KOHEN AVOCATS FIRM · PARIS 16
Nord Stream Sabotage Warrants Split Europe: Italy Surrenders, Poland Refuses, Croatia Holds the Next 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

Nord Stream Sabotage Warrants Split Europe: Italy Surrenders, Poland Refuses, Croatia Holds the Next Suspect

Speak to a lawyer 06 89 11 34 45

On 19 August 2026, German federal prosecutors announced that a second Ukrainian suspect in the Nord Stream pipeline sabotage had been arrested in the Croatian city of Pula and would be extradited to Germany. The man, identified as Vladimir Z., is accused of helping to plant the subsea explosives that tore through the Nord Stream 1 and 2 gas pipelines in the Baltic Sea on 26 September 2022. His arrest, reported by Reuters from Frankfurt, comes less than a year after two European courts gave opposite answers to the same German request: in October 2025, a court in Bologna authorised the surrender of the alleged mastermind Serhij K., while a court in Warsaw refused to hand over Volodymyr Z. and set him free. One pipeline, one European Arrest Warrant system, three national answers. For anyone who travels with an arrest warrant hanging over them, the lesson is stark: within the European Union, surrender is not automatic, and the court of the country where you are arrested decides your fate. This article reconstructs what Germany alleges, explains why Italy surrendered and Poland refused, follows the German Federal Court of Justice ruling of 15 January 2026 that kept the surrendered suspect in custody, and then gives the view from Paris: if a German sabotage warrant landed on the desk of a French chambre de l’instruction, what would French judges test before handing anyone over.

I. One Pipeline, Three Courts: Why Italy Surrendered While Poland Refused

A. Can a German Arrest Warrant Reach Saboteurs Arrested in Italy, Poland and Croatia

The facts alleged by the Office of the German Federal Public Prosecutor General (Generalbundesanwalt) are unusually detailed for a case still at the investigation stage. Because the acts, in the German analysis, may have undermined the internal security of the Federal Republic of Germany, and given the particular importance of the case, the investigation is conducted at federal level. Seven Ukrainian nationals are under investigation. According to the prosecutors’ findings, the group chartered a yacht, sailed into the Baltic Sea in September 2022 and placed explosive devices on the Nord Stream 1 and 2 pipelines. Several explosions damaged the two pipelines so badly that gas could no longer be transported from Russia to Europe. The arrest warrants issued by the Federal Prosecutor General charge anti-constitutional sabotage committed concurrently with causing an explosion and with destruction of buildings and structures, under Sections 88, 305 and 308 of the German Criminal Code. That legal characterisation matters far beyond Germany, because every executing court in Europe must fit those German charges into its own surrender test.

The instrument Germany used is the European Arrest Warrant, created by Council Framework Decision 2002/584/JHA of 13 June 2002 on the European arrest warrant. In French domestic law, the definition is written in terms that mirror the Framework Decision: “Le mandat d’arrêt européen est une décision judiciaire émise par un Etat membre de l’Union européenne, appelé Etat membre d’émission, en vue de l’arrestation et de la remise par un autre Etat membre, appelé Etat membre d’exécution, d’une personne recherchée pour l’exercice de poursuites pénales ou pour l’exécution d’une peine ou d’une mesure de sûreté privative de liberté.” That sentence, codified at Article 695-11 of the French Code of Criminal Procedure, contains the whole philosophy of the system: a judicial decision, issued in one Member State, executed in another, for prosecution or for enforcement of a sentence. But the same chapter that defines the warrant also organises the refusals, and the Nord Stream cases show how much room those refusals leave to national judges.

The Italian chapter of the story concerns Serhij K., described by German investigators as a mastermind of the operation. Arrested in Italy on the German warrant, he was the subject of a surrender order by the Bologna Court of Appeal. On 15 October 2025, the Italian Supreme Court (Corte di Cassazione) overturned that first order. The defect was procedural but revealing: the Bologna court had reclassified the offence described in the German warrant so as to apply the tighter procedural rules reserved for terrorist cases in extradition proceedings. The Supreme Court held that the executing authority had overstepped its role by rewriting the issuing authority’s legal characterisation. After remittal, the Bologna Court of Appeal reordered K.’s surrender to Germany on 23 October 2025. This time the court cured the procedural defect identified by the Supreme Court, but it did not substantively revisit the defence’s core objections: functional immunity, ne bis in idem and the risk of a violation of Article 3 of the European Convention on Human Rights. A second appeal was dismissed by the Italian Supreme Court on 19 November 2025, which confirmed the surrender order, and K. was effectively surrendered to Germany on 27 November 2025, as reported by Tagesschau. The Italian message is therefore a narrow one: defects in reasoning can quash a surrender, but once the reasoning is repaired, Italian courts will not turn themselves into a trial court for immunity or human-rights defences.

The Polish chapter went the opposite way. On 17 October 2025, the Regional Court in Warsaw refused to extradite 46-year-old Ukrainian Volodymyr Z. and lifted his pre-trial detention. Two pillars supported the refusal. First, double criminality: any action taken, the court reasoned, would have occurred within the framework of a just defensive war waged by Ukraine, so the suspect had committed no crime under Polish law, and the condition of Article 607r section 1(1) of the Polish Code of Criminal Procedure was not met. Second, jurisdiction: Germany, in the Warsaw court’s view, has no jurisdiction to prosecute any natural person for blowing up pipelines lying in international waters; only an international tribunal adjudicating armed conflicts at sea, or an ad hoc tribunal appointed by the United Nations to judge the incident, could do so. Polish Prime Minister Donald Tusk endorsed the outcome politically, stating that “the case is closed now” and that it was not in his country’s interest to prosecute or extradite the man. Whatever one thinks of the legal reasoning, the political signal was unmistakable.

Germany’s own highest criminal court then closed the circle for the surrendered suspect. On 15 January 2026, the Federal Court of Justice (Bundesgerichtshof) dismissed Serhij K.’s complaint against the arrest order that followed his extradition from Italy (case StB 60/25, press release of 15 January 2026). The court affirmed both the strong suspicion of an offence within the Federal Prosecutor General’s jurisdiction and the risk of flight. It rejected the defendant’s arguments based on immunity and on a so-called combatant privilege that would have justified the act as a lawful operation of war. And it answered the jurisdictional objection that had persuaded Warsaw: Germany has territorial jurisdiction, the court held, because the result of the offence, the pipelines being rendered inoperable, also occurred on German territory, where the pipelines end. Jurisdiction by result is a classical criminal-law construction, and Karlsruhe applied it without hesitation to a subsea explosion hundreds of kilometres from the German coast.

B. Why the Warsaw Court Saw a Defensive Act of War Where Bologna Saw Ordinary Sabotage

The two rulings look contradictory only if the European Arrest Warrant is imagined as an automatic conveyor belt. It is not. The Framework Decision abolished the old political-offence exception for intra-EU surrenders and sharply reduced the double-criminality check, but it kept a list of mandatory and optional refusal grounds, and national courts interpret them through their own criminal cultures. The Nord Stream split runs along exactly those fault lines, and each one deserves a close look, because each one would come back in any French surrender hearing.

The first fault line is double criminality. Under Article 2(2) of the Framework Decision, thirty-two categories of serious offences, including terrorism and sabotage, are surrendered without verification of double criminality when they carry a custodial penalty of at least three years in the issuing State. Everything else can still be refused if the act is not an offence under the law of the executing State. Warsaw treated the pipeline attack as falling outside any protected category and then asked whether the conduct was criminal under Polish law at all. Its answer, that a sabotage operation carried out in the context of Ukraine’s defensive war is not a crime in Poland, collapsed the legal test into a moral and political judgment about the war. Most commentators, including the eucrim analysis by Thomas Wahl of the Max Planck Institute for the Study of Crime, Security and Law, consider that step the most fragile part of the Warsaw reasoning: double criminality normally asks whether the bare facts, stripped of context, match a domestic offence such as causing an explosion or destroying property, not whether the judge sympathises with the motive. If a Polish court can declare a bombing non-criminal because the cause was just, the executing judge becomes the arbiter of foreign wars. That is precisely the drift the Framework Decision was designed to prevent, and it explains why the German and Italian courts refused to follow Warsaw down that road.

The second fault line is the political offence. In classic extradition law outside the European Union, the exception is explicit. French law still states it plainly: “Lorsque le crime ou le délit à un caractère politique ou lorsqu’il résulte des circonstances que l’extradition est demandée dans un but politique”, extradition is not granted, under Article 696-4 of the Code of Criminal Procedure. Within the European Arrest Warrant system, by contrast, there is no political-offence refusal ground: mutual trust between Member States is deemed to exclude it. The Warsaw court’s reasoning reintroduced the political exception through the back door of double criminality, by treating the war context as negating criminality itself. The Italian courts and the German Federal Court of Justice kept the front door shut: motive does not decriminalise, and a sabotage operation ordered, as German prosecutors now allege for the first time, by Ukrainian state agencies does not become a lawful act of war simply because a war is going on. The BGH expressly rejected combatant privilege, the international humanitarian law doctrine that shields lawful combatants from prosecution for acts of war. Privilege belongs to members of regular armed forces meeting strict conditions of distinction and command responsibility, not to a covert team operating from a chartered yacht. A French court would reason in the same terms: French law knows no justification of sabotage by patriotic motive, and the Cour de cassation has never accepted that the political context of an offence erases its criminal character in surrender proceedings.

The third fault line is immunity. The defence argued that operatives acting for a State in wartime enjoy functional immunity, the ratione materiae shield that covers official acts. The BGH dismissed the claim, and rightly so under settled international law: functional immunity does not cover acts that constitute international crimes, and in any event sabotage of civilian energy infrastructure in the exclusive economic zones of third States is not an official act capable of attracting immunity in the State of prosecution. The International Court of Justice drew the boundary in the Arrest Warrant case (Democratic Republic of the Congo v. Belgium, judgment of 14 February 2002): personal immunity protects only a narrow circle of incumbents, and immunity from jurisdiction never means impunity for grave crimes. For the French practitioner, the parallel is direct: French courts apply the same distinction, and a suspect arrested in France on a German sabotage warrant could not plausibly invoke the orders of a foreign power to block surrender.

The fourth fault line is jurisdiction over acts in international waters. The pipelines were hit outside any State’s territorial sea, in exclusive economic zones. Warsaw concluded that Germany had no business prosecuting and gestured toward an international maritime tribunal or a UN ad hoc tribunal. Karlsruhe answered with the doctrine of ubiquity: an offence is committed everywhere one of its constituent elements, including its result, occurs. German criminal jurisdiction therefore attached because the pipelines terminate in Germany and the damage, the loss of a strategic energy supply route, materialised there. The French reader will recognise the exact twin of Article 113-2 of the French Criminal Code: “La loi pénale française est applicable aux infractions commises sur le territoire de la République. 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.” Add the United Nations Convention on the Law of the Sea, which gives coastal States sovereign rights but not full sovereignty in the exclusive economic zone while preserving the criminal jurisdiction States exercise under their own law over effects felt on their territory, and the German construction looks solid. No international tribunal with jurisdiction over pipeline sabotage exists, and the UN Security Council has created none; in that vacuum, national courts with a genuine connecting factor prosecute. A French chambre de l’instruction faced with a German warrant would apply the same effects doctrine without difficulty, because French law has applied it for decades to frauds, cyberattacks and explosions whose authors never set foot in France.

There remains the Croatian sequel. Vladimir Z., arrested in Pula on 19 August 2026, now waits for the Croatian courts to decide what the Italian and Polish courts decided in opposite ways. Croatian surrender proceedings will test whether Zagreb follows the Italian line of strict mutual recognition or finds its own version of the Warsaw exception. Extradition from Croatia to Germany, announced by the prosecutors as the expected outcome, would place a second Ukrainian suspect in the hands of the Federal Prosecutor General and consolidate Karlsruhe, rather than Warsaw, as the judicial centre of gravity of the case. For practitioners, the file to watch in the autumn of 2026 is therefore the Pula file: its reasoning on double criminality and on jurisdiction will either confirm the Italian-German axis or give the Polish refusal a second ally.

II. What France Would Test Before Surrendering Anyone on a Foreign Sabotage Warrant

A. Would the Double-Criminality Check Block Surrender for an Attack in International Waters

Imagine the German warrant landing in France: a Ukrainian national, resident in Paris or arrested at Roissy on a Schengen alert, wanted by the Federal Prosecutor General for the same pipeline facts. The French procedure would open before the chambre de l’instruction of the cour d’appel, the only authority competent to rule on surrender. If the person consents, the court verifies consent and specialty and rules fast: Article 695-31 of the Code of Criminal Procedure provides that the chamber rules within seven days of the appearance, and that decision is not open to appeal. If the person refuses, as any suspect in a sabotage file would, the chamber examines each refusal ground in adversarial proceedings, with a right of appeal to the Cour de cassation. That is the forum where the four Nord Stream fault lines would be argued in French terms.

Double criminality would come first, and in France it would almost certainly fail as a defence. “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.” That is Article 695-23, and the paragraph that follows neutralises the objection for serious listed offences: “un mandat d’arrêt européen est exécuté sans contrôle de la double incrimination des faits reprochés lorsque les agissements considérés sont, aux termes de la loi de l’Etat membre d’émission, punis d’une peine privative de liberté d’une durée égale ou supérieure à trois ans d’emprisonnement ou d’une mesure de sûreté privative de liberté d’une durée similaire et entrent dans l’une des catégories d’infractions prévues par l’article 694-32.” Sabotage is expressly on that French list: category 32 of Article 694-32 reads “Sabotage”, alongside terrorism, trafficking in arms, ammunition and explosives, and arson. The German charges carry far more than three years’ imprisonment, so a French chambre de l’instruction would execute the warrant without any double-criminality control at all. The Polish shortcut, declaring the facts non-criminal because the war was defensive, would be unavailable: where the list applies, the statute says the legal characterisation belongs exclusively to the issuing authority.

Even outside the list, French case law reads double criminality in abstracto, comparing incriminating principles rather than demanding identical elements. The Cour de cassation confirmed the constitutional solidity of that mechanism in a QPC ruling on Article 695-23 itself: by judgment of 29 November 2022 (no. 20-86.216, published in the Bulletin, full text on the Cour de cassation website), the Criminal Chamber declined to refer the constitutional challenge to the Conseil constitutionnel in a case that, fittingly, concerned “qui a refusé sa remise aux autorités judiciaires italiennes en exécution d’un mandat d’arrêt européen pour une partie des faits et ordonné, pour le surplus, un supplément d’information”. The ruling keeps the 695-23 machinery intact, including its list-based waiver. Applied to Nord Stream facts, the French test would be straightforward: causing a subsea explosion and destroying energy infrastructure correspond in French law to destruction by explosive substance, terrorism-related sabotage and endangerment offences, each punishable well beyond the thresholds. No French court would hold that a defensive motive strips those qualifications of their criminal character.

Jurisdiction would be the second battle, and here French law would side with Karlsruhe. The defence would argue, as in Warsaw, that the explosions occurred in international waters, outside German sovereignty, and that Germany is prosecuting a foreign armed conflict. The French answer runs through ubiquity and effects. French courts claim jurisdiction over offences whose constituent acts or results touch French territory, and they enforce foreign warrants without re-trying the issuing State’s jurisdiction except in narrow cases. Nothing in Article 695-22, which lists the mandatory refusal grounds, authorises a French chamber to refuse surrender because it doubts the German theory of jurisdiction over international waters. The mandatory grounds are closed: amnesty extinguishing a French prosecution for the same facts, a final judgment for the same facts (ne bis in idem), minority under thirteen, and discriminatory purpose. A German prosecution for pipeline sabotage fits none of them. The optional territoriality ground, where France could take the case itself, would not help either: France has no closer title than Germany to prosecute a Baltic Sea explosion, and the Framework Decision’s logic is that the State with the genuine connecting factor, here the State where the gas was meant to arrive, tries the case. A Paris chamber would therefore do what Bologna did: verify, not second-guess.

B. Can the Defence Plead a Political Act, a Foreign Trial or Inhuman Detention in Paris

The remaining defences are the ones Bologna left unexamined and Warsaw embraced: the political character of the act, the risk of unfair or inhuman treatment after surrender, and ne bis in idem. Each has a precise French answer, and counsel should prepare them with realism rather than hope.

The political-act argument would be raised first and dismissed first. Within the European Arrest Warrant system, French law provides no political-offence refusal: the closed list of Article 695-22 contains nothing of the kind, because mutual confidence among Member States excludes questioning each other’s democratic character. The contrast with classic extradition is instructive. Outside the Union, Article 696-4 bars extradition for political offences, and Article 696 recalls that the statutory chapter applies only in the absence of an applicable international convention. Inside the Union, the Framework Decision is that convention, and it abolished the exception. The Court of Justice confirmed the architecture early on in Melloni (C-399/11): the Framework Decision exhaustively harmonises surrender conditions, and a Member State cannot add refusal grounds drawn from its constitution. A suspect who argues in Paris that blowing up a pipeline was a legitimate act of Ukrainian resistance would receive the Karlsruhe answer in French dress: motive is for the trial court at sentencing, not for the surrender judge. The Polish court’s elevation of motive into a ground of non-criminality has no equivalent in French surrender law.

The human-rights argument is the serious one. Since the Court of Justice judgments in Aranyosi and Căldăraru (C-404/15 and C-607/15) and Dorobantu (C-128/18), an executing court that faces evidence of systemic deficiencies in detention conditions in the issuing State must assess whether the individual faces a real risk of inhuman or degrading treatment, contrary to Article 3 of the European Convention on Human Rights, which prohibits torture and inhuman or degrading treatment or punishment The procedure is dialogical: request supplementary information, obtain assurances, and postpone, rather than immediately refuse, unless the risk cannot be dispelled. Germany’s prison system is not the subject of any systemic Article 3 finding, and the suspect surrendered from Italy in November 2025 has been detained under the supervision of the Federal Court of Justice, which reviews detention continuously. A French chamber seized of a German sabotage warrant would therefore ask for information about the detention facility and regime, receive satisfactory answers, and surrender. The more recent E.D.L. judgment (C-699/21) allows refusal only where surrender itself would expose the person to a real risk of a serious and irreparable deterioration of health, a narrow medical exception that pipeline-sabotage facts do not trigger. Counsel who wish to run an Article 3 defence against Germany must bring concrete, individualised evidence, not generalities about high-security detention, and even then the likely outcome is postponement with assurances, not release.

Ne bis in idem would complete the defence triptych. Article 695-22 refuses execution where the person has been finally judged for the same facts by the French authorities or by another Member State than the issuing State. No final judgment for the Baltic explosions exists anywhere: Poland refused surrender but tried nobody, Italy surrendered without trying, and the German investigation is pending. A Warsaw-style refusal is not an acquittal and creates no ne bis in idem shield. Only a genuine final decision on the merits, after a full trial, in France or in another Member State, would block a German warrant, and counsel should not confuse a Polish non-surrender with a verdict. The same article’s amnesty and minority grounds are plainly irrelevant to adult sabotage suspects.

Two practical consequences follow for anyone exposed. First, movement matters: within the Schengen area, the country of arrest decides, and the spread between Warsaw and Bologna proves that the same warrant can mean freedom in one capital and a prison cell in another. A person who knows a German warrant may exist should take verified legal advice before travelling, because the alert will surface at any border check. Second, preparation matters: the French surrender hearing moves quickly, the chambre de l’instruction decides on documents, and defences built on motive, jurisdiction or immunity need written evidence, expert material on detention conditions and precise legal argument from the first appearance. Our firm has already analysed the French side of the European Arrest Warrant refusal case law in detail, including detention-based refusals and the enforcement of foreign sentences, in our earlier analysis of EAW refusal for inhuman detention and enforcement of foreign sentences in France. The Nord Stream saga confirms that analysis: surrender judges verify, they do not retry, and only concrete, individualised grounds move them.

Conclusion

The Nord Stream extradition saga is the most instructive European Arrest Warrant file of the decade, because it displays the system’s strength and its fragility in a single case. Italy applied mutual recognition as written: repair the reasoning, then surrender. Poland applied mutual recognition as feared: reintroduce the political exception through double criminality and deny the issuing State’s jurisdiction over international waters. Germany’s Federal Court of Justice supplied the doctrinal backbone, strong suspicion plus flight risk, no combatant privilege, jurisdiction by result. And Croatia, holding the suspect arrested in Pula on 19 August 2026, now holds the tie-breaking vote. For the international observer, the autumn of 2026 will show whether Warsaw remains an outlier or becomes a model. For the French lawyer, the file is already decided in principle: a German sabotage warrant executed in France would pass the double-criminality list without control, survive the jurisdictional objection through the effects doctrine, defeat the political and immunity defences at the threshold, and yield on human-rights grounds only to concrete, individualised proof that no pipeline-sabotage file currently contains. The pipelines were destroyed in one night; the litigation over who judges the saboteurs will shape European surrender law for years.

📄 Official circular