KOHEN AVOCATS FIRM · PARIS 16
Alishov v. Azerbaijan: Can a Ten-Year Travel Ban Survive a Criminal Investigation?

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
Hassan Kohen, criminal defence lawyer in Paris
Hassan KOHEN
Attorney at the Paris Bar

Alishov v. Azerbaijan: Can a Ten-Year Travel Ban Survive a Criminal Investigation?

Speak to a lawyer 06 89 11 34 45

A fresh judgment from Strasbourg has put a sharp limit on the idea that a criminal investigation can keep a person inside a country indefinitely. In Alishov and Others v. Azerbaijan, notified on 25 August 2026, the European Court of Human Rights found violations of both the reasonable-time guarantee in Article 6 § 1 of the Convention and the right to leave a country under Article 2 of Protocol No. 4. Three Azerbaijani men had been investigated for large-scale fraud, placed under police supervision and effectively prevented from leaving Azerbaijan for around ten years, even though no investigative act had been carried out for more than a decade. The Court awarded each applicant €5,500 for non-pecuniary damage and €500 for costs. The ruling is not a French judgment, and it does not make every travel restriction unlawful. It does, however, provide a powerful benchmark for any French or European case in which a cross-border suspect is kept within a territory while the file stands still. French procedure allows a judge to restrict travel under judicial supervision, but it also provides tools to seek a narrower order, a temporary exemption, a lifting of the restriction, or an end to a dormant investigation. The practical question is no longer simply whether a travel ban has a legal basis. It is whether the measure remains necessary, individually justified and proportionate today.

I. What did the ECHR decide about Azerbaijan’s ten-year travel ban?

A. Why Alishov and Others v. Azerbaijan reached Strasbourg

The judgment joins three applications: no. 25527/23, Alishov v. Azerbaijan; no. 25578/23, Huseynov v. Azerbaijan; and no. 36156/23, Abbasov v. Azerbaijan. The case was decided by a Committee of the Third Section. It was adopted on 7 July 2026 and notified in writing on 25 August 2026. Because it was delivered by a Committee under the Convention, the judgment is final, subject only to editorial revision. The applicants were Azerbaijani nationals who lived in Baku and were represented by a lawyer based in Azerbaijan.

The underlying criminal proceedings were separate proceedings against the three applicants. They were instituted in 2015 under Article 178.3.2 of the Azerbaijani Criminal Code for fraud causing large-scale damage. The second and first applicants were remanded in custody on 15 May and 3 November 2015. The third applicant was placed under police supervision on 25 May 2015. When the first two men were released in 2016, police supervision remained in place. In practice, the preventive arrangement prevented all three from leaving Azerbaijan.

The chronology mattered more than the seriousness of the accusation. The first applicant was released on 21 October 2016 and the second on 30 May 2016. The criminal cases then remained open. The judgment records that no investigative actions had been carried out since the relevant period of release. In 2023, the applicants challenged the length of the proceedings and sought an end to the preventive measure. The domestic courts rejected their complaints: final decisions were given by the Baku Court of Appeal on 5 June 2023 for the first and second applicants and on 14 September 2023 for the third.

The applications to Strasbourg therefore did not concern a single brief airport alert. They concerned a continuing restriction attached to unresolved criminal proceedings. The Court treated the arrangement as a restriction on the right to leave Azerbaijan, even though the domestic authorities described the measure as police supervision. That functional approach is important for international cases. A state cannot avoid proportionality review by giving a measure a label that does not describe its practical effect.

The Court first examined the length of the criminal proceedings under Article 6 § 1. Its conclusion was direct: “the lengths of the proceedings were excessive”. The Court had no factual or legal reason to depart from that conclusion. The proceedings had remained pending for years, and the Government had not identified progress capable of explaining the delay. The finding does not mean that a complex fraud file must be closed on a fixed calendar date. It means that complexity must be connected to real investigative work and a reasonable procedural timetable.

The applicants also relied on Article 2 of Protocol No. 4, which protects freedom of movement and the right to leave any country. The Government argued that Article 9.3.1 of the Azerbaijani Migration Code supplied a legal basis for temporarily restricting the right to travel when a preventive measure applied. The applicants responded that the provision was discretionary and could not itself justify a restriction that lasted for years. The Court did not need to resolve the legal-basis dispute in the abstract. It held that the restriction failed the Convention requirements for other reasons, particularly its exceptional duration and the absence of progress.

The Court reviewed the familiar proportionality sequence. A restriction must be connected to a legitimate aim, such as maintaining public order or preventing crime. It must also be necessary in a democratic society. A state may have a legitimate interest in ensuring that a suspect remains available to investigators or the trial court. That interest does not create a permanent presumption that the person will flee. It must be supported by current facts and reassessed as the file evolves.

Article 41 produced a modest but concrete remedy. Each applicant received €5,500 for non-pecuniary damage and €500 for costs and expenses, with payment due within three months and default interest thereafter. The amounts are less important than the reasoning behind them. Strasbourg treated a long restriction on leaving one’s country as a distinct harm, not as a technical inconvenience absorbed into the general complaint about delay.

The judgment should also be read with procedural precision. It did not decide that the applicants were innocent of fraud. It did not order an acquittal, erase the Azerbaijani investigations, or impose a universal maximum period for travel restrictions. It found that the state had failed to maintain a fair balance after keeping the restrictions in place throughout the proceedings without showing investigative progress or special circumstances that justified the continuing interference.

B. Why “complex financial fraud” did not justify an open-ended restriction

The Court’s most useful passage concerns the Government’s reliance on the complexity of the alleged fraud. The Government argued that the proceedings involved complex financial fraud. Strasbourg accepted that complexity can matter when a court assesses a reasonable period. It rejected the idea that complexity, by itself, can explain an entire decade of immobility. The judgment says that “it was disproportionate to restrict the applicants’ freedom of movement” for around ten years.

That reasoning separates two questions that are often collapsed. The first is whether investigators may need time to collect records, trace money, obtain foreign evidence or analyse a network of participants. The second is whether the suspect must remain physically available in the country during every month of that work. The first may be answered yes while the second is answered no, or answered yes only with a narrower condition. A court must explain why the person’s departure would create a concrete risk that cannot be managed by reporting duties, a bond, a fixed address, a passport arrangement, a return undertaking or a limited territorial restriction.

The case also rejects an inactivity paradox. If the authorities do not act, the absence of progress cannot automatically become an argument for maintaining the restriction. A dormant file does not become more dangerous merely because it has grown older. On the contrary, the passage of time may weaken the connection between the original risk assessment and the present reality. Witnesses may already have been heard, records may have been secured, the person may have complied with every summons, and the alleged flight risk may have changed.

The Court referred to earlier cases in which restrictions lasting more than eight years had been capable, by their duration alone, of becoming disproportionate. It also noted that no special circumstances had been identified, such as previous attempts to flee justice, deliberate obstruction, or conduct that caused the proceedings to be delayed. The absence of those facts did not automatically require release at the start of the investigation. It did remove the factual foundation for treating the restriction as exceptional for an indefinite period.

The reasoning is particularly relevant to cross-border financial cases. Fraud investigations routinely involve bank accounts, corporate structures, digital records, beneficial ownership, tax material and witnesses in several states. The international character of the evidence can support a carefully designed restriction at an early stage. It can also create the opposite obligation: the court should explain which investigative act is pending, why the person must remain in the territory for it, and when the restriction will be reviewed. A generic reference to foreign cooperation is not a substitute for a timetable.

There is a second distinction between a travel restriction and detention. The applicants were not kept in prison for the whole period. They were under a preventive measure that left them outside custody but stopped them from leaving the country. A state cannot assume that the lower intensity of police supervision makes the duration irrelevant. The right to leave a country is protected in its own terms. A person who can work and sleep at home may still be unable to take a job abroad, visit family, attend a hearing, manage a company, or respond to an international emergency.

The judgment therefore offers a useful audit for any order that uses geography as security. A court should identify the risk in terms that can be tested: a demonstrated attempt to evade proceedings, a missing address, a particular relationship with foreign assets, an unresolved identity question, or a real risk of interference with a witness. It should identify the counterweight: the person’s residence, family, employment, compliance history, financial guarantees, legal representation and willingness to return. It should then select the least restrictive measure that can manage that risk.

For a person facing a cross-border investigation, the evidence to preserve is not limited to the order itself. The file should show every summons answered, every reporting obligation completed, every request for permission to travel, every passport deposit and every act of investigation of which the defence was informed. That chronology can demonstrate that the restriction is no longer serving its original purpose. It can also expose a mismatch between the reasons given at the beginning and the facts available when the court reviews the order.

Alishov also matters beyond Azerbaijan because it treats movement, delay and investigation as one proportionality problem. A court cannot review the right to leave in isolation from the state of the criminal file. Nor can it review delay without asking what burdens the delay has placed on the person. A ten-year restriction attached to active proceedings may present a difficult case. A ten-year restriction attached to no investigative act presents a much simpler one.

The decision has a final institutional message. Domestic judges remain the first decision-makers. Strasbourg intervenes after domestic remedies have been used and after the state has had the opportunity to correct its own process. That makes a timely national application important. A defendant who waits until a case reaches Strasbourg may obtain a finding and compensation, but not necessarily the immediate travel document, job opportunity or family visit that could have been requested through a domestic modification application years earlier.

II. How should French courts and defendants respond to a long-running cross-border restriction?

A. What French judicial control allows—and what a person can challenge

French criminal procedure begins from a different premise than the existence of a permanent travel ban. Article 137 of the Code of Criminal Procedure states: “Toute personne mise en examen, présumée innocente, demeure libre.” It then allows judicial-control obligations where the needs of the investigation or a security measure require them. Detention is described as exceptional when judicial control or electronic home detention would be insufficient. The structure matters: liberty is the starting point, and the burden is on the measure to remain justified.

Article 138 gives the judge of investigation or the judge of freedoms and detention a menu of obligations. The first is geographical: “Ne pas sortir des limites territoriales déterminées par le juge d’instruction ou le juge des libertés et de la détention.” Other obligations can require the person to remain at a specified residence except on stated terms, avoid particular places, report movements beyond set limits, attend periodic appointments, or surrender identity documents. A prohibition on leaving metropolitan France, a departmental boundary or a named area can therefore be a lawful judicial-control obligation. Its lawful existence does not answer whether its continued application is justified.

The order must be read line by line. A person may remain under judicial control while only one obligation, such as the territorial restriction, is amended. The return of a passport may be dealt with separately. A temporary permission to travel may not remove a continuing obligation to answer summonses or report to a police service. The practical document is the operative part of the order, together with its notification and any receipt for a passport or identity card. An informal assurance from an investigator or a lawyer’s telephone call cannot safely replace a written modification.

The judge has continuing power to adapt the measure. Under Article 139 of the Code of Criminal Procedure, the judge of investigation may, at any point, impose new obligations, remove all or part of the existing ones, modify them, or grant an occasional or temporary exemption. The exact expression “à tout moment” is important for a case affected by changing facts. A person does not have to wait for the entire investigation to end before requesting a targeted change. The request should identify the destination, dates, purpose, accommodation, return arrangements and safeguards that make the trip controllable.

Where the case no longer requires judicial control, Article 140 allows the lifting of the measure at any time, either on the judge’s own initiative, on the prosecutor’s submissions or at the person’s request after the prosecutor has been consulted. The judge must decide on the request within five days by a reasoned order. If the judge does not decide within that period, the person can refer the request directly to the chamber of investigation, which has its own statutory period. This is a procedural route, not a licence to travel while the request is pending.

The risk of ignoring the current order is serious. Article 141-2 permits an arrest or escort warrant and a referral to the judge of freedoms and detention for pre-trial detention when the person voluntarily evades judicial-control obligations. A departure that the person believes is reasonable can still be treated as a breach if the written order has not been changed. The safest sequence is to secure an express dispensation or modification, confirm notification and passport status, and preserve proof of compliance with all remaining obligations.

Where a prosecutor appeals a partial lifting of judicial control, the same document-first method applies: identify the obligation removed, the scope of the appeal and the practical status of the passport before any departure. The firm’s analysis of a prosecutor’s appeal against a lifting of a French travel restriction addresses that narrower procedural sequence.

French case law shows that territorial restrictions are used in practice but are not abstract labels. In Cass. crim., 7 August 2024, no. 24-83.215, the Court of Cassation ordered release under judicial control with a territorial condition limited to the Alpes-Maritimes, subject to exceptions for judicial summonses, alongside other obligations. In Cass. crim., 18 December 2024, no. 24-85.799, the published decision records a judicial-control order that barred the person from leaving metropolitan France and required reporting and identity-document obligations. These decisions do not establish that every national restriction is proportionate. They demonstrate that the court’s wording, territory, exceptions and associated safeguards matter.

A person can challenge a decision through the appeal mechanism in Article 186 of the Code of Criminal Procedure. The provision expressly covers decisions under Articles 139 and 140 and sets a ten-day period after notification or service. The calculation should be made from the procedural notification, not from a press report or a conversation with a relative. The appeal should identify the exact obligation challenged, the current state of the investigation, the person’s compliance history and the less restrictive safeguards proposed.

A cross-border alert is a separate issue. A French judicial-control order is not automatically an Interpol Red Notice, an extradition request or a European Arrest Warrant. Conversely, an Interpol notice does not replace the national legal process that permits arrest or surrender. For an EU request, Article 695-26 sets out how a European Arrest Warrant is transmitted and executed in France, including the role of the territorially competent prosecutor general and the treatment of privilege or immunity. The person’s existing judicial-control obligation and any foreign request must be analysed together but not conflated.

This distinction becomes critical when a defendant travels through several jurisdictions. Leaving France in breach of an order may lead to a domestic enforcement response even if the destination has no interest in the foreign investigation. A foreign police alert may create a separate detention risk even if the French order is lifted. The defence should obtain the current French order, check whether a national or Schengen signal exists, identify any foreign request and verify whether the person is subject to a passport surrender or reporting obligation that survives a territorial modification.

B. How to attack stagnation: reasonable time, evidence, and proportionality

Alishov’s strongest French application is the combination of a movement challenge with a stagnation challenge. The investigation is not a background fact. It is the reason given for the restriction, so its progress must be measurable. Article 81 of the Code of Criminal Procedure requires the investigating judge to carry out the information acts considered useful to establish the truth and to investigate both against and in favour of the person. A file that remains open without a concrete investigative purpose raises a different problem from a file in which foreign evidence is actively being collected.

French procedure contains a route for asking the judge to bring an information to an end. Article 175-1 allows the person under examination, the assisted witness or the civil party to request a referral to the trial court, an indictment or a declaration that there is no case to answer, including a severance where appropriate. The request may be made when no investigative act has been completed for four months. The judge has one month to accept the request or issue a reasoned decision that the information should continue. If the judge does not decide, or rejects the request, the applicant can refer the matter to the president of the chamber of investigation within five days under the statutory procedure.

This mechanism is not identical to a motion to lift a travel restriction. The two requests serve different purposes. The first seeks a decision about the dormant information. The second seeks to remove or narrow a security obligation. They can nevertheless reinforce each other. If the prosecutor argues that the travel restriction is needed because the investigation is active, the defence can request the acts said to justify that assertion. If the file has had no act for months, the Article 175-1 route makes the lack of activity a formal procedural issue rather than a rhetorical complaint.

Article 175-2 states that “la durée de l’instruction ne peut excéder un délai raisonnable” in light of the seriousness of the allegations, the complexity of the necessary investigations and the exercise of defence rights. If the information has not ended after two years, the judge must issue a reasoned order explaining its length, the reasons for continuing it and the prospects for resolution; that order must be renewed every six months. A ten-year investigation without acts would demand a particularly clear explanation under this logic, even though the article does not itself create an automatic dismissal at the two-year mark.

The French framework also requires the defence to separate a restriction from detention. Article 144-1 says that pre-trial detention cannot exceed a reasonable period in light of the alleged facts and the complexity of the investigations, and requires immediate release when the statutory conditions are no longer met. That text directly concerns detention, not every judicial-control obligation. Its proportionality language is still a useful reminder that the passage of time and investigative complexity must be assessed in concrete terms. A party should not cite detention rules as if they automatically terminate a travel restriction; the correct claim should be anchored in Articles 138, 139, 140, 175-1 and 175-2.

Recent French authority illustrates why the argument must be fact-specific. In Cass. crim., 24 June 2025, no. 25-82.919, the Court of Cassation upheld the refusal to lift a judicial-control obligation that had prohibited a person from leaving metropolitan France for nearly five years. The decision records the person’s foreign connections, the penalty faced and the fact that the information was still in progress. Its lesson is not that five years is always acceptable. Its lesson is that a court may maintain the restriction where its reasons are tied to current facts and the procedural stage. The same judgment therefore helps the defence: a bare reference to gravity or foreign ties should be tested against the actual state of the file.

An older but still useful reference is Cass. crim., 10 July 2013, no. 13-82.981, which concerned an application to lift a prohibition on leaving the national territory. The decision shows that the court looks at guarantees of representation, including family, residence, employment and economic ties, rather than treating the existence of an overseas connection as conclusive. The defence should therefore present positive facts, not only criticism: a stable address, a precise itinerary, a fixed return date, a professional or medical reason, secured representation, a financial guarantee where appropriate, and consent to a reporting appointment on return.

The chronology should also identify who caused each period of delay. A defendant cannot rely on a reasonable-time argument in the same way if the defence repeatedly requested adjournments, failed to appear or obstructed the investigation. The prosecution, however, cannot transform ordinary defence rights into evidence of flight risk. An application should distinguish delays caused by foreign authorities, delays caused by the investigating office, delays caused by expert work and delays caused by the defence. It should attach proof rather than offer a general assertion that “nothing has happened”.

Cross-border evidence calls for a similarly exact approach. The order should identify the foreign request or investigative act that allegedly requires the person’s presence. Is the person expected to give evidence? Is there a risk of contact with a witness? Is a bank account or a company based abroad? Could a summons by video, a return undertaking, a bond, a limited permission or an address for service solve the problem? If the answer is yes, the court should be invited to replace a country-wide restriction with that targeted safeguard.

The defence should not overread Alishov. Strasbourg did not create a rule that every restriction exceeding eight or ten years automatically violates Article 2 of Protocol No. 4. It assessed a particular combination: an effective ban on leaving the country, criminal proceedings open for around a decade, no investigative action for more than ten years, and no special flight or obstruction circumstances. A French court would examine its own record. But Alishov makes it harder to defend a long restriction with a formula that never changes as the file becomes older.

The prosecution also has a practical interest in a reasoned review. A carefully limited travel permission can secure cooperation better than a blanket ban. It can include a fixed return date, advance notice of itinerary, an address abroad, a continuing French lawyer, remote attendance, a guarantee or a specified reporting appointment. If a person is permitted to travel under conditions that are clear and verifiable, a later breach can be identified more reliably. Proportionality is therefore not only a defence argument; it is a way to make international cooperation workable.

For a case in Paris or Île-de-France, the correct court and clerk’s office must be identified from the information itself, not guessed from the person’s home address. The file may be before the investigating judge, the judge of freedoms and detention, the chamber of investigation or another court after referral. The notification date controls the appeal period. A request that is addressed to the wrong registry, omits the operative order or seeks a general “right to travel” instead of a precise modification may lose time that the statutory deadlines do not restore.

The final safeguard is behavioural. Until the written order is amended, the person must comply with it. A pending appeal, a press statement, a promise that a foreign border will not check the passport, or a belief that the investigation has gone quiet does not remove the obligation. The person should seek an express written decision, keep the receipt for every deposited document, attend every summons and record every request for permission. That discipline converts the Strasbourg principle into evidence that a French judge can use today.

Alishov therefore supplies a clear legal test for long-running international files: identify the legitimate aim, identify the current risk, identify the investigative act still required, measure the burden on movement, and explain why a narrower alternative cannot work. If the court cannot answer those questions, the restriction may have outlived its legal purpose even if the original order was valid.

Conclusion

Alishov and Others v. Azerbaijan does not abolish travel restrictions in criminal investigations. It establishes something more practical: a state must keep explaining why a person still cannot leave, especially when the criminal file has stopped moving. Serious allegations and international evidence can justify controls. They cannot, without more, turn a preventive measure into a decade-long substitute for a decision on the merits.

French law already contains the tools needed to apply that principle. Articles 138 to 140 permit a territorial obligation to be narrowed, temporarily suspended or lifted; Article 186 provides an appeal route; Articles 175-1 and 175-2 address dormant or excessively long investigations; and the Court of Cassation’s decisions show that current risks and guarantees of representation must be examined. The winning application is factual: it identifies the exact restriction, proves compliance, maps the delay, explains the cross-border need and proposes a workable alternative.

📄 Official circular