On 7 September 2026, a security review described the migrant crisis on NATO’s eastern flank as entering its fifth year: “tense and difficult but contained”, while warning that Minsk keeps adapting and probing European defences. Behind that assessment lies the first case in history in which one state accuses another of migrant smuggling before the International Court of Justice. On 19 May 2025, Lithuania sued Belarus in The Hague over what Vilnius calls a deliberate state policy of funnelling irregular migrants across its border in retaliation for European sanctions. In the summer of 2026 the case suddenly grew heavier: Poland filed a declaration of intervention on 30 June 2026, the European Union lodged written observations on 20 July 2026, and on the same day Russia filed its own declaration of intervention. The Court had already decided, by an Order of 17 July 2025, that the written pleadings would first address jurisdiction and admissibility, with Belarus filing its Memorial by 19 January 2026 and Lithuania its Counter-Memorial by 20 July 2026, and both pleadings were filed on time. For readers in France, this Baltic dispute matters directly. France applies the same European border rules, punishes people smugglers with its own criminal law, shelters asylum seekers who may have been instrumentalised, and its highest courts have drawn the line between solidarity and crime. This analysis explains what Vilnius must prove, where Minsk can still escape, and how Paris handles the same facts.
I. Can the ICJ Punish Belarus for State-Sponsored Migrant Smuggling?
A. How Lithuania Seized the Hague Court Despite the Reservation Trap
Lithuania’s vehicle is Article 20(2) of the 2000 Protocol against the Smuggling of Migrants by Land, Sea and Air, which supplements the United Nations Convention against Transnational Organized Crime. That provision is the compromissory clause: it opens the way to the Court for a dispute about the Protocol once negotiation has failed within a reasonable time and once six months have passed without agreement on organising arbitration after such a request. Both Lithuania and Belarus are parties to the Protocol, so the clause is in principle available. The difficulty is that Lithuania itself had ratified the Protocol with a reservation under Article 20(3), declaring that it did not consider itself bound by the Court’s jurisdiction under Article 20(2). According to Vilnius, the dispute arose as early as June 2021, while that reservation was still in force. Then, on 12 May 2023, Lithuania notified the United Nations Secretary-General that it was withdrawing the reservation. Belarus reacted on 31 July 2023 with what it called an interpretative declaration: Article 20(2) to (4), Minsk said, should not be read as binding Belarus to the Court’s jurisdiction against a state that withdraws such a reservation where the dispute arose or was the subject of settlement attempts before, on, or immediately after the withdrawal. Belarus framed the move as a clarification, arguing that giving a reservation withdrawal retroactive effect would unfairly favour states that lifted their reservations late over states that had accepted the Court’s jurisdiction from the outset.
Lithuania, Austria, Poland and the European Union formally objected, arguing in substance that the Belarusian text is a late reservation dressed up as interpretation, designed to modify Minsk’s treaty obligations and the effect of Article 20(4), which permits unconditional withdrawal of reservations. On 10 September 2024 Belarus extended its declaration to cover those objections and accused the European Union of biased and political assessments. Under the International Law Commission’s Guide to Practice on Reservations to Treaties, the character of a unilateral statement is determined by the legal effect its author purports to produce, so a text that seeks to exclude the Court’s jurisdiction where it would otherwise exist looks like a reservation rather than a clarification. If it is a reservation, its timing is fatal to it: Article 20(3) of the Protocol only permits reservations to Article 20(2) at signature, ratification, acceptance, approval or accession, and Belarus ratified back in 2003 without one. A late reservation is only possible where the treaty allows it or where no other contracting state objects, and here Austria and Poland expressly objected, with the European Union adding that the text is not even a reservation to Article 20(2) but effectively an impermissible reservation to Article 20(4). An impermissible reservation is null and void and devoid of any legal effect, while Lithuania’s own withdrawal is governed by the general rule that a reservation may be withdrawn at any time with full effect for the provision concerned, which suggests Article 20(2) became fully applicable between the two states from May 2023.
The remaining fight is temporal: can a 2023 withdrawal found jurisdiction over a dispute that crystallised in 2021? Belarus will argue that jurisdiction must have existed when the dispute arose and that retroactive consent offends good faith. Vilnius will answer that what counts is the existence of a dispute when the Court is seised, not the date of the underlying facts, provided the compromissory clause contains no explicit time limit. The Court’s recent case law gives both sides ammunition. In its judgment of 12 November 2024 on preliminary objections in Armenia v. Azerbaijan, the Court tied the temporal scope of its jurisdiction under the CERD compromissory clause to the date the Convention obligations took effect between the parties, meaning the later date on which the applicant became bound, and declined jurisdiction over earlier acts. But Judges Tomka and Charlesworth stressed in separate opinions that compromissory clauses covering any dispute normally carry no temporal condition, apply to whatever dispute exists when proceedings start, and look backwards absent an explicit limitation. That reading follows a century of authority: the Permanent Court said in Mavrommatis that doubtful jurisdictional grants cover disputes submitted after their creation and that the frequent treaty reservations for pre-existing disputes prove an explicit limitation is needed, and the Court confirmed in the Bosnia Genocide case that the Genocide Convention clause contains no time bar against hearing a dispute about earlier acts, so long as the dispute concerns the Convention and existed when the Court was seised. Lithuania’s strongest distinction is that, unlike staggered treaty accessions, the Smuggling Protocol was in force for both parties throughout: withdrawing the reservation merely activated a suspended jurisdictional mechanism for an existing dispute, while the substantive obligations bound Minsk at the time of the facts, which keeps Article 28 of the Vienna Convention on the Law of Treaties out of the picture for the merits.
The summer of 2026 then internationalised the case. Invoking Article 63 of the Statute, Poland declared intervention on 30 June 2026 and Russia followed on 20 July 2026, each as a party to the Protocol claiming that its construction is in question, with Moscow listing Articles 2, 3, 4, 6, 10, 11, 12, 15, 16 and 20. The Registry’s rule is straightforward: whenever the interpretation of a convention binding other states is at stake, each of those states may intervene, and the interpretation adopted in the judgment will then bind the intervener as well. Lithuania and Belarus were invited to file written observations on the Russian declaration under Article 83 of the Rules of Court. In parallel, the European Union furnished observations under Article 43(2) and Article 69(2) of the Rules, the procedural channel through which a public international organisation supplies information. The Court must now decide the jurisdiction and admissibility phase and rule on the Polish and Russian declarations, and whatever construction it gives of the Protocol will bind every intervener. That is why this case, filed by a small Baltic state, is watched far beyond Vilnius. Readers following the Court’s docket may also read our analysis of the September 2026 hearings on Germany’s preliminary objections in Nicaragua v. Germany and what France takes from that jurisdiction battle.
B. Did Minsk Really Commit Migrant Smuggling Under the Palermo Protocol?
Lithuania’s factual story is detailed. It describes a systematic state policy beginning in late 2020 and escalating afterwards, presented as retaliation for European Union sanctions imposed after the contested 2020 presidential election and the forced landing of a Ryanair flight in Minsk, with public statements by President Lukashenko accusing the European Union of waging hybrid war. State-linked entities allegedly did the logistics: the tourist operator Tsentrkurort, the airline Belavia and state-run hotels lured migrants with advertised tour packages and simplified visas, while designated hotels in Minsk housed them and organised transport carried them to the Lithuanian border. The Application then accuses Belarusian state officials directly, notably the ASAM border-guard unit, of escorting migrants to the frontier, handing over tools and instructions to breach it, and in some cases forcing them across, with violent mistreatment of those who tried to turn back. On that basis Vilnius alleges many breaches of the Protocol: actively facilitating, supporting and enabling migrant smuggling, failing to take the required border measures to prevent and detect smuggling and to secure travel documents, failing to exchange information, to cooperate with Lithuanian border agencies and to warn potential migrants against criminal networks, and failing to preserve and protect migrants’ rights and give them proper assistance. The alleged provisions are Articles 10, 11, 11(6), 12, 15 and 16: information exchange, border measures, document security, prevention, cooperation and protection of smuggled migrants. The harm pleaded is twofold: serious injury to Lithuanian sovereignty, security and public order, and grave injury to the smuggled migrants themselves, exposed to grave abuse while trying to reach Lithuanian territory.
The legal question is narrower than the outrage. The Protocol exists, in the words of its Article 2, to prevent and combat migrant smuggling and to promote cooperation among states parties to that end, while protecting the rights of smuggled migrants. It defines migrant smuggling as procuring the illegal entry of a person into a state of which the person is neither national nor permanent resident, in order to obtain a financial or other material benefit directly or indirectly. Three filters follow from Article 4, read with the parent Convention: the conduct must be an offence the state is required to criminalise under Article 6, it must be transnational in nature, and it must involve an organised criminal group. Each filter is contested terrain. The financial or other material benefit is normally the smuggler’s fee paid by or for the migrant, which separates smuggling from humanitarian help and from trafficking, where people are moved against their will for exploitation. Lithuania treats the benefit condition briefly, as a device to exclude humanitarians and family helpers, but Minsk’s lawyers will ask where the state apparatus obtained a financial benefit: geopolitical leverage is not a fee. Vilnius answers with state-owned companies selling packages, visas and flights, plus scholarship and preparatory works, yet proving that ministries, an airline and a border-guard unit together form an organised criminal group seeking material benefit is the evidentiary mountain of the case. A further caution comes from legal scholarship on the file: many acts Lithuania calls instrumentalisation are not unlawful as such, since states remain free to set generous visa conditions and no rule of international law obliges them to block transit migration, while bad intentions alone cannot make otherwise lawful conduct illegal. Issuing visas generously is lawful; escorting people to a fence with wire cutters and pushing them across is a different matter, and the case will turn on the coercive end of the spectrum, not the consular end.
Lithuania’s pleaded breaches then divide into failures to prevent and detect (Articles 11 and 12, border measures and travel documents), failures to cooperate (Articles 10, 11(6) and 15, information and law-enforcement cooperation), and failures to protect and assist migrants (Article 16). Russia’s intervention list overlaps almost exactly, adding Articles 2, 3, 4, 6 and 20, which signals that the definitional core, the criminalisation duty and the compromissory clause itself will all be construed with erga partes effect. For France, the interest is concrete: the same Protocol binds the European Union, and French criminal law expressly extends punishment to facilitation committed on the territory of another state party to the Schengen Convention or to the Palermo Protocol. The comparison with the other frontline docket is instructive: where Vilnius sues a state, Paris prosecutes networks, as our report on the Vietnamese smuggling network dismantled with France through a joint investigation team shows. The Hague case will decide whether the Protocol, written against mafias, can also catch a ministry.
II. What France Does When Weaponised Migrants Reach Its Borders
A. Who France Punishes and Who France Protects in a Smuggling Case
French criminal law punishes the facilitator and shields the migrant, and it does so in terms that mirror the Palermo Protocol. Article L.823-1 of the Code of entry and residence of foreigners and asylum law provides: “Sous réserve des exemptions prévues à l’article L. 823-9, est puni de cinq ans d’emprisonnement et de 30 000 euros d’amende le fait, pour toute personne, de faciliter ou de tenter de faciliter, par aide directe ou indirecte, l’entrée, la circulation ou le séjour irréguliers d’un étranger en France. Les dispositions du présent article sont applicables y compris lorsque les faits sont commis par une personne se trouvant sur le territoire d’un autre État partie à la convention signée à Schengen le 19 juin 1990.” Article L.823-2 extends the same five years and 30,000 euros to facilitation committed on the territory of another Schengen state or of another state party to the Palermo Protocol against migrant smuggling, with the Protocol branch applying from the Protocol’s publication in the Official Journal. Against that broad offence, Article L.823-9 carves three blocks of immunity. The first two cover close family: ascendants, descendants, siblings and spouses, including the spouse’s family and unmarried partners living openly in a marital situation. The third, the humanitarian clause (L.823-9), provides that prosecution is barred for “De toute personne physique ou morale lorsque l’acte reproché n’a donné lieu à aucune contrepartie directe ou indirecte et a consisté à fournir des conseils ou accompagnements juridiques, linguistiques ou sociaux, ou toute autre aide apportée dans un but exclusivement humanitaire.”
That clause is the legislative child of the Constitutional Council’s landmark decision on solidarity. Asked whether the old Article L.622-4 respected fraternity, the Council first recognised fraternity as a constitutional principle from which flows everyone’s freedom to help others for humanitarian reasons regardless of residence status. It then balanced that freedom against the constitutional objective of safeguarding public order through the fight against irregular immigration, and censored the exclusion of circulation aid by declaring the words limiting immunity to residence aid contrary to the Constitution. Finally it imposed a broad conforming interpretation, requiring the provision to cover any other act of help given for a humanitarian purpose. Parliament drew the consequences in the law of 10 September 2018, and the Court of Cassation enforced them. In the Herrou case, the criminal chamber held on 12 December 2018 (No. 17-85.736) that “que ne peut donner lieu à des poursuites pénales, sur le fondement des articles L. 622-1 à L. 622-3, l’aide à la circulation et au séjour irréguliers d’un étranger lorsque l’acte reproché n’a donné lieu à aucune contrepartie directe ou indirecte et a consisté à fournir des conseils ou accompagnements juridiques, linguistiques ou sociaux, ou toute autre aide apportée dans un but exclusivement humanitaire ;” and it applied the new, milder law retroactively to pending prosecutions. It confirmed the method on 26 February 2020 (No. 19-81.561): “Il résulte du premier de ces textes, dans sa version issue de la loi n°2018-778 du 10 septembre 2018, qu’est accordé le bénéfice de l’immunité pénale à toute personne physique ou morale ayant apporté une aide à la circulation ou au séjour irrégulier d’un étranger lorsque l’acte reproché, ne donnant lieu à aucune contrepartie directe ou indirecte, a consisté à fournir une aide apportée dans un but exclusivement humanitaire.” The same judgment censured reasoning that denied immunity for lack of spontaneity, individual action or militant motive, circumstances the Court called inoperative against an exclusively humanitarian purpose, while recalling that any judgment must contain reasons capable of justifying the decision. The extended offence is set by Article L.823-2 of the foreigners code, and the Council’s ruling here: Decision No. 2018-717/718 QPC of 6 July 2018.
France also separates smuggling from trafficking, exactly as the Protocol system requires. Smuggling serves the migrant’s border crossing for profit; trafficking exploits the person. Article 225-4-1 of the Criminal Code defines trafficking as: “La traite des êtres humains est le fait de recruter une personne, de la transporter, de la transférer, de l’héberger ou de l’accueillir à des fins d’exploitation dans l’une des circonstances suivantes :” including constraint, authority, and crucially “Soit par abus d’une situation de vulnérabilité due à son âge, à une maladie, à une infirmité, à une déficience physique ou psychique ou à un état de grossesse, apparente ou connue de son auteur ;” as well as exchange of payment or advantage. An instrumentalised migrant pushed through a forest fence at night is the textbook vulnerable person, and a French prosecutor faced with Belarus-style facts would weigh trafficking charges alongside smuggling charges, with the migrants treated as victims and witnesses rather than offenders. Our French-language analysis of the domestic offence, networks and immunity is here: délit d’aide, filières de passeurs et immunité humanitaire devant la chambre criminelle. The lesson for the Hague case is symmetrical: Vilnius itself pleads the migrants’ suffering, and any court that finds state smuggling must keep the smuggled at the centre as victims, which is precisely what Article 16 of the Protocol demands. The definition of trafficking is here: Article 225-4-1 of the Criminal Code.
B. Can France Refuse You Entry If Another State Sent You?
Instrumentalisation does not extinguish the right to seek asylum in France, and French border procedure is built to examine claims individually even in a mass influx. Article L.352-1 of the foreigners code provides: “La décision de refuser l’entrée en France à un étranger qui se présente à la frontière et demande à bénéficier du droit d’asile ne peut être prise que dans les cas suivants :” followed by an exhaustive statutory list, including cases where another state is responsible under the Dublin III Regulation. The logic is that of non-refoulement: the fact that Minsk, or any other capital, organised your journey says nothing about whether you personally fear persecution, and French law requires an individual examination before any refusal. Article L.352-1 is the shield at the gate. Behind it stands refugee status itself: Article L.511-1 recognises refugee quality by reference to persecution for freedom, the United Nations High Commissioner for Refugees mandate and the Geneva Convention definitions, which the French asylum office and the asylum court apply file by file.
Between arrival and decision, France uses the waiting zone, and here judicial control is tight. A foreigner held at the border is placed in a waiting zone, and Article L.342-1 provides: “Le maintien en zone d’attente au-delà de quatre-vingt-seize heures à compter de la décision de placement initiale peut être autorisé, par le magistrat du siège du tribunal judiciaire statuant sur l’exercice effectif des droits reconnus à l’étranger, pour une durée qui ne peut être supérieure à huit jours.” In practice, the first four days belong to the administration, any extension belongs to a judge who verifies that rights were effective: interpreter, lawyer, doctor, communication, and the possibility of filing asylum from inside the zone. Article L.342-1 is the provision families call about when a relative is blocked at Roissy. If entry is refused, remedies run to the administrative judge in extreme urgency, and if admission as an asylum seeker follows, the file goes to the asylum office with an appeal to the asylum court, where instrumentalisation background can support vulnerability findings rather than discredit the claim. Detention beyond the legal caps, a refusal without individual examination, or the return of a trafficking victim without protection measures are the three arguments Paris lawyers raise first, and they mirror Lithuania’s own Hague pleading that instrumentalised migrants suffer grave abuse.
France therefore plays both roles that the Hague case separates: it represses smugglers with an extraterritorial offence and it receives the smuggled with an individual procedure. Poland’s intervention in the Hague proceedings shows the frontline playbook of documenting, litigating and legislating at once, and France’s Mediterranean and eastern exposures give Paris the same incentives. For anyone caught in this machinery, a relative held in a waiting zone, an asylum seeker refused at the border, a volunteer worried about prosecution, or a trafficking victim afraid to testify, the decisive steps are fast: lawyer within hours, asylum request in writing before any removal, medical evidence of coercion preserved, and every notification deadline counted in hours not days. The procedure rewards the quick and punishes the late, which is why early advice changes outcomes.
Conclusion
Lithuania v. Belarus is the most ambitious attempt yet to turn the Palermo Protocol against a state that allegedly wields migration as a weapon. Vilnius must first survive the reservation battle: its 2023 withdrawal against Minsk’s 2023 interpretative declaration, with the Armenia v. Azerbaijan temporal precedent and the Mavrommatis line pulling in opposite directions. It must then prove the hardest element in the Protocol, that a state apparatus acted for financial or other material benefit as an organised criminal group, rather than merely behaving cynically within its sovereign migration powers. The construction the Court adopts will bind Poland and Russia as interveners and guide the European Union’s own enforcement, which is why the summer 2026 interventions matter as much as the 2025 filing. France will not be a party, but French law already answers the questions the case raises: broad punishment of facilitators including acts abroad, constitutional immunity for genuine humanitarians, trafficking protection for the coerced, individual asylum examination at the border and judicial control of waiting zones. Whatever The Hague decides on jurisdiction, admissibility and the merits, that French balance of repression and fraternity is the standard against which Europeans will judge the judgment.