KOHEN AVOCATS FIRM · PARIS 16
Duterte Defence Asks ICC to Vet Digital Evidence: What the E-Court ‘Expert Verification’ Fight Means Before the 30 November Trial

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 Defence Asks ICC to Vet Digital Evidence: What the E-Court ‘Expert Verification’ Fight Means Before the 30 November Trial

Speak to a lawyer 06 89 11 34 45

On 3 September 2026, the defence team of former Philippine president Rodrigo Duterte filed a request before Trial Chamber III of the International Criminal Court asking for an amendment to the Court’s e-Court Protocol. The request is narrow in form and significant in substance: the defence wants a new metadata field, called “Expert Verification”, attached to every piece of digital evidence the prosecution intends to use at trial. The filing, signed by defence counsel Peter Haynes, KC, describes a prosecution disclosure set of more than 25,000 items, with a further 200 gigabytes of material announced, and warns that the record is being “flooded” with documents that are duplicated, illegible, or impossible to trace to a human author. The trial on charges of crimes against humanity is scheduled to open on 30 November 2026, and the next status conference is set for 16 September.

This dispute is about much more than one high-profile case. It crystallises the central procedural question of modern international criminal justice: how does a court verify the authenticity of evidence that was born on a phone, uploaded to a platform, collected by anonymous informants through a portal such as the OTP’s OTPLink, and disclosed as a file? The answer matters to journalists, companies, and individuals far beyond The Hague, because the same screenshots, metadata, and chain-of-custody questions arise in French criminal and civil courts every day. This article explains what the Duterte defence is asking for, why the ICC’s existing framework leaves room for the request, and how a French lawyer reads the problem through the tools of the Code de procédure pénale and the Code civil.

I. Why is Duterte’s defence asking the ICC to change its e-Court Protocol before the 30 November trial?

A. What does the 3 September 2026 filing actually request from Trial Chamber III?

The procedural context is now fixed. Pre-Trial Chamber I confirmed the charges against Mr Duterte on 23 April 2026, committing him to trial on three counts of crimes against humanity under Article 7(1)(a) of the Rome Statute, the Court’s founding treaty, which France published domestically by decree reproduced on Légifrance. The judges found substantial grounds to believe that murders and attempted murders were committed as part of a widespread and systematic attack against the civilian population in the Philippines between 1 November 2011 and 16 March 2019, in the context of the so-called war on drugs. The ICC’s public case page for The Prosecutor v. Rodrigo Roa Duterte confirms that the opening of the trial is scheduled for 30 November 2026 before Trial Chamber III, and that 539 victims have been authorised to participate in the proceedings through a single team of common legal representatives.

Against that backdrop, the defence filing of 3 September 2026 does not attack the charges. It attacks the plumbing of the evidence. The e-Court Protocol is the Court’s internal technical rulebook governing how electronic documents are disclosed, numbered, and displayed during trial: each item receives an evidence record number, or ERN, and metadata fields describe its provenance. The defence reports that the prosecution has already disclosed more than 25,000 items and plans to disclose a further 200 gigabytes. According to the filing, individual pages of a single document have been disclosed independently, “with distinct and inconsecutive ERNs”, audio-visual material has been produced without transcripts or translations, and many items are illegible.

The remedy proposed is a new compulsory metadata field in the e-Court system, “Expert Verification”. In the defence’s own words, “this field would confirm whether or not the item will be accompanied by an expert report if it is to be submitted to the Chamber as evidence”, and it “would operate similarly to the witness fields in the E-Court Protocol, which list forms of metadata to be provided on an ongoing basis during the trial phase”. The request is therefore administrative in appearance — a column in a database — but its legal effect would be to force the prosecution to declare, item by item, whether a qualified expert stands behind the digital evidence or not.

The filing is equally precise about what such an expert report should contain. The defence argues that “any such expert report ought to include, at a minimum, information on: the source of the item (both the sender/uploader and the creator of the piece of evidence, if known); its chain of custody and whether the item is the original or a copy; its metadata and any other technical aspects, such as the hash value generated at the time of capture; the steps taken to verify the material and to check for editing or manipulation”, together with the qualifications and experience of the person who carried out the verification. Anyone who has commissioned a forensic report on a phone extraction or a platform download will recognise this checklist: it is the standard content of a serious digital forensics deliverable.

Two features of the prosecution’s evidence collection make the request especially pointed. First, part of the material was gathered through OTPLink, the Office of the Prosecutor’s online portal that allows anyone to submit information about ICC investigations, potentially anonymously. Second, the defence points to “screenshots of text messages that are unaccompanied by a statement or testimony from an individual who authored any of the messages or participated in the exchange”. A screenshot detached from its author is, in evidentiary terms, a floating image: it shows what someone claims a screen displayed, not what the device actually contained.

B. Why does the defence say the disclosure set is unfit for a fair and expeditious trial?

The filing does not merely complain about volume; it describes quality failures that, if established, would compromise the Chamber’s ability to assess the evidence at all. Defence counsel writes that “based on a small sample of documents from the same dataset which has already been received by the Defence, it is clear that there are major issues with this collection of documents; it includes many duplicates and documents of very poor quality which, notwithstanding its volume, makes meaningful review immensely challenging”. The concern is expressed in deliberately institutional language: the defence warns of the risk of “flooding the record” with unreliable or irrelevant material, “detrimental for the parties and the Chamber’s search for truth”.

Disclosure overload is a known pathology of modern international trials. When a prosecution team discloses tens of thousands of items in a format that fragments documents into separately numbered pages, the defence must spend its preparation time reconstructing the record instead of testing it. The 30 November 2026 trial date creates a hard deadline: every week spent identifying duplicates or chasing missing transcripts is a week not spent preparing cross-examination. The defence request can therefore be read as a case-management measure as much as an evidentiary one — an attempt to make the prosecution’s database legible before the first witness is called.

The deeper claim is about authenticity. Digital items collected from social media, messaging applications, or anonymous upload portals do not authenticate themselves. A photograph may have been cropped, a video may have been re-encoded, a text message may have been fabricated on a second phone. Without metadata — the technical footprint recording when and how a file was created — and without a documented chain of custody showing who held the file from capture to courtroom, the Chamber would be asked to trust the surface of the image. The defence filing states that without the requested verification information it would be “very difficult for the Chamber to come to a view on the evidence’s relevance and admissibility, pursuant to Article 69(4) of the Statute”.

That reference to Article 69(4) is the legal hinge of the whole request. The provision authorises the Court to “rule on the relevance or admissibility of any evidence, taking into account, among other things, the probative value of the evidence and any prejudice that such evidence may cause to a fair trial”. The defence is in substance arguing that admissibility cannot be seriously assessed in the dark: the Chamber needs to know, for each contested digital item, whether anyone technically competent has verified what it is and where it comes from. The filing stops short of asking for exclusion; it asks for a flag. That measured remedy may make the request harder for the prosecution to resist.

The timing also matters politically. The Duterte proceedings have generated an intense flow of online commentary, including organised misinformation about the case, and the prosecution itself has been collecting open-source material in that polluted information environment. When the evidentiary record is built partly from the same platforms where disinformation circulates, the distinction between a genuine artefact and a manufactured one is not academic. It is the difference between proof and propaganda.

There is also a fairness dimension that goes to the heart of the trial’s calendar. The confirmation of charges in April 2026 was decided on the basis of a documentary record that the defence could only sample, and the trial stage will multiply that record several times over. If the Chamber were to wait until final submissions to confront authentication problems, the defence would have spent months building its case on a database whose integrity it could not assess. Procedural economy argues for resolving the question now, at the threshold of the evidentiary phase, rather than in the shadow of the judgment. A ruling on the “Expert Verification” request before the 30 November opening would give both parties a stable framework and would spare the Chamber the far costlier exercise of untangling contested files during the hearings themselves.

II. How do international and French law approach the authentication of digital evidence, and what can litigants learn from this dispute?

A. What does the ICC framework provide on relevance, admissibility, and expert evidence?

The Rome Statute gives ICC judges an unusually free hand on evidence. Unlike common law systems with elaborate exclusionary rules, the Court proceeds on the basis of free assessment: Article 69(4) of the Statute leaves relevance and admissibility to the Chamber’s discretion, weighed against probative value and fairness, and Article 64(9) confers on the Trial Chamber the power to rule on “any other relevant matters” for the conduct of a fair and expeditious trial. The Rules of Procedure and Evidence complete this architecture, notably by rejecting any strict corroboration requirement: a Chamber may in principle rely on a single piece of evidence, which makes the reliability of each individual item even more decisive.

This freedom has a cost. When no formal authentication rule exists, everything depends on the parties’ ability to expose weaknesses in provenance. Cross-examination of a witness can test a human memory, but a file cannot be cross-examined. The e-Court Protocol’s existing metadata fields — typically the date, author, source, and ERN of a document — record administrative information about the disclosure, not forensic information about the artefact. A field stating that a file was uploaded on a given day says nothing about whether the file was manipulated before upload. The gap identified by the Duterte defence is precisely this: the protocol describes the envelope, not the content’s integrity.

The Court has tools to respond. Under its general powers, a Trial Chamber can order the prosecution to provide additional information about its evidence, appoint its own experts, or attach reduced weight to unverified items at the judgment stage. The defence’s “Expert Verification” field would operationalise those powers at the disclosure stage rather than at final deliberations. If granted, every party and every judge would be able to filter the 25,000-item record by a simple criterion: verified by a named expert, or not. The October and November pre-trial weeks will show whether Trial Chamber III prefers this systematic solution or a case-by-case approach.

The victims’ participation scheme adds a further layer of complexity. With 539 authorised victims represented through common legal counsel, evidentiary disputes in this case will not remain a duel between prosecution and defence. Every authentication ruling will affect the victims’ ability to present their views and concerns, and every delay in clarifying the record will postpone the moment when their representatives can make meaningful submissions. Efficient verification upstream therefore serves not only the accused’s right to a fair trial but also the victims’ right to a trial that actually takes place within a reasonable time.

For observers, the broader lesson is that the ICC is converging with a question that domestic systems resolved long ago in their own way: digital evidence requires an audit trail. International tribunals have repeatedly faced contested videos, intercepted communications, and satellite imagery; the novelty in the Duterte case is the industrial scale of open-source material and the anonymity of some of its sources. Whatever the Chamber decides, the filing has already framed the debate that will accompany the trial from its opening on 30 November 2026.

B. How would a French court handle the same problem, and what should parties do in practice?

French criminal procedure starts from a principle very close to the ICC’s free assessment: under Article 427 of the Code de procédure pénale, “hors les cas où la loi en dispose autrement, les infractions peuvent être établies par tout mode de preuve et le juge décide d’après son intime conviction” — offences may be established by any means of evidence, and the judge decides according to intimate conviction. But this freedom is immediately disciplined by practice. When digital evidence is contested, French courts expect a technical demonstration of integrity: hash values computed at seizure, sealed exhibits, a documented chain of custody from the investigating officer to the hearing, and, where the dispute justifies it, a judicial expert.

The cornerstone of that technical demonstration is the court-appointed expertise. Under Article 156 of the Code de procédure pénale, “toute juridiction d’instruction ou de jugement” may order an expertise when a technical question arises, at the request of the prosecution, the parties, or on its own motion. The expert is sworn, registered on an official list, and works under the parties’ adversarial scrutiny; the defence can comment on the expert’s operations, request a counter-expertise, and challenge the report at the hearing. A French examining magistrate confronted with the Duterte defence’s concerns — duplicated files, missing metadata, anonymous uploads — would have a ready-made procedural answer: appoint an expert to verify the collection, with a mission letter covering exactly the points the ICC filing lists.

French investigators also dispose of specific powers to capture computer data. Under Article 706-102-1 of the Code de procédure pénale, a technical device may be used to record computer data displayed on a screen or entered on a keyboard, and Article 60-1-2 with Article 77-1-2 of the same Code organise access to data stored in another computer system. Each of these mechanisms comes with reporting obligations designed precisely to preserve integrity: the officer must record the operations performed so that the file produced in court can be linked to the data as it existed on the device.

On the civil side, French law addressed the probative force of electronic documents directly. Article 1353 of the Code civil places the burden of proof on the claimant — “celui qui réclame l’exécution d’une obligation doit la prouver” — and Article 1366 provides that “l’écrit électronique a la même force probante que l’écrit sur support papier”, electronic writing having the same probative force as paper, provided the person can be duly identified and the writing is stored under conditions safeguarding its integrity. Article 1367 adds that an electronic signature consists of a reliable identification process guaranteeing its link with the act to which it attaches. Identification plus integrity: French civil law has reduced the authentication of digital evidence to these two cumulative conditions, and they map exactly onto the Duterte defence’s demands about source, chain of custody, and hash values.

The practical lessons extend well beyond The Hague. A company producing internal chat logs in an employment dispute, a spouse presenting screenshots in family litigation, or a victim documenting online harassment faces the same three questions the defence has put to the ICC prosecution: who created this file, what happened to it between capture and courtroom, and has anyone technically competent verified that it was not edited? Parties should preserve original files rather than screenshots, keep the source device, record the date and method of capture, and consider a bailiff’s report — the French constat d’huissier — or a private forensic report before the evidence is challenged. In criminal matters, a defence lawyer should request a judicial expertise early, because once the file is built on unverified digital items, dislodging them at trial becomes far harder.

One point deserves emphasis for clients who collect their own evidence. The reflex of capturing a screen with a telephone camera, or forwarding a message thread by email, destroys precisely the metadata that gives the item its probative force. The original file carries embedded information — creation date, device identifier, editing history — that a photograph of a screen does not. Where the stakes justify it, a forensic capture performed by a professional, with a hash value computed at the moment of collection, transforms a contestable image into a document whose integrity can be demonstrated. The difference in court is frequently decisive, and it is the same difference the Duterte defence is asking the ICC to systematise through its e-Court amendment.

The Duterte filing is, in this sense, a symptom of a general movement. As courts everywhere absorb ever-larger volumes of platform-native evidence, the procedural systems that survive will be those that force verification upstream, at the moment of disclosure, rather than leaving authenticity to be guessed at judgment. The ICC’s answer to the 3 September request will be watched by every practitioner who litigates with digital evidence — in Paris as in The Hague.

Conclusion

The Duterte defence has not asked Trial Chamber III to exclude the prosecution’s digital evidence; it has asked the Court to say, for each item, whether an expert has verified it. That modest procedural demand raises the defining evidentiary question of the digital age: a file is only as good as its provenance. With more than 25,000 disclosed items, anonymous OTPLink uploads, and a trial opening fixed for 30 November 2026, the Chamber’s response will shape not only this case but the working standards of international criminal justice. French law, with its combination of free proof under Article 427 of the Code de procédure pénale and its disciplined machinery of judicial expertise and civil-law authentication under Articles 1353, 1366 and 1367 of the Code civil, shows that free assessment and technical rigour are not opposites: the second is what makes the first safe. Anyone facing digital evidence in a French or international proceeding should apply the same discipline the Duterte defence is demanding at the ICC.

Our firm regularly assists clients whose cases turn on the integrity of digital evidence — screenshots, platform records, phone extractions — in France and in cross-border proceedings. Related analyses: extradition under French law and the guarantees of the person claimed, Interpol red notices, arrest and remedies in France, and how to seize the International Court of Justice.

📄 Official circular