At the heart of this process lies a fundamental procedural mismatch. According to public reporting, the OIOS was mandated to conduct fact-finding, not adjudication. The ad hoc Judicial Panel was tasked with legally characterising OIOS’ findings, but had no authority to address their deficiencies considering the seriousness of this matter, including the failure to assess witness credibility or resolve material narrative inconsistencies.
Association of International Criminal Law Prosecutors (AICLP)
No, I am not suggesting that there is an ongoing cover-up in L’Affaire Khan, nor that there is institutional indifference at the International Criminal Court (ICC). Nor am I attributing the process for handling allegations of sexual assault, witness interference, and retaliatory conduct attributed to ICC Prosecutor Karim A. A. Khan KC to incompetence on the part of the Bureau (all laid out in the Rules of Procedure and Evidence (RPE) and the Staff Regulations, save for the standard of proof issue).
But as a postscript to my last post, in which I suggested that the ICC finds itself at an inflection point, this episode underscores something more structural. If the ICC is to maintain credibility as a judicial institution, it must ensure that its internal processes are not only fair but also seen to be fair, particularly when allegations implicate those at the highest levels. That, in turn, requires confronting an uncomfortable possibility: that investigative cultures, whether at the ICC or within the broader UN system, may at times reflect institutional instincts for protection rather than accountability.
Against that backdrop, recent contributions from former insiders merit attention. Two former OIOS investigators, Messrs. Mostafa Hassan and Peter Gallo, have set out, in strikingly candid terms, their experiences with the investigative process. What emerges is not merely a critique of isolated failings but a broader concern about competence and rigor, and, at times, an institutional predisposition that risks privileging the protection of the office over the protection of complainants.
This is not entirely surprising. What is striking is the extent – the depth and apparent normalization – of these shortcomings. Their accounts raise serious questions about the reliability of investigative outputs and, by extension, the evidentiary value of the resulting files. If the foundation is uncertain, any conclusions drawn from it – whether measured against a “clear and convincing” standard or “beyond reasonable doubt”– are necessarily constrained.
This leads to a more fundamental point. Investigators are, by design, fact-gatherers, not fact-finders. That function belongs to judges. When investigations are incomplete, contested, or structurally limited (as they often are in cases of this nature), the more reliable course is not to treat the investigative file as dispositive but to subject the evidence to testing. Not on paper, but through an adversarial process.
Let me give an example of how an approved, adversarial process could work.
For individuals occupying positions beyond ordinary staff, the process could be recalibrated to reflect a more judicial character. One option would be to vest a judge with broader investigative authority, enabling a more structured and coherent development of the evidentiary record. Consideration might also be given to applying differentiated standards of proof, particularly where the alleged conduct rises to the level of criminality. At a minimum, the parties should be afforded the opportunity to comment meaningfully on the evidence before any decision is reached. But even a more structured, quasi-inquisitorial model has its limits. From the perspective of an adversarial lawyer, the essential safeguard remains the ability to confront witnesses. Without that, the process risks reducing judges to the role of investigative actors, assembling a dossier rather than testing its contents. And once that line is crossed, a further step becomes necessary: a separate, independent panel tasked with conducting hearings in which the evidence is properly examined. Hence, however refined the investigative phase may be, it cannot substitute for adversarial scrutiny.
If I understand Hassan correctly, his position is that, since a panel of independent judges has already reviewed the file (such as it is), the Assembly of States Parties (ASP) should rely on the panel’s report rather than the underlying investigation, which it would be doing were it to conduct the disciplinary proceedings or to determine the outcome, as opposed to binding itself to the panel’s report. There is force in that view where the universe of options is limited to these two imperfect choices. But it rests on a constrained premise that these are the only available options. They are not.
Indeed, the difficulty is more immediate. Judges cannot meaningfully assess allegations of this gravity – whether framed in terms of criminal conduct or serious professional misconduct – if they are confined to the four corners of an investigative file, without the ability to hear witnesses, assess credibility, or resolve inconsistencies. To ask them to do so is to ask them to approximate adjudication without the tools of adjudication – using them as a fig-leaf to give the appearance of a judicially independent review.
If, as Hassan and Gallo suggest, investigative processes may themselves be vulnerable to incompleteness or distortion (not necessarily in this case, but as a general proposition), then reliance on such a record, without more, raises obvious concerns. And if the choice is between accepting an untested investigative file or deferring to a panel constrained by that same file, then the latter may indeed be the better course. But it is still a second-best solution.
Nor am I persuaded that the Bureau or the ASP are bound by the panel’s conclusions. The text of Rule 26bis of the ICC Rules of Procedure and Evidence makes that clear:
Rule 26bis
-
-
- The ad hoc panel shall provide its non-binding advice to the competent decisionmaker(s), as set out in articles 46 and 47, and rules 29 and 30, on legal characterization whether the factual findings by the Independent Oversight Mechanism or the external investigator, as applicable, establish serious misconduct or serious breach of duty, as defined under rule 24, or misconduct of a less serious nature, as defined under rule 25, or establish no misconduct or breach of duty under either rule 24 or rule 25.
-
-
-
- Following the transmission of the ad hoc panel’s final report under sub-rule 4 above, the competent decision-maker may disagree with the ad hoc panel’s advice and the procedure in rule 29, sub-rule 7, or rule 30, sub-rule 5, shall apply. (emphasis added)
-
The Rule frames the panel’s role as advisory, not determinative. It leaves discretion to the decision-makers regarding how to proceed. Properly understood, the issue is not whether the ASP can depart from the panel’s advice, for it plainly can, but how it should exercise that discretion to preserve institutional legitimacy.
There comes a point when the distinction between misconduct and fitness for office ceases to sound like a legal abstraction and begins to carry real institutional weight. Even if the allegations, narrowly framed, fail to meet the relevant evidentiary threshold, the ASP is still confronted with a more awkward question: whether Prosecutor Khan retains the confidence necessary to lead the OTP.
That is not a disciplinary inquiry. It does not hinge on proof beyond a reasonable doubt or on parsing regulatory breaches. It is, in essence, a question of institutional confidence; one that sits adjacent to, but ultimately outside, the ICC’s formal legal framework. It may be uncomfortable. It may be contested. But it is difficult to argue that it falls outside the ASP’s authority.
Politics inevitably complicates matters. The contours of disagreement among States Parties are already emerging, with a discernible North–South undertone. There are suggestions – sometimes thinly veiled – that certain European states are pressing too hard, perhaps to appease a Trump administration that has openly taken umbrage at the ICC’s arrest warrants for the Israeli Prime Minister and former Minister of Defense. The not-so-subtle counter-narrative is that Khan, having shown resolve in pursuing accountability in Gaza, is now being targeted for it, paired, somewhat paradoxically, with claims of his indispensability, as though the OTP’s functioning were uniquely contingent on his person.
At the same time, the discourse has been muddied by more speculative claims: that the allegations themselves are the product of a coordinated effort by foreign intelligence services. These claims remain unproven. But in the current climate, proof is almost beside the point; repetition alone is enough to shape perception.
Inside the OTP, the mood is hardly more settled. There are calls–some discreet, some less so–for Khan’s departure. Others push back, invoking what they call the “silent majority”—a phrase borrowed, perhaps a little too knowingly, from John le Carré’s The Tailor of Panama, where narratives have a way of acquiring a life of their own. Whether silent or simply strategic, the effect is the same: an office divided against itself.
The cumulative damage is not trivial. The ICC’s standing has taken a hit. The OTP risks appearing less like a prosecutorial authority and more like a contested political arena. The overarching result is an institution under visible strain: credibility dented, authority questioned, and internal cohesion fraying.
This inevitably brings the question back to the ASP. Having elected Khan to a nine-year term in one of the most consequential roles at the ICC and in international criminal justice, does the ASP not also retain the authority to revisit that choice if circumstances so demand?
It should. And if, after deliberation, a two-thirds majority of the 125 States Parties were to conclude that a change in leadership is necessary, would that decision not carry its own legitimacy? Not as a finding of wrongdoing, but as a judgment about the institution’s future.
The distinction matters. But so does the willingness to act on it.
Returning to the investigative question, Gallo’s observations sharpen the point. His account – measured yet unflinching – highlights both the structural limits of OIOS and the risks of overstating what its work can establish. As he puts it:
In reality OIOS is never required to adjudicate; that is left to the UN Office of Human Resource Management or other decision-maker. The major difference here appears to be that the 3-judge panel was more diligent in reviewing the report, recognizing that unresolved inherent contradictions and other shortcomings in it were such that it failed to meet even the lower “clear and convincing” standard that is normally required.
The 3-judge panel was however limited by their Terms of Reference and – faced with the binary question of whether the investigation delivered evidence ‘beyond reasonable doubt’ – they reached the only conclusion they were permitted to reach; it did not.
That is being hailed, by Khan’s legal counsel and other acolytes, as being grounds for the dismissal of the charges against him – but the investigation did not establish that he was entirely innocent. On the contrary, it appears to have established (albeit to a slightly lower standard of proof) that:
-
-
-
- that Khan did engage in some degree of “non-consensual sexual contact”,
- that a number of ICC staff members pressured the victim to withdraw her complaint, and
- that Khan retaliated against two officials who first reported the victims allegations.
-
-
We are therefore left with [sic] not one but three elephants in the room, because.
-
-
-
- Any “non-consensual sexual contact” is, by definition, a criminal offence. It is also a tort.
- Pressuring ICC staff members to withdraw a misconduct complaint constitutes ‘harassment’ and is a clear violation of section 2(c) of the ICC Admin Instruction on ‘Addressing Discrimination, Harassment, including Sexual Harassment, and Abuse of Authority‘ (ICC/AI/2022/003 dated 6 April 2022); and
- Retaliating against ICC staff members who reported misconduct is violation of section 4.1(p) of the ICC Admin Instruction on ‘Unsatisfactory Conduct and Disciplinary Proceedings‘ (ICC/AI/2022/002 dated 14 March 2022)
-
-
This account appears to align closely with the two-page summary of the investigative report as described by The Guardian. Referring to it as the “watchdog’s final report,” the newspaper quotes the investigation as finding that “evidence indicated” Khan’s conduct toward the woman had “escalated over time, resulting in him engaging in nonconsensual sexual contact with her.” It also reports that during interviews with investigators, Khan “would not confirm” whether he had a sexual relationship with the staffer, and that the investigation concluded it had “identified evidence establishing a factual basis” for the alleged victim’s claims against Khan. It recommended the ASP “consider appropriate action based on the evidence.” According to the document, Khan told investigators he had “never engaged in any prohibited conduct” in relation to the woman. His lawyers have said the summary – which has been circulated to ICC member states – was “not accepted by the prosecutor as fair or accurate.”
The panel’s report, also seen by The Guardian, sits uneasily alongside this. It ‘unanimously concluded that the UN watchdog’s findings “do not establish misconduct or a breach of duty under the relevant legal framework”’, while at the same time acknowledging that the inquiry had failed to establish “where the truth lies,” leaving “many unresolved factual disputes” and rendering any legal assessment “almost destined for fruitlessness.” The panel ultimately stated that it could not make “a definitive proclamation on the existence or absence of the alleged misconduct” and, constrained by its mandate, found only that the evidence did not meet the “beyond reasonable doubt” threshold.
Assuming the accuracy of these excerpts, the picture that emerges is not one of resolution but of limitation. The investigation appears to have produced material pointing in one direction; the panel, lacking the means to test it, could not meet the evidentiary threshold required for legal findings. That is not exoneration. But neither is it proof.
More fundamentally, it reflects a flawed process. A system that gathers evidence but does not test it, and that asks judges to characterize findings they cannot probe, should not be expected to yield definitive answers. On that point, The Guardian’s reporting appears entirely consistent with Gallo’s observations about where the panel’s report has left us and with the broader concern I and others have raised: without adversarial scrutiny by judges who hear testimony and confrontation by the parties before assessing the evidence (even if gathered under the direction of an investigative judge, as I’ve noted above), serious allegations risk being left in a state of suspended conclusion.
All of this brings the issues into sharper focus, providing the linchpin for this postscript’s thesis.
As much as L’Affaire Khan has cast a shadow over the ICC – particularly at a moment of external pressure – it also presents an opportunity. The process, imperfect and wanting as it seems, has exposed what is too often obscured: the limits of internal investigative mechanisms when confronted with allegations of this gravity.
The ICC has long been aware that sexual harassment and abuse of authority have existed, at least within parts of the OTP. The Independent Expert Review of 2020, commissioned by the ASP, said as much. Whether this case reflects a systemic problem or an isolated failure remains to be seen. But what is increasingly difficult to maintain is that the current framework is sufficient.
What is required is not simply investigation, but adjudication: a process that allows evidence to be tested in a structured, adversarial setting. That, in turn, points to a more durable solution: the appointment of an independent ad hoc counsel, vested with both investigative authority and the resources to conduct a thorough inquiry, and tasked with presenting the evidence before a panel of independent judges.
Such a process would serve multiple purposes. It would insulate the proceedings from institutional pressures. With the proper modalities, it could ensure that both the complainant and the accused receive full procedural protections. And, critically, it would enhance public confidence that allegations of this seriousness are neither minimized nor managed, but properly examined.
Simply, disciplinary proceedings should be adversarial, transparent where possible, and reasoned in their outcome. In this instance, that would not be possible; it would require changing the RPE and staff Regulations, possibly the Statute, and moving the goalposts after the event.
As an aside, I pose a question cautiously, mainly out of curiosity: Given that some of the alleged conduct reportedly occurred in the Netherlands and, if established, would constitute serious criminal offenses, why was the matter not reported to the Dutch prosecuting authorities? Diplomatic immunity would not appear, at least on its face, to be an insurmountable barrier. I may be mistaken, but the question seems worth asking.
And Gallo notes, the complainant may yet pursue civil remedies irrespective of the outcome, though the prospect of such litigation is costly, time-consuming, exhausting, and the financial rewards are uncertain.
In sum, at the outset, the problems in L’Affaire Khan present themselves in stark procedural terms: a system designed to find facts, not to test them; to characterize conclusions, not to interrogate their foundations. The OIOS was tasked with fact-finding, not adjudication. The ad hoc Judicial Panel was asked to assign legal meaning to those findings, yet was constrained from addressing their evident deficiencies, most notably the inability to assess credibility or resolve material inconsistencies in the narrative. That mismatch is not merely a technical flaw. It goes to the integrity of the process itself.
In the end, however, the issue extends well beyond the confines of this case. What the ICC now faces may not be an existential crisis, but it is unmistakably an inflection point – one that offers a rare and consequential opportunity to recalibrate its internal processes and reaffirm the standards it seeks to project. The procedural limitations revealed at the outset are not isolated; they are symptomatic of a framework that, when tested under pressure, risks producing outcomes without assurance that they were earned.
The ICC’s credibility has always rested on its willingness and ability to hold others accountable. But once credibility is tested, it demands symmetry. It requires that the same rigor, transparency, and discipline be applied when scrutiny turns inward. No one in a position of power should be above scrutiny, above the law, or acting with impunity. It should not matter who the office holder is, how politically connected they may be, how valuable to the institution, or how important their past contributions. The same rules should apply to everyone.
This moment, then, is not simply about resolving a set of allegations. It is about demonstrating that accountability is not selective, that process is not contingent, and that institutional integrity is not negotiable. Regardless of how L’Affaire Khan is ultimately resolved, the greater test lies in what follows: whether the ASP seizes the opportunity to establish a framework that is not only fair in outcome but also unimpeachable in design.
Institutions that demand accountability cannot afford ambiguity when the lens turns inward.
About Author
