Governance as Legitimacy: A Further Reflection on the Removal of the ICC Prosecutor (Part II)

[Ezequiel Jimenez Martinez is an Adjunct Professor at the Pompeu Fabra Law Faculty, and a Senior Fellow at the Center for International Law Research and Policy. He is the author of Governing the International Criminal Court: the History and Practice of the Assembly of States Parties to the Rome Statute (Brill, 2025).

Sergey Vasiliev is a Professor of International Law at the Open University of the Netherlands and co-convened, with Prof. Niels Blokker (Leiden), the project on the governance of international courts and tribunals.]

Part I located the emergent literature studying the governance capabilities of international criminal tribunals to show how, in the context of the International Criminal Court, the actions taken by the Assembly in relation to the Office of the Prosecutor had a tangible effect on its operations.

The Removal Act

Shany and Varaki defend the outcome of the Khan case with reference to the ASP plenary’s acceptance of the Bureau’s interpretative position: the panel of judicial experts’ conclusion was merely advisory, Khan’s right to be heard had been exhausted before the July special session, the Bureau was competent to qualify the conduct under Article 46(1), and the plenary ‘seems to have accepted‘ its qualification. Acceptance is doing heavy lifting in this view of the relationship between the ASP and its Bureau.

Article 112(3)(c) explicitly confines the Bureau to assisting ‘the Assembly in the discharge of its responsibilities’, a formulation repeated in Rule 29(1) of the Assembly’s own Rules of Procedure. An organ whose statutory function is to assist the Assembly cannot, consistently with that function, have the power, even an informal one, to effectively bind the Assembly by its determination that the Prosecutor engaged in serious misconduct or to overshadow and displace the ASP plenary’s own determination of this matter. Such a practice would make the Assembly subordinate to the Bureau – a complete inversion of their proper relationship as contemplated by the Statute and the ASP Rules. 

Yet at the Special Session, the Assembly adopted a single decision, by secret ballot and by an absolute majority of 82 States Parties, finding that the Prosecutor had committed serious misconduct and removing him from office It did so instead of first holding a vote, requiring a two-thirds majority of States Parties present and voting, on whether the serious misconduct finding should be upheld (Rules 63 and 66 of the Assembly) and, if so, holding a second vote on removal by an absolute majority under Article 46. The two steps were collapsed into one ballot measured against a single threshold. By resolving the question of the applicable majority through practice rather than through a reasoned determination of the formal legal requirements, the Assembly has established a precedent whereby the removal of an elected official is a single-stage political act taken at the lowest threshold mentioned anywhere in the applicable legal framework. This precedent has the systemic effect of making the ouster of ICC Prosecutors less onerous for States Parties. The next case may not be so numerically forgiving. 

The Bureau recommended this course of action, and the Assembly voted on it, in an institutional context in which the removal of the Prosecutor is already procedurally less exacting than the removal of other elected officials, such as judges. Given the fundamental role the Prosecutor plays within the Rome Statute system, their removal, as with the removal of judges or the Registrar, signifies a loss of confidence, a breakdown in trust, and a finding that conduct sufficiently grave to warrant removal has been committed. However, while Article 46(2)(a) permits the Assembly to remove a judge by secret ballot and a two-thirds majority of States Parties, after a two-thirds majority of judges has recommended removal, Article 46(2)(b) sets the threshold for the Prosecutor at an absolute majority of States Parties, voting by secret ballot, without any prior qualified recommendation from another organ. Incomplete safeguards and ambiguous drafting create scope for the abuse of power. This vulnerability materialized in the proceedings against the third Prosecutor in three distinct respects, which may cast doubt on whether the protection of the Court’s legitimacy was the ASP’s sole or main objective.

First, the Presidency of the Assembly decided to sideline the Independent Oversight Mechanism in favour of an external investigation. Article 112(4) envisages the establishment of this mechanism for, among other purposes, the investigation of the Court, and the original framework designated it as the only body competent to investigate wrongdoing by elected officials. On 11 November 2024, having consulted the Bureau, the President of the Assembly announced that the investigation into allegations of misconduct against Karim Khan would be conducted by the United Nations Office of Internal Oversight Services (OIOS). No underlying legal reasoning was made public to justify this departure from the Rules of Procedure and Evidence and the Independent Oversight Mechanism’s mandate as established in 2024. The consequence was foreseeable: the OIOS investigation, conducted under UN administrative law, applying a lower standard of proof, and concluded in December 2025, could not yield a definitive factual conclusion on the allegations within the Rome Statute legal and disciplinary framework. The Bureau then forwarded the case to an ad hoc panel of judicial experts, selected by the Bureau and versed in administrative, criminal, and labour law, to carry out a legal characterization of the OIOS factual findings. Having analysed the OIOS report and underlying materials, the ad hoc panel concluded three months later that it was unable to find that the Prosecutor had committed serious misconduct or a breach of duty under the relevant standard of proof. The latter was, if controversially, a high one: beyond reasonable doubt, as derived from the case law of the International Labour Organization’s Administrative Tribunal, under whose jurisdiction the International Criminal Court falls.

Second, States Parties amended the applicable rules while the process was under way. In December 2025, the Assembly adopted by consensus, and with unprecedented speed, amendments to the Rules of Procedure and Evidence and to its own Rules governing the removal and sanctioning of elected officials. While these amendments represented a qualitative improvement and implemented the Independent Expert Review recommendations, in particular Recommendation 108, this came five years too late, after the fact-finding had been concluded but before the ad hoc panel began deliberating on the legal characterization of the OIOS findings. The Assembly thereby created, mid-proceedings, a new and untested disciplinary framework and applied it immediately to the ongoing case. Generally, best practice for any type of investigation, certainly in the context of removing the Court’s elected officials, calls for stability of the applicable norm, predictability of process, and due process that respects the basic principle of legality. 

Third, the Bureau decided, in its first-ever contested vote, to disregard the ad hoc panel of judicial experts’ conclusion that the materials did not establish serious misconduct or a breach of duty. The authority of the Bureau to reject the legal advice of the ad hoc panel is contested. Einarsen argued that, under Rule 81 of the Assembly’s Rules, the President could only request the Bureau to refer the matter to the full Assembly only if the panel had found serious misconduct. If an ad hoc panel returns a recommendation of less serious misconduct or no misconduct, as in this case, Rule 82 applies, under which the Bureau can still disagree with the ad hoc panel’s finding, but is compelled, under Rules 29 and 30 of the Rules of Procedure and Evidence, to allow the full Assembly to determine whether serious misconduct has occurred and, on the basis of that decision, what consequences should follow, including removal under Article 46. Here, the Bureau itself selected the three jurists, defined their mandate to exclude additional fact-finding, and then set aside their unanimous conclusion that the materials before them did not establish serious misconduct beyond reasonable doubt. In other words, a diplomatic forum, operating potentially on the basis of political considerations rather than necessarily in strict accordance with the rule of law, decided that it was better suited to draw its own factual and legal conclusions, effectively adding a new layer to the procedure by requesting submissions from the parties – a step not explicitly authorized by the amended rules. 

In sum, the procedural rules governing disciplinary action by States Parties against the Prosecutor are, at best, unclear and, at worst, highly controversial. This situation further erodes the systemic safeguards necessary to protect the independence of the Prosecutor as an organ of the Court, as well as, prospectively, that of any such elected official, in a context of protracted and unprecedented external political pressure on, and interference with, the Court. 

Wanting High Moral Character

Shany and Varaki close with observations on ethical leadership, virtue ethics, and Camus’s decency. We are in complete agreement with those lofty aspirations. The difficulty is identifying to whom those expectations are addressed. High moral character under Article 42(3) is not self-executing. It is a legally binding requirement that States Parties must operationalise through elections and management oversight. However, examining how the Assembly has applied Article 42(3) in its electoral practice raises serious doubts about whether it has treated the provision as such. In fact, both the elections of Moreno Ocampo (2003) and Bensouda (2011) were marred by questions of integrity and transparency. 

The third election repeated the pattern at a greater cost. The lessons-learnt report commissioned afterwards found that ‘politics, and lobbying, still played an influential role’, and recorded the view of several States Parties that removing the political element was impossible or even undesirable. The Assembly’s inability or unwillingness to arrange for a depoliticised, merit-based election has, on each occasion, put at risk the legitimacy and operational capability of the Court, in particular the Office of the Prosecutor, where the officeholder’s perceived impartiality, demonstrable credentials, and conduct befitting the role are critical. In fact, what the politicised electoral process ultimately produced was the first-ever removal of a Prosecutor. If anything, this may be adduced as showing the limits of any vetting apparatus designed to determine whether candidates satisfy the high moral character requirement.

Conclusion: Where Next? 

Del Ponte’s line about colliding with the edge of the political universe in which a tribunal is allowed to function may at first sight be a fitting epigraph. It does, however, raise the question of where that universe begins and where it ends. Del Ponte used the phrase to refer to external political constraints on prosecutorial and judicial work. However, the Assembly of States Parties is emphatically not an external constraint; it is the Court’s injugovin, constituted under Article 112 of the Statute, and an integral part of the ICC’s institutional system. It owes a non-waivable duty of care to the judicial institution it governs, including shielding it from undue external pressures, which it can only do if it does not itself succumb to them. To fold its conduct into the political universe in which the Court is allowed to function, or to allow the adverse external context to shape the way in which the ASP exercises its disciplinary powers vis-à-vis the Court, even if ostensibly for the sake of preserving legitimacy, is to accept extra-legal constraints on the Court and the use of governance authority for ulterior purposes. That would transgress the rule-of-law limits within which judicial governance must be performed – a collision with the edge of the legal universe of the Rome Statute system.

While proposals for amendments have been circulated to strengthen the textual safeguards in the Rome Statute, the Assembly must critically reflect on how to improve its governance practice and be candid about how its political nature and operations impede the necessary progress in that direction. We believe that the Assembly should commission a structured lessons-learnt exercise, drawing upon a painstaking review of this experience and an audit of practices in comparable international organizations, and create a standing independent legal advisory function for the Bureau, institutionally embedded in the Secretariat, with authority to issue opinions and recommendations. The ASP could also mandate a periodic expert audit of the quality of its governance of the ICC and adopt an enforceable code of conduct for State Party delegates holding office in the Bureau and its subsidiary bodies. Such measures would go a long way towards improving the governance mechanisms, including the disciplinary process. Ultimately, it is the quality of that process, not the perceived character of the officeholder, that determines whether the Assembly can credibly claim to govern the Court in a legitimate manner.

Whether States Parties will undertake that work is a separate question. The Independent Expert Review is a sobering precedent for what happens when reform commitments are not fully implemented in a timely fashion. The recurring tendency in the ASP’s governance practice thus far has been to treat each crisis as sui generis, devise an ad hoc procedure, and draw no general conclusions for the future. The removal of the third Prosecutor will confirm the same pattern unless States Parties choose otherwise. They remain the masters of the Rome Statute, and the independence of the Court can only be assured if the Assembly’s governance performance is up to the task. For the time being, however, there are reasons to doubt whether the ASP’s purported attempt to salvage the ICC’s legitimacy—as some State Party representatives may view the dismissal—has been or will prove to be a successful one. Given how the process leading up to it was organised and communicated to the outside world, the result may prove to be just the opposite of what was intended. Only time will tell.