Formal, Functional, and Intermediate Approaches to Reparations Liability: Situating the ICC’s 15 December 2017 Lubanga Reparations Decision

On 15 December 2017, the International Criminal Court (ICC) Trial Chamber II found Thomas Lubanga Dyilo, former President and Commander-in-Chief of the UPC/FPLC, responsible for reparations in the amount of USD 10,000,000 — the largest ICC reparations order issued to-date. The Lubanga case was the first to reach the reparations stage — yet controversy surrounding procedural requirements delayed the Chamber’s determination of Lubanga’s monetary liability. Last month’s decision answered some of these procedural questions, and raised new ones. This piece breaks down Trial Chamber II’s 15 December 2017 decision, and situates it alongside Trial Chambers’ recent assessments of monetary liability in the Katanga and Al Mahdi cases. We suggest that we have now seen ICC Trial Chambers assess defendants’ monetary liability for reparations via formal, functional, and intermediate approaches.

Lubanga was convicted on 14 March 2012 of enlisting and conscripting children under the age of 15, and using them to actively participate in hostilities from 1 September 2002 until 13 August 2003. On 7 August 2012, Trial Chamber I delivered the ICC’s first-ever order for reparations, authorising only collective reparations. On 3 March 2015, the Appeals Chamber overturned part of the Trial Chamber’s decision and issued an amended order for reparations, giving a newly constituted Trial Chamber II (composed of Judges Brichambaut, Herrera Carbuccia and Kovács) the confined tasks of a) determining the amount for which Lubanga was responsible, and b) monitoring and overseeing the implementation of the order. In its Judgment and order, the Appeals Chamber did not identify the number of victims who suffered harm as a result of Lubanga’s crimes. Nor had Trial Chamber I provided a figure in its original Judgment, although it found the crimes were widespread.

As explained in an article published last year, heated procedural debates soon emerged, as Trial Chamber II and the Trust Fund for Victims (TFV) clashed in their understandings of their respective mandates: while the Chamber believed it needed to identify and “approve” victims entitled to reparations as a prerequisite to determining Lubanga’s monetary liability, the TFV believed this was unnecessary, and something the TFV should do during implementation (the TFV had estimated there were 3,000 potentially eligible victims). Similarly, while the Trial Chamber believed that it needed to determine the extent of the harm caused to victims to establish Lubanga’s liability, the TFV thought that the extent of the harm was already described adequately in the Judgment, Sentencing Decision, and decisions on victims participation. However, in what appeared to be a change of its original position, the Trial Chamber acknowledged mid-proceedings that the victims identified by the TFV were a sample, but did not comprise the totality, of victims potentially eligible for reparations, namely those who suffered harm as a result of the crimes for which Lubanga was convicted. This shift proved foundational to the Trial Chamber’s 15 December 2017 decision.

In the midst of this Lubanga debate, Trial Chamber II (with the same composition as in Lubanga) issued a reparations order in the case against Germain Katanga. Trial Chamber II pursued in Katanga what we term a “formal” means of calculating liability, akin to civil claim proceedings: first, the Chamber identified a bounded set of 297 victims (out of 341 applicants) entitled to reparations (¶¶33, 168-80). Then, the Chamber — without resort to experts — added up the monetary value of the harm suffered by each of these identified victims, to reach a total monetary value of the overall harm of USD 3,7 million (¶¶181, 190-239). The Chamber found, however, that Katanga was criminally responsible for only USD 1 million, as an amount deemed proportionate to both the harm caused and the specific circumstances of Katanga’s participation in the commission of the crimes (¶264).

Months later, Trial Chamber VIII (composed of Judges Pangalangan, Kesia-Mbe Mindua, and Schmitt) approached monetary liability for reparations differently in its reparations order in the case against Al Mahdi, issued shortly after the Judgment. Trial Chamber VIII assessed the value of harm “suffered by or within the community of Timbuktu” (¶56) using what we term a “functional” approach. The Trial Chamber rejected the arguments that harm and associated liability could only be determined on the basis of the 139 individual victim applications before the Chamber (¶¶5, 59), and that the Chamber needed to identify and approve the victim beneficiaries (¶143). Instead, the Chamber engaged with expert reports to enable it to “reasonably approximate” costs that established Al Mahdi’s monetary liability of 2.7 million euros (¶¶109-34) and delegated the identification of the victims to the TFV (¶144-46), moving the process forward. In both Katanga and Al Mahdi, the Chambers awarded a combination of individual and collective reparations. Both orders have been appealed.

Last week, the Lubanga Trial Chamber took a different, “intermediate” approach. Despite its initial opposition to the TFV’s requests, the Chamber largely endorsed the TFV’s position that the Chamber need not identify all victims — nor assess their specific harm — to quantify Lubanga’s monetary liability. The Chamber endorsed aspects of both Katanga and Al Mahdi reparation orders to conclude that Lubanga was liable for USD 10 million.

As in Katanga but unlike Al Mahdi, the Trial Chamber assessed whether each of 472 individual victim applications were entitled to reparations. It determined that, on a balance of probabilities, only 425 victims out of this pool of identified victims had suffered harm resulting from the crimes for which Lubanga stood convicted, and were thus entitled to access collective reparations (¶¶194, 239). However, and unlike in Katanga, the Chamber did not assess or quantify these victims’ specific harms (¶185). Instead, it determined that all direct and indirect identified victims had suffered an “average harm” comprising elements of material, physical and psychological harm, and estimated the value of this harm to be USD 8,000 “ex aequo et bono” (meaning, “according to the right and good”, or “from equity and conscience”). In so finding, the Chamber relied on the submissions by the Legal Representatives of the Victims and the Office of Public Counsel for the Victims and the Chamber’s own assessment in Katanga (¶245-59).

The Chamber further found that these 425 victims were only a “sample” of the overall number of still-unidentified victims of Lubanga’s crimes (¶¶191, 235, 240); determined that unidentified-yet-eligible victims were in the hundreds or thousands (¶¶231, 244); and entrusted the TFV with identification of further victims. The Trial Chamber reached this conclusion by relying on Trial Chamber I’s findings in the Lubanga Judgment and Sentencing Decision (which indicated that the crimes were widespread) (¶238); estimates provided by the Office of Public Counsel for the Victims and one of the two Legal Representatives that victims numbered 1,000-1,500 (the other Legal Representative estimated 20,000 – 25,000 victims, and Lubanga estimated 200) (¶¶200-12); and figures provided by the Democratic Republic of the Congo (¶¶195-99). The Chamber also relied on open source data (mostly from the United Nations and other international and non-international organisations) which estimated 2,451 to 5,938 direct victims (¶¶213-31 and annex III, where the Chamber explains the “reasonable hypothesis” considered). Underreporting was likely due to the long time elapsed since the crimes, length of the proceedings, geographic dispersion of victims and/or stigmatization, among other reasons (¶236).

In a further surprise move, given Trial Chamber II’s approach to calculating liability to-date, the Chamber — noting the gravity and widespread nature of the crimes as well as the extent of Lubanga’s participation — determined Lubanga’s monetary liability for reparations to be USD 3.4 million for harm suffered by the 425 identified victims, plus USD 6.6 million for harm suffered by non-identified victims, totalling USD 10 million in overall liability (¶¶268-80).

Unlike in Katanga, the Chamber appears to have found Lubanga monetarily responsible for the totality of harm suffered by at least the 425 identified victims (425 x USD 8,000 amounts to USD 3.4 million) — regardless of the Chamber’s recognition that several persons are potentially responsible for the crimes tried before the Court (¶276). Although the Chamber did note that no other persons have been found guilty of crimes causing the victims’ harm in this case, its decision was confined to Lubanga’s liability (¶277). It is unclear whether the Chamber may have considered Lubanga’s “essential” contribution as co-perpetrator under article 25(3)(a), taken alongside the specific circumstances of the case, so fundamental as to support monetary responsibility for the totality of the harm caused to the identified victims — as compared to Katanga’s contribution as an accessory under article 25(3)(d). It is also unclear whether the Chamber held Lubanga monetarily responsible for the totality of harm suffered by non-identified potential victims since the Chamber did not estimate their individual harm.

Further, like the Katanga (¶246) and Al Mahdi (¶114) Trial Chambers, the Lubanga Trial Chamber held the defendant’s indigence irrelevant to his overall liability for reparations (¶269). The TFV must now indicate whether it can cover the bill — yet this will prove difficult, given the TFV’s recent indication that it only has 5.5 million euros. The registry will monitor Lubanga’s financial situation in the unlikely event that he obtains funds to repay the TFV. In the meantime, the TFV continues with the implementation of the reparations order: it is currently searching for a partner to implement service-based collective reparations programs, and has already identified a partner to implement symbolic reparations.

The above shows that the ICC is developing a reparation system that incorporates divergent methods of calculating a convicted person’s monetary liability. To some extent, different approaches seem dependent on the case-specific particularities such as the nature of the crimes and ensuing harm; geographical and temporal scope of the crimes; number of victims; and, possibly, the legal background and pragmatism of each bench. It remains to be seen what approach Trial Chamber III will adopt in the Bemba case, now at the reparations stage, with over 5,000 victims participating in proceedings. Considering the characteristics of the case, and noting that the Chamber has already engaged experts, the Chamber will likely follow the functional approach in determining Bemba’s liability, and delegate to the TFV and/or Registry the identification of victims. In any event, this Lubanga decision may not be the last in the case, since Lubanga and the victims have a right to appeal the Chamber’s determination of Lubanga’s liability. Stay tuned!