Some Reflections on the Bemba Appeals Chamber Judgment

[Alexander Heinze is an Assistant Professor at the Department of Foreign and International Criminal Law, Institute for Criminal Law and Justice, Georg-August-Universität Göttingen. He is the author of the book “International Criminal Procedure and Disclosure” (Duncker & Humblot, 2014).]

In less than a month’s time, the International Criminal Court (ICC) will celebrate its 20th anniversary. Distinguished speakers will be calling into mind that the ICC was first and foremost a Court for victims – and then they will be expressing their condemnation of the recent acquittal of Jean-Pierre Bemba from the charges of war crimes and crimes against humanity, as did the recent blog posts here, here and here. It goes without saying that every aspect of the Majority Judgment, the Separate Opinions and the Dissenting Opinion deserves its own publication. However, since the ICC Prosecutor already issued a press statement strongly criticizing the Majority Judgment, and the above mentioned blogposts – written by authors who are or were affiliated with the ICC’s Office of the Prosecutor – bemoan, without waiting for the publication of Judge Eboe-Osuji’s Separate Opinion, a “lack of clarity, retroactive application of new law, and negative consequences in this and future cases”, “extremely negative consequences for the institution”, or fuel resentments by stressing that “the Court ruled that he could join his family in Belgium while he awaits sentencing in that case” (citing BBC News and adopting the report’s almost exact wording instead of referring to the actual Court decision, where Bemba’s family ties in Belgium are provided merely as a side information in a half sentence), the purpose of this blog entry is to bring the discussion back on the legal track.

With regard to the standard of review, the minority employs an approach that is well known before International Criminal Tribunals, i.e. an Appeals Chamber would only interfere where the Trial Chamber’s appreciation of the facts was wholly unreasonable, that is, where it “cannot discern how the Chamber’s conclusion could have reasonably been reached from the evidence before it” (Dissenting Opinion, para 9). While this approach carries the idea of a margin of deference to the factual findings of the Trial Chamber, the majority views this idea “with extreme caution” (Majority Judgment, para. 38). Even though the majority does support the standard of reasonableness, it clarifies that this standard “is not without qualification” (para. 41; more critically Separate Opinion Eboe-Osuji, para. 72). Thus, the majority opines it may interfere with the factual findings whenever the failure to interfere may occasion a miscarriage of justice. It had the right to deviate from the conventional approach, due to the fact that the ICC “may” but does not have to “apply principles and rules of law as interpreted in its previous decisions” (Art. 21(2) ICC-Statute – Judge Eboe-Osuji justifies the deviation also through a contextual interpretation of Art. 83(1) ICC-Statute, which seems to stretch the ratio of that provision a little too far). The application of the concrete standard of review is then a policy decision (in the same vein Separate Opinion Eboe-Osuji, para. 46) and there are good arguments for both sides (consistency of Appeals Chamber decisions and not turning appeals proceedings into a second trial vs. protection of the rights of the accused and avoiding to “consign the fate of a convicted person to the undoubted good faith of the Trial Chamber whose verdict of conviction has been appealed” (Separate Opinion Eboe-Osuji, para. 11)). There is another argument in favour of the broad approach of the majority: the ICC Appeals Chamber is not a national appeals court. There is no external review mechanism with regard to the rights of the Defendant as we know it in national jurisdictions, where a convicted person can still make an application to a regional human rights body for an infringement of his or her fair trial rights. There is also no Constitutional Court Chamber, as we know it from the Kosovo Specialist Chambers (see my blogpost here). The Appeals Chamber is very much the end of the road. It is therefore illogical when the Prosecutor in her press statement uses this fact as an argument for a “cautious approach to appellate review”. Shouldn’t it be the opposite? When there is no further appellate or revisionary body, the standard of review of the Appeals Chamber needs to be designed in a way that protects the rights of the accused most effectively. Judge Eboe-Osuji’s reasons why he thought following the conventional standard of appellate was especially inappropriate are illuminating. Everyone who evaluates the Majority Judgment as a particularly grave form of arbitrary decision making should read them in total. It is telling that even Judge Eboe-Osuji’s he leaned towards referring the case back to the Trial Chamber, he evaluated the evidential analysis by the Trial Chamber as so flawed that he preferred an acquittal over an inconclusive judgment.

The controversy around the standard of review continues in the question of how the charges must be framed and confirmed so that a conviction does not exceed the charges (Article 74(2) ICC-Statute). Bemba was convicted partly based on individual acts of murder, rape and pillaging committed against particular victims at specific times and places that had not been confirmed in the Confirmation Decision. The Trial Chamber argued that the Pre-Trial Chamber’s Confirmation Decision was broad enough so that new allegations could be included without a new Confirmation Decision. The question therefore was: How concrete must the charges be and how much evidence must the Prosecutor provide? Is it acceptable to formulate them in such a broad fashion that – exaggerated – the Prosecutor can add criminal acts whenever she wishes to do so, with leave of the Trial Chamber after the trial has begun? The minority answers this in the affirmative, which is nothing unusual from a domestic law point of view. The prosecutor enjoys discretion in formulating the charges and that rightly so. The burden of proof in the confirmation proceedings is “substantial grounds to believe that the person committed the crime” (Article 61(5) ICC-Statute) – a standard that is lower than the burden required for a conviction (“beyond reasonable doubt”). Nevertheless, even in national proceedings the opening of an investigation against a suspect and the subsequent charging creates a considerable stigma. This is all the more true in an international context where everything, especially worldwide media attention, is potentiated. The least that can be expected in that situation is a clear and fairly strict formulation (and confirmation) of the charges. The majority therefore rightly states: “Simply listing the categories of crimes with which a person is to be charged or stating, in broad general terms, the temporal and geographical parameters of the charge is not sufficient” (para. 110). Judge Eboe-Osuji concurs with this opinion, even though he disagrees with Bemba that a Trial Chamber was in principle precluded from amending the indictment after the commencement of trial, employing a teleological and contextual interpretation of Art. 74(2) ICC-Statute (in conjunction with Article 69(9)) (Separate Opinion Eboe-Osuji, paras. 118-134).

Taking both procedural questions together, it is certainly fair to say that they address the interpretation of procedural rules that can work both ways with the better arguments for the majority. This also applies to the main substantial question, the requirements of command responsibility. To make this crystal clear: No one affiliated with the implementation of International Criminal Law has an interest that leadership-level- or mid-level perpetrators go free despite their criminal responsibility while their subordinates commit unimaginable atrocities. However, the concept of individual criminal responsibility for violations of humanitarian and human rights norms is universally recognized. There should be no situation in international criminal adjudication today where this self-evident principle needs special emphasis. And yet, the catchphrase “crimes did not commit themselves” is carried like a torch through social media and even the blogosphere (see here) to light fires of outrage and open resentment to the Appeals Chamber and its majority Judges. Again, even the Prosecutor could not resist the temptation of an emotional rejection of the Appeals Chamber’s majority view, instead of entering a self-critical analysis of its work.

In contrast to the minority opinion, it found that Bemba took all necessary and reasonable measures to prevent or repress the commission of crimes by applying a rather realistic set of criteria: Bemba was a “remote commander” with “non-linear command” in a foreign country (Majority Judgment, para. 171) and therefore faced limitations in controlling his troops. The minority disagrees: Even though Bemba as a remote commander faced “logistical difficulties” (Dissenting Opinion, para. 57), the evidence showed that it was possible to overcome these difficulties. The majority generally was of the view that a commander cannot be expected to do the impossible, taking any preventive measure, no matter how unrealistic they would be. Instead, the formula of the majority is rather clear, echoing the principle of individual criminal responsibility:

“The trial chamber must specifically identify what a commander should have done in concreto. Abstract findings about what a commander might theoretically have done are unhelpful and problematic, not least because they are very difficult to disprove” (Majority Judgment, para. 170).

Moreover, what a commander should have done in concreto is not for the accused to show, but for the Trial Chamber. When determining whether the measures the commander had taken were necessary and reasonable, the motive to counter public allegations and rehabilitate the public image of the subordinates does not intrinsically render the measures “any less necessary or reasonable” (Majority Judgment, para. 177; Separate Opinion Eboe-Osuji, para. 16) – even though the commander is required to act in good faith in adopting such measures and must show that he “genuinely” tried to prevent or repress the crimes in question or submit the matter to the competent authorities. Again, these legal aspects require much closer analysis than this, as has been done with the latter aspect (Bemba’s motivation) here.

Considering the remarks of the majority and reading the separate opinion, it becomes clear that the Defense’s call for a more realistic approach to command responsibility that takes into account the specific operative situation of the commander has been heard, at least by the majority of the chamber. Judge Eboe-Osuji’s very detailed and thorough engagement with the arguments brought forward by Defense Counsel show how much of an impact the actual hearing had on the majority’s opinion (read, for instance, para. 247: “It was wise, indeed, of Ms Gibson to have readily accepted the endangerment rationale on behalf of the Defence during the oral hearings. Notably, however, co-counsel Mr Newton, was not as forthcoming. […]). The separate opinion touches upon further controversial criteria of command responsibility that are certainly noteworthy from an academic point of view, even though they are not part of the majority judgment but rather an obiter, since there was apparently no consensus between the judges in that regard. For instance, the separate opinion of Judges van den Wyngaert and Morrison provides an excellent analysis of the subjective element in Article 28, explaining why it matters whether the accused is charged with/convicted of “knowing” the crimes of his/her subordinates or “should have known” the crimes – both standards trigger different obligations for the commander and must therefore be proven differently, what made the Trial Chamber’s notice under Regulation 55 of the Regulations of the Court to characterize the charges from “knowledge” to “should have known” defective (Separate Opinion, para. 39). The minority, by contrast, saw no harm in that, based on a “unitary standard for the mental element” (Dissenting Opinion, para. 266).

The fundamental differences about a possible causation element mirror the unclear legal nature of Art. 28. This unclear legal nature is best captured by Judge Eboe-Osuji, whose journey through the familiar classifications of Art. 28 as dereliction of duty on the one hand and accomplice liability on the other hand leads him to the rather surprising destination of command responsibility as “endangerment liability”: Since armed conflicts are “notoriously dangerous”, Art. 28 seeks to “protect innocent victims from the risk of the excesses that is so notoriously a feature of armed conflicts” (para. 243). Thus, it was not necessary that the defendant’s conduct caused the actual; proof of the creation of danger was sufficient “in the sense that the commander’s complicity in the subordinates’ crimes originated when (s)he created or fostered the danger of the subordinates’ criminality (by training them, arming them and/or deploying them, so as to be able to commit those crimes), and that complicity was consummated when (s)he failed to exert authority (that was effectively available to be exerted) properly to prevent or repress the crime (including through punishment), when s(he) knew (or should have known) that the subordinates were committing or about to commit such crimes.” (para. 251). This is indeed a novel approach, reminiscent of the theory that complicity was a crime of endangerment (as proposed in Germany by Herzberg, Goltdammer’s Archiv für Strafrecht 1971, pp. 1 et seq.). It is for another publication to enquire whether this theory can and should be transferred to command responsibility. The Dissention Opinion and Judges van den Wyngaert and Morrison in their Separate Opinion follow the conventional path of this debate: While the former justifies the causation element in convincing fashion, the latter rejects it based on the argument that it was “not possible that an omission after a fact has occurred (that is, failure to refer criminal behavior to the competent authorities) causes this fact”. This might be true but is a too easy escape route. In case the crimes have already occurred, it is indeed hard to construct an omission. However, viewing the omission with regard to future crimes that might be committed as a result of it, a “hypothetical” or “quasi”-causation can indeed be construed. Unsurprisingly, the separate opinion rejects the view that the commander’s omission can increase the risk of the commission of crimes, as it was held by the Judge Steiner in her Separate Opinion to the conviction decision (Separate Opinion Steiner, para. 19). This rejection seems to be based on a misunderstanding of that view when Judges van den Wyngaert and Morrison opine that the failure of the commander to act does not increase the risk but “[t]he responsibility of the commander is precisely to decrease the risk that his/her subordinates will commit crimes. Failing to reduce a risk can hardly be seen as causing the manifestation of said risk.” However, the theory of increasing risk (see esp. Ambos, in: Stahn (ed.), The Law and Practice of the ICC, 2015, pp. 603 et seq.) is exactly about the question, whether the ex ante formulated norm, even when viewed ex post, still appears to contain a prohibition that reduces the actual harm. Thus, of course the norm in itself is about harm reduction, while the act might increase it. The minority view – by contrast – takes this into account and justifies the causation element in convincing fashion.

In sum, in a very narrow decision, the majority of the Chamber did what it is supposed to do – review a Trial Chamber decision on the basis of the law. It goes without saying that it would be inappropriate to applaud this decision as a victory of criminal doctrine and procedural sophistication. There are no winners. Thousands of victims will have been left in shock by the Appeals Chamber judgment. However, it would also be inappropriate to bemoan the decay of everything the ICC stands for. Not sacrificing the rights of the accused on an altar of grand gestures by the world community (such as the “need to send a clear signal globally that such abhorrent crimes must not go unpunished” [ICC Prosecutor] or a “promise of accountability” [Amann]) is certainly a decision that should find – despite its controversy – support.