KOHEN AVOCATS FIRM · PARIS 16
Duterte at The Hague: ICC Confirms Crimes-Against-Humanity Charges and Admits Secret-Witness Testimony — What France Would Do

Hassan KOHEN acts in urgent matters, from the police station to the assize court. Initial strategic analysis: €80 incl. VAT, personal reply within 24 hours.

100% confidential · Legal professional privilege · No obligation

Article generated by artificial intelligence, through a process designed and supervised by the firm

Source: French Court of Cassation – Judilibre and Légifrance open data

Paris Bar Police custody, investigation, assize court CNB listing avocat.fr
Maître Hassan KOHEN, avocat au Barreau de Paris
Hassan KOHEN
Attorney at the Paris Bar

Duterte at The Hague: ICC Confirms Crimes-Against-Humanity Charges and Admits Secret-Witness Testimony — What France Would Do

Speak to a lawyer 06 89 11 34 45

On 2 October 2026, three judges of the International Criminal Court in The Hague handed the prosecution in the case against former Philippine president Rodrigo Duterte a significant pre-trial victory: the written statement of a protected witness, identified only as P-1100, will be admitted as evidence at trial under Rule 68(3) of the Court’s Rules of Procedure and Evidence. The decision, reported in Manila on 3 October 2026, lands barely two weeks after Duterte appeared in person before the Court for the first time on 16 September 2026, and as the judges put the finishing touches to the trial arrangements, including an amended e-court protocol reported on 5 October 2026. For an international audience, the headlines write themselves: a former head of state, detained at The Hague since March 2025, facing trial for the crimes against humanity of murder and attempted murder over a “war on drugs” campaign that killed thousands. But the questions that matter for anyone exposed to the same machinery — officials, intermediaries, witnesses, families — are procedural, and they are the same questions a Paris extradition and international criminal law practice answers every week: what exactly did the judges confirm, what does the admission of a masked witness’s written statement change for the defence, and what would happen if a file like this one landed on a French judge’s desk? This analysis answers all three, with the decisions and statutes in hand.

I. The Duterte trial at The Hague: confirmed charges and the October 2026 witness ruling

A. From the March 2025 arrest warrant to confirmed crimes-against-humanity charges

The public record of the case, published by the Court itself under case number ICC-01/21-01/25, is unusually complete, and it sets the frame for everything that follows. An arrest warrant was issued on 7 March 2025. Philippine authorities arrested Duterte in Pasay City on 11 March 2025, and he was flown to the Netherlands, where he was surrendered into ICC custody on 12 March 2025 and has remained detained ever since. The confirmation-of-charges hearing ran from 23 to 27 February 2026, and on 23 April 2026 Pre-Trial Chamber I committed him to trial. These dates are not decoration: in international criminal procedure, each step — warrant, surrender, confirmation — is a judicial filter, and each filter that the prosecution passes narrows the defence’s room for manoeuvre at trial.

The confirmation decision itself is the single most important document in the file to date. The Chamber stated its conclusion in careful terms: according to the Court’s press release of 23 April 2026, Pre-Trial Chamber I found substantial grounds to believe that Duterte bore responsibility for the crimes against humanity of murder and attempted murder under article 7(1)(a) of the Rome Statute. The confirmed crimes were, in the Chamber’s characterisation as reported by the Court, committed as part of a widespread and systematic attack against a civilian population on Philippine territory over the period from 1 November 2011 to 16 March 2019, in the context of the anti-drugs campaign. The Chamber added that its ruling rested on the evidence and arguments put forward by the Prosecutor and the Defence at the confirmation hearing of 23 to 27 February 2026, together with the submissions and documents of the parties and participants, victims included. Three features of this ruling deserve emphasis. First, the legal characterisation: murder and attempted murder as crimes against humanity under article 7(1)(a) of the Rome Statute, which means the prosecution does not have to prove a genocide-style plan to destroy a group, but rather a widespread or systematic attack on a civilian population with the accused’s knowledge. Second, the time window: 1 November 2011 to 16 March 2019, a period that starts while Duterte held local office in Davao and runs through his presidency, which explains why the prosecution’s narrative ties the Davao Death Squad period to the national anti-drugs campaign. Third, the standard of proof at this stage: “substantial grounds to believe” is deliberately lower than the trial standard of beyond reasonable doubt, so confirmation tells the defence what it must demolish, not what has been proved.

Readers familiar with cross-border arrest practice will notice a structural difference with ordinary extradition that conditions the whole defence strategy. Duterte was not extradited between two states; he was arrested by his own national authorities and surrendered to an international court under a cooperation regime. There was no requested-state judge testing double criminality, specialty or human-rights bars the way a Paris chambre de l’instruction tests a foreign extradition request before the Government may sign a surrender decree. The filters were internal to the ICC system: the warrant judge, then the confirmation judges. Anyone arrested in France on the basis of an international alert, by contrast, goes through French filters first — a point developed in Part II, and already examined on this site in the guide to arrest in France on an Interpol diffusion and the fight for release and deletion. The practical lesson of the Duterte timeline is that once surrender to The Hague has happened, the battle moves entirely inside the ICC’s procedural world: detention reviews, disclosure fights, and evidentiary rulings like the one of 2 October 2026.

Detention is the first of those inside battles, and Duterte lost it. As reported from Manila, Trial Chamber III ruled that he should remain detained, finding a likelihood that he could abscond or obstruct, endanger the investigation or the Court’s proceedings. Pre-trial detention at the ICC, like detention pending surrender in France, is not a punishment; it is a risk-management measure, and the risks the judges named — flight and obstruction — are the classic ones. For the families and counsel who follow these cases, the message is stark: advanced age, former office and health arguments, all of which were aired at the September 2026 status conference, did not outweigh the judges’ assessment of the incentive to flee and the capacity to interfere with witnesses. That assessment now colours everything, because a detained accused prepares a complex international-crimes trial from a cell in Scheveningen, through counsel, with limited access to witnesses on the ground — while the prosecution refines exactly the kind of insider testimony the 2 October ruling admitted.

B. September–October 2026: in-person appearance, the P-1100 decision and the road to trial

The fortnight that brought this case back to the front pages began on 16 September 2026, when the former president appeared before the ICC in person for the first time during the third status conference. Status conferences are case-management hearings: the judges take stock of disclosure, witness protection, translation, health and scheduling. According to the Manila press reports of the hearing, the chamber heard argument on Duterte’s health and his ability to stand trial — the standard fitness-to-plead-and-stand-trial question that arises in every atrocity trial involving an elderly accused — while keeping him in detention. No fitness ruling removing the case from the trial track was reported, and the machinery moved forward. Two weeks later came the evidentiary ruling that is the real subject of this analysis.

In a decision dated 2 October 2026, the Trial Chamber granted the prosecution’s application under Rule 68(3) of the ICC Rules of Procedure and Evidence for the written statement of witness P-1100 to be received as evidence covering part of what would otherwise have been his examination-in-chief, as reported by the Manila Times. The witness is unidentified in the public filings: substantial passages about his background and evidence are redacted, and while Manila commentators have speculated about self-declared former death-squad members and police officers, the reporting is explicit that there is no indication in the public filing that P-1100 is any of the named figures. The prosecution’s case for admission, as reported, was that the testimony bore on the allegation of a policy of neutralising suspected criminals through violent crime including murder, and that it could illuminate the working methods attributed to the Davao Death Squad as well as Duterte’s personal criminal responsibility, knowledge and intent as alleged. In other words, P-1100 is presented as an insider bridge between the Davao-period killings and the presidential-period campaign — the evidentiary hinge of the whole case.

The Chamber’s reasoning, as reported, is a textbook application of the Rule 68(3) safeguards, and defence teams in every ICC-related file should study its structure. The judges held that admitting the earlier testimony would not prejudice Duterte’s rights, since the defence would still be able to question P-1100 on his evidence, including the witness’s knowledge, credibility and the basis for his statements. The Chamber added the proposition that future Rule 68(3) litigation will revisit: prior recorded testimony under that provision supplements rather than replaces live evidence, with the defence retaining a full opportunity to examine P-1100 on anything arising from his account. Admission is therefore conditional, not final: the ruling is preliminary and remains subject to P-1100 appearing before the Chamber and consenting to the use of his statement, with the defence and the judges entitled to examine him. The Chamber also showed it had read the statement critically rather than rubber-stamping it: it noted that parts of the text stated opinions or conclusions and directed that the defence be able to test the basis of those passages in questioning, and it stressed that the testimony would be weighed on its evidentiary value and the witness’s personal knowledge.

Three practical consequences follow. First, the prosecution has secured the right to lead part of its most sensitive evidence in chief in writing, which shortens and de-risks the examination of a witness whose live testimony might be fragile, evasive or dangerous to give in full. Second, the defence has not lost the confrontation right, but the confrontation is displaced: instead of meeting the evidence for the first time as it unfolds live, counsel must now dismantle a fixed written narrative through cross-examination, attacking personal knowledge, internal consistency and the inferential leap from “policy” and “modus operandi” to the accused’s own knowledge and intent. Third, the ruling confirms the trial’s centre of gravity: individual criminal responsibility proved through insider evidence about a system, not through crime-base testimony alone. That is the pattern of every modern atrocity trial, from The Hague to Paris, and it is why the French comparison in Part II matters: France tries the same category of cases, with the same dependence on insider witnesses, but under different evidentiary rules — and the differences decide outcomes.

The last reported item completes the picture of a trial entering its final approach: on 5 October 2026 the Court amended its e-court protocol for the case, the technical framework governing electronic filing, disclosure and courtroom presentation of evidence. Protocol amendments sound administrative; in atrocity trials they are not. They govern how thousands of pages of disclosure reach the defence, how redacted and protected material is handled, and how prior recorded testimony like P-1100’s statement is physically introduced into the record. Combined with the September fitness discussions and the October evidentiary ruling, the signal is consistent: the Chamber is clearing the procedural runway, and the trial will be fought on the admissibility and weight of insider evidence. Anyone advising a suspect, a witness or a government in the orbit of such a case should read that signal the way Paris counsel reads a chambre de l’instruction scheduling order: the time for preliminary challenges is ending, and the evidence battle has begun.

II. The same file in Paris: immunity, surrender and trial under French law

A. A former president has no immunity shield for crimes against humanity, in The Hague or in Paris

The first question any French lawyer asks about a Duterte-type file is the immunity question, and French law answered it at the highest level barely a year ago. On 25 July 2025, the Assemblée plénière — the most solemn formation of the Cour de cassation — ruled in the case of a former governor of the Syrian central bank, prosecuted in France for complicity in crimes against humanity and war crimes, that functional immunity is no bar to such prosecutions. The Court’s holding, published as Assemblée plénière, 25 July 2025, appeal no. 24-84.071, states that “le principe d’immunité fonctionnelle de juridiction en matière pénale ne peut pas être opposé en cas de poursuites des chefs de génocide, crimes contre l’humanité et crimes de guerre”. The ruling reserves only treaty-based immunities, notably diplomatic and consular instruments: for a former official facing atrocity charges, there is no residual official-acts shield. The same decision explains why Duterte’s former presidency never generated an immunity debate at the ICC at all: “Le principe de l’immunité ne peut être opposé devant les juridictions internationales, dont les statuts écartent expressément toute immunité, et qui agissent au nom de la communauté internationale”. Before the ICC, article 27 of the Rome Statute renders official capacity irrelevant; before the French courts, the plenary ruling now does the equivalent work for genocide, crimes against humanity and war crimes.

The substantive definitions France would apply to Duterte’s alleged conduct sit in the Penal Code, and they mirror the ICC charges closely enough that double-characterisation would present no difficulty. Article 212-1 of the Penal Code provides: “Constitue également un crime contre l’humanité et est puni de la réclusion criminelle à perpétuité l’un des actes ci-après commis en exécution d’un plan concerté à l’encontre d’un groupe de population civile dans le cadre d’une attaque généralisée ou systématique : 1° L’atteinte volontaire à la vie ; 2° L’extermination ; 3° La réduction en esclavage”, followed by deportation, imprisonment, torture, sexual violence, persecution, enforced disappearance and apartheid acts. Wilful killing carried out under a concerted plan against a civilian population within a widespread or systematic attack — the ICC’s confirmed “widespread and systematic attack” and the alleged “neutralize” policy — maps directly onto this text, and the penalty, life imprisonment, matches the gravity the ICC confirmation found. Genocide, defined next door in article 211-1 of the Penal Code as acts “en exécution d’un plan concerté tendant à la destruction totale ou partielle d’un groupe national, ethnique, racial ou religieux”, is not charged here and would not fit: the Duterte file is a crimes-against-humanity file, not a genocide file, and keeping the two qualifications distinct is the first duty of rigour in this field. Conviction would also carry the full range of additional penalties under article 213-1 of the Penal Code, from loss of civic rights to bans on public office — a detail that matters for former officials whose remaining political capital is often their most valuable asset.

Had an ICC arrest-and-surrender request concerning a Duterte-like suspect reached France, it would have travelled through a dedicated statutory channel, not through ordinary extradition. Article 627 of the Code of Criminal Procedure lays down the principle, in force on the date of writing: “Pour l’application du statut de la Cour pénale internationale signé le 18 juillet 1998, la France participe à la répression des infractions et coopère avec cette juridiction dans les conditions fixées par le présent titre”. The operational provision is article 627-4 of the same Code: “Les demandes d’arrestation aux fins de remise délivrées par la Cour pénale internationale sont adressées, en original et accompagnées de toutes pièces justificatives, aux autorités compétentes en vertu de l’article 87 du statut qui, après s’être assurées de leur régularité formelle, les transmettent au procureur général près la cour d’appel de Paris et, dans le même temps, les mettent à exécution dans toute l’étendue du territoire de la République”. Paris is therefore the single entry point: the procureur général at the Paris Court of Appeal, with urgent requests executable immediately by the local prosecutor. For counsel, the corollary is the release remedy in article 627-9: “La mise en liberté ou la mainlevée ou la modification du contrôle judiciaire ou de l’assignation à résidence sous surveillance électronique peut être demandée à tout moment à la chambre de l’instruction de la cour d’appel de Paris”. Duterte’s failed detention challenge at The Hague has its Paris mirror in this provision: release pending surrender to the ICC is applied for at any time before the Paris investigating chamber, which rules in public, reasoned session — and the same risk calculus of flight and witness interference decides it. The parallel analysis of how surrender-related detention is fought in Franco-American files, where Article 3 of the European Convention on Human Rights does much of the work, is set out in this site’s guide to blocking surrender to the United States from France on Article 3 grounds.

B. Trying atrocity crimes in France: universal jurisdiction, insider witnesses and the defence playbook

France does not only surrender suspects to the ICC; it tries atrocity crimes itself, and the jurisdictional key is article 689-11 of the Code of Criminal Procedure, in force on the date of writing. Its opening words are worth quoting because they contain the whole architecture: “Hors les cas prévus au sous-titre Ier du titre Ier du livre IV pour l’application de la convention portant statut de la Cour pénale internationale, ouverte à la signature à Rome le 18 juillet 1998, peut être poursuivie et jugée par les juridictions françaises toute personne soupçonnée d’avoir commis à l’étranger l’une des infractions suivantes : 1° Le crime de génocide défini au chapitre Ier du sous-titre Ier du titre Ier du livre II du code pénal ; 2° Les autres crimes contre l’humanité définis au chapitre II du même sous-titre Ier”, followed by war crimes. The provision then locks French prosecution into the complementarity logic that also governs the ICC: French courts step back where an international or national court claims the case, since prosecution is possible only “si aucune juridiction internationale ou nationale ne demande la remise ou l’extradition de la personne”, with the Paris anti-terrorism prosecutor tasked to verify “l’absence de poursuite diligentée par la Cour pénale internationale”. Applied to a Duterte file, the consequence is clean: with the ICC seized and the trial running, France would surrender rather than prosecute; without the ICC in the picture, and the suspect present on French soil, Paris could open its own case for the same killings.

The second plenary ruling practitioners must know sets the strictest limit on that French fallback jurisdiction. On 12 May 2023, again in Assemblée plénière (appeal no. 22-82.468), the Court policed the double-criminality condition of article 689-11 in a Syrian war-crimes file. Two holdings bind future cases. First, foreign criminalisation cannot be conjured from good intentions: “l’incrimination des faits par la loi étrangère, au sens de l’article 689-11 du code de procédure pénale, ne saurait résulter de la volonté affichée par un État de lutter contre des infractions”. Second, treaties alone do not do the work either: “en l’absence de renvoi par la loi syrienne aux incriminations des instruments internationaux et de prévision d’une peine par lesdits instruments, le principe de légalité criminelle fait obstacle à ce que des faits soient considérés comme punis, au sens de la loi française, par la législation de l’État étranger à travers de telles normes”. The prosecution was saved only because, in substance, “il ressort de ses énonciations que les faits pour lesquels M. [W] a été mis en examen, sous la qualification de crimes de guerre et complicité, étaient punis, en substance, par la législation syrienne au travers d’infractions de droit commun et de celle d’implication d’enfants dans les hostilités”. Transposed to a Philippines-type file, the lesson is that a French universal-jurisdiction case for overseas killings stands or falls on proof that the foreign law punished the underlying acts in substance as ordinary crimes — murder, attempted murder — at the time, not on the grandeur of the international labels. Where two states claim the same suspect at once, the priority analysis follows the method explained in this site’s guide to competing extradition requests and how France decides priority.

The deepest France–The Hague divergence, and the one the P-1100 ruling brings into focus, concerns insider-witness evidence. French criminal procedure is built on orality and adversarial debate: the cour d’assises and criminal courts hear witnesses live, and prior statements are read only within strict limits, with the accused’s right to confront shaping every inch of the trial. The ICC’s Rule 68(3) mechanism — admitting a written statement in place of part of the examination-in-chief, provided the witness appears, consents, and submits to questioning by the defence and the bench — has no exact French equivalent, and that contrast cuts both ways for the defence. In Paris, counsel confronting a P-1100 equivalent would attack the statement’s journey into the file from the investigation stage: conditions of the first interviews, pressure or promises, translation chains, inconsistencies across hearings, and the leap from observed facts to conclusions about the accused’s knowledge and intent — exactly the “opinions or conclusions” passages the Trial Chamber flagged for testing. At The Hague, the same attacks are channelled into the cross-examination that the Chamber guaranteed “on any issues relating to his testimony”, plus a weight-phase argument that a redacted insider whose background the public cannot verify deserves corroboration before a conviction may rest on his word. Either way, the defence syllabus is identical: isolate what the witness personally saw from what he inferred, heard or was told; map every inconsistency between the written statement and earlier accounts; and force the prosecution to corroborate the hinge points — policy, chain of command, the accused’s personal knowledge — with documents, intercepts or second witnesses.

For anyone who might one day be the subject, the associate or the witness in such a file, the operational checklist that Paris practice has distilled from ICC-surrender and atrocity cases holds in five lines. First, an ICC arrest-and-surrender request executed in France triggers immediate Paris proceedings: verify the formal regularity of the request, apply for release under article 627-9 without waiting, and build the detention file (ties, guarantees of representation, supervised residence) from day one, because flight-risk findings harden fast. Second, immunity arguments for former officials end where the 2025 plenary ruling draws the line: no functional-immunity shield for genocide, crimes against humanity or war crimes, in Paris or The Hague. Third, in any French universal-jurisdiction exposure, test double criminality concretely against the foreign ordinary criminal law as it stood at the time of the facts, following the 2023 plenary method — labels do not self-prove. Fourth, treat insider-witness risk as the centre of the file: secure independent counsel for witness-status advice early, preserve every prior statement and its context, and never assume a redacted protective regime at the ICC means the evidence cannot be tested. Fifth, manage competing proceedings from the start — ICC, territorial state, third states — because complementarity and priority decide which court actually judges, and the answer is rarely the one the headlines assume.

Conclusion

The Duterte case enters its trial phase as the archetype of twenty-first-century international justice: a former head of state, surrendered by his own country, detained at The Hague, facing confirmed charges of murder and attempted murder as crimes against humanity for a documented period from 1 November 2011 to 16 March 2019, with the decisive evidence now expected from protected insiders whose written words a Trial Chamber admitted on 2 October 2026 under Rule 68(3) — usable only if the witness appears, consents and faces questioning. France’s answer to the same file, written across two Assemblée plénière rulings and a full cooperation title in the Code of Criminal Procedure, is symmetrical: no immunity for atrocity crimes, surrender through Paris with release remedies before the investigating chamber, and a domestic universal jurisdiction that is real but disciplined by complementarity and strict double criminality. The Hague will now test whether P-1100’s statement survives confrontation; Paris practitioners will be watching, because the method for breaking — or crediting — an insider witness is the same in every courtroom that tries the gravest crimes.

📄 Official circular