KOHEN AVOCATS FIRM · PARIS 16
Three EU Courts, Three Different Answers on the Same German Warrants: Would Paris Surrender a Nord Stream Suspect?

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

Three EU Courts, Three Different Answers on the Same German Warrants: Would Paris Surrender a Nord Stream Suspect?

Speak to a lawyer 06 89 11 34 45

On 24 September 2026, a Croatian court approved the extradition to Germany of a Ukrainian man arrested in the northern town of Pula on a German European Arrest Warrant for the undersea explosions that tore through the Nord Stream 1 and Nord Stream 2 gas pipelines on 26 September 2022. The suspect, identified by German prosecutors as Vladimir Z. and described as a trained scuba diver, was reportedly working on a Hollywood dramatisation of the attack when he was arrested in August. His lawyer, Ljubo Pavasovic Viskovic, immediately announced an appeal, so the ruling is not final and it remains unclear when, or whether, the surrender will take place. The case is the latest chapter of the most divisive extradition saga in recent European criminal justice: the same German investigation has already produced three opposite answers from three European Union courts. In October 2025, the Regional Court in Warsaw refused to hand over a Ukrainian suspect and set him free, while the Italian courts authorised, then confirmed after remittal, the surrender of another suspect who was transferred to Germany in November 2025. In January 2026, the German Federal Court of Justice upheld arrest and jurisdiction in Karlsruhe. The question this article answers is the one an international reader, a suspect, or a foreign colleague would ask a French criminal lawyer today: if a Nord Stream suspect were arrested in France, what exactly would a Paris court test before surrendering him to Germany, and why would the French answer likely differ from the Polish one?

I. Would France Surrender a Nord Stream Suspect to Germany?

The short answer is that a Paris chambre de l’instruction would apply a far narrower test than the Warsaw court did, and that nearly every ground on which Poland refused the German warrant is unavailable, or much harder to use, under French surrender law. Understanding why requires walking through the procedure exactly as it would unfold in Paris, then through the double-criminality question that decided the Polish case.

A. How Does a European Arrest Warrant Work in a Paris Court?

A person arrested in France on a German European Arrest Warrant is brought before the procureur general, who notifies the warrant, and then before the chambre de l’instruction of the cour d’appel, which alone decides on surrender. The hearing before that chamber has a strictly defined purpose. In a decision of 29 July 2026, the criminal chamber of the Cour de cassation restated it in these terms: “L’audition de la personne recherchée en vertu d’un mandat d’arrêt européen par la chambre de l’instruction, en application de l’article 695-30 du code de procédure pénale, ne vise qu’à constater son identité, à recevoir ses observations sur le déroulement de la procédure dont elle fait l’objet et à lui permettre de consentir ou non à sa remise, de sorte que cette formalité est préliminaire à tout débat.” The chamber does not retry the German case. It verifies identity, hears observations, records consent or refusal, and checks the legal conditions for execution.

Time limits are short and strict. Article 695-31 of the Code of criminal procedure provides that where the person consents to surrender, the chambre de l’instruction rules within seven days of the appearance, and that decision cannot be appealed. Where the person refuses consent, the chamber rules within twenty days, and that decision can be challenged before the Cour de cassation by the procureur general or the person sought. In a sabotage case of this magnitude, refusal of consent is certain, so the Paris timetable would be twenty days plus cassation review, exactly the path the Romanian case decided on 29 July 2026 followed through the Paris chambre de l’instruction, a first cassation ruling, a rehearing on remand, and a second rejection of the appeal.

The refusal grounds a Paris court may use are exhaustively listed by statute, and the list is what separates French surrender law from the Polish reasoning. Refusal is mandatory only in the cases set by article 695-22 of the Code of criminal procedure: amnesty in France, a final decision for the same acts by the French authorities or those of another Member State with the sentence executed or no longer enforceable, minority under thirteen at the time of the acts, and discrimination. On that last point the statute provides: “S’il est établi que ledit mandat d’arrêt a été émis dans le but de poursuivre ou de condamner une personne en raison de son sexe, de sa race, de sa religion, de son origine ethnique, de sa nationalité, de sa langue, de ses opinions politiques ou de son orientation sexuelle ou identité de genre, ou qu’il peut être porté atteinte à la situation de cette personne pour l’une de ces raisons.” Beyond these cases, article 695-24 of the Code of criminal procedure lists six optional grounds: ongoing or discontinued French proceedings for the same acts, French nationality or residence combined with enforceability of the sentence in France, commission of the acts wholly or partly on French territory, extraterritorial offences France could not itself prosecute abroad, a final third-state decision for the same acts, and limitation where France could have prosecuted.

What is strikingly absent from both lists is the political-offence exception. In classic extradition outside the European Arrest Warrant, article 696-4 of the Code of criminal procedure still provides that extradition is not granted where the offence has a political character, and in particular “lorsqu’il résulte des circonstances que l’extradition est demandée dans un but politique”. That shield, which has structured French extradition law for over a century and which readers of this firm’s extradition analyses will recognise from classic France-to-third-state files such as the Megaupload extradition file examined here, does not exist in European Arrest Warrant proceedings. The Framework Decision that created the warrant abolished the political-offence exception between Member States, and French transposition follows it. A Paris court therefore cannot refuse a German warrant on the ground that sabotaging a Russian gas pipeline in wartime was a political act. It must fit any refusal into articles 695-22 or 695-24, or into the fundamental-rights clause of article 1(3) of Framework Decision 2002/584/JHA. That single structural difference already explains most of the gap between Warsaw and Paris.

B. Can a Paris Judge Check Double Criminality for Pipeline Sabotage?

The Warsaw refusal turned on double criminality: the Polish court held that conduct carried out within a just defensive war on behalf of Ukraine was not a crime under Polish law, so the condition of article 607r paragraph 1 point 1 of the Polish Code of Criminal Procedure was not met. In Paris, that door is largely closed before it is even opened. Article 695-23 of the Code of criminal procedure states the principle plainly: “L’exécution d’un mandat d’arrêt européen peut également être refusée si le fait faisant l’objet dudit mandat d’arrêt ne constitue pas une infraction au regard de la loi française.” But it immediately adds the exception that carries the whole European Arrest Warrant system: no double-criminality check applies where the conduct is punishable in the issuing State by at least three years of imprisonment and falls into one of the listed categories of offences. The same article reserves the legal characterisation of the facts and the applicable penalty exclusively to the issuing judicial authority.

That list is set by article 694-32 of the Code of criminal procedure, and pipeline sabotage sits inside it twice over: terrorism appears at point 2 and sabotage expressly at point 32, alongside trafficking in arms and explosives, organised crime, and offences within the jurisdiction of the International Criminal Court. The German charges described in the European legal press fit comfortably: the Federal Prosecutor General is pursuing anti-constitutional sabotage concurrently with causing an explosion and destroying structures, under sections 88, 305 and 308 of the German Criminal Code, each carrying penalties well above the three-year threshold. A Paris court would therefore, in principle, execute the German warrant without any double-criminality review at all, and it would be bound by Germany’s characterisation of the facts.

Even if a Paris chamber did examine correspondence with French law, it would find it without difficulty. Article 411-9 of the Criminal Code punishes sabotage against the fundamental interests of the nation in these terms: “Le fait de détruire, détériorer ou détourner tout document, matériel, construction, équipement, installation, appareil, dispositif technique ou système de traitement automatisé d’informations ou d’y apporter des malfaçons, lorsque ce fait est de nature à porter atteinte aux intérêts fondamentaux de la nation, est puni de quinze ans de détention criminelle et de 225 000 euros d’amende.” The same article raises the penalty to twenty years where the act is committed to serve the interests of a foreign power or foreign-controlled organisation. Alongside it, article 421-1 of the Criminal Code defines terrorist acts to include destruction, damage and explosives offences where intentionally connected with an individual or collective enterprise aimed at seriously disturbing public order through intimidation or terror. Whether characterised as sabotage, destruction by explosive substance, or terrorism, conduct consisting of chartering a yacht, diving to pipelines on the seabed and detonating charges that permanently disable them corresponds to several French offences carrying fifteen years or more. The Polish conclusion that such conduct is simply not criminal finds no equivalent in French statute, and the European Arrest Warrant system would not require the Paris court to ask the question in the first place.

Two objections sometimes raised against this analysis should be addressed precisely, because the defence teams in Warsaw, Bologna and Zagreb all raised versions of them. First, the argument that a defensive war against aggression justifies the act. French criminal law recognises justifications, but within strict statutory frames: self-defence and necessity suppose an imminent, proportionate response to an unjustified attack, assessed act by act, and combatant privilege belongs to members of armed forces in international armed conflict, not to private groups chartering civilian boats. No French court has ever extended these concepts to cover a clandestine cross-border sabotage operation planned over months, and the German Federal Court of Justice expressly rejected both immunity and combatant privilege for the surrendered suspect in its ruling of 15 January 2026. A Paris chambre de l’instruction, which does not judge the merits, would have even less room: justification defences belong to the trial court, and the issuing State’s characterisation binds the executing court for listed offences. Second, the argument that the pipelines’ Russian ownership or the absence of French victims changes the criminal analysis. Ownership affects civil reparation, not the existence of destruction and explosives offences, and French sabotage law protects installations rather than owners. On double criminality, then, the Paris answer is the opposite of the Warsaw answer, for reasons written into the statutes themselves.

II. Why Did Poland Refuse While Italy and Croatia Said Yes?

With the French test established, the divergence between the three courts that have ruled on the German warrants becomes legible, and each ruling illuminates one question a Paris court would have to decide if a suspect were arrested on French soil. Poland refused on substance, Italy surrendered after correcting procedure, Germany confirmed jurisdiction and detention, and Croatia has now joined the surrender camp subject to appeal. The three defences that split Europe map neatly onto the three controls Paris would run: justification and immunity, jurisdiction over acts at sea, and fundamental rights with limitation.

A. Does a “Just Defensive War” Defence Work Before French Judges?

The Warsaw decision of 17 October 2025 deserves to be read carefully, because it is the most thorough judicial endorsement of the defensive-war theory and therefore the measure of what Paris would refuse to follow. The Regional Court in Warsaw held, first, that any action by the 46-year-old Ukrainian suspect would have occurred within a just defensive war waged on behalf of Ukraine, so that he had committed no crime under Polish law and double criminality failed. Second, it held that Germany had no jurisdiction to prosecute any person for blowing up the pipelines, assigning jurisdiction instead to an international tribunal for armed conflicts at sea or an ad hoc tribunal appointed by the United Nations. The court lifted pre-trial detention, and Polish Prime Minister Donald Tusk declared the case closed, adding that it was not in his country’s interest to prosecute or extradite the man. As analysed by the Max Planck Institute’s European criminal law review, the ruling raised fundamental questions about the relationship between the European Arrest Warrant and public international law: competence over acts in international waters, the scope of double-criminality review, justification as an anti-war operation, immunity, the political-offence exception, and fair-trial safeguards.

The Italian sequence tested the opposite proposition: that an executing court must surrender even where the defence raises international-law objections, provided the procedure is correct. The Bologna Court of Appeal first authorised the surrender of Serhij K., identified by German investigators as a suspected mastermind. The Italian Supreme Court quashed that decision on 15 October 2025, not on justification or jurisdiction, but on a procedural ground: Bologna had reclassified the offence in the German warrant so as to apply the stricter procedural rules reserved for terrorism cases, exceeding the executing authority’s powers in what the Framework Decision makes a mandatory-surrender procedure for listed offences. On remittal, Bologna reordered surrender on 23 October 2025, curing the procedural defect without revisiting the defence’s substantive grounds of functional immunity, ne bis in idem, and risk of treatment contrary to article 3 of the European Convention on Human Rights. The Supreme Court dismissed the second appeal on 19 November 2025, and the suspect was effectively surrendered on 27 November 2025. The lesson for Paris is direct: within the warrant system, executing courts control regularity, not merits, and international-law defences that go to criminal responsibility are for the issuing State’s trial court.

Karlsruhe then closed the merits debate at the pre-trial stage. On 15 January 2026 the Federal Court of Justice dismissed the surrendered suspect’s challenge to his arrest, affirming both the strong suspicion of offences within the Federal Prosecutor General’s jurisdiction and the flight risk, and rejecting immunity and combatant-privilege arguments. For a Paris court, this German ruling would carry a specific procedural weight beyond its reasoning: French surrender proceedings do not reassess strong suspicion, and a post-surrender judicial confirmation by the issuing State’s highest criminal court makes any fundamental-rights objection based on the arbitrariness of the prosecution practically impossible to sustain.

French law adds a final lock that the Polish court never had to turn. Where a first warrant has been refused and the issuing State sends a second one, the defence cannot simply invoke the authority of the refusal. In its 29 July 2026 decision, the Cour de cassation approved a Paris chambre de l’instruction that had authorised surrender under a second Romanian warrant after refusing the first, holding that the earlier refusal concerned a distinct warrant so that, for lack of an identical object, its authority could not be raised against examination of the new one. The Court stated: “l’arrêt susmentionné en date du 29 novembre 2023 portait sur un mandat d’arrêt européen distinct, de sorte que, faute d’objet identique, son autorité ne peut être opposée à l’examen du mandat d’arrêt européen objet de la présente instance.” It added a proposition of general importance for every successive-warrant case, including any future Nord Stream file: “la procédure relative à l’exécution d’un mandat d’arrêt européen émis en vue de l’exécution d’une décision de condamnation ne constitue pas une poursuite nouvelle pour les faits objet de cette condamnation, de sorte que le principe ne bis in idem ne lui est pas applicable.” Relying on the Court of Justice’s ruling of 29 July 2024 in Breian, case C-318/24, the chamber confirmed that a final refusal in one Member State does not deprive the issuing authority of the right to maintain its request where Union law conditions are met. Applied to Nord Stream, the Polish refusal binds no one: neither Germany, which may reissue, nor Croatia, which has now ruled the other way on a second suspect, nor France, which would examine any warrant presented to it on its own terms.

B. Which Court Has Jurisdiction Over a Pipeline Blown Up at Sea?

The Warsaw court’s second pillar, denied jurisdiction, is the more technically serious objection, and it is the one a Paris defence team would press hardest. The explosions occurred near the Danish island of Bornholm, outside any State’s territorial waters, on infrastructure connecting Russia to Germany. If Germany truly lacked jurisdiction, the warrant would collapse at its foundation, and the Framework Decision’s optional refusal grounds would give Paris the tools: article 695-24 of the Code of criminal procedure allows refusal where the acts were committed wholly or partly on French territory, or where the offence was committed outside the issuing State and French law would not authorise prosecution for an offence committed abroad. The Croatian appeal will likely test the mirror provision of Croatian law, and a French court seized of the same file would have to reason through the same jurisdictional map.

The German answer, endorsed in Karlsruhe, rests on the effects doctrine: Germany has territorial jurisdiction because the result of the offence, pipelines rendered permanently inoperable, occurred on German territory where the pipelines end. French law knows exactly this reasoning. Article 113-2 of the Criminal Code provides: “L’infraction est réputée commise sur le territoire de la République dès lors qu’un de ses faits constitutifs a eu lieu sur ce territoire.” French courts routinely apply this to cross-border fraud, cyberattacks routed through foreign servers, and environmental damage originating abroad but realised in France. A Paris chambre de l’instruction asked to second-guess German jurisdiction would therefore find the German theory orthodox rather than exotic: the executing court does not verify the issuing State’s jurisdiction as a condition of surrender, and the effects-based reasoning Germany uses is the same France uses for its own prosecutions. Only if the acts had been committed partly on French territory could Paris convert jurisdiction into an optional refusal ground, and nothing in the published investigations places any constitutive act, chartering, loading, diving or detonation, in France.

Two remaining controls complete the Paris checklist, and both illustrate how narrow the executing court’s review is. Limitation first: refusal is possible where France could have prosecuted and time has run, but French case law gives limitation defences minimal purchase in cross-border files. The Cour de cassation held on 23 May 2023 that extradition requests themselves interrupt limitation, ruling: “Il résulte de ces textes que sont interruptifs de prescription les actes qui ont pour but de constater une infraction, d’en rassembler les preuves ou d’en rechercher les auteurs. Il en est de même lorsque de tels actes sont accomplis à l’étranger.” In that case, Ukrainian, Kazakh and Russian extradition requests of August 2013 had interrupted limitation for offences committed between 2005 and 2009, even though one request had been refused by diplomatic letter for its political purpose and another’s authorising decree annulled by the Conseil d’Etat on the same ground. For Nord Stream, the lesson is symmetrical: successive German warrants, Italian proceedings, and the Croatian file all constitute acts aimed at finding the perpetrators, resetting any limitation clock a defence might invoke, and a political purpose behind a request does not strip it of interruptive effect.

Fundamental rights second: article 1(3) of the Framework Decision preserves the executing court’s duty to refuse surrender where there is a real risk of inhuman or degrading treatment, in particular from detention conditions, and article 3 of the European Convention on Human Rights stands behind it. In the 29 July 2026 decision, the Cour de cassation recorded that the Paris judges had examined and set aside, in reasoning distinct from the passages challenged on appeal, the alleged violations of fundamental rights invoked against detention in Romania and the fairness of the conviction. The control is real but exacting: general criticism of a Member State’s justice system never suffices, and the defence must prove an individual, real risk. Germany’s rule-of-law record, the Federal Court of Justice’s own review of detention in January 2026, and the specialised jurisdiction of the Federal Prosecutor General would make that burden, in a Paris courtroom, close to insurmountable. Health-based objections meet the same fate unless supported by medical evidence showing that detention itself, not merely trial abroad, would cross the article 3 threshold.

That leaves the Croatian appeal as the live proceeding to watch. The defence of Vladimir Z. will presumably replay the Warsaw and Bologna arguments before the higher Croatian court: reclassification of the offence, double criminality, jurisdiction, and the political character of an attack on Russian infrastructure during an ongoing war. The first-instance approval suggests the Croatian judges, like their Italian colleagues, treat these as merits for the German trial court rather than bars to surrender. If the appeal confirms, the transfer could follow within weeks under the Framework Decision’s surrender deadlines, and Germany would then hold two suspects, one surrendered from Italy and one from Croatia, while the Polish suspect remains free in a Union where warrants circulate freely but bind no executing court. That patchwork, three answers to the same investigation, is precisely what the warrant system was designed to prevent, and precisely what it produces when courts disagree on who may judge an act of sabotage committed at sea in wartime.

Conclusion

The Croatian ruling of 18 September 2026 does not end the Nord Stream extradition saga; with the appeal announced, it extends it. But the pattern of decisions now allows a confident forecast of the French position. Arrested in France on an identical German warrant, a Nord Stream suspect would face a chambre de l’instruction that hears him under article 695-30, records his refusal within twenty days, and authorises surrender unless one of the closed statutory lists provides otherwise. Double criminality would not be reviewed, because sabotage and terrorism sit on the article 694-32 list and carry penalties far above three years, and French sabotage and terrorism law would in any event correspond to the German charges. The defensive-war justification that freed the Warsaw suspect has no procedural vehicle in Paris, where the political-offence exception does not exist for European warrants and justification defences belong to the trial court. Jurisdiction would follow the effects doctrine France itself applies. Successive warrants would not be blocked by the Polish refusal, limitation would have been interrupted by the very requests the defence attacks, and fundamental-rights objections would confront the German Federal Court of Justice’s own detention review. None of this prejudges guilt: the German trial court alone will decide whether the yacht, the divers and the explosives prove the charges, and the presumption of innocence remains intact. What it means is procedural: of the three European answers given so far, the French one would almost certainly resemble Bologna and Zagreb, not Warsaw. Anyone concerned by a German warrant in this investigation, whether named in the files or unexpectedly arrested on a trip through France, should therefore prepare the only defences a Paris court can actually hear: precise evidence on French territorial links, on individual detention risks documented medically, and on discrimination in the issuing decision, rather than the grand theories of war and jurisdiction that persuaded Poland and failed everywhere else.

📄 Official circular