A Dutch court is due to rule on 28 August 2026 in the DeSoto case, a prosecution arising from alleged conduct during the 1994 genocide in Rwanda. The case matters beyond the Netherlands because it shows what happens when a state handles a serious international-crimes file through its own courts, while witnesses, records, victims and the underlying events remain spread across borders. At the time of publication, the Dutch judiciary still describes the date as a planned ruling rather than an outcome. It says the defendant is charged with looting and destruction as warcrimes, and with incitement to and the commission of genocide against Tutsis in Rwanda in April 1994. The court also announced an English-language livestream because of expected public interest. The legal question for France is precise: can material developed in a Dutch trial support a French investigation, and when could French courts lawfully open a separate case? The answer is not that a foreign verdict automatically creates French jurisdiction. France must identify a statutory gateway, the exact conduct and the procedural status of the Dutch case. It must also protect the presumption of innocence, preserve the reliability of evidence and prevent two proceedings from punishing the same person twice for the same facts.
I. What is the DeSoto Rwanda genocide trial and why does it matter in France?
A. Why is the Netherlands trying a suspect it did not extradite?
The first point is institutional. The DeSoto file is being heard by the District Court of The Hague as a Dutch criminal case. The official English case page records the arrest of a 64-year-old man in February 2024 on suspicion of involvement in an attack and looting spree in Gatobotobo, Rwanda, during the genocide against Tutsis and moderate Hutus in 1994. It records the June 2026 substantive hearing and states that the court intends to deliver its ruling on 28 August 2026. The court’s separate notice uses a deliberately careful formulation: “The court is currently aiming to render a verdict on 28 August 2026.” That language is a procedural timetable, not a finding of guilt. The Dutch judiciary’s international-crimes case page and its English DeSoto trial notice should therefore be read as primary descriptions of the procedural posture, not as a substitute for the judgment.
A domestic prosecution can become necessary when the state in which an accused is found cannot, or does not wish to, transfer that person to the state where the alleged conduct occurred. The legal solution is not to treat nationality as proof of innocence or guilt. It is to decide which court can lawfully hear the evidence and whether that court can provide a fair trial. A state may have an interest in avoiding impunity while also refusing to send its own national to a foreign jurisdiction if the applicable rules or constitutional safeguards prevent surrender. It can then investigate conduct that occurred abroad under its own criminal law. The result is a cross-border criminal process: the forum is Dutch, the alleged conduct is Rwandan, and much of the factual material may have been collected in Rwanda or by cooperation with other states.
That distinction is important for French readers because “extradition” and “prosecution” are different legal operations. Extradition transfers a person to another jurisdiction so that the requesting state can investigate or try the person. A domestic prosecution asks the state where the person is present to investigate and judge the alleged conduct itself. The first is principally a cooperation decision; the second is a question of criminal jurisdiction, evidence and trial rights. A French lawyer examining the DeSoto consequences must not describe a Dutch judgment as an enforcement of a Rwandan arrest request, and must not assume that the Dutch court has created a warrant that France must execute. The judgment will be a national judicial decision whose evidential and procedural effects must be assessed under the law of any later forum.
The timing adds a public-interest dimension. The Dutch judiciary made the hearing available in Dutch, English and Kinyarwanda and explained that the live stream was offered because of expected public interest. The official Dutch livestream page states that the planned ruling is at 10:00 CEST on 28 August 2026. A scheduled ruling in a genocide case naturally attracts international media attention, but speed creates legal risks. Headlines may collapse charges into convictions, mix the alleged attack in Gatobotobo with other events in Rwanda, or report an appeal-stage decision as if it were final. A French submission built from those headlines would be fragile. The first file should contain the court’s own judgment or procedural decision, the charge description, the dates, the identity material and the status of any appeal.
The Dutch case also illustrates the difference between a state’s obligation to investigate serious allegations and the burden required for conviction. A prosecutor can open an investigation on a reasonable suspicion. A court can order investigative measures or hold a trial after a threshold review. A conviction requires the prosecution to prove each relevant element to the standard required by Dutch law, while the accused retains the right to challenge witnesses, documents, translations and expert conclusions. France cannot use the fact that a Dutch court reached a procedural stage as proof that the alleged conduct occurred. Nor can a future acquittal be summarised as a judicial declaration that every historical allegation was false. The exact reasons, factual findings and scope of the judgment will control.
The non-extradition dimension still matters even if the Dutch court acquits. A domestic proceeding may be the only available route in the state where the suspect is located. It may also generate a record that the state of territorial origin could not safely or practically create. That record can include witness interviews, maps, photographs, call data, military or administrative records, expert reports and translations. None of those materials travels automatically to France. They become potentially useful only after a lawyer identifies how they were obtained, who controlled them, what guarantees accompanied the interviews and whether a French court can test them in an adversarial procedure.
The case should therefore be understood as a live test of the “place of prosecution” problem. Rwandan territory supplies the factual setting. The Netherlands supplies the investigating authority, the accused’s current procedural forum and the trial safeguards. France may become relevant only if it has its own jurisdictional connection, a person or witness located there, evidence that must be preserved there, or a legally sufficient basis to investigate an international crime. A French institution does not acquire power merely because the story is important to French public opinion or because a French lawyer can explain the international-crimes framework.
For victims and witnesses, the distinction has immediate consequences. They may have participated in the Dutch case, supplied information to investigators or watched the hearing from another country. They may also hold documents or memories that were not used in the Dutch trial. A French approach must ask what objective is sought: recognition of the Dutch judgment, protection of a witness, preservation of a record, support for a French investigation, a request for mutual legal assistance, or a civil claim connected with a specific French jurisdictional basis. Those objectives require different documents and different authorities. A general request to “open a case in France” is less useful than a schedule showing the facts, the person, the evidence, the Dutch procedural status and the precise French gateway relied upon.
The official allegation itself must remain accurately framed. The Dutch judiciary writes that “The defendant is charged with looting and destruction as warcrimes, and incitement to and the commission of genocide against the Tutsis in Mbazi, Rwanda, in April 1994.” That is a charge description. It does not establish that every act alleged amounts to genocide, that every participant shared the required intent, or that looting and destruction automatically form part of the same legal offence. A French analysis should preserve those distinctions. It should also use the Dutch spelling of the case name, DeSoto, consistently so that later searches do not confuse this file with a different Rwanda investigation.
B. What must a court prove when witnesses and records are abroad?
International-crimes trials are difficult because the facts are old, the physical scene has changed and many witnesses have experienced trauma, displacement or pressure. The legal qualification adds another layer. Under Article 211-1 of the French Penal Code, genocide involves a concerted plan directed at the total or partial destruction of a national, ethnic, racial or religious group, or a group defined by another arbitrary criterion, together with specified acts against members of that group. The provision lists intentional attacks on life, serious physical or psychological harm, destructive living conditions, measures intended to prevent births and forcible transfer of children. The historical label “genocide” therefore does not remove the need to prove the protected group, the conduct, the plan and the relevant intent.
A different legal route may concern crimes against humanity. Article 212-1 of the French Penal Code requires one of the listed acts to have been committed in execution of a concerted plan against a civilian population in the context of a widespread or systematic attack. The list includes murder, extermination, enslavement, deportation or forcible transfer, serious deprivation of liberty, torture, sexual violence, persecution, enforced disappearance, apartheid and other analogous inhumane acts. The article is useful as a comparison because it shows why a French judge would separate the underlying act from the contextual element. A witness who proves a seizure of property may be important, but that fact alone does not prove a genocidal plan or a widespread or systematic attack.
French criminal procedure begins from a broad evidence principle but places a firm adversarial limit on it. Article 427 of the Code of Criminal Procedure states that, outside statutory exceptions, offences may be established by any mode of proof and that the judge decides according to intimate conviction. It immediately adds that the judge may rely only on evidence brought before the court and discussed contradictorily. A document produced in The Hague may therefore be highly valuable without being self-proving in Paris. Its author, translation, collection method, completeness and relationship to the accused must remain open to challenge.
In practical terms, a French evidence file should be built in layers. The first layer is provenance: who created the document, when, for what purpose, from which device or archive and under whose custody it remained. The second is authenticity: whether the version is original, certified, digitally verified or merely a later copy. The third is translation: which language was used by the witness, whether the interpreter was independent, and whether names, dates, place names and military terms were translated consistently. The fourth is relevance: which element of which offence the material tends to prove. The fifth is contradiction: how the accused could test the material and whether any witness can be heard or questioned lawfully.
Witness evidence requires particular care. A survivor’s account can be central, but a lawyer should preserve the distinction between direct observation, information received from another person, later identification and inference. A witness who identifies a commander from a photograph years later may be reliable, mistaken or influenced by later publications; the legal answer depends on the full interview history and the surrounding corroboration. A record that places a vehicle at a location may establish presence but not the person who gave an order. A military document may show formal hierarchy but not effective control. These are not technical objections designed to delay justice. They are the steps by which a court turns historical material into findings that can withstand appeal.
The charge of genocide also makes pattern evidence important but dangerous. A court may need to examine a series of attacks, the treatment of a protected group, orders, speeches, checkpoints, detention sites, lists, transport, communications and the conduct of subordinates. Yet the existence of a genocide in Rwanda does not prove that every person accused in every locality participated in it. French counsel should map each item to the individualised conduct alleged against the defendant. The same discipline applies to war-crimes allegations: the existence of a conflict, the protected status of victims or property, the relation between the act and the conflict, and the accused’s mental element must not be assumed from a general historical narrative.
Material generated by a Dutch judgment can be used in several ways. The reasoning may identify which witnesses the court found reliable, which documents it rejected, which translation dispute mattered and which factual propositions remained unresolved. The judgment may also contain a chronology that helps investigators compare a French file with the Dutch record. But a French authority must obtain the complete decision, not rely on a media summary. It must know whether the decision is final, whether an appeal is pending, which findings are necessary to the result and whether the court made observations that were not essential to the operative part. A fact mentioned in a judgment is not automatically a fact legally established for every future proceeding.
There is a further danger in importing a foreign record without its context. An interview may have been conducted under a Dutch statutory regime, with a specific warning, interpreter and disclosure process. A French prosecutor or investigating judge must know that setting before evaluating the material. A statement may be usable as an investigative lead while not being sufficient at trial. A foreign expert report may identify a mass grave or a pattern of destruction while still requiring French-appointed expertise. A document can be authentic yet incomplete. A video can be genuine yet edited. A database can be accurate yet fail to establish who entered the data. Cross-border litigation rewards files that expose those limits rather than hiding them.
The 2023 case law of the French Cour de cassation, Assembly Plenary, appeal No. 22-82.468, 12 May 2023, offers a helpful illustration of fact-sensitive reasoning. In discussing habitual residence, the Court referred to a “un lien de rattachement suffisant avec la France
” and required an assessment of duration, reasons for presence, intention to remain and family, social, material or professional links. The point is methodological: a label is not evidence. A person’s actual connections must be proved through documents and testimony. The same approach should govern a foreign evidence file. A document is not “international evidence” in the abstract; its origin, content, reliability and connection to the alleged act must be demonstrated.
That is where the French perspective can add value to the international story. The relevant question is not whether France can pronounce on the entire history of Rwanda from a newspaper report. It is whether a specific French authority can lawfully preserve, request, analyse or litigate a specific piece of evidence linked to a defined suspect and a defined offence. The difference between those two questions is the difference between a serious cross-border file and a political statement.
II. Could France open its own case after the Dutch verdict?
A. When can France use universal jurisdiction for genocide?
French jurisdiction must be tested before French evidence strategy. The general treaty-based gateway in Article 689-1 of the Code of Criminal Procedure provides that, under the international conventions referred to in the following provisions, a person who committed an offence abroad may be prosecuted and tried by French courts if that person is found in France. The official text begins: “En application des conventions internationales visées aux articles suivants, peut être poursuivie et jugée par les juridictions françaises, si elle se trouve en France, toute personne
” who committed one of the listed offences outside the Republic. This is a gateway tied to a convention and presence in France; it is not a free-standing power to try every international allegation that attracts attention.
For genocide, other crimes against humanity and war crimes, the more specific provision is Article 689-11 of the Code of Criminal Procedure. The current version lists genocide, the other crimes against humanity defined in the Penal Code and war crimes defined by Articles 461-1 to 461-31. It requires the suspect to have a habitual residence in France, defined by a sufficient connecting link with France. It also provides that prosecution must be requested by the National Anti-Terrorism Prosecutor and that no international or national jurisdiction must be seeking the person’s surrender or extradition. A Dutch trial and a later French file must therefore be compared against the current statutory wording, the accused’s actual location and residence, and any active request. A French reader should not confuse residence in the Netherlands with habitual residence in France.
The current text matters because older case law arose under earlier versions of Article 689-11. The Cour de cassation, Criminal Chamber, appeal No. 21-81.344, 24 November 2021, addressed the then-applicable double-incrimination analysis in a Syria case. The judgment was a ruling on the French jurisdictional condition in the statutory version before later legislative changes. It cannot be quoted as though it rewrote the current text. Its continuing lesson is that jurisdictional conditions must be read from the version legally applicable to the proposed proceedings, not from a generic summary of “universal jurisdiction”.
The Assembly Plenary’s appeal No. 22-80.057, 12 May 2023, provides another historical warning. The Court dealt with the relationship between French universal jurisdiction and the absence of a request for surrender or extradition under the version then at issue. The official reasoning describes universal jurisdiction as “une alternative au mécanisme de coopération pénale qu’est l’extradition
”. The sentence is useful when analysing why a state may prosecute rather than transfer a person. It does not mean that France may choose a parallel prosecution whenever another state is already trying the same facts. The current statute, the identity of the conduct and the status of the foreign proceeding must be examined together.
The same Assembly Plenary decision also explains that, under the former double-incrimination condition, the foreign and French legal labels did not have to be identical. The companion appeal No. 22-82.468, 12 May 2023, states that the condition did not require identical qualifications and that underlying offences such as murder, rape or torture could matter. That reasoning is a valuable caution against stopping an investigation because a foreign statute uses a different name. It is not a reason to omit the current Article 689-11 analysis. The change in statutory wording, the date of the alleged conduct, the accused’s residence and any competing request remain decisive. A lawyer should cite the historical judgment with its legal version and not turn a contextual holding into a universal shortcut.
The Paris architecture is also specific. Article 628 of the Code of Criminal Procedure places crimes against humanity, war crimes and connected offences within a dedicated procedural title. Article 628-1, in the version in force from 25 July 2026, gives the National Anti-Terrorism Prosecutor, the Paris investigating judge and the Paris assize court concurrent jurisdiction for offences within Article 628, and allows the prosecutor and investigating judge to exercise their functions throughout the national territory when they are competent. That does not mean every file belongs in Paris on demand. It means that once the statutory scope and a lawful French jurisdictional basis are established, the specialised national architecture must be used.
Venue rules do not repair a missing jurisdictional gateway. Article 693 of the Code of Criminal Procedure identifies possible venues for offences committed abroad, including the accused’s residence, last known residence, place where the person is found and the victim’s residence, subject to special rules such as Article 628-1. A French witness or a French victim may be relevant to the location of an investigative act, but the existence of a French address does not by itself prove that Article 689-11 applies. The file must distinguish territorial venue, personal jurisdiction, treaty-based jurisdiction and the special conditions for international crimes.
The temporal dimension must also be handled carefully. The French law of 26 December 1964 states in its unique article that crimes against humanity, as defined by the United Nations resolution and the Charter of the International Military Tribunal, are imprescriptible by their nature. The current text is available from Légifrance’s official version of Law No. 64-1326. The historical Cour de cassation judgment, Criminal Chamber, appeal No. 75-93.296, 30 June 1976, concerned the effect of that principle and international instruments on alleged crimes committed decades earlier. A lawyer should not use the word “imprescriptible” to skip every temporal question: the offence definition, the legal characterisation, the applicable jurisdictional provision and the transitional rules still require a precise analysis.
For DeSoto, the basic scenario is therefore narrow. If the defendant remains in the Netherlands and has no habitual residence in France, Article 689-11 does not create a French trial simply because a Dutch judgment is newsworthy. If a person connected with the alleged conduct later establishes a sufficient link with France, a new jurisdictional analysis may become possible, but the facts must be documented. If French authorities have a separate French-nationality or territorial gateway, that route must be identified rather than hidden behind a general reference to universal jurisdiction. If an international or national jurisdiction requests surrender, the current Article 689-11 condition must be checked before any French prosecution is presented as an alternative.
Finally, a concluded foreign trial triggers a separate protection against duplicate prosecution. Article 692 of the Code of Criminal Procedure provides that, in the cases covered by the preceding chapter, no prosecution may be brought against a person who shows that the person was finally judged abroad for the same facts and, in the event of conviction, that the sentence was served or is time-barred. The test is not “has any Dutch judge mentioned this person?” It is finality, identity of facts and the statutory consequences of the foreign decision. That is why the operative part, factual findings and appeal status of the DeSoto ruling will matter more than the first news alert.
B. How should victims and counsel act on the evidence?
The most useful French response to the DeSoto judgment may be a disciplined evidence and jurisdiction package rather than an immediate demand for a second trial. The package should begin with an event chronology: dates, locations, alleged acts, the defendant’s role, the witnesses who saw each act, the documents that place the defendant at the relevant location and the Dutch procedural step in which each item appeared. It should then separate facts established by the Dutch court from allegations, investigative leads and matters that the judgment left open. This format lets the National Anti-Terrorism Prosecutor or another competent authority decide whether there is a French basis to act without asking that authority to reconstruct the case from press coverage.
Second, preserve the original material before making a public accusation. Keep native digital files, metadata, original photographs, letters, notebooks, audio recordings and earlier translations. Record who supplied each item and every later transfer. Do not edit a video to create a short clip for social media and then treat that clip as the original. Do not translate a witness statement several times without keeping the source language. Do not circulate a document containing personal details when a sealed or redacted version can protect a witness. A French court will care about reliability, but victims also need protection from retaliation, identification and public pressure.
Third, obtain the Dutch record through a lawful channel. Article 694 of the Code of Criminal Procedure supplies the French framework for transmitting and executing requests for mutual legal assistance when no international instrument provides a different route. It directs French requests to foreign judicial authorities through the Ministry of Justice and sets out the corresponding route for foreign requests to France, while allowing urgent direct transmission to competent authorities subject to the statutory conditions. In a France-Netherlands file, European or bilateral instruments may supply more specific mechanisms. The practical rule is to identify the correct instrument and authority before asking a court for a document, a witness interview or a certified copy.
The request should be targeted. It may seek the certified judgment, the charge sheet, the procedural status, a particular exhibit, a transcript, a translation record or permission to interview a witness. It should explain why the material is relevant to a defined French investigation and what safeguards are requested. “Send the whole case” is often slower and less persuasive than “send the final decision, exhibit 42, the chain-of-custody record for the photograph dated 25 April 1994, and the interpreter declaration for witness X.” Narrow requests also make it easier to identify whether the material can be used as evidence, as an investigative lead or only as background.
Fourth, map the accused’s status. The file should record nationality, current residence, habitual residence, travel, family and professional connections, any arrest warrant, any surrender or extradition request, and whether the Dutch proceeding is pending, final or under appeal. The 2023 Cassation case on habitual residence shows why those facts cannot be replaced by a single address. The Court treated residence as a “faisceau d’indices
” rather than a checkbox. In a French submission, rental records, tax documents, employment, school enrolment, family life and the reasons for presence may all be relevant, but none is conclusive in isolation.
Fifth, give the French prosecutor a charge-by-charge analysis. For a genocide theory, use Article 211-1 and identify the protected group, the alleged acts, the concerted plan and the evidence of intent. For crimes against humanity, use Article 212-1 and identify the civilian population, the widespread or systematic attack and the connection between the individual acts and the plan. For war crimes, identify the conflict, the protected persons or property, the act, the relevant law of armed conflict and the relation to the conflict. The fact that Dutch prosecutors used a charge heading does not determine the French qualification. Conversely, the fact that a French lawyer prefers a different qualification does not invalidate the Dutch case.
Sixth, protect the defence side of the record. A French file that includes only inculpatory material may be challenged as selective and may produce avoidable delay. Keep prior inconsistent statements, exculpatory documents, alternative explanations for travel or presence, and any Dutch finding that a witness or exhibit was unreliable. The presumption of innocence is not a public-relations sentence. It changes how the evidence must be described, how a report is drafted and how a lawyer speaks to authorities. It also protects victims: a carefully tested case is more likely to result in findings that remain credible after an appeal.
Seventh, distinguish investigation from enforcement. France may be asked to preserve a witness’s statement, transmit a request, analyse a document or assess jurisdiction. None of those steps authorises France to enforce a Dutch sentence or to detain a person outside French territory. If the Dutch court convicts, the judgment may have consequences in the Netherlands and may be relevant to cooperation elsewhere, but it does not become a French conviction by translation. If the court acquits, the reasons and finality must be examined before anyone says that France is barred from all inquiry. Article 692 supplies a specific final-judgment rule; it is not a slogan about foreign proceedings.
Eighth, use the Paris procedural route only when the threshold is met. Article 628-1 places the relevant prosecution and investigation functions within the Paris national framework, but Article 689-11 still requires its own conditions. A report should therefore identify the legal gateway in its first pages, explain why the suspect’s location and residence satisfy it, state whether any international or national request exists, and show why the proposed French investigation does not duplicate a final foreign judgment. If those facts are missing, a lawyer may still advise on preservation or mutual assistance, but should not promise that a French criminal case will open.
The prosecutor’s decision-making framework also matters. Article 40-1 of the Code of Criminal Procedure describes the options available when facts reported to the prosecutor appear to constitute an offence and no legal obstacle prevents public action: prosecution, a legally available alternative procedure or dismissal when particular circumstances justify it. International-crimes files involve additional statutory requirements and institutional actors, but the principle remains practical. A report should make the next decision easier by identifying the alleged offence, the person, the dates, the evidence, the jurisdictional link, urgent risks and the requested procedural act.
There is a useful internal comparison for readers who want to understand the evidence problem without treating one case as a precedent for another: the firm’s French universal-jurisdiction analysis of overseas evidence explains why preservation, individual attribution and the jurisdictional link must be separated. The DeSoto file is different in facts and forum. The comparison is useful because it keeps the central discipline visible: a historical crime, a foreign defendant and an international headline do not remove the need to prove who did what, where, when and under which legal gateway.
Victims should also decide what outcome they seek from France. They may want to assist a French investigation, obtain protection, have a witness heard, preserve a document, challenge a decision not to proceed or understand whether a French person or asset creates a separate route. Each objective affects the relevant deadline, authority and supporting material. A victim who has already testified in the Netherlands should preserve the testimony’s exact status and not assume that a second statement will improve the case. A new interview can clarify a point, but it can also create an apparent inconsistency if the earlier wording and interpretation are ignored.
Media organisations and civil-society groups face the same obligation to separate source material from legal conclusions. Publishing the Dutch court’s schedule is different from publishing a claim that the defendant committed genocide. Publishing a witness’s account is different from proving the account. Publishing a French jurisdictional theory is different from showing that Article 689-11’s residence and prosecutorial conditions are satisfied. A careful source trail protects the public record and makes it possible for a court, prosecutor or defence team to verify what was actually said.
The likely next step after the ruling is therefore not a single French headline but a three-column review: what the Dutch court found, what the Dutch court did not decide, and what France could lawfully do with the remaining material. If the judgment is appealed, the status column changes. If a witness moves to France, the evidence and protection columns change. If a new French jurisdictional fact emerges, the legal analysis changes. If no French gateway exists, France may still cooperate with the Netherlands while declining to open its own prosecution. That is not inaction; it is the allocation of criminal authority between states.
Conclusion
The DeSoto ruling is important because it turns a 1994 Rwandan fact pattern into a present-day test of domestic accountability, evidence transfer and the limits of universal jurisdiction. The Dutch court’s planned date and charge description are verified procedural facts; the outcome must not be anticipated. Whatever the result, France will need the complete decision, its final status and the underlying evidential record before drawing consequences. A Dutch trial may produce material that helps a French authority preserve evidence, assess a suspect’s role or decide whether a statutory gateway exists. It does not, by itself, create French jurisdiction, a French conviction or an automatic arrest duty.
The French route is built from exact elements: a lawful jurisdictional basis under Articles 689-1 or 689-11, a qualifying offence under the Penal Code, the Paris procedural architecture where applicable, reliable and challengeable evidence, a lawful cooperation request and a careful check against a final foreign judgment under Article 692. The most responsible cross-border strategy is consequently both ambitious and restrained: preserve what may disappear, request what can be obtained lawfully, document the connecting facts and describe the allegation as an allegation until a competent court decides.
We can help you preserve evidence, assess French jurisdiction and choose the right cross-border procedure.