June 12, 2026
by Marilena I. Katsogiannou
The intersection of alternatives to prosecution and fair trial guarantees has long occupied a contested space in European criminal procedure. In B.G. v. France (application no. 70945/17, 19 March 2026), the European Court of Human Rights (ECtHR) confronted this tension in its most acute form: a minor who had reported a sexual assault found herself the subject of a rappel à la loi – i.e. a formal reminder of the law issued by a prosecutor’s delegate, carrying an explicit finding that she had committed the criminal offence of false accusation – without having admitted the alleged facts and without any possibility of judicial scrutiny. The Court found, unanimously, a violation of Article 6 § 1 of the European Convention on Human Rights (ECHR) in its criminal limb.
The judgment is significant on two levels. First, it brings alternatives to prosecution within the criminal limb of Article 6 § 1 where they form part of an indivisible procedural sequence attributing criminal responsibility. Second, it identifies stereotyped reasoning on sexual consent as a fair trial deficiency, extending to Article 6 § 1 a line of reasoning hitherto developed under Articles 3 and 8.
Though the rappel à la loi was abolished in France by Article 14 of Law no. 2021-1729 of 22 December 2021 and replaced, from 1 January 2023, by the avertissement pénal probatoire (Article 41-1 1° of the Code de procédure pénale) – a measure now conditional upon the person’s acknowledgment of guilt and issued only by the public prosecutor or a designated delegate – the reasoning of B.G. is not confined to that measure. It speaks to any alternative to prosecution across the Council of Europe’s forty-six member States that attributes criminal responsibility outside the framework of a fair trial.
The applicant, B.G., was born in 2000 and was sixteen at the material time. On 27 May 2016, she alleged that L.A., a seventeen-year-old fellow pupil, had compelled her to perform oral sex in the school toilets by threatening to disclose their relationship to a mutual friend. She reported the incident to the police on 3 June 2016, accompanied by her mother.
When transmitting the file to the prosecutor, the police appended a note observing that the applicant ‘had not screamed, had not resisted, and had not clearly expressed her refusal’, and that she had subsequently maintained contact with L.A. by text message. A gynaecological examination on 20 June 2016 revealed no physical injury – a circumstance which, under the Court’s settled case-law since M.C. v. Bulgaria, cannot in itself be treated as indicative of consent. On 6 October 2016, the prosecutor decided not to prosecute on the ground of ‘insufficient characterisation’ of the offence.
On 6 September 2016 – a month before the prosecutor’s decision – L.A.’s mother had filed a complaint against B.G. for false accusation of rape. Following the decision not to prosecute the original complaint, on 8 August 2017, the prosecutor initiated proceedings against B.G. by way of a rappel à la loi: an alternative to prosecution under French law, by which a public prosecutor’s delegate formally reminds the person concerned of their legal obligations. In response, the applicant’s parents invoked a 2004 ministerial circular restricting the use of such measures to facts ‘simple, established, acknowledged, or at least not seriously contestable’, and requested either no further action or formal prosecution so that the facts could be examined before a court. The request was refused.
On 20 September 2017, B.G. appeared before the prosecutor’s delegate, accompanied by her parents and her lawyer. The record stated explicitly that the applicant ‘had committed’ the offence of false accusation. B.G. refused to accept the terms of the measure and maintained throughout that her original account was true. As a consequence of the rappel à la loi, she was entered into the judicial antecedents database (TAJ) for five years – the reduced period applicable to minors – with the record accessible in the context of administrative checks for certain public sector positions.
The Government raised two preliminary objections. First, it argued that the rappel à la loi did not constitute a ‘criminal charge’ within the autonomous meaning of Article 6 § 1, characterising it as a mere administrative warning falling outside the criminal limb of the Convention. Applying the Engel criteria and adopting the substantive rather than formal approach consistent with Deweer v. Belgium, the Court joined the objection to the merits. Although it did not formally enumerate which of the three Engel criteria proved decisive, the Court’s reasoning shows that the second – the nature of the offence – weighed most heavily: the measure, by its own terms, entailed a formal attribution of criminal responsibility, a feature the Government itself acknowledged (para. 44). Crucially, the Court declined to assess the rappel à la loi in isolation. It characterised the entire procedural sequence – from the decision not to prosecute the rape complaint to the issuance of the rappel à la loi – as an indivisible whole (paras. 47−48).
Second, the Government contended that the applicant had suffered no significant disadvantage within the meaning of Article 35 § 3 (b), pointing to the five-year TAJ entry as the sole consequence. The Court rejected the objection, finding that the formal judicial attribution of criminal responsibility, the entry in the police database, the continuing possibility of prosecution within the limitation period, and the potential chilling effect on future complainants collectively amounted to a significant disadvantage (para. 50).
On the substance, the Court identified four distinct deficiencies in the domestic proceedings. The Court examined the complaint exclusively under Article 6 § 1, treating that provision as lex specialis in relation to Article 13. First, the prosecutorial authorities had proceeded on the basis that the applicant had made a false accusation without adequate reasoning, despite the existence of two competing accounts and the applicant’s consistent refusal to acknowledge the alleged facts (para. 68); and they had failed to accord equal weight to the respective statements of B.G. and L.A., without explaining why the applicant’s account was considered unreliable (para. 69).
Second, and most significantly for the broader development of the Court’s case law, the reasoning applied by the police when transmitting the file – centred on the absence of screaming, physical resistance, or explicit verbal refusal – reflected a stereotyped understanding of consent incompatible with the Court’s own jurisprudence as articulated in H.W. v. France and E.A. v. France (para. 72). The Court stated expressly that ‘such an assessment does not rest on a correct understanding of the notion of consent as defined by the Court’s case-law’ (para. 72).
Third, the applicant had been deprived of the opportunity to contest the characterisation of her complaint as false before a court with full Article 6 § 1 guarantees – a possibility explicitly provided for under Article 226-10 of the French Criminal Code, as recognised in Klouvi v. France (para. 74).
Fourth, she had at no stage waived her fair trial rights: her consistent refusal to acknowledge the alleged facts precluded any inference of waiver, distinguishing the case from R. v. the United Kingdom, where the minor had admitted the relevant facts and the measure had an explicitly educational character (para. 75).
The most structurally innovative aspect of B.G. lies in the Court’s methodological choice to assess the procedural sequence as an indivisible whole rather than examining the rappel à la loi in isolation.
The orthodox position, reflected in the Court’s earlier case law, treated alternatives to prosecution as falling outside the criminal limb of Article 6 § 1 on the basis that they did not determine a ‘criminal charge’ in the Engel sense. The rappel à la loi, in particular, had been characterised in domestic French law as a mere reminder of legal obligations, carrying no formal finding of guilt. B.G. departs from this characterisation by insisting that the legal nature of a measure cannot be assessed in abstraction from its functional context. This is the natural extension of the Deweer principle that the concept of ‘criminal charge’ is autonomous and substantive rather than formal. Where a rappel à la loi is the direct and inseparable consequence of a prior decision not to prosecute – and where it carries an explicit attribution of criminal responsibility to a person who has never admitted the alleged facts – the two decisions must be assessed together. The formal label attached to the second measure is insufficient to exclude it from the guarantees of Article 6 § 1. The 2004 ministerial circular invoked by the applicant’s parents already required, as a matter of French law, that rappels à la loi be confined to facts ‘simple, established, acknowledged, or at least not seriously contestable’ – a domestic safeguard whose disregard reinforces the Court’s diagnosis of a structural, not merely conventional, deficit.
The implications of this methodological choice extend well beyond France. The critical variable identified by B.G. is not the form of the measure but whether, in its concrete application, it offers the person concerned a genuine right of opposition triggering full judicial scrutiny. Comparable measures elsewhere may survive an Article 6 § 1 challenge precisely because they preserve such a route: the cautions of England and Wales (formal police warnings conditional upon the suspect’s acknowledgment of guilt); the Dutch strafbeschikking [a prosecutorial penalty order under Article 257a of the Wetboek van Strafvordering, subject to a right of opposition (verzet) triggering full proceedings]; and the German conditional discontinuance (Absehen von der Verfolgung unter Auflagen und Weisungen) under § 153a of the Code of Criminal Procedure, which may not be imposed without the accused’s consent. Similar mechanisms exist in other Council of Europe jurisdictions: in Greek law, the conditional abstention from criminal prosecution of misdemeanours and the abstention from criminal prosecution of juveniles (Articles 48 and 46 of the Greek Code of Criminal Procedure); and in Cypriot law, the diversion programmes for children (προγράμματα αποδικαστικοποίησης) established by the 2021 ‘Children in Conflict with the Law’ (Articles 35-40 and 59 of Law 55(I)/2021), which redirect cases away from prosecution subject to the child’s meaningful participation in the relevant decisions and a return to ordinary proceedings if the programme is not pursued. Their compatibility with Article 6 § 1 will depend, in each case, on whether the person concerned retains a genuine right to contest the underlying facts before a court. The rappel à la loi, by contrast, offered none. National systems that deploy alternatives to prosecution without this minimum threshold of contestability now face a markedly strengthened Article 6 § 1 challenge.
The second dimension of B.G. that merits close analysis is the Court’s explicit identification of stereotyped reasoning on sexual consent as a fair trial deficiency under Article 6 § 1 – a notable doctrinal development.
The foundational case law on stereotyped reasoning in sexual offences – M.C. v. Bulgaria, J.L. v. Italy, X v. Cyprus, and the more recent E.A. v. France – has been developed primarily under Articles 3 and 8, engaging the State’s positive obligations to investigate and prosecute sexual offences effectively. This jurisprudence resonates with broader international standards, notably the Istanbul Convention (Articles 36 and 49) and the CEDAW Committee’s General Recommendation No. 35 (paras. 26 and 29), both of which require that the assessment of sexual offences be free from gender stereotypes and grounded in a substantive understanding of consent. In those cases, the stereotype critique addressed the substantive inadequacy of the State’s response to sexual violence. B.G. relocates the critique, at least partially, to the procedural domain of Article 6 § 1: the police notation that the applicant ‘had not screamed, had not resisted, and had not clearly expressed her refusal’ was identified not merely as a substantive error but as a fair trial deficiency, infecting the prosecutorial reasoning that led to the rappel à la loi and depriving the applicant of a fair assessment of her complaint.
The doctrinal boundaries of this development merit attention. The Court’s reasoning does not collapse the distinction between Articles 3, 8 and 6 § 1; the critique under Article 6 § 1 is engaged specifically because the stereotyped reasoning was deployed in a procedural context that directly affected the applicant’s fair trial rights – namely, the decision to subject her to a measure attributing criminal responsibility. It does not follow that every investigative failure informed by gender stereotypes will automatically give rise to an Article 6 § 1 claim. The nexus between the stereotyped reasoning and the formal attribution of criminal liability appears to be the operative requirement, protecting the doctrinal coherence of the Court’s framework and forestalling an overbroad reading of B.G. as a wholesale transposition of Articles 3 and 8 jurisprudence into the procedural domain.
It is striking, in this connection, that the case was examined exclusively under Article 6 § 1. The factual circumstances – a decision not to prosecute a minor’s rape complaint following stereotyped police reasoning – might plausibly have engaged the procedural limb of Article 3. The contrast with X v. Cyprus is particularly telling: in that case, on closely analogous facts – a young woman whose gang-rape complaint was discontinued amid prejudicial gender stereotypes and who was subsequently prosecuted for false reporting – the Court found violations of Articles 3 and 8. Whether the different doctrinal pathway in B.G. reflects the way the application was framed or a deliberate methodological choice by the Court is not made explicit, but it leaves open the question whether future cases of this nature might more naturally find their doctrinal home at the intersection of Articles 3 and 6 § 1.
A further dimension of the judgment that deserves attention is the Court’s recognition of the potential chilling effect of the domestic practice on future complainants. In assessing whether the applicant had suffered a significant disadvantage for admissibility purposes, the Court included among the relevant factors the ‘deterrent character’ of the measure (para. 50). The chilling effect operates here on two levels simultaneously: it features in the admissibility analysis as part of the significant-disadvantage assessment, and it informs the substantive reasoning on the unfairness of the procedure. The Court thus ‘locks in’ the systemic significance of the issue already at the threshold stage – a technique visible elsewhere in the Court’s case-law concerning sexual violence and gender stereotyping, where structural concerns and individual harm have intersected (see, for instance, M.C. v. Bulgaria and X v. Cyprus, cited above). The concern is particularly acute in the context of minors: B.G. was sixteen, had consistently maintained her account, and had explicitly refused to accept the terms of the measure. The signal to domestic authorities is unambiguous: alternatives to prosecution must not be deployed in a manner that exposes complainants in sexual violence cases to the prospect of being themselves prosecuted for false accusation – a deployment that generates precisely the chilling effect the Court identified, particularly where the complainant is a minor and the underlying facts remain genuinely disputed.
B.G. marks a doctrinally significant – if implicit – shift in the personal scope of Article 6 § 1. The provision has traditionally operated as a shield for the accused: it is the person charged with a criminal offence who benefits from the guarantees of a fair trial. The judgment refines this orthodoxy by extending the procedural protection of Article 6 § 1 to a particular category of complainant – one who, having reported a sexual offence, is subsequently transformed into a de facto accused through the operation of an alternative to prosecution attributing criminal responsibility for a false accusation. The extension is narrower than it might first appear, but no less significant for that: B.G. does not establish that complainants, as such, enjoy Article 6 § 1 rights in respect of the investigation of their own complaints; that domain remains governed by the positive obligations framework of Articles 3 and 8. What it establishes is that where the State’s response to a sexual violence complaint generates a discrete formal attribution of criminal responsibility to complainants themselves, the procedural guarantees of Article 6 § 1 are engaged for their benefit. The extension is conditioned on the absence of waiver: B.G.’s consistent refusal to acknowledge the alleged facts secured her the provision’s full guarantees; the position may differ where a complainant accepts the terms of such a measure without contest. The development reflects a broader convergence between the positive obligations framework developed under Articles 3 and 8 and the procedural guarantees of Article 6 § 1, coalescing into a more integrated standard of protection at the precise point where the State’s response to a sexual violence complaint produces adverse criminal consequences for complainants themselves.
B.G. v. France is more than a judgment about a now-abolished French procedural measure. Its most durable contribution lies in its methodology: by assessing the procedural sequence as an indivisible whole, the Court establishes that the criminal limb of Article 6 § 1 cannot be evaded through the formal characterisation of prosecutorial measures as ‘alternatives’ to proceedings. What matters is not the ‘label’ but the substance – whether the measure, in its concrete operation, attributes criminal responsibility to a person who contests the underlying facts and is denied meaningful judicial recourse. This principle applies with equal force to the caution regimes of England and Wales, the Dutch strafbeschikking, the German Absehen von der Verfolgung unter Auflagen und Weisungen, the conditional abstention from prosecution under Greek law, the Cypriot diversion programmes for children in conflict with the law, and any comparable mechanism whose architecture lacks a genuine right of judicial opposition.
The judgment’s engagement with stereotyped reasoning on sexual consent as a fair trial deficiency is equally significant, extending to Article 6 § 1 a line of reasoning previously confined to the positive obligations framework of Articles 3 and 8, and expanding the procedural standing of a particular category of complainant – those transformed into de facto accused by the State’s own response to their report. Together with the recognition of the chilling effect on future complainants, these strands signal a more integrated standard of protection at the intersection of fair trial guarantees and the rights of victims of sexual violence.
Strasbourg Observers is an academic blog that discusses recent developments at the European Court of Human Rights. The editorial team of Strasbourg Observers is based at the Human Rights Centre of Ghent University and the Centre for Government and Law of Hasselt University.
Strasbourg Observers would not be possible without the valued work of countless guest bloggers. Visit the “Contact” page on our website to find out how to become one!
