December 04, 2025
by Vandita Khanna
After three recent judgments on racial profiling (analysed here and here), Seydi and Others v France has brought the issue back before the European Court of Human Rights (ECtHR). On 26 June 2025, the Fifth Section found France to have violated Article 14 (prohibition of discrimination) in conjunction with Article 8 (right to respect for private life) of the European Convention on Human Rights (ECHR) for subjecting Karim Touil to racial profiling. At the same time, it rejected claims of racial discrimination alleged by five other applicants of North African or sub-Saharan origin. This judgment, which continues the story of judicial engagement with racial profiling in Europe, reveals the significant challenges in proving racial profiling as discrimination in European human rights law.
The facts can be traced to 2011-12 when thirteen young French men of North African or sub-Saharan origin were subjected to police identity stops in public (known as contrôles au faciès). To challenge their stops as discriminatory, they initiated civil proceedings before the Tribunal de Grande Instance and, when unsuccessful, appealed before the Cour d’Appel. Subsequently, the Cour de Cassation confirmed that five of them had been subjected to discriminatory checks. Of the remaining unsuccessful applicants, six approached the ECtHR.
The six applicants’ identity checks unfolded as follows. Mounir Seydi was subjected to an identity check as he, accompanied by a friend, was exiting a metro station. Bocar Niane, wearing a jumper with the hood over his head and walking briskly, was stopped by four police officers when leaving his parents’ home. Dia Abdillahi was subjected to an identity check and search while walking from a post office with his cousin. Lyes Kaouah and Amine Mohamed Dif were checked and searched by around 15 police officers when sitting and talking on the stairs of a building where the former lived. Finally, Karim Touil was subjected to three identity checks in a span of ten days – once when he was with a friend, once when he was on a street with two other friends, and a third time in front of a hotel with several friends. In the last of these incidents, a police officer insulted and slapped Touil, before taking him into a police van and arresting him.
All six applicants alleged a violation of the prohibition of discrimination (Article 14) taken in conjunction with the right to respect for private life (Article 8), right to effective remedy (Article 13), and the freedom of movement (Article 2 of Protocol 4).
Drawing from its previous jurisprudence on racial profiling, the Court characterised the complaint in Seydi as fallingunder Article 14 read with Article 8 ECHR (para 58).
The Court first examined the complaint under the procedural aspect of Article 14 in conjunction with Article 8. Here, it considered that the French domestic courts had fulfilled their obligation to examine whether discriminatory grounds played a role in the identity checks. Since the applicants had benefited from a careful examination of their allegations of racial profiling and the domestic courts had conducted a balanced, comprehensive, and reasoned assessment, France had fulfilled the duty to investigate (paras 97-98).
Next, the Court examined the complaint under the positive obligation to put in place an adequate and effective legal and administrative framework (first insisted upon in racial profiling cases by Judge Krenc in Muhammad v Spain, paras 9-12). French domestic law, specifically Article 78-2 of the Code of Criminal Procedure, empowers the police to conduct three types of checks: ‘spontaneous’ or ‘routine’ checks (under the first paragraph), checks by requisition issued by the public prosecutor (under the second paragraph), and ‘public order’ or ‘preventive’ checks to prevent disturbances to public order (under the third paragraph). The applicants had submitted that the legislative framework was too vague and allowed for arbitrary decision-making by police officers. The ECtHR however found the law (which had previously been declared constitutional by domestic courts) to provide a detailed framework for law enforcement authorities to carry out checks (para 102). Additional safeguards outlining obligations of the police and gendarmerie further strengthened the legal framework, according to the Court. As a result, the domestic legal and administrative framework was found to be compatible with the requirements of Article 14 read with Article 8 (para 105).
In examining whether each identity check revealed any discrimination, the Court clarified that it was for the applicants to establish a prima facie case of discrimination before the burden could shift to the State to offer a justification.
For five applicants – Mounir Seydi, Bocar Niane, Dia Abdillahi, Amine Mohamed Dif, and Lyes Kaouah – the Court found no violation of Article 14 in conjunction with Article 8. This was because (i) the identity checks to which these applicants were subjected had at least one identifiable legal basis under Article 78-2 of the French Code of Criminal Procedure and (ii) none of the applicants had provided a comparator to disclose a difference in treatment. In the absence of a coexistence of sufficiently strong, clear and concordant inferences capable of creating a presumption of discriminatory treatment (para 122), the Court thus dismissed the claims of these five applicants.
In contrast, in respect of Karim Touil, the Court considered that several factors coexisted to make out prima facie evidence of discrimination and to shift the burden onto the respondent State accordingly. These factors included the absence of any legal basis for one of the three identity checks to which Touil had been subjected within a span of ten days as well as official reports and statistical data on the broader practice of racial profiling in France. The State provided no objective and reasonable justification for any of the three checks and, therefore, the Court (by a majority of 6:1) found France to have violated Article 14 in conjunction with Article 8 in respect of Touil. Judge Mourou-Vikström dissented on this finding largely because Touil had not provided a comparator to evidence a difference in treatment.
Finally, the ECtHR returned a finding of non-violation regarding the right to effective remedy under Article 13 bearing in mind that the applicants were able to bring their grievances before judicial authorities and that the domestic courts had adjusted the burden of proof.
Seydi can be considered significant for its condemnation of racial profiling as discrimination in Touil’s case. This finding of a violation in respect of Touil must, however, be read alongside the Court’s finding of a non-violation in respect of the five other applicants. Indeed, a holistic reading of Seydi confirms that proving racial discrimination remains challenging in European human rights law. In what follows, I will focus on three elements of this challenge, namely the characterisation of the facts as direct discrimination, the requirement to produce a comparator, and the resistance to statistical evidence.
Racial profiling is often described as a persistent and pervasive practice that cannot be easily squeezed into the legal categories of direct and indirect discrimination. In Seydi, although the applicants argued for both direct and indirect discrimination (para 76), the Court in a first step treated the complaint under direct discrimination alone. This move in itself carries doctrinal implications: under the ECHR direct discrimination has become more difficult to establish than indirect discrimination. As Barbara Havelková explains, indirect discrimination may have ‘outpaced’ direct discrimination in the Court’s case law possibly because ‘[f]inding structural bias can be more palatable than finding racist motives’ (see e.g. here, p. 98). On a conceptual plane, the choice to characterise racial profiling as potentially directly discriminatory (which is not a foregone conclusion – see e.g. Memedova and Others v North Macedonia) requires the Court to zoom in on the individual circumstances of differential treatment (allegedly) meted out by individual police officers. As a result, the focus shifts away from the structural, dispersed, and pervasive nature of the practice.
Second, the ECtHR’s insistence that the five applicants (barring Touil) provide a comparator to establish prima facie discrimination appears inconsistent with its Article 14 jurisprudence and does not account for evidentiary difficulties in the specific context of untraceable identity checks, such as the ones practiced in France. Discrimination under Article 14 is commonly defined as ‘treating differently, without an objective and reasonable justification, persons in relevantly similar situations’ (e.g. Zarb Adami v Malta, para 71). Despite the definitional emphasis on comparison, the Court is inconsistent in applying comparisons and often skips the comparator inquiry. This is arguably because not all forms of discrimination warrant a comparator analysis. In some cases (as argued here), the objection to discrimination is not that someone has been treated unfairly compared to others, but that the treatment in itself is demeaning. In the case of the applicants in Seydi, subjecting racialised individuals to identity checks on streets and in public places (within a context of statistical evidence that reports racial profiling in France) could in itself result in indignity, stereotyping, and stigmatisation. In Alexandra Timmer’s words, the ‘wrongs of stereotyping are not comparative in nature: they do not derive from a comparison with another group that has been treated better.’[1] That in the present case the Court insisted on comparators for five applicants but not for Touil further confirms that a comparative assessment may not always be necessary.
From an evidentiary perspective, showing a point of comparison (that those who did not belong to ‘visible minorities’ were not subjected to an identity check) is all the more difficult in the context of identity checks, which (as submitted by the applicants in para 35) often occur in the absence of any witnesses (Seydi, para 71). Even so, as the Court has previously acknowledged in Wa Baile v Switzerland, the respondent state alone is in a position to provide information related to how many others had been subjected to such a check on that day (Wa Baile, para 134). It remains unclear why the Court in Seydi departed from this line of reasoning (despite the applicants’ submissions to this effect in para 72) and imposed a disproportionate burden on the applicants to produce a comparator. The disproportionate nature of this evidentiary burden is compounded by the fact that identity checks in France are not traceable because there is no written record of the number of checks conducted and of whom in any given day (para 72). Taken cumulatively, the requirement to produce a comparator obstructs the ability of applicants to evidence prima facie discriminatory treatment in Seydi.
Third, the ECtHR’s treatment of statistical evidence in Seydi sits at odds with its general approach to statistical evidence in proving discrimination. The Court in Seydi acknowledged that it has previously relied on statistical data to establish a difference in treatment but held nonetheless that statistics, of themselves, are insufficient to reveal a practice of discrimination (para 110). It cited Di Trizio v Switzerland(para 85) to support this holding. However, para 85 of Di Trizio says the opposite: specifically, although ‘in the past … statistics could not in themselves disclose a practice which could be classified as discriminatory’, ‘in more recent cases on the question of discrimination …, the Court relied extensively on statistics produced by the parties in finding a difference in treatment between two groups in similar situations’. The striking reversal of statements in Seydi reflects the Court’s reluctance to rely on statistical evidence to establish prima facie discrimination in cases of racial profiling. Despite noting a wide range of reports and decisions of national and international bodies that substantiate the nature of discriminatory checks and confirm that they are not isolated cases, the Court nonetheless clarified that its ‘sole concern’ was to determine whether each identity check was motivated by the applicant’s skin colour or perceived membership of a visible minority (para 113). The dissenting opinion of Judge Mourou-Vikström goes further in calling for stricter evidentiary standards:
“Il n’apparaît pas souhaitable qu’en se fondant sur des données statistiques constantes ayant valeur de postulat, des contrôles d’identité, fussent-ils répétés, tendus et accompagnés d’incidents, soient considérés comme discriminatoires, alors qu’aucun élément de nature à traduire une différence de traitement n’est apportée par des personnes qui allèguent subir une telle discrimination.”
The individualisation of what has been often described as a structural and systemic problem constructs racial profiling as a product of a ‘few bad apples’. In that the previous judgment on racial profiling, Wa Baile, is considered to have implicitly endorsed an understanding of racial profiling as structural discrimination, Seydi appears to take a step back.
Against this backdrop, the finding of discrimination regarding Touil’s checks is striking. Proof of prima facie discrimination may follow from the coexistence of sufficiently strong, clear and concordant inferences or of similar unrebutted presumptions of fact. In Touil’s case, the Court admitted that Touil had not provided a comparator but the frequency of his checks, along with the absence of an identifiable legal basis for one of the three checks, the insulting and violent behaviour of the police officers towards Touil, and the statistical data on racial profiling in France, taken together sufficed to reverse the burden of proof. The facts in Touil’s case suggest the extreme situations that would, per the Court, meet the (high) threshold for prima facie racial discrimination.
The Court frequently stresses that racial discrimination is ‘a particularly invidious kind of discrimination and, in view of its perilous consequences, requires from the authorities special vigilance and a vigorous reaction’ (e.g. Seydi, para 87). Paradoxically, the seriousness with which the Court takes racial discrimination seems to make it more difficult for applicants to prove it. Seydi reinforces this pattern in the Court’s case law. The political sensitivity attached to the label of racial profiling – which was acknowledged in the dissenting opinion and is certainly not confined to France (see e.g. the recent response by Italy to a report by the European Commission against Racism and Intolerance) – suggests that the ECtHR may tread cautiously going forward. In such a situation where the Convention and the Court may not be fully equipped to tackle racial profiling, the idea of a European Convention against Racial Profiling (which could include, inter alia, a duty to collect data, provide training, and involve affected communities in developing anti-racial profiling strategies) remains very much alive.
[1]Alexandra Timmer, ‘Judging Stereotypes: What the European Court of Human Rights Can Borrow from American and Canadian Equal Protection Law’ (2015) 63 American Journal of Comparative Law 239, 252.
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!
