding of liability’. ⁵⁹ In the Katanga case, the victims’ legal representatives tried to argue that children born after the massacre should have access to collective rehabilitative psychological services, given the transmission of transgenerational trauma of parents to children. ⁶⁰ However, the Court rejected this on the ground that it lacked a proximate connection to the massacre on 24 February 2003. ⁶¹ The consequence is that the universe of victims is smaller for the ICC. This contrasts with the jurisprudence of other tribunals, such as the IACtHR, which has recognized that even children who are posthumously born have a right to reparation, they can suffer material and psychological harm, and are eligible for different forms of reparation including compensation and rehabilitation. ⁶² The Chamber in the Ntaganda case nuanced this approach by recognizing that children born as a result of rape and sexual violence should be considered as direct, rather than indirect, victims given the ‘particular harm’ they suffer. ⁶³ However, other children of victims of rape or sexual slavery not born as a result of the crime would be considered indirect victims. This is a reasonable distinction, given the additional stigma children born as a result of rape face.

Victim eligibility at the ICC has not only been based on the idea that individuals have rights, but also that there might be a collective entitlement to reparation. In identifying what constitutes a group of victims for collective reparations, Trial Chamber II in the Katanga case set down that:

a group or category of persons may be bound by a shared identity or experience, but also by victimization by dint of the same violation or the same crime within the jurisdiction of the Court. Collective reparations may, therefore, benefit a group, including an ethnic, racial, social, political or religious group which predated the crime, but also any other group bound by collective harm and suffering as a consequence of the crimes of the convicted person. ⁶⁴

This is in line with international human rights law and transitional justice practice, which not only recognize the cultural community link that certain victims have, but also that there might be other grounds to claim collective reparation such as the particular harm suffered. Indeed, the Inter-American Court has awarded collective reparation not only in cases involving indigenous groups, ⁶⁵ but also in other cases where such a cultural link did not exist, but experienced similar mass atrocities. ⁶⁶ This has been followed in other transitional justice contexts. In Morocco, for example, the idea of collective reparation intended to remedy collective punishment of communities in terms of economic deprivation and a lack of development. ⁶⁷

It is not surprising that different bodies are engaging with collective reparations and trying to frame them widely, given the lack of resources and the complex challenges in identifying and delivering reparations to victims. Indeed, a community-based approach was proposed by the TFV and other participants in the Lubanga case and accepted by the Trial Chamber in the first instance, as a way to broaden the benefits and utility of reparations beyond individual awards, as well as to minimize costly verification procedures. ⁶⁸ However, community-based reparations would allow those who perpetrated and supported the recruitment and conscription of child soldiers to benefit. ⁶⁹ The Appeals Chamber rejected this broad interpretation of community-based reparations, re-affirming that reparation can be individual and collective, and requiring that reparations be awarded against a person for the crimes of which he/she was convicted. ⁷⁰ In the Al Mahdi case the community of Timbuktu was recognized as an eligible group of victims for the purposes of collective reparations. The Chamber also acknowledged that Mali and the international community suffered harm as a result of the destruction of the UNESCO World Heritage site in Timbuktu, awarding each a symbolic €1 each. ⁷¹ That said, the community in Timbuktu has a cultural and spiritual connection to the site, as recognized as the primary victim and prioritized in reparations to maximize its effects. ⁷²

Being identified as a victim eligible for reparations depends on various issues beyond charges and the conviction of the perpetrator even if these are essential. In most cases victims have to apply for reparation at the Court to be eligible. ⁷³ This is not an easy process particularly in the types of cases of which the Court is exercising jurisdiction on, where there is insecurity or the presence of the Court or of intermediaries are not permanent. In the Lubanga case, while the Appeals Chamber has held that ‘reparations need to be responsive to the operational reality in which they take place’, ⁷⁴ victims have strongly contested the need for the defence to screen their applications, given that the rules suggest that this only has to be done in individual claims for reparations. ⁷⁵ This responsibility of the Court to protect the rights of the convicted person against false claims has discouraged victims from coming forward to engage with collective reparation programmes out of fear for their own safety. ⁷⁶ Ullrich has termed this the ‘blame cascade’, where victims continue to be marginalized even in reparation processes. ⁷⁷ Thus, victims are treated with mistrust, but unlike other reparation processes the Court has to balance the rights of the convicted person against victims who can benefit from a reparation order.

The Bemba case illustrates some of these challenges. The ICC received 5,760 applications for participation and/or reparation, 5,229 of these individuals were eligible to participate at trial. ⁷⁸ Approximately 3,700 victims also requested reparation. ⁷⁹ Up to 5,000 did not have sufficient documentation and many did not fill in an application due to misunderstanding or being displaced. ⁸⁰ Moreover, it is difficult for some victims to understand why certain people were included and others excluded from reparations. This reflects the complexity of trying to fulfil the right to reparation through a criminal trial, which is necessarily limited at the ICC with its strict temporal, subject-matter, personal, and territorial jurisdiction. Victims and civil society have, on a number of occasions, challenged the confines of the charges against accused persons, given their implications for reparations, from broadening the charges against Mr Lubanga beyond child soldiers, ⁸¹ crimes committed by Bemba’s troops in the DRC, ⁸² and gender based violence in the Al Mahdi case, ⁸³ with little success. That said, the Court in the Lubanga case at least encouraged the Trust Fund to consider using its assistance mandate to offer services to victims of sexual and gender-based violence, given that Mr Lubanga was not convicted of this crime. ⁸⁴ Who is considered an eligible victim has also shaped the scope of reparation measures that the ICC can deliver.

2.3 Reparation measures

The ICC has categorized reparations into individual and collective measures. ⁸⁵ The Rome Statute refers to three modalities of reparation – restitution, compensation, and rehabilitation. ⁸⁶ The Court has also recognized that ‘other types of reparations, for instance those with a symbolic, preventative or transformative value, may also be appropriate’. ⁸⁷ The Court has discretion on whether or not to order reparations, on an individual or collective basis or both. ⁸⁸ This approach to forms of reparation creates an opportunity for the Court to respond to the diverse harms suffered by individual victims and communities. This is in line with the practice of the IACtHR. By contrast the Extraordinary Chambers in the Courts of Cambodia explicitly limit reparations to collective and symbolic measures, ⁸⁹ given the millions of victims and reluctance of the Cambodian government to set up a complementary reparations body. ⁹⁰ The rest of this section discusses compensation and collective measures as the Court tries to navigate operational realities of reparations, where it is dealing with indigent perpetrators, the limited voluntary funds of the TFV and a multitude of victims.

2.3.1 Compensation

In terms of individual reparations, in particular in the form of compensation, there were conflicting views in the initial cases of the Court. In the Lubanga case the working presumption was that with the convicted person being indigent, compensation would be inappropriate, ⁹¹ despite victims’ wishes for such a measure. ⁹² The Trial Chamber erroneously did not set Mr Lubanga’s liability and was, instead, relying on the TFV to deliver some form of redress to victims at the discretion of its Board of Directors, using its own resources. ⁹³ The Appeals Chamber held that reparations must be ordered against a convicted person, but while compensation was acknowledged as a form of reparation, the Chamber considered that ‘determining the nature and/or size of the reparation award is an appropriate task of the Trust Fund …’. ⁹⁴ The driving concern here seems to be the Court wanting to minimize community tension by awarding reparations to child soldiers from the Hema community, and not those they victimized, creating a hierarchy of victimhood. ⁹⁵ However, such a paternalistic approach undermines victims’ right to reparation before the ICC, by assuming the Chamber or the Trust Fund knows better what victims need. ⁹⁶ This was further compounded in the Ntaganda case, which adopted a ‘do no harm’ principle to avoid community tension and secondary victimization, but this is too strongly influenced by humanitarian assistance where victims have needs, not rights. ⁹⁷ The Court also set out the ‘no over-compensation’ principle to avoid double redress for Lubanga victims who were also in the Ntaganda case, but it strikes of a certain detachment from victims’ lived experiences that any reparations could ‘over’ compensate them. ⁹⁸ Yet, this also assumes that collective measures can be neutral or apolitical in societies where resources are scarce, communities continue to be vulnerable, face insecurity, and experience everyday social violence with other communities years after the end of hostilities. ⁹⁹ In practice reparations may need to be individual anyway given ongoing insecurity and social distancing required by the pandemic.

In the Katanga case the Trial Chamber was acutely aware that the majority of victims wanted compensation rather than collective measures. ¹⁰⁰ The victims’ legal representatives and defence had agreed to a €1 compensation amount as symbolic recognition of their suffering before the decision, as a way to focus more on psychological support for survivors. ¹⁰¹ The Chamber found that the harm caused to the community amounted to $3.7 million, but given that Katanga was convicted for aiding and abetting, he was only liable to $1 million. For victims this meant that despite the Court recognizing they had suffered what would be valued between $2,000–$14,000 per individual victim, they would only be awarded $250. The Chamber justified the $250 amount as a way to provide a ‘personal and symbolic acknowledgement’ to victims and to regain their ‘self-sufficiency and to make decisions for themselves on the basis of their needs’. ¹⁰²

Given how long these proceedings had taken and their inevitable cost, as well as victims’ loss, such a token amount may feel like an insult or cheapening of their suffering. Moreover, all victims were eligible for the symbolic award of $250 for moral harm, making no distinction between those who lost an immediate family member and those who lost minor material items. The Chamber held that this amount was ‘not intended as compensation for the harm in its entirety …[but] may provide some measure of relief for the harm suffered by the victims’. ¹⁰³ However, this suggests that support would enable victims to become self-sufficient. Yet, as most of the victims had lost a family member, witnessed horrendous atrocities, and lost their homes and goods, such an amount was seen as just virtue-signalling with some of the victims dying while waiting for more substantive redress including rehabilitation and educational support for their children. ¹⁰⁴ Nonetheless, even though the amount was small, victims in the Katanga case were able to invest the money, use it for their children’s education or pay off debts. ¹⁰⁵

In Al Mahdi the Chamber limited compensation for moral harm to those individuals who exclusively relied on income from businesses they ran beside the destroyed mausoleums and those who were descendants of the saints. However, it was difficult to establish who was a descendant as there was no clear kinship or documentation to this effect. Where there were documents, they related to the male descendants and not the female ones, which, if used, would produce a gendered outcome. ¹⁰⁶ In its draft implementation plan the TFV proposed that including female-based lines would increase the number of victims, but where these could not be identified, any remaining money would be placed into the fund for collective moral reparations. ¹⁰⁷ The Chamber approved the TFV approach to ensuring compensation was ‘real’ rather than symbolic, in providing comparable values based on similar awards in Malian law for cultural destruction and economic loss. ¹⁰⁸ Nevertheless, compensation has been an easier form of reparation to deliver to victims, ¹⁰⁹ compared to the collective measures in the first three cases of the ICC, where insecurity in Ituri and Mali inhibited more public-facing implementation. In a separate opinion in the Lubanga case, Judge Eboe-Osuji argued that individual reparations approach should be ‘exhausted … before restoring’ to collective reparations, suggesting that these issues are not settled. ¹¹⁰

2.3.2 Collective reparation

In terms of collective measures, the Court has ordered a variety of measures influenced by context, the particular harm caused in the case and the views of the parties, and the TFV. In Al Mahdi, the Court recognized that ‘the number of victims and the scope of the consequential economic loss make a collective award more appropriate’, ¹¹¹ but it also recognized that certain individual victims were entitled to reparation. It indicated that collective reparation ‘should be aimed at rehabilitating the community of Timbuktu in order to address the economic harm caused’. ¹¹² Moreover, it suggested that the measures could include ‘raising programmes to promote Timbuktu’s important and unique cultural heritage, return/resettlement programmes, a “microcredit system” that would assist the population to generate income, or other cash assistance programmes to restore some of Timbuktu’s lost economic activity’. ¹¹³ The TFV draft implementation plan also proposed an Economic Resilience Facility (ERF) to support economic initiatives proposed by members of the Timbuktu community. The ERF would be a microcredit centre to support cultural activities and the affected tourism industry, ¹¹⁴ as well as creating a safe space for women and girls. ¹¹⁵

In Katanga the Trial Chamber ordered, for victims of the Bogoro massacre, collective measures that included housing assistance, education assistance, income-generating activities, and psychological rehabilitation. ¹¹⁶ Notably, victims in Katanga rejected collective symbolic measures as unsuitable, pointless or with the potential to cause unrest. ¹¹⁷ In the Al Mahdi case the victims rejected the TFV-proposed symbolic measures of memorialization, use of Mr Al Mahdi’s apology, and a re-sanctification of the buildings, as inappropriate. ¹¹⁸ Due to the ICC’s distance from the affected community, such measures may feel too top-down on an affected community and detract from the local meaning of redress. This reflects a fundamental contrast to transitional contexts where reparation claims are driven from below by victims, exhibiting their agency to articulate and shape measures to their needs. ¹¹⁹ Moreover, it reflects the frustration of victims with short-term, development or symbolic measures, ¹²⁰ when they expect substantive redress. In contrast in the Lubanga case collective reparations took the form of service-based and symbolic measures – including construction of community centres and a mobile programme to reduce stigma and discrimination against former child soldiers – along with service-based collective measures such as physical and psychological rehabilitation, vocational training, and income generating activities.

Nearly nine years after the first Lubanga reparation order, the ICC approved the implementation plan of the TFV for collective reparations for services, with symbolic measures unable to be complemented by partners on the ground due to insecurity. ¹²¹ One important development was that, although individual compensation was rejected for child soldiers due to risk of community tension, the reality that many of them are left physically or psychologically disabled (or elderly indirect victims) means that they are unable to engage in collective measures of income generating. They have, instead, been given a ‘periodic subsistence payment’ or pension, which is effectively compensation in all but name. ¹²² In the Ntaganda case the Court ordered ‘collective reparations with individualized components’, which included compensation, restitution, rehabilitation, and satisfaction. ¹²³ It seems the Court in perpetuating the individual/collective division of reparations is tying itself in semantic knots to meet victims’ expectations against the limitations of the Court’s capacity and liability of the convicted person. Really the Court is slowly recognizing that victims need to be dealt with through a collective administrative process rather than as individual judicial claims in the face of contexts that have a background of ongoing violence. The Court should recognize that collective/individual measures are not a dichotomy and focus on appropriate measures to remedy victims’ harm so that the process is more transparent and clearer for victims to understand.

Collective reparations present further challenges in their operationalization. For example, in the Katanga case the Chamber and the TFV have started to develop a flexible approach to maximizing the benefits for victims. For instance, measures such as educational support may not be needed by some victims who do not have children, and so the Trust Fund allows concentrating their benefits on housing and/or income generating activities and vice-versa where there are a number of children in a household. ¹²⁴ This method enables victims to apply their symbolic compensation awards to bolster collective measures, such as obtaining further housing benefits or a more valuable cow. ¹²⁵ This flexible approach is to be welcomed, as it recognizes victims’ agency to shape reparations to their needs, rather than representing the one-size-fits-all approach. Nevertheless, victims who are refugees in European countries or the US will not be beneficiaries of collective reparations due to the increased cost involved for the TFV, which undermines victims’ rights for the sake of ease of service delivery. ¹²⁶ Moreover, in the face of ongoing violence victims may decide not to engage in housing benefits or spend their compensation for fear it will be wasted when they are displaced again. This speaks to the limits of the Court’s reparation mandate when it comes to guarantees of non-repetition.

In human rights law combining individual and collective measures has been considered a key means to ensure an effective remedy of victims’ harm. ¹²⁷ Similarly domestic reparation programmes, such as in Peru and Colombia, individual measures, and collective reparations are awarded. ¹²⁸ In these contexts the major challenge is effective implementation. In relation to compensation, in contrast with the restrictive approach applied so far by the ICC towards this form of reparation, regional human rights courts have most often awarded compensation for moral and pecuniary damages for victims of gross human rights violations and not in small numbers. There has been a similar experience in transitional justice practice where, although the amount awarded might not be as high as regional human rights courts, it remains higher than compensation paid so far by the ICC and is complemented with other measures. ¹²⁹ The tension between different forms of reparations is reflected in the operationalization of such measures by the Trust Fund.

3. Operationalizing reparations and the Trust Fund

Delegates at the Rome Conference did not make a serious effort to work out how reparations would be implemented in practice, given that the Trust Fund was only supposed to operate as a repository of funds, rather than as an implementation body on the ground. As such, the TFV was envisaged with a dual mandate: its reparation mandate of using collected resources through fines or forfeitures of the convicted person as assets for reparations and supplementing them with any other funds; ¹³⁰ and its assistance mandate of providing ‘physical or psychological rehabilitation or material support for the benefit of victims and their families’. ¹³¹ Despite this, the Trust Fund has become a key actor of the reparations process, exceeding any expectations created during the drafting process of the Rome Statute. Its role is necessary for the implementation and delivery of reparations to victims, so that it is separate from the Court, despite it often having an assistance mind-set and the legitimacy issues that this raises. ¹³² Moreover the Independent Experts Review (IER) found that the TFV has ‘ineffective oversight … absence of a fundraising strategy … significant budgetary underperformance, which combined with delays in the judicial process, eroded (potential) donors’ confidence in the TFV’. ¹³³ These problems point to long-term problems in implementing reparations through the TFV. It is also grossly inefficient compared to other reparation implementation bodies, which normally have an administration running cost of 1–10 per cent of total budget to maximize resources to victims, ¹³⁴ yet in 2019 the TFV operational costs were €2.8 million and it only received €2.67 million in voluntary donations, reflecting 51 per cent. ¹³⁵

In the first decision on reparations in Lubanga the Trial Chamber held that the TFV was well placed, given its assistance programmes in the DRC, to determine appropriate forms of reparations and implement them. ¹³⁶ The Appeals Chamber turned this delegation of reparation procedure over to the TFV, requiring the Trial Chamber to monitor implementation, determine the harm victims suffered, and the corresponding appropriate modalities of reparations. ¹³⁷ There have been tensions between the Chambers and the Trust Fund in working out who is responsible for identifying victims in implementing awards. ¹³⁸ In some part this tension is the result of the practical needs of the TFV to enable its work on the ground, which requires independence so as to not compromise its assistance mandate. Yet it has resulted in reparations at the ICC taking an overtly humanitarian composition.

The TFV has no experience in operating a reparation programme and it has only recently begun to implement reparations in the case of Lubanga. ¹³⁹ This means that most of its work of the Fund has been the implementation of assistance. The TFV believes that its assistance mandate does have a reparative dimension that distinguishes it from development and other humanitarian work. ¹⁴⁰ However, the reparatory effect or impact of its policy programmes or other measures cannot be conflated with reparations. It is at best unclear, and at worst misrepresentation, how funding a radio programme on peace-building or community dialogue or a mobile book library in DRC or Uganda amounts to anything close to reparation for international crimes. This piecemeal approach might offer some basic support, but it is too ethereal to provide a meaningful remedy to victims’ suffering.

This is apparent in the Lubanga case, where services, as forms of collective reparations, are directed at socio-economic programmes for child soldiers, such as vocational training, improving agricultural techniques, and micro-finance associations. ¹⁴¹ The TFV proposed conducting interviews and evaluations of victims to determine their interests in programmes, assess their literacy and numeracy skills, as well as providing them with relevant training to their chosen area, but it is questionable how transferrable these skills will be if their small business goes bankrupt. This speaks to the appropriateness of reparations addressing their harm, whether they missed years of school education and family support, in which a small business investment is not going to be a long-term remedy. ¹⁴² This goes to the heart of what the ICC can modestly do and should feasibly achieve as ‘reparation’ for victims before it.

The focus on income generating activities is a reflection that victims have to be economically productive to get over their suffering. This might reduce the impetus on the state to make an effort to redress victims’ suffering, as they are left ‘self-responsible’ in mitigating their harm, and reduces such ‘reparations’ to more development. ¹⁴³ This language of ‘self-sustaining’ ¹⁴⁴ has been used by the Court to ensure that victims and their families can benefit in the long-term, but there is little evidence to suggest that such measures will be more than a short-term solution. ¹⁴⁵ Our own research in Nepal with child soldiers found that many such businesses that were set-up with training and investment became unprofitable after a couple of years, forcing them to move to the capital or Gulf states for a more sustained income. ¹⁴⁶ While this may be down to poor planning and lack of long-term monitoring and follow-up, the TFV does not have the capacity to provide sustained support for victims, a problem that is exacerbated by the difficult conflict conditions in which the TFV is working. For example, consultations in Northern Mali as part of the reparations process in the Al Mahdi case, given ongoing conflict, were ‘a nearly impossible task to accomplish’. ¹⁴⁷

Moreover, the Trust Fund relies more on rehabilitation and reconciliation activities, based on its ability to fund organizations that can deliver large-scale benefits to a range of victims under its assistance mandate. ¹⁴⁸ This, perhaps, overstates the difference the TFV has made in individual lives. According to its 2018 report, 104,000 victims had directly benefited from its assistance mandate in the DRC and Uganda, and 350,000 family and community members indirectly benefited over a ten-year period. ¹⁴⁹ Yet on average it only receives €3–4 million annually, and between 2008 and 2016 it spent €12 million on assistance to victims. ¹⁵⁰ It is questionable what the tangible benefit is for all victims, whether it is long-term, and associated with the Court, or providing mixed messages. ¹⁵¹

Given the role of the TFV in implementing reparations, victims appear to move between having an entitlement enforceable by a Court, to being more of a victim-consultee. ¹⁵² Dixon suggests that without the Chambers being involved in monitoring implementation of any reparation plan, and providing a space for victims to challenge the measures adopted by the TFV, such measures may amount to ‘long-delayed assistance, stripped of the meaning that makes reparations powerful beyond their material value’. ¹⁵³ As Fisher points out, reparations have a two-fold function in providing practical means to alleviate victims’ harm, as well as an expressive or symbolic dimension that recognizes their suffering and affirms their dignity, which is often absent in assistance. ¹⁵⁴ Nevertheless, the Court has tried to address this problem. For example, in the Al Mahdi case, the Court ordered that ‘victims know that it is aimed at repairing the harm suffered’ and that victims have to be involved in the design of collective reparations with the TFV. ¹⁵⁵ The Trust Fund in their subsequent implementation plan listened to victims in providing the most appropriate forms of reparations within its limited budget and capacity. Structural issues remain (i) the sustainability of the current jurisprudence of the Court on reparations in the face of limited resources, (ii) ongoing violence, and (iii) growing caseload. We may be expecting too much of the TFV which is allocating limited money from a bank account to organizations on the ground in situations before the Court to redress victims’ harm. Ultimately there is a growing frustration amongst victims of the length reparations are taking at the Court, the convoluted disagreements and the ‘divergence’ of the TFV from victims’ preferences in operationalizing such measures, which although separate is impacting on the legitimacy of the ICC. ¹⁵⁶ To the extent that the Independent Expert Report on the ICC reparation scheme finds it has ‘not delivered fair, adequate, effective and prompt reparations to victims of crimes under the jurisdiction of the Court’. ¹⁵⁷

4. The future of reparations at the ICC

The ICC is not the place to restore social justice or act as an administrative reparation programme for millions of victims. However, it has a reparation mandate that cannot be simply an afterthought. As Ferstman finds, the current Court’s approach to reparations is fundamentally problematic, as it lacks a ‘common vision about what successful reparations look like, and at best lukewarm commitment to doing what would be necessary to achieve anything beyond tokenism’. ¹⁵⁸ The growing unease with the current status quo of reparations in the Court signals the need for change. This has been exposed in light of the number of recent acquittals and termination of trials on a no-case basis.

The final submission of the victims in the Bemba case makes for sober reading: after 15 years waiting for reparations and justice at the ICC victims felt a ‘deep disappointment and hopelessness for not receiving justice … [leaving them] filled with scepticism and distrust towards the Court’ ¹⁵⁹ with their expectations raised for redress, especially when engaged on reparations by the expert team. ¹⁶⁰ Their legal representatives wanted the Chamber to establish principles of victims’ harm that could be used to claim reparations before other legal fora ¹⁶¹ but the Chamber rejected this. In the acquittal appeal decision in the Bemba case, Judges van den Wyngaert and Morrison, while recognizing that victims have a right to reparations, considered that the rights of defendants are at the heart of ICC proceedings:

It is emphatically not the responsibility of the International Criminal Court to ensure compensation for all those who suffer harm as a result of international crimes. We do not have the mandate, let alone the capacity and the resources, to provide this to all potential victims in the cases and situations within our jurisdiction … [what we suggest] is that we stop viewing the International Criminal Court’s reparation procedures as (part of) a mechanism to restore social justice and to heal the wounds of societies that have been torn apart by aggression, genocide, crimes against humanity or war crimes. Only if we do that will it be possible to manage victims’ expectations and can we relieve International Criminal Court prosecutors and judges from potential pressure that is currently imposed upon them to secure convictions at all cost. ¹⁶²

By contrast, in the Ruto and Sang case, where the Chamber terminated all trial proceedings against the accused, Judge Eboe-Osuji argued in his dissenting opinion that the ICC should allow victims to voice their views and concerns, even though the case was terminated, as the Court can still make a decision on reparation principles. ¹⁶³ More recently, in the Al-Rahman (aka Al-Kushayb) case, the defendant, at the pre-trial stage, while affirming his innocence, expressed ‘regret’ that reparations for victims will be conditional on his conviction, with a long period for them to wait. Instead, he proposed alternative principles for reparations including for the Court and TFV to process reparations from the outset of a case. ¹⁶⁴

Such cases and statements on Article 75(1) point to the need to have a fundamentally different procedure as well as a different institutional design to deal with reparations at the ICC. This would require an internal overhaul of the working of the ICC. The Court should aim to narrow the gap between the rhetoric or uninformed expectations of victims and the reality of reparations at the Court. We concur with Judge Eboe-Osuji who suggests that the reparation process should be ‘delinked’ from the criminal trial. ¹⁶⁵ As proposed in the filing of Mr Al-Rahman, a principled reparations order under Article 75(1) may offer this opportunity, as it does not require an order to be made against an accused, but rather to determine the principles on scope and extent of damage, loss or injury and appropriate forms of reparations. ¹⁶⁶ Such an order as suggested by victims’ representatives in the Bemba case could be used before other forums or fundraised through donor voluntary contributions facilitated by a reformed TFV. A principled reparation order would provide official recognition of victims’ harm and a legal basis to seek redress through domestic courts. Acknowledgment is crucial to provide an opportunity for vindication of victims’ suffering and means-making in understanding of the impact of violence on themselves. ¹⁶⁷ However, symbolic reparations are not sufficient in themselves to provide a sense to victims that their harm mattered as much as the crime did, and to alleviate their suffering.

We would argue that each Chamber involved in a case should have, right from the outset, a team working on victims’ participation and reparation alongside VPRS and people at the field office. The Court should also have a multidisciplinary team of experts on reparations in house to support the work of each Chamber and VPRS. Chambers should have a pool of experts such as a psychologist, a physician, an anthropologist, an economist, and a lawyer with practical expertise on victims’ consultation on harms and reparations, determining different types of harm (including collective), and identifying adequate forms of redress. The team should include at least one expert on gender violence, including sexual violence. The Court would then only appoint experts on reparation to address issues beyond in-house expertise.

There is an urgent need to also look beyond the ICC and consider reparations in holistic terms, by strengthening domestic and international developments on redress and reparation mechanisms, rather than placing all expectations on a Court that will never be able to adequately deliver reparations. ¹⁶⁸ Facing this challenge requires a clear international commitment by states to ensure that reparations for victims of international crimes are a priority. The Assembly of States Parties offers an important political space to raise the level of the discussion on reparation and to challenge state behaviour. So do other political spaces at the United Nations, for example, the Human Rights Council or the Security Council, as well as within regional international organizations such as the Organization of American States, the African Union or the Council of Europe which can seek such principled reparation orders implemented. ¹⁶⁹

The serious implementation deficit facing human rights courts, societies transitioning from mass atrocities and the ICC, indicates that states’ commitment to reparation, and operationalization are crucial to fulfil the right of victims to adequate, prompt and effective reparation. In the case of the ICC and the TFV, it is critical to consider how to create harmonious and reparative complementary mechanisms that work at the international, regional and national level to enable reparations and that prevent new harm of victims as a result of forum shopping or resentment across victims and communities. ¹⁷⁰

In relation to considering reparations holistically, the ICC should use reparative complementarity to ensure that domestic systems not only respond with diligent criminal investigations but also with reparation for victims of such atrocious crimes. ¹⁷¹ At the end of the day, the ICC will remain a Court of last resort and the bulk of the work on reparation should be done domestically and by other relevant actors, but greater co-ordination and vision is needed. Reparative complementarity in positive terms is a subsidiarity effort within states to end impunity by creating an effective remedy for victims to seek reparations for international crimes, thereby complementing reparations at the Court. ¹⁷² Such an approach could avail of complementarity where victims could seek implementation before national, regional or other bodies for reparations. A reparative complementarity approach should be more closely examined by the Court in how reparations can be provided through domestic mechanisms to widen the benefits of any reparations ordered by the Court, but always in a manner that is harmonious and consistent with international responses, and not one that would create further harm to victims by setting competing orders on reparation. ¹⁷³

In light of the Dominic Ongwen conviction in February 2021 there may be an opportunity for reparative complementarity in Uganda, where the government released its transitional justice policy in June 2019, including a programme for reparations. Uganda has an experienced, knowledgeable, and organized civil society and victim groups that can substantively engage in the development and design of a reparation programme. The Court should encourage state parties to take greater ownership in using reparations to end the pervading effects of impunity on victims and broaden out the benefits to all victims of international crimes, rather than just those before the ICC. The Court and the Assembly of State Parties should consider how they can add value to the transitional justice policy’s implementation on reparation in Uganda with the Ongwen case, building on the links the Field Office and the TFV have already developed there. ¹⁷⁴ This could be through donor support of the setting up of an administrative reparations body, funding, capacity building or delivering particular programmes such as identification documents for children born of war or recovery of the remains of those missing. This may tie in well with the work of other international organizations, such as the International Committee of the Red Cross, which has expertise in the forensic recovery of those missing in war. ¹⁷⁵ There are three concerns, however, that will continue to constrain the work of the Court in the years to come and where reflection is needed to ensure that the Court and the TFV are able to deliver on their mandates: security; funding; and implementation.

4.1 Security

Without adequate security on the ground, implementation of reparations is highly problematic. Insecurity has also been exacerbated since the COVID-19 pandemic began, making it even harder for the Court and the TFV to reach victims. ¹⁷⁶ Victims are in urgent need of prompt, adequate, and effective redress, which the pandemic has only compounded as many are unable to support themselves through livelihoods income, or access healthcare (due to shielding for underlying health problems), or at risk from increased domestic and sexual violence from being confined at home. Virtual or phone contact with victims or radio broadcasts are still possible, but some victims are unlikely to have access to direct communication with intermediaries or the Court through such means. These problems will be a persistent issue in situations facing insecurity and fragility like those in Libya, Burundi, the Central African Republic II, and Darfur, for example. This, bearing in mind reparative complementarity, may require creating better local and international networks with civil society and inter-governmental organizations to increase the reach to victims, but also in terms of delivery of reparations through such bodies where the state or the TFV are unable or unwilling to do so during ongoing insecurity. Victims can be displaced across borders or internally making their tracing over numerous spates of violence difficult, but measures could be put in place to ensure more direct provision of reparations to victims, such as mobile phone cash transfers. ¹⁷⁷

4.2 Funding

Given that all persons convicted by the Court so far have been declared indigent, a key challenge is how to secure the financial resources to implement the various forms of reparation, individual and collective, that are ordered by the Court. Voluntary contributions to the Trust Fund do not make viable the reparation mandate of the Court, among other reasons because reparation should not depend on the will of states or other actors to contribute to redress but also because, as Mégret suggests, this is in part a problem of demand and supply. ¹⁷⁸ It should be a process through which reparation is provided bearing in mind that redress should aim to wipe out the harms caused to the victims by the convicted person. As the Trust Fund has indicated, its funds ‘have already run short of even complementing all of the existing three reparations orders where the convicted persons were found indigent, totalling close to 15 million US dollars’. ¹⁷⁹ The Court was funded over €1.7 billion between 2005 and 2021, but only secured five convictions for international crimes. ¹⁸⁰ Most of this money has been spent on institutional and staff costs, with very little support reaching victims. ¹⁸¹ A possible response to the lack of voluntary contributions could be that the Assembly of State Parties consider creating a dedicated budget line for reparations in future budgets of the Court, creating more certainty and sustainable funding, rather than the TFV having to go around cap-in-hand seeking funding each year.

4.3 Implementation

Ordering reparations by a court that is based in the Hague with the aim of making them operational, is no small challenge. The Court needs a very clear understanding of the situation on the ground, the harms caused to victims, the cultural, religious and political features of the context and of the victims, as well as a well-grounded appreciation of what is possible in such situations and with those victims. Implementation can be helped by experts on reparation appointed by the Court as occurred in Al Mahdi, Bemba and Ntaganda, but it needs to be more sustainable and enable the ICC to develop its own institutional expertise. However, the Trust Fund has faced challenges in terms of implementation in terms of the ‘significant increase in workload, TFV Fund to directly deliver reparations awards; complexity induced by legal proceedings and contextual challenges; intense collaborative working relationships [with diverse actors] and … the need to intensify and diversity resource mobilization’. ¹⁸² How the TFV and the Court can operationalize and monitor compliance with reparations on the ground is one of the key questions.

One solution may be to reorient the assistance mandate to be interim relief, in that it would be targeted to those victims in a case, rather than a situation. This would help to focus redress on those victims before the Court, enabling harm to be mitigated at an earlier stage, instead of only after a conviction. This would not impact upon the rights of the accused, as victims’ rights to assistance and reparations are autonomous of any responsible individual being identified, prosecuted or convicted. ¹⁸³ While arguably this blurs the line between assistance/reparations and victims of the situation/case, it ultimately is an issue for judges to discern that receipt of interim relief does not go to the guilt of the accused person. Judges in domestic jurisdictions often can separate such issues, given the different burdens of proof in civil and criminal cases. It seems incongruous that judges of the ICC as legal professionals cannot do the same.

5. Conclusion

Reparations for international crimes are intended to respond to the harm suffered by victims. While the inclusion of reparations in the Rome Statute of the ICC was intended to offer more victim-centred justice within an international criminal body, it is confined within the criminal justice paradigm. For example, when considering the scope of liability of the convicted person, or when deciding if victims can still participate or apply for reparation at the ICC. Nevertheless, in contrast to ten years ago where the Court has no reparation decisions, today, despite a very late start, the picture of reparations at the ICC is becoming clearer. The Court has identified key substantive and procedural principles and others will continue to emerge, with six new ones being endorsed in the Ntaganda case. Judges have played a key role in experimenting and blending different legal principles and practices to make reparations work at the Court. They all denote a clear tendency at the ICC, as far as that is appropriate, to adopt international standards on reparation. However, major challenges persist.

We need to consider how to measure the success of the ICC reparations mandate in the future. This will require going beyond current developments of the Court. First, the Court should consider if its current architecture is adequate to implement its reparation mandate. As suggested in this article, the ICC should strengthen each Chamber responsible for a case by having dedicated experts on victims’ participation and reparation, working alongside VPRS and the TFV. Field offices should also require more funding and staffing to better capture data on victims’ harm and their views and concerns on reparations. In addition, a multidisciplinary pool of experts on victims’ participation and reparation should be created within the Court to assist Chambers and VPRS.

Second, the Court needs to strike a better balance between the rights of the accused and those of victims. This requires thinking of integrating reparations at the start of situations, from early investigations into assets of suspects to be frozen, as well as informing victims of the likelihood of reparations at the Court, their scope and comparable timeframe in other cases. This will allow victims to make more informed choices on whether or not to seek reparation before the Court and to consider opportunities for redress elsewhere. A principled reparations order provides a good basis to position reparations from the outset of the case, rather than the end, which could help inform interim relief measures.

Third, the Court needs to carry out its work as part of a landscape of international and domestic responses to reparation for victims and engage, in this sense, in reparative complementarity, particularly given the lack of resources and its narrow mandate. The ICC has limited capacity to deliver redress that is going to appear symbolic and a dilution of victims’ rights. Reparations are a complex and costly issue, which require long-term commitment by the state to adequately resolve. It cannot be a top-down, external exercise, if it going to be sustainable in the long-term. The Court and the TFV need to consider how, when, and with whom to operationalize reparations as ordered by the Court in the very difficult contexts they are dealing with, and as a consequence, they also need to consider how to monitor and measure compliance with this complex mandate.

These are issues that should have been raised 20 years ago, but some lessons are being slowly learned and the awareness and importance of the right to reparations of victims is increasingly becoming centre-stage. Decisive collective action is needed to make reparation a reality. The ICC is just one element on that process, but a very important one given the visibility it gives to crimes and victims. Otherwise, if we continue with the current arrangement of trying to tilt at windmills and not realistically implement reparations, victims will seek redress elsewhere. So, rather than having competing bodies ordering reparations worldwide, such bodies, should engage in better co-ordination and dynamics of dignification and reparation for victims. The ICC can play a pivotal role in ensuring that all of these bodies work as a system through principled orders.

The ICC is often the only justice actor in situations of ongoing conflict; conducting trials is not enough to end impunity. The ICC was established to end impunity for international crimes and deliver justice for victims, of which reparations are a key part. Reparations at the ICC should not be an afterthought of the trial, or a rhetorical exercise to legitimize the Court’s existence, nor should it be dismissed as quixotic or an enemy of the Court’s core work. We need to recognize that the failure to include reparations at the ad hoc tribunals, curtailed its legitimacy amongst affected communities. More than 20 years on from the adoption of the Rome Statute, we should rethink how to better implement reparations on the ground in domestic programmes and at the Court, so that it is not an international project enriching only lawyers in the Hague but alleviating the suffering of those bearing the consequences of such atrocities.