On 16 September 2026, at 10:00 CET in The Hague, the Kosovo Specialist Chambers sentenced former Kosovo president Hashim Thaçi to 25 years in prison for war crimes committed as the top commander of the Kosovo Liberation Army during the territory’s violent break from Serbia in the 1990s. Two of his co-accused received 25 and 18 years, the third 13 years, while all four men were acquitted of six counts of crimes against humanity because the judges found that the prosecution had failed to prove a widespread or systematic attack against the civilian population. Thaçi, 58, stood in silence in a suit and tie as Presiding Judge Charles Smith read out the sentence, having denied every charge. Thousands of supporters watched the verdict live on a giant screen in Pristina, stunned by a penalty almost nobody in Kosovo had expected, while scuffles broke out between demonstrators and police outside the tribunal in The Hague and Albania’s prime minister posted a single word: “Unbelievable.” This article explains, from the standpoint of a French criminal practice that handles extradition, surrender and universal-jurisdiction files every week, what the judges decided and why, what happens next on appeal and enforcement, and what the verdict changes for suspects, fugitives and victims who find themselves on French soil. The first part dissects the judgment itself. The second part sets out what French courts can and cannot do with the same facts: prosecute them in Paris under universal jurisdiction, surrender a suspect to an international court, or enforce a foreign sentence.
I. Why former Kosovo president Hashim Thaçi was sentenced to 25 years for war crimes
A. Which charges of murder, torture and illegal detention did the judges retain, and why did the crimes against humanity counts fail?
The case the judges decided on 16 September 2026 is The Specialist Prosecutor v. Hashim Thaçi, Kadri Veseli, Rexhep Selimi and Jakup Krasniqi, the flagship trial of a court most of the public had never heard of before this week. The indictment had been confirmed on 26 October 2020 and amended three times, most recently on 30 September 2022, with the operative version filed on 27 February 2023. Each of the four accused faced six counts of crimes against humanity — persecution, imprisonment, other inhumane acts, torture, murder and enforced disappearance — and four counts of war crimes — illegal and arbitrary arrest and detention, cruel treatment, torture and murder. After their arrest, the four men were transferred to the detention facilities of the Specialist Chambers in The Hague on 4 and 5 November 2020, where three of them have now spent almost six years in custody. The trial itself opened on 3 April 2023, the prosecution closed its case on 15 April 2025, counsel for the 156 participating victims presented their case in July 2025, and closing statements ran from 9 to 18 February 2026 after 134 witnesses had testified, 125 called by the prosecution, two by victims’ counsel and seven by the defence. Those bare figures matter for anyone facing an international prosecution: from indictment to verdict, the proceedings lasted nearly six years, and the evidentiary record was so voluminous that the Trial Panel formally extended its deliberations to produce a fair, comprehensive and reasoned assessment of a voluminous evidentiary record.
On the substance, the verdict draws a sharp line that every defence lawyer in cross-border cases should study. The judges convicted the accused of war crimes — including murder, torture and illegal detention committed as senior commanders of an armed group — and the factual findings reported from the courtroom are strikingly precise: arbitrary and illegal detention of 385 people, torture of 303 people and cruel treatment of 49 people, with the court stressing that none of the victims was actively participating in hostilities at the time. Presiding Judge Charles Smith summarised the prosecution’s proven theory of a joint criminal enterprise as a plan he called “as simple as it was harmful”: detention facilities set up across every KLA-controlled area, identification of anyone perceived as opposing the movement, then arrest, detention and, where deemed necessary, killing. In other words, the panel accepted that detention centres across KLA-controlled areas were used to lock up perceived opponents — including political rivals, suspected collaborators and members of other ethnic groups — and that detainees were tortured and in some cases killed. That finding explains the headline sentences and also explains why the defence narrative of a clean liberation war collapsed in the courtroom.
Yet the same panel threw out all six crimes-against-humanity counts against all four accused, holding that the prosecution had failed to show that these crimes formed part of a widespread or systematic attack directed against the civilian population as a whole. The distinction is not academic hairsplitting; it is the central fault line of international criminal law. A war crime punishes a serious violation of the laws of war committed in connection with an armed conflict against protected persons, while a crime against humanity requires, in addition, that the acts form part of a broader attack that is either widespread in scale or systematic in organisation. The judges found the first element proven beyond reasonable doubt and the second not proven, so the same killings, tortures and detentions were punished as war crimes but not as crimes against humanity. For practitioners, the lesson is concrete: in The Hague as in Paris, the legal characterisation of mass violence turns on context evidence — orders, structures, scale, patterns — and a prosecution that proves the bodies but not the system loses the heaviest counts. The acquittal on these counts also caps the historical narrative the judgment can carry: the court did not declare that the KLA ran a systematic campaign against civilians, only that its commanders ran a network of abusive detention sites whose victims the law of war protected.
Two further features of the trial deserve attention because they recur in every major atrocity case France handles. First, the judges openly stated that the proceedings unfolded against severe witness intimidation, recalling witnesses who recanted earlier statements, lied to the court or rewrote their previous accounts on the stand. That finding echoes the French experience in comparable files, where diaspora pressure, clan loyalties and fear routinely hollow out testimony years after the facts, and it explains the panel’s reliance on documentary material, insider evidence and victim testimony that survived cross-examination. Second, the judgment was delivered in a court that applies Kosovo law with international judges and sits in the Netherlands precisely because witnesses and indictees could not be tried safely in Pristina — a structural admission, built into the design of the Specialist Chambers, that some trials can only be fair when they are physically removed from the territory where the crimes occurred. Anyone who assumes that distance from the crime scene weakens a case should read this verdict: relocation did not prevent convictions, and the six-year detention of the accused before judgment shows how seriously international judges take the risk of flight and interference.
B. Can Thaçi appeal the 25-year sentence, and where would he actually serve it?
The individual penalties announced on 16 September 2026 are severe by any European standard: 25 years for Hashim Thaçi, 25 years for Jakup Krasniqi, 18 years for Kadri Veseli and 13 years for Rexhep Selimi, with the judgments for Veseli and Selimi expressly including credit for the time already spent in custody since November 2020. Thaçi resigned the presidency in November 2020 immediately after the indictment was confirmed — he had served as president from 2016 to 2020 after earlier stints as prime minister and foreign minister — and he has therefore already spent almost six years behind bars before even reaching a first-instance verdict. Under the Specialist Chambers’ procedural framework, modelled on international criminal procedure, the convicted men have a full right of appeal, and the verdict itself states that they may challenge both conviction and sentence before the Court of Appeals Panel. An appeal would re-examine the legal characterisation of the facts, the evaluation of intimidated witnesses and, crucially, the sentence, because 25 years for commanders who denied everything and showed no remorse sits at the upper end of what comparable tribunals impose for detention-centre cases without a proven extermination policy. Defence teams will almost certainly attack the joint-criminal-enterprise findings and the reliability of recanting witnesses, while the prosecution may cross-appeal the acquittals on crimes against humanity, so the final word in this saga is still years away and the four men are likely to remain in detention in The Hague throughout the appellate phase.
Where a confirmed sentence would actually be served is a separate question that draws directly on the French lawyer’s daily practice of sentence enforcement across borders. The Specialist Chambers have no prison of their own: like the International Criminal Court, they depend on agreements with cooperating states willing to host convicted persons, and enforcement follows the tribunal’s designation of a state of enforcement after the judgment becomes final. For the families and for states, the designation matters enormously, because conditions of detention, family visits, early-release regimes and even the language of prison life differ from one host country to another. French readers should understand the mirror image of this mechanism, because France regularly receives foreign convicts for enforcement and transfers its own nationals abroad under bilateral and European arrangements, and the same logic of judicial supervision, credit for pre-trial detention and conditional release applies. Practically, this means that the 25-year figure announced in the courtroom is a starting point, not a calendar: time already served counts, appellate review can reduce it, and enforcement-state law will govern the remainder. Anyone advising indictees before international courts should therefore negotiate and litigate with the enforcement phase in mind from day one, because the state where a client will grow old in prison is often as important as the number of years pronounced.
The political aftershock is the third dimension of the verdict, and it explains why this story leads every international bulletin this week. In Pristina, thousands of people who revere the KLA as the army that liberated Kosovo watched the sentence in disbelief, shouting slogans for Thaçi and the guerrilla movement; in The Hague, police scuffled with hundreds of supporters who tried to march on the tribunal while listening to an audio stream of the judgment; and Albania’s prime minister publicly branded the outcome “Unbelievable.”. The anger is inseparable from the asymmetry many Kosovars perceive: Serbian forces killed more than 10,000 civilians, many of them women and children, during the 1998-1999 war that left an estimated 13,000 dead and hundreds of thousands displaced, yet only a handful of Serbian commanders have ever faced trial, while four of the most revered Albanian figures now carry 25, 25, 18 and 13-year sentences. International lawyers must take that grievance seriously without confusing it with a legal argument: selective-enforcement complaints, however legitimate politically, do not acquit an accused whose own crimes are proven, and the panel’s careful acquittal on crimes against humanity shows a bench that refused to convict beyond the evidence even under intense public scrutiny. The deeper precedent is institutional: a court created by a constitutional amendment of 3 August 2015, grafted onto every level of the Kosovo judiciary yet staffed entirely by internationals and seated abroad, has just demonstrated that it can convict a former head of state — the first such conviction in its short history and a signal to every hybrid tribunal that political stature is no shield.
II. Could France prosecute the same facts, surrender a suspect to The Hague, or enforce the sentence?
A. Can French courts prosecute foreign war crimes and crimes against humanity committed abroad?
The short answer is yes, under strict conditions that the Cour de cassation has clarified in two landmark plenary rulings of the last three years — and the Thaçi verdict is the perfect illustration of when Paris can act and when it must step aside. French universal jurisdiction over atrocity crimes rests on article 689-11 of the Code of criminal procedure, which provides that “peut être poursuivie et jugée par les juridictions françaises toute personne soupçonnée d’avoir commis à l’étranger” genocide, crimes against humanity or war crimes. Three locks guard that door, and each of them speaks directly to files like the Kosovo saga. First, the suspect must have a habitual residence in France, defined as a sufficient connecting link assessed through the duration of presence, its conditions and reasons, the person’s stated will to settle or remain, and family, social, material or professional ties — “La personne soupçonnée doit avoir une résidence habituelle sur le territoire français”, in the exact words of the statute. Second, only the anti-terrorism prosecutor in Paris may initiate the prosecution: “La poursuite de ces crimes ne peut être exercée qu’à la requête du procureur de la République antiterroriste”. Third, and most decisively for competing international cases, France must defer whenever another court claims the suspect: the prosecutor has to verify that the International Criminal Court is not already prosecuting and that no competent international court has requested surrender and no state has requested extradition — “si aucune juridiction internationale ou nationale ne demande la remise ou l’extradition de la personne”. In plain terms, if a Kosovo commander lives quietly in Lyon and neither The Hague tribunal nor any state claims him, Paris may prosecute; the moment an international court demands his surrender, the French file yields.
The Cour de cassation has given those locks real teeth in decisions every cross-border practitioner should know by number. On 12 May 2023, the full Assembly of the court, in judgment no. 669 dismissing appeal no. 22-82.468, upheld the Paris investigation into a Syrian suspect pursued for torture, enforced disappearances and war crimes committed within Jaysh al-Islam in Eastern Ghouta, ruling “REJETTE le pourvoi” and confirming that Syrian law punished the substance of the acts through ordinary criminal offences. On the residence question — the suspect argued that three months in France for a limited stay could not amount to habitual residence — the court endorsed the investigating chamber’s finding that the evidence showed a settled stability of residence over more than three months, so that “le critère de résidence habituelle [était] ainsi rempli”. Then, on 25 July 2025, the Assembly confirmed the same architecture in judgment no. 684 dismissing appeal no. 24-84.071, again ruling “REJETTE le pourvoi” in an investigation opened for complicity in crimes against humanity and war crimes together with aggravated money laundering and criminal association. The message to suspects and victims alike is unambiguous: France’s highest court twice validated Paris investigations into atrocities committed abroad by foreigners against foreigners, provided the residence link and the prosecutor’s exclusive initiative are established. A Kosovar commander who settled in France after 1999 could therefore be investigated in Paris on exactly the legal basis the Thaçi judges applied in The Hague — but only while no international tribunal claims him.
The substantive definitions France would apply mirror the Hague categories so closely that the verdict reads like a case study for French law students. Article 211-1 of the Criminal Code defines genocide as acting “en exécution d’un plan concerté tendant à la destruction totale ou partielle d’un groupe national, ethnique, racial ou religieux” through killing, grave bodily harm and comparable acts. Article 212-1 punishes with life imprisonment any act “commis en exécution d’un plan concerté à l’encontre d’un groupe de population civile dans le cadre d’une attaque généralisée ou systématique”, listing murder, extermination, enslavement, deportation and torture among others. And article 461-1 provides that war crimes are offences “commises, lors d’un conflit armé international ou non international et en relation avec ce conflit, en violation des lois et coutumes de la guerre”, committed against protected persons or property. Read the Hague findings through that lens and the parallel is exact: the detention, torture and killing of 385, 303 and 49 non-combatants in connection with a non-international armed conflict fits the French war-crimes definition squarely, while the collapsed crimes-against-humanity counts correspond precisely to the French requirement of an attack that is “généralisée ou systématique” — the very element the Hague panel found unproven. Time does not protect perpetrators either, because article 7 of the Code of criminal procedure states that “L’action publique des crimes mentionnés aux articles 211-1 à 212-3 du code pénal est imprescriptible”. A quarter-century after the Kosovo war, a prosecution in France for genocide or crimes against humanity would therefore still be legally possible — for war crimes under articles 461-1 and following, limitation questions would need case-by-case analysis, which is exactly why early advice from a lawyer matters before memories fade and witnesses scatter.
B. How does France surrender suspects to international courts, and what defences can stop a transfer?
Suppose the suspect is not a quiet resident but a wanted man: an international court or a foreign state asks France to hand him over. Two different tracks then exist, and confusing them is the most common mistake in extradition files. The first track is cooperation with the International Criminal Court, governed by a dedicated title of the Code of criminal procedure whose opening provision states that “Pour l’application du statut de la Cour pénale internationale signé le 18 juillet 1998, la France participe à la répression des infractions et coopère avec cette juridiction”, as article 627 puts it, for acts constituting genocide, crimes against humanity or war crimes within the meaning of the Rome Statute. Surrender to the ICC — lawyers say “remise”, not extradition, because no sovereignty negotiation is involved — follows a judicial procedure before the Paris courts with limited but real avenues of challenge, and the person sought can contest identity, the scope of the warrant and the conditions of transfer. The Thaçi case shows the mechanism in action from the other side: arrested by the Specialist Prosecutor, the four accused were transferred to The Hague detention facilities on 4 and 5 November 2020, a surrender operation that required the cooperation of the arresting authorities, medical and security arrangements, and years of pre-trial detention management. Anyone arrested in France on an ICC warrant goes through the same machinery in reverse, and the first 48 hours — lawyer, interpreter, challenge to the arrest record, application for release — shape the entire file.
The second track is classic extradition to a foreign state, governed, absent a treaty providing otherwise, by the Code of criminal procedure itself: “En l’absence de convention internationale en stipulant autrement, les conditions, la procédure et les effets de l’extradition sont déterminés par les dispositions du présent chapitre”, in the words of article 696. The procedure runs in two phases that every requested person must understand: a judicial phase before the investigating chamber of the court of appeal, which examines the legal conditions and gives a reasoned opinion, and then, only if that opinion is favourable, a governmental phase ending in a decree authorising surrender, itself open to challenge before the administrative courts. Custody is the immediate battlefield, and the Cour de cassation polices it closely: on 9 August 2023, the criminal chamber, in decision no. 23-83.192, ruled “REJETTE le pourvoi” in extradition proceedings brought at Ukraine’s request, where the dispute concerned the lifting of judicial supervision ordered against the requested person. The ruling is a reminder that even when surrender is on the table, the requested person keeps concrete rights — to contest detention, to seek supervision measures, to argue health, family life and the risk of inhuman treatment in the requesting state, and to demand diplomatic assurances whose breach can later be litigated. In files connected to the Balkans, the Caucasus or the Middle East, those arguments regularly decide outcomes, because European courts will block a transfer where prison conditions, the risk of torture or the absence of a fair retrial make surrender unlawful.
Three practical defences emerge from this framework that apply well beyond Kosovo. First, speciality and double criminality: France does not surrender for conduct that is not punishable on both sides or for charges added after the fact, and the precise counts of the foreign warrant must be mapped against French definitions article by article. Second, human-rights bars: evidence that the requesting justice system tortures, that detention centres resemble the ones described in the Thaçi judgment, or that the accused would face a flagrantly unfair trial can stop an extradition even where every formal condition is met. Third, forum strategy: where France itself can prosecute under universal jurisdiction, the defence can argue that a trial in Paris — with French procedural safeguards, family nearby and no transfer — serves justice better than surrender, turning article 689-11’s subsidiarity clause from a limit into a shield. Victims have a mirror strategy: the 156 participating victims in the Thaçi trial show the power of organised civil-party action before international courts, and French law offers victims resident in France a forum to trigger investigations, claim evidence preservation and seek reparation, including against assets located in France. Whether you are the person sought or the person harmed, the decisive step is the same: instruct counsel in the requested state immediately, before the first hearing locks in statements, custody and timelines that will bind the file for years.
Conclusion
The 25-year sentence imposed on Hashim Thaçi on 16 September 2026 will be debated in Pristina, Belgrade, Brussels and The Hague for a decade, but its legal content is already clear and already useful. A hybrid court applying Kosovo law with international judges has proven that detention-centre crimes — 385 people arbitrarily detained, 303 tortured, 49 cruelly treated — can be punished as war crimes even when the prosecution cannot establish the widespread or systematic attack that crimes against humanity require. The acquittals on the six heaviest counts are not a failure of justice; they are justice working as it should, refusing to upgrade convictions beyond the evidence while still imposing some of the longest sentences in the history of hybrid tribunals. For France, the verdict lands in a legal landscape that is ready to receive it: universal jurisdiction validated twice by the full Assembly of the Cour de cassation, imprescriptible prosecution for genocide and crimes against humanity, a dedicated ICC-cooperation regime, and an extradition procedure whose judicial phase the criminal chamber supervises file by file. The appeal will now test whether the convictions survive, the enforcement phase will test where and how four ageing commanders serve their time, and the victims’ claims will test whether international sentences can carry reparation with them. What no appeal can undo is the precedent: former presidents can be indicted, detained for six years, tried for three, convicted and sentenced — and every suspect with a residence, a bank account or a transit stop in France should draw the obvious conclusion and take legal advice before history knocks on the door.