Making Sense of Command Responsibility
In this fourth opinion piece on the Bemba Appeals Judgment (find the previous ones here, here and here), Amnesty’s International Justice Team considers the majority’s controversial findings on command responsibility in the context of the Trial Chamber’s problematic application of Article 28.
Like a number of other commentators, we find problematic the explanations provided by the majority of the Appeals Chamber for two putative errors it emphasised in the Trial Chamber’s analysis of whether Jean Pierre Bemba had “failed to take all necessary and reasonable measures in response to MLC crimes in the CAR”.
Firstly, in our view, the dissenting Judges and other commentators are right to point out that the majority’s assessment that the Trial Chamber’s “preoccupation with Mr Bemba’s motivations appears to have coloured its entire assessment of the measures that he took” is not clearly supported by the Trial Judgment. Moreover, evidence of a commander’s motivations can be relevant in assessing the reasonableness of certain measures that have been taken. Secondly, as we will examine in more detail in our next opinion piece on Wednesday, the majority’s finding that the Trial Chamber “paid insufficient attention to the fact that the MLC troops were operating in a foreign country with the attendant difficulties on Mr Bemba’s ability, as a remote commander, to take measures” raises a number of issues that are not adequately addressed in the Appeals Judgment.
Of equal concern, other serious problems with the Trial Chamber’s analysis of whether Jean Pierre Bemba had failed to take all necessary and reasonable measures were only partly addressed by the Appeals Chamber. These should also be considered in the post-Judgment analysis.
The Trial Chamber’s failure to address alleged omissions in the charges
A plain reading of Article 28 clearly establishes command responsibility as a mode of liability to address specific omissions by commanders to (1) prevent crimes under the Statute; or (2) repress crimes under the Statute; or (3) submit allegations of such crimes to competent authorities for investigation and prosecution. The Office of the Prosecutor set out specific allegations of such omissions in the Amended Document Containing the Charges, although arguably they could have been more detailed. However the Trial Chamber did not address these alleged omissions specifically. In an adversarial criminal trial, these surely should have been the primary focus of its analysis.
Instead the Trial Chamber started by conducting a critical analysis of measures that it recognized had been taken (most of which related to repressing and submitting crimes to competent authorities) rejecting them all as less than what was “necessary and reasonable” in the circumstances. It then proceeded to list a number of very general measures that Mr Bemba “could have” taken that did not fully address (and in some instances went beyond) the allegations in the Amended Document Containing the Charges. The Trial Chamber further failed to clearly demonstrate that the prosecution had proved each of these omissions beyond a reasonable doubt.
To its credit, the majority of the Appeals Chamber picked up on this issue and strongly rejected the Trial Chamber’s approach. It stated:
Simply juxtaposing the fact that certain crimes were committed by the subordinates of a commander with a list of measures which the commander could hypothetically have taken does not, in and of itself, show that the commander acted unreasonably at the time. 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. Indeed, it is for the trial chamber to demonstrate in its reasoning that the commander did not take specific and concrete measures that were available to him or her and which a reasonably diligent commander in comparable circumstances would have taken. It is not the responsibility of the accused to show that the measures he or she did take were sufficient.
The majority clarified that an accused person must be informed of the factual allegations on the basis of which the Prosecutor seeks to establish their failure to take all necessary and reasonable measures. However, the Trial Chamber’s reliance on apparently hypothetical failures receives little prominence in the Appeals Chamber’s concluding analysis and decision to overturn the conviction (paras 189-194) compared to the two “errors” relating to Mr Bemba’s motivations and remote location. The majority may have been on stronger ground if they had relied on this.
The Trial Chamber’s conflation of a commander’s duties
Whilst there appears to be general agreement that a breach of any of the three duties to take “all necessary and reasonable measures” in Article 28 may attract criminal liability, all of the chambers involved in this case appear to conflate or confuse them. In our view the distinction between the duties is relatively simple, although they are of course closely related and in many instances overlap:
- The duty to prevent relates to measures taken to prevent crimes that have not yet occurred, such as training troops to respect human rights and international humanitarian law, as well as ordering them not to commit crimes.
- The duty to repress relates to measures to stop the commission of further crimes when crimes are being or have been committed, including issuing orders to stop crimes, disciplining those responsible or withdrawing them from positions or locations where they are able to commit further criminal acts.
- The duty to submit the matter to the competent authorities for investigation and prosecution relates specifically to taking measures to ensure effective criminal accountability for crimes that have been committed.
Regrettably, both the Pre-Trial Chamber and the Trial Chamber appeared to conflate the duty to repress with the duty to submit the matter to the competent authorities for investigation and prosecution. The Pre-Trial Chamber in the Confirmation of Charges decision even appears to suggest that they are alternative, rather than distinct obligations, stating that submitting the matter to competent authorities “remedies a situation where commanders do not have the ability to sanction their forces”. However, impunity often exists because militaries and armed groups get away with conducting insufficient or sham disciplinary proceedings instead of genuine criminal trials. Such processes often charge persons suspected of horrific violence with minor disciplinary offences such as “disobeying orders” and as a result fail to ensure accountability for serious crimes, including in particular sexual and gender based violence. For the ICC, which was created with the aim of putting an end to impunity for the most serious crimes of concern to the international community, to suggest that such measures to repress crimes are sufficient without submitting serious criminal conduct to competent authorities is deeply problematic. Instead of the addressing this flawed interpretation of Article 28, the Appeals Chamber in this case appears to give significant weight to commissions and court-martials established by the MLC that in most cases were not mandated to charge murder and sexual violence.
Moreover, none of the Chambers in this case appear to have considered the argument in Amnesty International’s 2009 amicus curiae brief to the Pre-Trial Chamber that mechanisms established by non-state actors cannot normally be considered as “competent authorities”. In our view, a commander who submits alleged crimes to such mechanisms has not taken reasonable and necessary measures to submit allegations of crimes to competent authorities in accordance with Article 28. It is concerning that, as we have seen in other contexts, the ICC seems willing to accept that any type of justice effort at the domestic level – no matter how flawed it is – will suffice.
Failure to address specific allegations of failures to prevent crimes
While the OTP alleged in the Amended Document Containing the Charges that Jean Pierre Bemba was responsible for inadequate training of MLC troops in international humanitarian law and failed to issue clear and efficient orders prior to, and during, the operation to ensure that the crimes were not committed by the MLC troops, these allegations of failures to prevent crimes were not clearly addressed by the Trial Chamber. Since the focus of its examination of the “necessary and reasonable measures” test in Article 28 focussed on measures that had been taken, the Judgment only lists hypothetical prevention measures that the accused could have taken. However, in considering separately the question of causation, the Trial Chamber found that Jean Pierre Bemba:
failed to take any measures to remedy. . . deficiencies in training, either prior to deployment of the troops or in response to the consistent reports of crimes occurring from the earliest days of the 2002-2003 CAR Operation.
This failure to expressly consider and determine alleged failures to take measures to prevent crimes was not identified in the majority’s Appeal Judgment. As a result, a clear determination of whether Jean Pierre Bemba failed to take all necessary and reasonable measures to prevent the crimes committed by the MLC in the CAR is missing in both the Trial and the Appeal Judgment.
Opaque application of causation
Although the question of causation is technically separate to the Appeals Chamber’s consideration of the ground of appeal of whether Mr Bemba took all necessary and reasonable measures, the majority appeared to criticise the Trial Chamber for not linking “Mr Bemba’s putative failure to take adequate measures to any of the specific acts… which he was ultimately convicted of.” Unfortunately, this only further muddies the waters on the long debated question of whether and how causation should be applied to the definition of command responsibility in Article 28.
Amnesty International argued in its amicus curiae brief that such a requirement was not necessary given that the duty to repress and submit crimes to competent authorities arises after crimes have been committed. Applying the same logic, the Pre-Trial Chamber reached the reasonable conclusion that “causality only relates to the commander’s duty to prevent the commission of future crimes”. It also held that, as command responsibility focuses on omissions, it is only necessary to prove that the commander’s omission increased the risk of the commission of the crimes charged in order to hold them criminally accountable under article 28(a).
The Trial Chamber endorsed the increased risk test but did not expressly concur with its conclusion that causation only extends to the duty to prevent. Confusingly the Judgment noted instead the “core principle of criminal law” that “a person should not be found individually criminally responsible for a crime in the absence of some form of personal nexus to it”.
In this context, the majority of the Appeal Chamber’s statement, read together with Judges Van den Wyngaert and Morrison’s acknowledgment that the judges in the Appeals Chamber were divided on this issue and Judge Osuji-Eboe’s detailed position on causation under Article 28 in their separate opinions creates further uncertainty on this key issue.
At the end of the first trial applying the unique definition of command responsibility in Article 28, the ICC’s interpretation and application of this provision is more confused than ever. In our view, the Trial Chamber misapplied the test of whether a commander failed to take all necessary and reasonable measures and the Appeals Chamber failed to fully identify and resolve the problems. In fact, it further muddied the waters by emphasising two controversial reasons in its decision to overturn the conviction. There must be some allowance for ICC Judges to work through their interpretation and application of the unique provisions of the Rome Statute. However, the ICC must also be conscious of the potential damage caused to its credibility when it is unable to interpret and apply key provisions of the Statute in a convincing manner.