Dr. Marina Lostal, University of Essex, School of Law
NB: The author has been involved in the work of reparations at the ICC, first as a Court-appointed expert in the Al Mahdi case in 2017 and, between 2018 and 2020, as a reparations expert at the TFV. However, this post is written in the author’s personal capacity.
On 8 July 2019, Trial Chamber VI of the International Criminal Court (ICC) sentenced Bosco Ntaganda to 30 years of imprisonment after finding him guilty of five counts of crimes against humanity and thirteen counts of war crimes committed between 2002-2003 in Ituri (DRC). The Appeals Chamber confirmed both the sentence and verdict on 30 March 2021.
A few weeks earlier, on 8 March 2021, Trial Chamber VI issued the Ntaganda Reparations Order. An order for reparations is the document that sets out the foundations for the implementation of measures in favour of victims and, for that purpose, must be (1) directed against the convicted person; (2) establish the convicted person’s amount of liability; (3) identify – or set the criteria to later identify – the victims of the case; (4) determine the harm suffered; and (5) the type and modalities of reparations appropriate to address such harm. Aside from this case-specific goal, and pursuant to article 75(1) of the Rome Statute, an order for reparations can be used to amend or develop the legal framework applicable to reparations through the enunciation of principles.
The Ntaganda Reparations Order is the fourth issued by the Court following those of Lubanga (2012-2015), Katanga (2017), and Al Mahdi (2017). It is, however, the first one that expands and modifies the reparations principles that had been initially laid down in the Lubanga case (i.e. Lubanga reparations principles). In so doing, the Ntaganda reparations principles (in paras 28-103 of the Reparations Order) have effectively modified the legal framework applicable to reparations at the ICC, and brought it closer to a victim-centred approach where the emphasis is increasingly placed in the goal of repairing the harm caused to victims.
This post gives an overview of the five elements of the Ntaganda Reparations Order, and discusses some of the changes that it has brought to the Court’s reparative framework.
Against the convicted person
Article 75(2) of the Rome Statute requires the order for reparations to be made against the convicted person. However, nothing in the legal framework demands such conviction to be final. The Ntaganda Trial Chamber decided to issue an order for reparations while the verdict and sentence were on appeal; a practice that remains in keeping with the legal framework but that, nonetheless, is unusual and, in this author’s view, potentially problematic.
Never before had a Trial Chamber issued a proper order for reparations until the judgment had become final (cf. Ntaganda Reparations Order, footnote 10). In Bemba, Trial Chamber III held off until the Appeals Chamber reached a decision. Following Mr Bemba’s acquittal, Trial Chamber III found that “no reparations order [could] be made against Mr Bemba under Article 75 of the Statute” (Final decision on the reparations proceedings, para. 3).
The longer the lapse of time between a pre-appeal order for reparations and an appeal judgment, the more problematic this practice could become. If the conviction is partially or completely reversed on appeal, the order for reparations would have to follow suit by reducing the monetary sum and cutting off entire pools of victims. In practice this would imply, at the very least, damage-controlling any outreach messages already made and, in the worst case, halting or reversing any implementation work taken up.
The Ntaganda Trial Chamber justified issuing a Reparations Order ahead of the appeal judgment because the mandate of two of its judges was coming to an end before the date foreseen for the appeals judgment (para. 5). It also argued that issuing the Reparations Order earlier could “contribute to more expeditious reparations proceedings” (ibid.) pursuant to the victims’ right to receive prompt reparations. However, given that the appeal judgment was rendered only 23 days after the Reparations Order, the latter reason remains unconvincing. It would have been perhaps more appropriate to argue the opposite: that because the appeal judgment was expected only shortly after, any potential reversal of some or all of the charges would not have impacted victims because the implementation work was unlikely to have started. Be that as it may, since the Appeals Chamber confirmed the trial sentence and verdict, no amendments were needed to the Ntaganda Reparations Order.
Mr Ntaganda’s liability has been set at USD 30 million (para. 247); the highest so far (Lubanga USD 10 million, Katanga USD 1 million, and Al Mahdi EUR 2.7 million). In arriving at this sum, the Reparations Order has finally clarified the primary factors that should be taken into account in determining someone’s monetary liability: the harm and the costs to repair it (para. 98). Modes of liability, gravity of the crimes, or mitigating factors are irrelevant for the determination of the sum (ibid.). This has changed the Lubanga principle on liability, according to which the convicted person’s liability for reparations had to be:
“proportionate to the harm caused and, inter alia, his or her participation in the commission of the crimes for which he or she was found guilty, in the specific circumstances of the case”Lubanga Reparations Principles, para. 21
The application of this principle had led to contrasting outcomes. In Lubanga, the Court “appears to have found Lubanga monetarily responsible for the totality of harm suffered”, USD 10 million, irrespective of the fact that there were other perpetrators. By contrast, in Katanga, the Trial Chamber first calculated the sum-total of the harm caused in the attacks against Bogoro (i.e. USD 3,752,620) and then, apportioned Mr Katanga with USD 1 million based on his participation and mode of responsibility (Katanga Reparations Order, para. 264). In Al Mahdi, the Trial Chamber considered that EUR 2.7 million was a fair assessment of the Mr Al Mahdi’s liability alone, and did not deem it “necessary to decide whether the figures […] constitute[d] the sum-total of harm suffered” (Al Mahdi Reparations Order, para. 111).
The internal discrepancies with the Lubanga principle on liability became even more tangible when, in its judgment against the Katanga Reparations Order, the Appeals Chamber both endorsed and rejected the principle in the same paragraph. It said the fact that a convicted person’s liability for reparations had to be proportionate to the harm caused and, inter alia, issues of participation
“d[id] not mean, however, that the amount of reparations for which a convicted person is held liable must reflect his or her relative responsibility for the harm in question vis-à-vis others who may also have contributed to that harm”.Katanga Appeals Judgment on Reparations, para. 175
The Ntaganda Trial Chamber decided to unequivocally amend the Lubanga liability principle as follows:
“[i]n determining the amount of the convicted person’s liability, the primary consideration should be the extent of the harm and the costs to repair it. Other criteria, such as modes of liability, gravity of the crimes, or mitigating factors are not relevant to this determination”.Ntaganda Reparations Order, para. 98
In other words, the Ntaganda Trial Chamber has embraced a victim-centred approach where the objective is repairing the harm caused to the victims, not as caused by the offender. This is also more in line with the non-punitive nature of reparations proceedings where, as a consequence, the degree of culpability of the individual should be inconsequential to questions of monetary liability. Accordingly, the Ntaganda Reparations Order found:
“Mr Ntaganda liable to repair the full extent of the harm caused to the direct and indirect victims of all crimes for which he was convicted, regardless of the different modes of liability […] and regardless of whether others may have also contributed to the harm” (para. 218).
When there are other perpetrators, two things must be taken into account: first, victims cannot be over-compensated over time (paras 99-100). Second, the existence of responsibility in solidum where the perpetrator who bore the cost of reparation can “seek to recover from the co-perpetrators their proportionate share” (para. 221).
Accordingly, concerning the child soldier victims that Mr Ntaganda shared with Mr Lubanga, the Ntaganda Reparations Order stated that “the reparation programmes implemented in the Lubanga case […] should be understood to repair the victims’ harm on behalf of both” (para. 220). Even though the measures in favour of these former child soldiers would be implemented under the umbrella of the Lubanga Reparations Order, Mr Ntaganda remains jointly and severally liable to repair in full the harm suffered by these child soldiers, that is, he also bears the obligation to reimburse the Trust Fund for Victims in that respect (para. 221).
Identify – or set the criteria to later identify – the victims of his crimes
There is no doubt that the Ntaganda reparations will deal with a very large pool of victims, but the estimated number of prospective eligible victims remains far from clear. The Trial Chamber conceded that the “exact number [of victims] may never be determined given the passage of time, the widespread and systematic nature of the crimes committed” (para. 8). Current figures range from a starting base of 2,500 according to the Registry (paras. 232 and 234); 3,500 direct victims according to the Court-appointed experts (para. 232), to 100,000 people as per common legal representative 2 (para. 233). Given the impossibility to identify victims by name at this stage, the Trial Chamber has opted for establishing their eligibility criteria instead (para. 105).
The Trial Chamber kept the distinction between direct and indirect victims where the former “are those whose harm is the result of the commission of a crime for which the defendant was convicted” (para. 34); and the latter refers to “those who suffer harm as a result of the harm suffered by the direct victims” (para. 35). This would include family members of direct victims, even those who were not born at the time of the commission of the crime but that are able to prove transgenerational harm, that is, the transmission of noxious effects of trauma from one generation to the next (paras 36, 73 and 183(d)(vi))
Aside from the two known existing categories of direct victims (viz. victims of the attacks and child soldiers), the Ntaganda Reparations Order added a third one: children born out of rape and sexual slavery (para. 122). This carries an important practical implication: the offspring of these children born out of rape and sexual slavery could, in turn, qualify as indirect victims on account of transgenerational harm. Therefore, the damage Mr Ntaganda caused could transcend to, not only one, but two generations that were unborn at the time of the commission of the crimes.
Determine the harm suffered
The Ntaganda Reparations Order acknowledged that the crimes of Mr Ntaganda had led to mass victimisation (paras 149-169) and provided a detailed list of the, often multidimensional, types of harm suffered per victim category (para. 183).
As to questions of evidence, the Ntaganda Reparations Order has endorsed the same legal tests used in previous reparations proceedings. This is the balance of probabilities to establish someone’s identity and the occurrence of harm (para. 136), and the but/for and proximate cause tests to link the harm to the crime in question (para. 132).
The balance of probabilities requires presenting some sort of documentation or, in its absence, a statement signed by two credible witnesses (para. 137). However, a special rule will apply to victims of sexual and gender-based violence where “the victim’s coherent and credible account shall be accepted as sufficient evidence to establish their eligibility” (para. 139). In addition, factual presumptions will be relied on for certain types of victims. This means that material, physical and psychological harm would be assumed for, say, rape victims without any need to scrutinize the specific harm in question (paras 144-147).
Type and modalities of reparations
Reparations can be individual or collective, and both types can be awarded concurrently (rule 97(1) RPE). However, the Ntaganda Reparations Order has crystallised a middle-of-the-way category known as “collective reparations with individualised components”. These reparations, which are the only ones that will feature in the Ntaganda case (paras 7 and 186), are collective in nature but they result in individual benefits.
The Ntaganda Reparations Order seems to suggest that collective reparations with an individualised component may adopt any given modality, even monetary compensation (para. 86), something that had so far been framed as an individual form of reparation (Al Mahdi Reparations Order, paras 81 and 90). Why then label reparations as collective when they end up leading to concrete individual benefits?
Aside from the different expressive function embedded in these two terms, in the ICC legal framework, there are certain practical advantages in calling something “collective” reparations. One should be mentioned here. The Trust Fund for Victims (TFV) may prioritize funding collective reparations over individual ones (TFV Observations, para. 124). On its side, the Ntaganda Reparations Order has established that certain victims need priority treatment, such as “individuals who require immediate physical and/or psychological medical care, […] victims who are homeless” (para. 214). If the Ntaganda Trial Chamber had not categorised reparations as “collective” there could be, in principle, a clash between the prioritisation policy of the TFV, and that of the Ntaganda Reparations Order where the victims in need of tailored and urgent care would come last in their access to funding.
How would these reparative measures will look in practice? The ball is now in the TFV’s court as it is in charge of submitting two draft implementation plans, one urgent for priority victims by 8 June 2021, and the rest by 8 September 2021. What is clear is that the future draft implementation plans of the TFV needs to be comprehensive and include all modalities of reparations – compensation, restitution, rehabilitation, satisfaction and symbolic – or justify their exclusion (para. 212) and, of course, adopt a very victim-centred approach. Moreover, Mr Ntaganda is considered indigent and the Trial Chamber has, as in all other cases to date, invited the TFV to complement the award (para. 257). If so, the TFV would need to conduct an enormous fundraising effort, the biggest so far (see TFV Financial Overview, pp. 17-19, 25), where State parties are expected to play a major role so that reparations plans are turned into reality.
This piece was first published on the Blog of the European Journal of International Law and is reproduced here with permission and thanks. The original post can be accessed here.