Basu v. Germany and Muhammad v. Spain: Why the first European Court of Human Rights’ judgments on racial profiling in identity checks are disappointing  

Racial profiling by the police is a mounting concern in Europe. It has been defined by the European Commission against Racism and Intolerance (ECRI) as “the use by the police, with no objective or reasonable justification, of grounds such as race, colour, language, religion, nationality or national or ethnic origin in control, surveillance or investigation activities” (ECRI General Recommendation 11, §1). Numerous international and European institutions, including the UN Human Rights Committee (Rosalind Williams Lecraft v Spain), the UN Committee on the Elimination of Racial Discrimination (CERD), the Council of Europe Commissioner for Human Rights, and the European Parliament, have condemned this practice as discriminatory and called upon states to take measures to tackle it. In addition to being contrary to human rights, it is said to be ineffective, because it relies on stereotypes and gross generalizations rather than factual grounds or reasonable suspicion, and counterproductive as it damages the relationship between the police and certain communities, by stirring feelings of injustice and stigmatization. Nonetheless, a wealth of studies and reports show that this phenomenon remains widespread across Europe (see i.a. Commissioner for Human Rights, Ethnic Profiling: A Persistent Practice in Europe, 2019; Parliamentary Assembly of the Council of Europe, Ethnic Profiling in Europe: a Matter of Great Concern, 2020; EU Fundamental Rights Agency, Your Rights Matter: Police Stops, 2021).

In Basu v. Germany and Muhammad v. Spain, decided on 18 October 2022, the European Court of Human Rights (ECtHR) was for the first time confronted with alleged racial profiling in police identity check. The Court had previously dealt with some racial profiling cases in relation to other police powers (see Timishev v. Russia, on a ban on persons of Chechen origin crossing an administrative border, and Lingurar v. Romania, on a police raid targeting a Roma village, both resulting in a finding of discrimination). Identity checks however raise specific questions. It is an area that is particularly prone to racial profiling as the police often enjoys wide discretion in deciding who to stop and check. For the same reason, discrimination is especially difficult to prove in this context. Clarifications from the ECtHR on states obligations under the European Convention on Human Rights (ECHR) in this regard but also on the issue of proof were therefore eagerly awaited.

The applicants complained that they had been subject to an identity check by the police allegedly based on their race or ethnic origin. In Basu, the incident took place on a train just after crossing the Czech Republic border. In Muhammad, it occurred while the applicant was walking on a street in Barcelona. Both claimed that they were the only dark skin persons in the surroundings and the only ones to be checked. They filed complaints for discrimination before administrative courts and the Constitutional Court but their lawsuits were dismissed. The Court arrived at different conclusions in each case: in Basu, it found a breach of Article 14 (non-discrimination) combined with Article 8 (right to respect for private life) of the European Convention on Human Rights (ECHR) for failure to carrying out an effective investigation. In Muhammad, by a small majority of 4 to 3, it found no violation.

While they contain some valuable developments (I), these judgments are on the whole disappointing, for three reasons: the Court’s review of the effectiveness of the investigation in Muhammad appears too superficial; in both judgments, the Court refrains from examining the legal framework regulating police activities; and it disregards the special difficulties involving in proving discrimination in identity checks (II).

1. Two valuable developments

The Court makes two important points in these judgments. First, it establishes that while some special circumstances must be present for an identity check to fall within the ambit of the right to respect for private life, this condition is fulfilled “if the person concerned has an arguable claim that he or she may have been targeted on account of specific physical or ethnic characteristics” (Basu, §25; Muhammad, §50). This is a crucial point in view of the fact that Article 14 prohibits discrimination only in the enjoyment of the rights set forth in the Convention. Thus, it only applies if the facts complained of fall within the ambit of a substantive ECHR right – in this case the right to privacy. To be sure, Protocol 12 to the Convention does recognize a free-standing right to non-discrimination. However, not every state party to the Convention has ratified it. Germany, in particular, has not.

Second, extrapolating from its case law on alleged racist violence, in particular police violence (i.a. Nachova and others v. Bulgaria, §§160-161 and B.S. v. Spain, §58), the Court holds that, once a person has an arguable claim to have been targeted in an identity check because of her race or ethnic origin, state authorities have a duty under Article 14 to conduct an effective and independent investigation to ascertain whether the agent’s act was discriminatory. (For more on these points, see Mathias Möschel’s post on Strasbourg Observers).

2. Three major shortcomings

Ascertaining whether there has been an effective investigation

Applying these principles to the facts of the cases, in Basu, the Court found that state authorities failed to comply with their duty to investigate the incident. First, the internal investigation carried out by the Federal Police could not be deemed independent. Second, the administrative courts which handled the applicant’s complaint had not taken the necessary evidence and had rejected the complaint on formal grounds, denying that the applicant had a legitimate interest in a decision on the lawfulness of his identity check (§37).

In Muhammad, by contrast, the Court found no violation of state’s procedural obligation under Article 14. This conclusion is intriguing. The applicant and the police had given entirely diverging accounts of the incident. The applicant submitted that there was no particular reason to check him and that when he asked the police officer if he was checked because of his skin color, the officer replied in the affirmative. The police, however, claimed that the applicant had been checked because he had been disrespectful towards them. Yet, the administrative courts which examined the applicant’s complaint for discrimination relied exclusively on the statements of the police officers. The applicant’s request that a friend of him who was present at the scene and an expert witness on racial profiling be heard was rejected. So was his demand to examine video-recording of surveillance camera in the area. Such evidence was deemed unnecessary or irrelevant by the courts, which concluded that the applicant had not substantiated his argument. To be sure, criminal proceedings had also been instituted following a criminal complaint by the applicant. But these proceedings only concerned whether an offense had been committed during the identity check and not the alleged discrimination. And the evidence referred to by the applicant had not been taken into account either during these proceedings, which were discontinued for lack of evidence. In view of these elements, it is difficult to follow the court when it concludes that state authorities have fulfilled their duty to investigate which, according to the ECtHR’s previous case-law, entails the obligation to “do what is reasonable in the circumstances to collect and secure the evidence, explore all practical means of discovering the truth and deliver fully reasoned, impartial and objective decisions” (Muhammad, § 66). Both Judge Zünd and Judge Krenc strongly criticize this aspect of the judgment in their dissenting opinions.

Sidelining the issue whether the legal framework regulating police activities was adequate

Various international institutions have stressed that a key factor in countering racial profiling is the establishment of an adequate legal framework that circumscribes police powers to ensure that they are exercised on the basis of objective criteria related to an individual’s conduct rather than stereotypes and bias against certain groups (ECRI General Recommendation No. 11, §3; CERD General Recommendation No. 36, §§ 38-39). In previous cases, the ECtHR has recognized that state authorities have a positive obligation to set up an adequate legal framework affording effective safeguards against arbitrariness and abuse by the police (eg Giuliani and Gaggio v. Italy, § 209; Nachova and others v. Bulgaria, §§96-97), including in stop-and-search practice (Gillan and Quinton v. United Kingdom, §§77-87; Vig v. Hungary, §§51-63). 

Yet, in both Basu and Muhammad, the Court’s majority chooses not to examine whether the legal framework in place in Germany and Spain in relation to police power to perform identity check was adequate to prevent racial profiling, despite the issue being raised by the applicants. This choice is deplored by Judge Pavli in his partly dissenting opinion in Basu (§§9-12) and Judge Krenc in his dissent in Muhammad. The ECtHR’s judgments contrast strikingly in this regard with a decision issued a few days later, on October 25, 2022, by the Superior Court of Quebec: the latter held that legal provisions conferring the police discretionary power to practice roadside interception without having to establish a real motive or a suspicion of offense entailed a breach of multiple rights protected by the Canadian Charter of rights and freedoms, including the right to equal protection, as this discretionary power resulted in an overrepresentation of black persons stopped without real motive (Luamba c Procureur Général du Québec et Procureur général du Canada, 25 October 2022).

The problem of proof

Proving racial profiling raises significant hurdles. Police officers will rarely admit that they have acted on the basis of racial or ethnic prejudice. What is more, as highlighted by Judge Pavli, “discrimination in this context may not be driven necessarily by a police officer’s individual and conscious attitude or hostility against a particular racial or ethnic group; it may also be the result of biased (or at least permissive) internal police guidelines, practices or attitudes, whether formalized or merely tolerated by the hierarchy” (dissent in Basu, §7). Canadian Courts have acknowledged that racial profiling can be conscious or unconscious (R v Le, §76; Luamba, §§40 & 424-464). They recognized accordingly that it can rarely be established by direct evidence; usually “if racial profiling is to be proven it must be done by inference drawn from circumstantial evidence” (Luamba, §45).

The ECtHR’s approach to proving a substantive violation of Article 14 in both Basu and Muhammad does not seem to do justice to the complexity of the phenomenon at stake.

First, the Court suggests that what has to be established is whether there was “any racist motives” behind the identity check (Basu, §36); whether it was “motivated by animosity against citizens who shared the applicant’s ethnicity” (Muhammad, § 99). This is an overly narrow understanding of racial profiling: a person is racially profiled if she is checked because of her race or ethnic origin, regardless of whether police officers acted on the basis of racial hatred or were just following routinized practices of the police institution.

Second, in previous case-law, the Court has recognized that “proof may follow from the coexistence of sufficiently strong, clear and concordant inferences or of similar unrebutted presumption of fact” and that “Convention proceedings do not in all cases lend themselves to a strict application of the principle affirmanti incumbit probation.” In particular, “where the events in issue lie wholly, or in large part, within the exclusive knowledge of the authorities, the burden of proof may be regarded as resting on the authorities to provide a satisfactory and convincing explanation” (D.H. and others v. the Czech Republic, §§178-179; El Masri v FYROM, §§151-152). In relation to discrimination, the Court has accepted that once the applicant has established a prima facie case of discrimination, which can result among others from statistics showing that a measure or a practice has a disproportionate prejudicial impact on a racial or ethnic group, the burden of proof should shift to the government to prove that the difference in treatment is based on a legitimate ground (D.H., §§180 & 188-189). In Basu and Muhammad, the Court recognized that the applicants had an “arguable claim” of having been subject to racial profiling. Moreover, the applicants put forward international institutions and NGOs reports highlighting the prevalence of profiling practices by the police in Germany and Spain and arguably indicating a systemic pattern in both countries. The Court however refuses to rely on these two elements taken together to establish a presumption of discrimination (see Judge Pavli’s critique in his dissent; Bea Streicher and Mathias Möschel’s posts on Strasbourg Observers; and, more generally, Ch. Roberts, ‘Reversing the Burden of Proof Before Human Rights Bodies’, 25 International Journal of Human Rights (2022) 1682). The Court did not find international reports relevant, stating in Muhammad that “its sole concern” was to ascertain whether the identity check imposed on the applicant “was motivated by racism” (§100)

Lastly, the Court declares in Basu that because state authorities failed to comply with their duty to carry out an effective investigation into potential discrimination, it is unable to determine whether the applicant was the victim of a substantive violation of Article 14 (§38). This however amounts to requiring the applicant to pay the cost of state authorities’ failure to comply with their obligation to investigate. As emphasized by Judge Pavli, it creates a “perverse incentive” for national authorities to refrain from investigating potential incidents of police racial profiling. It also greatly limits the chances of victims to demonstrate discrimination in identity checks (dissent, Basu, §4 & 18).