Jacques B. Mbokani has a Ph.D. in law from the Université catholique de Louvain and is a professor at the University of Goma and at the ULPGL/Goma. He is also an external consultant to OSISA-DRC and former ICC (OTP) intern. He has previously written about command responsibility in the Bemba case.

The original version of this article appeared in French here. The views and opinions expressed in this commentary do not necessarily reflect the views and opinions of the Open Society Justice Initiative.

The court has decided! On June 8, 2018, the Appeals Chamber of the International Criminal Court (ICC) decided to acquit Jean-Pierre Bemba of war crimes and crimes against humanity charges. As we know, in 2016, Bemba was convicted (unanimously) as a military commander and sentenced to 18 years in prison by the Trial Chamber III. This conviction, now overturned, followed killings, rapes, and looting attributed to Bemba’s soldiers during their intervention in the Central African armed conflict of 2002-2003. The acquittal has been harshly criticized, even within the panel of the Appeals Chamber, which decided the outcome by a narrow majority of 3/2. In this respect, one can note the disproportion between the length of the appeals judgment (80 pages) and that of the two minority judges’ dissenting opinion [pdf] (269 pages). Anyone who does not have a personal interest in this case and who expected clarification from the Bemba judgment on the definition of crimes against humanity and the conditions for command responsibility would find himself confused by the harsh division of the appeals judges, each side being so sure of itself that it blames the other for making “serious errors.”

After carefully examining the appeals judgment and the various opinions of the judges, it seems that this misunderstanding has its origin in the very general manner that the pre-trial chamber (PTC) confirmed the charges against Mr. Bemba. This chamber confirmed the charges against Bemba for crimes against humanity and war crimes because it was convinced that Bemba’s soldiers had committed a very “high number” of killings, rapes, and pillaging in the Central African Republic, as reported by indirect sources, such as reports from non-governmental organizations (NGOs) and the media. To reach this conclusion, it relied, for illustrative purposes, on a handful of individual criminal acts, namely one murder, 11 cases of rape, and six acts of pillaging (while dismissing two allegations of murder and 35 allegations of rape). In support of its method, the PTC cited the jurisprudence of the ad hoc tribunals, which, according to the PTC, admitted as evidence reports of a “high number” of crimes, without the need to specify the exact number of victims. However, as indicated in a paper published in the last special edition of the Quebec Journal of International Law (QJIL), this case law related to situations that were different from the one in the Bemba case. They concerned direct witnesses who had survived mass killings and who could not specify the number of victims with precision because they were very numerous. In the Bemba case, no direct witness reported massacres or mass rapes. It was rather a collection of individual accounts (sometimes anonymous and hearsay-based) made by indirect sources (NGO reports and media). This method, however, was admitted in the trial phase and explains: (i) the difference between the number of individual criminal acts found in the confirmation of charges and the one found in the trial judgment; and (ii) the disproportion, in the trial phase, between the number of victims allowed to participate in the proceedings (5,229) and the number of criminal acts established beyond any reasonable doubt (3 murders, 28 rapes, and 21 pillaging).

On appeal, however, and as we anticipated in the article published in the QJIL, the aforementioned method proved very problematic on three aspects. First, should Bemba’s knowledge of the criminal behavior of his soldiers be demonstrated only in relation to the individual criminal acts established beyond reasonable doubt or should it be established in relation to the “high number” of crimes reported by indirect sources, with the double risk of (i) confusing the knowledge of an allegation of a fact with the knowledge of a fact as such; and (ii) charging Bemba the knowledge of the criminal acts that the chamber itself has not established beyond a reasonable doubt? Second, should the rule of proportionality between the measures that should be taken by the commander in charge be assessed with reference only to criminal acts established beyond any reasonable doubt or should it be assessed according to a “high number” of crimes reported by indirect sources? It is clear that the measures that a commander would take in order to prevent or punish only the limited crimes established beyond any reasonable doubt would necessarily be perceived as “largely insufficient” to prevent or punish the “high number” of crimes reported by indirect sources. Third, should the interpretation of the concept of a “widespread” attack in crimes against humanity be based on this handful of criminal acts established beyond any reasonable doubt, or should it take into account the “high number” of crimes whose existence remains controversial, with the risk of confusing the widespread nature of an “attack” and that of individual criminal acts?

As we can see, the majority and minority judges provide radically opposite answers to these three questions. However, this controversy could probably have been avoided if the PTC refrained from confirming the charges in a very general manner. In this respect, it is regrettable that neither of the parties appealed the confirmation of charges decision, despite its ambiguities, as we pointed out almost seven years ago. If the Appeals Chamber had confirmed that approach contained in the confirmation decision at an earlier stage, it would have certainly not decided the way it did on June 8.

With regard to the Appeals Chamber’s standard of review specifically, the discrepancy between the judges’ opinions is striking. According to the minority opinion, shared by the prosecutor, the Bemba appeals judgement deviated from the so-called “conventional” standard of review of an appeal against a conviction decision. However, it is important to note that Article 21(3) of the Rome Statute, a provision included for the first time in the statute of an international criminal court, requires that the application and interpretation of the Rome Statute be consistent with internationally recognized human rights. This peculiarity of the Rome Statute, therefore, calls for caution when applying the jurisprudence of the ad hoc tribunals (whose statutes do not include for such provision). One of the fundamental human rights, which is also set out in Article 14(5) of the International Covenant on Civil and Political Rights, is “to have the conviction and sentence reviewed by a higher tribunal.” While the restrictive approach advocated by the minority judges with regard to the scope of review of the Appeals Chamber may be in line with the jurisprudence of the ad hoc tribunals, there is a danger that it will divert from the way in which international human rights bodies have interpreted the right to have the conviction reviewed by a higher tribunal (see General comment no. 32 of the Human Rights Committee, § 48). In any event, this very restrictive approach to the powers of the Appeals Chamber, reducing it to a sort of “court of cassation,” would likely be favorable to the prosecutor only when she wins the trial in first instance but not when she loses it.