The Standard of Proof in ICC Disciplinary Proceedings

For obvious reasons, I am not going to comment on the substance of the investigation or any of the recent media commentary on it. In this post, I simply want to address a narrow legal issue that is drawing attention on social media: what is the standard of proof that applies in disciplinary proceedings at the ICC? The answer is clear and unequivocal: proof beyond a reasonable doubt.

There seems to be significant confusion over this point. Ken Roth has tweeted, for example, that “[o]ne question the judges will have to answer in clearing International Criminal Court chief prosecutor Karim Khan of sexual misconduct allegations is why they applied ‘beyond a reasonable doubt.’ That’s a criminal standard, not usually for workplace issues.”

The key to the answer is that the ICC is subject to the jurisdiction of the International Labour Organisation Administrative Tribunal (ILOAT), not the United Nations Appeal Tribunal (UNAT). That jurisdiction is not controversial, as indicated by the Court’s formal response to the Independent Expert Review’s recommendation, R120, that the Court should move from ILOAT to UNAT:

In light of recommendation R120, the Court will produce an assessment detailing the costs that moving from the ILOAT to the UNAT would have for the Court, as well as the legal, policy and practical considerations. The Court notes for instance the resort to ILOAT is specifically enshrined in the Staff Regulations of the Court, and any amendment to those regulations would have to be sanctioned by the ASP. A thorough assessment will therefore be of assistance in fully analysing the proposed recommendation.

The ICC has yet to act on recommendation R120, although in 2022 it produced a Report of the Court on the Internal Justice System (IJS Report) that compared the pros and cons of making the switch. The IJS Report discusses a number of differences between ILOAT and UNAT. One of the key differences the Court identified is the “standard of proof in disciplinary cases.” Here are the key paragraphs of the IJS Report, worth quoting in full:

46. The ILOAT requires international organizations “to prove allegations of misconduct beyond a reasonable doubt before a disciplinary sanction can be imposed”. This is the highest standard of proof, and it has been pointed out that the same formulation is used in certain national legal systems to establish the burden of proof in criminal proceedings.

47. The ILOAT has held that, “[f]or a finding of misconduct to withstand scrutiny, each of the elements of the alleged misconduct must be proved beyond a reasonable doubt.” Furthermore,

[t]he standard of beyond reasonable doubt concerns both the finding of specific facts and the overall level of satisfaction that the case against the staff member has been made out. In relation to the proof of any essential relevant fact, the person or body charged with the task of assessing the evidence and making a decision in the context of determining disciplinary proceedings must be satisfied beyond reasonable doubt that a particular fact exists.

[snip]

50. In contrast, the UNAT has expressly rejected the ILOAT’s “beyond a reasonable doubt” standard and instead applies a standard of “clear and convincing evidence”:

We will not follow the ILOAT in holding that the standard of proof in disciplinary cases is beyond a reasonable doubt. While it is correct that beyond a reasonable doubt is the standard at the ILOAT, this has never been the standard at the United Nations. In disciplinary cases we have required that when a disciplinary sanction is imposed by the Administration, “the role of the Tribunal is to examine whether the facts on which the sanction is based have been established, whether the established facts qualify as misconduct, and whether the sanction is proportionate to the offence”. But we have not as yet set an exact standard for the quantum of proof required. Disciplinary cases are not criminal. Liberty is not at stake. But when termination might be the result, we should require sufficient proof. We hold that, when termination is a possible outcome, misconduct must be established by clear and convincing evidence.

Whatever might eventually happen, the ICC currently remains subject to ILOAT. To impose a sanction on a member of staff or elected official in a disciplinary proceeding, therefore, “each of the elements of the alleged misconduct must be proved beyond a reasonable doubt.”

Danya Chaikel has suggested on Twitter that ILOAT does not apply the “beyond a reasonable doubt” standard in harassment cases. In her view, “ILOAT jurisprudence makes clear that assessing a harassment complaint solely through the lens of possible disciplinary proceedings is an ‘error of law.’ It requires a distinct assessment, applying a less onerous standard than BRD.”

This is not correct, and Chaikel’s statement concerning “possible disciplinary proceedings” (my emphasis) explains why: ILOAT specifically distinguishes between harassment claims and misconduct investigations based on harassment allegations. Here, for example, is what ILOAT said in Judgment No. 4207 concerning the IAEA’s internal justice system:

14. A claim of harassment and a report of misconduct based on an allegation of harassment are distinct and separate matters. A claim of harassment is a claim addressed to the organization the resolution of which only involves two parties, the organization and the reporter of the harassment. In contrast, a report of alleged misconduct, based on an allegation of harassment, triggers the Appendix G procedures, a process that is directed at the culpability of the staff member in question and potentially the imposition of a disciplinary measure. In this process, the two parties are the organization and the staff member in question. In this process, the reporter of the misconduct, a potential victim of the harassment, is a witness and not a party in the proceedings.

Judgment No. 4207 involved a woman claiming damages from the IAEA for sexual harassment, not reporting sexual harassment to initiate a disciplinary investigation of her harasser. Because she was not asking for a disciplinary investigation, ILOAT held that the IAEA should have applied a “less onerous standard” than “beyond a reasonable doubt” to her harassment claim (emphasis added):

20. Having regard to the distinction mentioned in consideration 14, above, between a claim of harassment and a report of misconduct based on an allegation of harassment, the DDG-MT’s decision concerning the complainant’s claim of harassment is fundamentally flawed. The DDGMT proceeded on the assumption that an allegation of harassment by the aggrieved staff member must not only be borne out by specific acts, the burden of proof being on the reporter of the harassment, but must also prove that the alleged perpetrator of the harassment acted with intent. This in turn resulted in the DDG-MT incorrectly applying the “beyond a reasonable doubt” standard of proof in his consideration of the complainant’s claim of harassment. It is noted that the Tribunal has specifically rejected this assumption that intent on the part of the alleged perpetrator is required in order to establish harassment (see, for example, Judgments 2524, consideration 25, 3233, consideration 6, and 3692, consideration 18, and the case law cited therein). The Tribunal’s case law states that the applicable standard of proof for a finding of harassment in a case such as this is not “beyond a reasonable doubt” but a less onerous standard (see Judgment 3725, consideration 14)

ILOAT thus ordered the IAEA to pay the woman “an award of moral damages in the amount of 25,000 euros for the harassment, the inexplicable, unreasonable delay and the significant harm she suffered due to this delay.”

Chaikel doesn’t mention Judgment No. 4207 in her tweet, but she does cite three other ILOAT judgments: No. 5023, No. 4839, and No. 4820. None of those cases, however, involved a disciplinary proceeding initiated by “a report of misconduct based on an allegation of harassment.” All three involved individuals making a “claim of harassment” and seeking monetary damages from the international organisation that employed them. Indeed, ILOAT specifically noted the different standards of proof that applied to the two types of proceedings in Judgment No. 4389, which involved the International Organization for Migration:

6. IOM’s legal framework does not specify the applicable standard of proof for a finding of harassment. Regarding this point, the Tribunal’s case law states that, while the standard of proof required to impose disciplinary measures on an individual charged with misconduct is that of “beyond a reasonable doubt”, the applicable standard of proof for a finding of harassment is a less onerous standard.

Finally, it is worth noting two ILOAT judgments that make clear that an international organisation cannot impose a disciplinary measure on an individual accused of harassment unless it proves the allegations beyond a reasonable doubt. In Judgment No. 4047, an employee of the European Patent Organisation (EPO) challenged “the decision to impose on her with immediate effect the disciplinary measure of dismissal for serious misconduct.” The misconduct in question was, inter alia, “a harassment campaign directed against an elected member of the Central Staff Committee.” ILOAT began by clarifying that the applicable standard of proof was “beyond a reasonable doubt,” despite that standard normally applying only in criminal cases:

6. Overall, the case law of the Tribunal is clear and consistent. It was recently referred to in Judgment 3863, consideration 8 (see, also, Judgment 3882, consideration 14, as another recent example), in which the Tribunal said: “[A]ccording to the well-settled case law of the Tribunal, the burden of proof rests on an organisation to prove allegations of misconduct beyond a reasonable doubt before a disciplinary sanction can be imposed (see, for example, Judgment 3649, consideration 14). It is equally well settled that the ‘Tribunal will not engage in a determination as to whether the burden of proof has been met, instead, the Tribunal will review the evidence to determine whether a finding of guilt beyond a reasonable doubt could properly have been made by the primary trier of fact’ (see Judgment 2699, consideration 9).” It is legally irrelevant, for the purposes of the Tribunal’s judicial determination of the complaint, that, as the EPO points out in the reply, the same formulation is used in the English common law to establish the standard of proof in criminal proceedings.

The Tribunal then said more specifically that, because the employee denied that she had engaged in harassment, EPO could not lawfully terminate her unless it proved she had engaged in the harassment beyond a reasonable doubt:

8. While this denial may not apply to each and every aspect of the charges against the complainant, it does make clear that she broadly put in issue both the fact that she engaged in the alleged conduct and the motivation attributed to her for her conduct. This position adopted by the complainant imposed an obligation on the EPO to assess the material by applying the test of beyond reasonable doubt.

The second relevant ILOAT judgment, No. 5119 — issued just last month — involved the International Telecommunication Union (ITU) firing a staff member for sexual harassment and abuse of power after a disciplinary investigation. ILOAT began (para. 5) by reiterating that “beyond a reasonable doubt” was the applicable standard of proof:

With respect to the burden of proof applicable in disciplinary proceedings, it is furthermore recognized that “the burden of proof rests on an organisation to prove the allegations of misconduct beyond reasonable doubt before a disciplinary sanction can be imposed” (see, for example, Judgment 4749, consideration 5). In Judgment 4362, considerations 8 and 10, the Tribunal relevantly stated the following in this respect:

“8. The standard of proof of beyond reasonable doubt does not exist to create an insuperable barrier for organisations to successfully prosecute disciplinary proceedings against staff members. Indeed it should not have that effect. What is required is discussed in many judgments of the Tribunal. Rather the standard involves the recognition that often disciplinary proceedings can have severe consequences for the affected staff member, including dismissal and potentially serious adverse consequences on the reputation of the staff member and her or his career as an international civil servant, and in these circumstances it is appropriate to require a high level of satisfaction on the part of the organisation that the disciplinary measure is justified because the misconduct has been proved. The likelihood of misconduct having occurred is insufficient and does not afford appropriate protection to international civil servants. It is fundamentally unproductive to say, critically, this standard is the ‘criminal’ standard in some domestic legal systems and a more appropriate standard is the ‘civil’ standard in the same systems involving the assessment of evidence and proof on the balance of probabilities. The standard of beyond reasonable doubt derived from the Tribunal’s case law as it has evolved over the decades, serves a purpose peculiar to the law of the international civil service.”

The Tribunal then set aside the staff member’s termination, holding that it had to return the case to the ITU because the organisation’s internal appeal body had failed to make clear that its decision upholding termination was based on the “beyond a reasonable doubt” standard and not on the lesser standard that applied during internal ITU investigations:

29. An internal appeal body has a duty to address pleas of substance (see, for example, Judgments 4534, consideration 17, 4169, consideration 5, and 4063, consideration 5). Yet, the Appeal Board report shows that while it apparently identified what it considered as being potential issues on the question of the alleged conflict of interest, it did not resolve these issues as part of its remarks. Similarly, regarding the evidence and the burden of proof, the Appeal Board noted the applicable standard of preponderance of evidence under the ITU Investigation Guidelines (Service Order No. 19/10) and the beyond reasonable doubt standard applied by the Disciplinary Chamber, and even expressed having some doubts on the evidence from both sides. Still, it refrained from explaining what to conclude from these different standards and what these doubts were and amounted to, or from resolving any discrepancies that may have existed in its mind.

31. This leaves the Tribunal with no other alternative but to annul the impugned decision and return the matter to ITU so that the internal appeal process be redone and a proper report be issued for consideration by the Secretary-General. While the complainant argued that in the event of such a finding, the Tribunal should not return the matter to the organisation and rather do itself the necessary analysis of the investigation and Disciplinary Chamber reports to resolve and decide whether the burden of proof was satisfied and the sanction was proportionate, the Tribunal considers that this would be ill-advised in the circumstances of this case.

Personally, I agree with Ken and Danya that “beyond a reasonable doubt” is too onerous a standard to use in a non-criminal context. But our opinion is irrelevant. The ICC is subject to the jurisdiction of ILOAT and ILOAT jurisprudence is unequivocal: an international organisation like the Court cannot impose a disciplinary sanction on someone unless it proves every element of the alleged misconduct beyond a reasonable doubt.