GRAND CHAMBER CASE OF D.H. AND OTHERS v. THE CZECH REPUBLIC (Application no. 57325/00) JUDGMENT STRASBOURG 13 November 2007 This judgment is final but may be subject to editorial revision. In the case of D.H. and Others v. the Czech Republic, The European Court of Human Rights (Second Section), sitting as a Grand Chamber composed of: Sir Nicolas Bratza, President, Mr B.M. Zupančič, Mr R. Türmen, Mr K. Jungwiert, Mr J. Casadevall, Mrs M. Tsatsa-Nikolovska, Mr K. Traja, Mr V. Zagrebelsky, Mrs E. Steiner, Mr J. Borrego Borrego, Mrs A. Gyulumyan, Mr K. Hajiyev, Mr D. Spielmann, Mr S.E. Jebens, Mr J. Šikuta, Mrs I. Ziemele, Mr M. Villiger, judges, and Mr M. O'Boyle, Deputy Registrar, Having deliberated in private on 17 January and 19 September 2007, Delivers the following judgment, which was adopted on the last mentioned date: PROCEDURE 1. The case originated in an application (no. 57325/00) against the Czech Republic lodged with the Court on 18 April 2000 under Article 34 of the Convention for the Protection of Human Rights and Fundamental Freedoms (“the Convention”) by eighteen Czech nationals (“the applicants”), whose details are set out in the annex to this judgment (“the Annex”). 2. The applicants were represented before the Court by the European Roma Rights Centre based in Budapest, Lord Lester of Herne Hill, Q.C, Mr J. Goldston, of the New York Bar, and Mr D. Strupek, a lawyer practising in the Czech Republic. The Czech Government (“the Government”) were represented by their Agent, Mr V.A. Schorm. 3. The applicants alleged, inter alia, that they had been discriminated against in the enjoyment of their right to education on account of their race or ethnic origin. 4. The application was allocated to the Second Section of the Court (Rule 52 § 1 of the Rules of Court). Within that Section, the Chamber that would consider the case (Article 27 § 1 of the Convention) was constituted as provided in Rule 26 § 1. 5. By a decision of 1 March 2005, following a hearing on admissibility and the merits (Rule 54 § 3), the Chamber declared the application partly admissible. 6. On 7 February 2006 a Chamber of that Section composed of the following judges: Mr J.-P. Costa, President, Mr A.B. Baka, Mr I. Cabral Barreto, Mr K. Jungwiert, Mr V. Butkevych, Ms A. Mularoni and Ms D. Jočienė, judges, and also of Mrs S. Dollé, Section Registrar, delivered a judgment in which it held by six votes to one that there had been no violation of Article 14 of the Convention, read in conjunction with Article 2 of Protocol No. 1. 7. On 5 May 2006 the applicants requested the referral of their case to the Grand Chamber in accordance with Article 43 of the Convention. On 3 July 2006 a panel of the Grand Chamber granted their request. 8. The composition of the Grand Chamber was determined according to the provisions of Article 27 §§ 2 and 3 of the Convention and Rule 24 of the Rules of Court. At the final deliberations, Mr K. Traja and Mr J. Casadevall, substitute judges, replaced Mr C. Rozakis and Mr P. Lorenzen, who were unable to take part in the further consideration of the case (Rule 24 § 3). 9. The applicants and the Government each filed written observations on the merits. In addition, third-party comments were received from various non-governmental organisations, namely the International Step by Step Association, the Roma Education Fund and the European Early Childhood Research Association; Interights and Human Rights Watch; Minority Rights Group International, the European Network Against Racism and the European Roma Information Office; and the Fédération internationale des ligues des droits de l'Homme (International Federation for Human Rights – FIDH), each of which had been given leave by the President to intervene in the written procedure (Article 36 § 2 of the Convention and Rule 44 § 2). The respondent Government replied to those comments (Rule 44 § 5). 10. A hearing took place in public in the Human Rights Building, Strasbourg, on 17 January 2007 (Rule 59 § 3). There appeared before the Court: (a) for the Government Mr V.A. Schorm, Agent, Ms M. Kopsová, Ms Z. Kaprová, Ms J. Zapletalová, Mr R. Barinka, Mr P. Konůpka, Counsel; (b) for the applicants Lord Lester of Herne Hill, Q.C., Mr J. Goldston, Mr D. Strupek, Counsel. The Court heard addresses by Lord Lester of Herne Hill, Mr Goldston and Mr Strupek, and by Mr Schorm. THE FACTS I. THE CIRCUMSTANCES OF THE CASE 11. Details of the applicants' names and places of residence are set out in the Annex. A. Historical background 12. According to documents available on the Internet site of the Roma and Travellers Division of the Council of Europe, the Roma originated from the regions situated between north west India and the Iranian plateau. The first written traces of their arrival in Europe date back to the fourteenth century. Today there are between eight and ten million Roma living in Europe. They are to be found in almost all Council of Europe member States and indeed, in some Central and East European countries, they represent over 5% of the population. The majority of them speak Romani, an Indo-European language that is understood by a very large number of Roma in Europe, despite its many variants. In general, Roma also speak the dominant language of the region in which they live, or even several languages. 13. Although they have been in Europe since the fourteenth century, often they are not recognised by the majority society as a fully-fledged European people and they have suffered throughout their history from rejection and persecution. This culminated in their attempted extermination by the Nazis, who considered them an inferior race. As a result of centuries of rejection many Roma communities today live in very difficult conditions, often on the fringe of society in the countries where they have settled, and their participation in public life is extremely limited. 14. In the Czech Republic the Roma have national-minority status and, accordingly, enjoy the special rights associated therewith. The National Minorities Commission of the Government of the Czech Republic, a governmental consultative body without executive power, has responsibility for defending the interests of the national minorities, including the Roma. As to the number of Roma currently living in the Czech Republic, there is a discrepancy between the official, census-based, statistics and the estimated number. According to the latter, which is available on the website of the Minorities Commission of the Government of the Czech Republic, the Roma community now numbers between 150,000 and 300,000 people. B. Special schools 15. According to information supplied by the Czech Government, the special schools (zvláštní školy) were established after the First World War for children with special needs, including those suffering from a mental or social handicap. The number of children placed in these schools continued to rise (from 23,000 pupils in 1960 to 59,301 in 1988). Owing to the entrance requirements of the primary schools (základní školy) and the resulting selection process, prior to 1989 most Roma children attended special school. 16. Under the terms of the Schools Act (Law no. 29/1984), the legislation applicable in the present case, special schools were a category of specialised school (speciální školy) and were intended for children with mental deficiencies who were unable to attend “ordinary” or specialised primary schools. Under the Act, the decision to place a child in a special school was taken by the head teacher on the basis of the results of tests to measure the child's intellectual capacity carried out in an educational psychology centre and was subject to the consent of the child's legal guardian. 17. Following the switch to the market economy in the 1990s, a number of changes were made to the system of special schools in the Czech Republic. These changes also affected the education of Roma pupils. In 1995 the Ministry of Education issued a directive concerning the provision of additional lessons for pupils who had completed their compulsory education in a special school. Since the 1996/97 school year, preparatory classes for children from disadvantaged social backgrounds have been opened in nursery, primary and special schools. In 1998 the Ministry of Education approved an alternative educational curriculum for children of Roma origin who had been placed in special schools. Roma teaching assistants were also assigned to primary and special schools to assist the teachers and facilitate communication with the families. By virtue of amendment no. 19/2000 to the Schools Act, which came into force on 18 February 2000, pupils who had completed their compulsory education in a special school were also eligible for admission to a secondary-school education, provided they satisfied the entrance requirements for their chosen course. 18. According to data supplied by the applicants, which was obtained through questionnaires sent in 1999 to the head teachers of the 8 special schools and 69 primary schools in the town of Ostrava, the total number of pupils placed in special schools in Ostrava came to 1,360, of whom 762 (56%) were Roma. Conversely, Roma represented only 2.26% of the total of 33,372 primary-school pupils in Ostrava. Further, although only 1.8% of non-Roma pupils were placed in special schools, in Ostrava the proportion of Roma pupils assigned to such schools was 50.3%. Accordingly, a Roma child in Ostrava was 27 times more likely to be placed in a special school than a non-Roma child. According to data from the European Monitoring Centre for Racism and Xenophobia (now the European Union Agency for Fundamental Rights), more than half of Roma children in the Czech Republic attend special schools. The Advisory Committee on the Framework Convention for the Protection of National Minorities observed in its report of 26 October 2005 that, according to unofficial estimates, the Roma represent up to 70% of pupils enrolled in special schools. Lastly, according to a comparison of data on fifteen countries, including countries from Europe, Asia and North America, gathered by the OECD in 1999 and cited in the observations of the International Step by Step Association, the Roma Education Fund and the European Early Childhood Research Association[1], the Czech Republic ranked second highest in terms of placing children with physiological impairments in special schools and in third place in the table of countries placing children with learning difficulties in such schools. Further, of the eight countries who had provided data on the schooling of children whose difficulties arose from social factors, the Czech Republic was the only one to use special schools. The other countries concerned almost exclusively used ordinary schools for the education of such children. C. The facts of the instant case 19. Between 1996 and 1999 the applicants were placed in special schools in Ostrava, either directly or after a spell in an ordinary primary school. 20. The material before the Court shows that the applicants' parents had consented to and in some instances expressly requested their children's placement in a special school. Consent was indicated by signing a pre-completed form. In the case of applicants nos. 12 and 16 the dates on the forms are later than the dates of the decisions to place the children in special schools. In both instances, the date has been corrected by hand, on one occasion is accompanied by a note from the teacher citing a typing error. The decisions on placement were then taken by the head teachers of the special schools concerned after referring to the recommendations of the educational psychology centres where the applicants had undergone psychological tests. The applicants' school files contained the report on their examination, including the results of the tests with the examiners' comments, drawings by the children and, in a number of cases, a questionnaire for the parents. The written decision concerning the placement was sent to the children's parents. It contained instructions on the right to appeal, a right which none of the applicants exercised. 21. On 29 June 1999 the applicants received a letter from the school authorities informing them of the possibilities available for transferring from special school to primary school. It would appear that four of the applicants (nos. 5, 6, 11 and 16 in the Annex) were successful in aptitude tests and thereafter attended ordinary schools. 22. In the review and appeals procedures referred to below, the applicants were represented by a lawyer acting on the basis of signed written authorities from their parents. 1. Request for a reconsideration of the case outside the formal appeal procedure 23. On 15 June 1999 all the applicants apart from those numbered 1, 2, 10 and 12 in the Annex asked the Ostrava Education Authority (Školský úřad) to reconsider, outside the formal appeal procedure (přezkoumání mimo odvolací řízení), the administrative decisions to place them in special schools. They argued that their intellectual capacity had not been reliably tested and that their representatives had not been adequately informed of the consequences of consenting to their placement in a special school. They therefore asked the Education Authority to revoke the impugned decisions, which they maintained did not comply with the statutory requirements and infringed their right to education without discrimination. 24. On 10 September 1999 the Education Authority informed the applicants that, as the impugned decisions complied with the legislation, the conditions for bringing proceedings outside the appeal procedure were not satisfied in their case. 2. Constitutional appeal 25. On 15 June 1999 applicants nos. 1 to 12 in the Annex lodged a constitutional appeal in which they complained, inter alia, of de facto discrimination in the general functioning of the special education system. In that connection, they relied on, inter alia, Articles 3 and 14 of the Convention and Article 2 of Protocol No. 1. While acknowledging that they had not appealed against the decisions to place them in special schools, they alleged that they had not been sufficiently informed of the consequences of placement and argued (on the question of the exhaustion of remedies) that their case concerned continuing violations and issues that went far beyond their personal interests. In their grounds of appeal, the applicants explained that they had been placed in special schools under a practice that had been established in order to implement the relevant statutory rules. In their submission, that practice had resulted in de facto racial segregation and discrimination that were reflected in the existence of two separately organised educational systems for members of different racial groups, namely special schools for the Roma and “ordinary” primary schools for the majority of the population. That difference in treatment was not based on any objective and reasonable justification, amounted to degrading treatment and had deprived them of the right to education (as the curriculum followed in special schools was inferior and pupils in special schools were unable to return to primary school or to obtain a secondary education other than in a vocational training centre). Arguing that they had received an inadequate education and an affront to their dignity, the applicants asked the Constitutional Court (Ústavní soud) to find a violation of their rights, to quash the decisions to place them in special schools, to order the respondents (the special schools concerned, the Ostrava Education Authority and the Ministry of Education) to refrain from any further violation of their rights and to restore the status quo ante by offering them compensatory lessons. 26. In their written submissions to the Constitutional Court, the special schools concerned pointed out that all the applicants had been enrolled on the basis of a recommendation from an educational psychology centre and with the consent of their representatives. Furthermore, despite having been notified of the relevant decisions, none of the representatives had decided to appeal. According to the schools, the applicants' representatives had been informed of the differences between the special-school curriculum and the primary-school curriculum. Regular meetings of teaching staff were held to assess pupils (with a view to their possible transfer to primary school). They added that some of the applicants (nos. 5 to 11 in the Annex) had been advised that there was a possibility of their being placed in primary school. The Education Authority pointed out in its written submissions that the special schools had their own legal personality, that the impugned decisions contained advice on the right of appeal and that the applicants had at no stage contacted the schools inspectorate. The Ministry of Education denied any discrimination and noted a tendency on the part of the parents of Roma children to have a rather negative attitude to school work. It asserted that each placement in a special school was preceded by an assessment of the child's intellectual capacity and that parental consent was a decisive factor. It further noted that there were 18 educational assistants of Roma origin in schools in Ostrava. 27. In their final written submissions, the applicants pointed out (i) that there was nothing in their school files to show that their progress was being regularly monitored with a view to a possible transfer to primary school, (ii) that the reports from the educational psychology centres contained no information on the tests that were used and (iii) that their recommendations for placement in a special school were based on grounds such as an insufficient command of the Czech language, an over-tolerant attitude on the part of the parents or an ill-adapted social environment. They also argued that the gaps in their education made a transfer to primary school impossible in practice and that social or cultural differences could not justify the alleged difference in treatment. 28. On 20 October 1999 the Constitutional Court dismissed the applicants' appeal, partly on the ground that it was manifestly unfounded and partly on the ground that it had no jurisdiction to hear it. It nevertheless invited the competent authorities to give careful and constructive consideration to the applicants' proposals. (a) With regard to the complaint of a violation of the applicants' rights as a result of their placement in special schools, the Constitutional Court held that, as only five decisions had actually been referred to in the notice of appeal, it had no jurisdiction to decide the cases of those applicants who had not appealed against the decisions concerned. As to the five applicants who had lodged constitutional appeals against the decisions to place them in special schools (nos. 1, 2, 3, 5 and 9 in the Annex), the Constitutional Court decided to disregard the fact that they had not lodged ordinary appeals against those decisions, as it agreed that the scope of their constitutional appeals went beyond their personal interests. However, it found that there was nothing in the material before it to show that the relevant statutory provisions had been interpreted or applied unconstitutionally, since the decisions had been taken by head teachers vested with the necessary authority on the basis of recommendations by educational psychology centres and with the consent of the applicants' representatives. (b) With regard to the complaints of insufficient monitoring of the applicants' progress at school and of racial discrimination, the Constitutional Court noted that it was not its role to assess the overall social context and found that the applicants had not furnished concrete evidence in support of their allegations. It further noted that the applicants had had a right of appeal against the decisions to place them in special schools, but had not exercised it. As to the objection that insufficient information had been given about the consequences of placement in a special school, the Constitutional Court considered that the applicants' representatives could have obtained this information by liaising with the schools and that there was nothing in the file to indicate that they had shown any interest in transferring to a primary school. The Constitutional Court therefore ruled that this part of the appeal was manifestly ill-unfounded. II. RELEVANT DOMESTIC LAW AND PRACTICE A. The Schools Act 1984 (Law no. 29/1984 – since repealed by Law no. 561/2004, which came into force on 1 January 2005) 29. Prior to 18 February 2000 section 19(1) of the Schools Act provided that to be eligible for secondary-school education pupils had to have successfully completed their primary-school education. Following amendment no. 19/2000, which came into force on 18 February 2000, the amended section 19(1) provided that to be eligible for secondary-school education pupils had to have completed their compulsory education and demonstrated during the admission procedure that they satisfied the conditions of eligibility for their chosen course. 30. Section 31(1) provided that special schools were intended for children with “mental deficiencies” (rozumové nedostatky) that prevented them from following the curricula in ordinary primary schools or in specialised primary schools (speciální základní škola) intended for children suffering from sensory impairment, illness or disability. B. The Schools Act 2004 (Law no. 561/2004) 31. This new Act on school education no longer provides for special schools in the form that had existed prior to its entry into force. Primary education is now provided by primary schools and specialised primary schools, the latter being intended for pupils with severe mental disability or multiple disabilities and for autistic children. 32. Section 16 contains provisions governing the education of children and pupils with special educational needs. These are defined in subsection 1 as children suffering from a disability, health problems or a social disadvantage. Section 16(4) provides that for the purposes of the Act a child is socially disadvantaged, inter alia, if it comes from a family environment with low socio-cultural status or at risk of socio-pathological phenomena. Subsection 5 provides that the existence of special educational needs is to be assessed by an educational guidance centre. 33. The Act also makes provision, inter alia, for educational assistants, individualised education projects, preparatory classes for socially disadvantaged children prior to the period of compulsory school education and additional lessons for pupils who have not received a basic education. C. Decree no. 127/1997 on specialised schools (since repealed by Decree no. 73/2005, which came into force on 17 February 2005) 34. Article 2 § 4 of the Decree laid down that the following schools were available for pupils suffering from mental disability: specialised nursery schools (speciální mateřské školy), special schools, auxiliary schools (pomocné školy), vocational training centres (odborná učiliště) and practical training schools (praktické školy). 35. Article 6 § 2 stipulated that if during the pupil's school career there was a change in the nature of his or her disability or if the specialised school was no longer adapted to the level of disability, the head teacher of the school attended by the pupil was required, after an interview with the pupil's representative, to recommend the pupil's placement in another specialised school or an ordinary school. 36. Article 7 § 1 stipulated that the decision to place a pupil in or transfer a pupil to, inter alia, a special school was to be taken by the head teacher, provided that the pupil's legal guardians consented. Article 7 § 2 provided that a proposal for a pupil to be placed, inter alia, in a special school could be made to the head teacher by the pupil's legal guardian, the pupil's current school, an educational psychology centre, a hospital or clinic, an authority with responsibility for family and child welfare or a health centre. In the event of the pupil not receiving a place in a special school, the head teacher was required by Article 7 § 3 to notify the pupil's legal guardian and the competent school authority or the municipality in which the pupil was permanently resident of the decision. The education authority was then required, after consulting the municipality, to make a proposal regarding the school in which the pupil would receive his or her compulsory education. Article 7 § 4 required the educational psychology centre to assemble all the documents relevant to the decision and to make a recommendation to the head teacher regarding the type of school. D. Decree no. 73/2005 on the education of children, pupils and students with special educational needs and gifted children, pupils and students 37. Article 1 of the Decree provides that pupils and students with special educational needs are to be educated with the help of support measures that go beyond or are different from the individualised educational and organisational measures available in ordinary schools. 38. Article 2 provides that children whose special educational needs have been established with the aid of an educational or psychological examination performed by an educational guidance centre will receive special schooling if they have clear and compelling needs that warrant their placement in a special education system. E. Domestic practice at the material time 1. Psychological examination 39. The testing of intellectual capacity in an educational psychology centre with the consent of the child's legal guardians was neither compulsory nor automatic. The recommendation for the child to sit the tests was generally made by teachers – either when the child first enrolled at the school or if difficulties were noted in its ordinary primary-school education – or by paediatricians. 40. According to the applicants, who cited experts in this field, the most commonly used tests appeared to be variants of the 'Wechsler Intelligence Scale for Children' (PDW and WISC III) and the 'Stanford-Binet Intelligence test'. Citing various opinions, including those of teachers and psychologists and the head of the special-schools department at the Czech Ministry of Education in February 1999, the applicants submitted that the tests used were neither objective nor reliable, as they had been devised solely for Czech children, and had not recently been standardised or approved for use with Roma children. Moreover, no measures had been taken to enable Roma children to overcome their cultural and linguistic disadvantages in the tests. Nor had any instructions been given to restrict the latitude that was given in the administration of the tests and the interpretation of the results. The applicants also drew attention to a 2002 report in which the Czech schools inspectorate noted that children without any significant mental deficiencies were still being placed in special schools. 41. In the report submitted by the Czech Republic on 1 April 1999 pursuant to Article 25 § 1 of the Framework Convention for the Protection of National Minorities, it was noted that the psychological tests “are conceived for the majority population and do not take Romani specifics into consideration”. The Advisory Committee on the Framework Convention noted in its first report on the Czech Republic, which was published on 25 January 2002, that while these schools were designed for mentally handicapped children, it appeared that many Roma children who were not mentally handicapped were placed in them owing to real or perceived language and cultural differences between Roma and the majority. The Committee stressed that “placing children in such special schools should take place only when it is absolutely necessary and always on the basis of consistent, objective and comprehensive tests”. In its second report on the Czech Republic published on 26 October 2005 the Advisory Committee observed: “Tests and methods used to assess children's intellectual abilities upon school enrolment have already been revised with a view to ensuring that they are not misused to the detriment of Roma children”. However, it noted with concern that “revision of the psychological tests used in this context has not had a marked impact. According to unofficial estimates, Roma account for up to 70% of pupils in the [special] schools, and this – having regard to the percentage of Roma in the population – raises doubts concerning the tests' validity and the relevant methodology followed in practice”. 42. In its report on the Czech Republic published on 21 March 2000, the European Commission against Racism and Intolerance (ECRI) noted that channelling of Roma children to special schools was reported to be often quasi-automatic. According to ECRI, the poor results obtained by these children in the pre-school aptitude tests could be explained by the fact that in the Czech Republic most Roma children did not attend kindergarten education. ECRI therefore considered that the practice of channelling Roma/Gypsy children into special schools for the mentally-retarded should be fully examined, to ensure that any testing used was fair and that the true abilities of each child were properly evaluated. In its next report on the Czech Republic, which was published in June 2004, ECRI noted that the test developed by the Czech Ministry of Education for assessing a child's mental level was not mandatory, and was only one of a battery of tools and methods recommended to the educational guidance centres. 43. In his final report on the Human Rights Situation of the Roma, Sinti and Travellers in Europe dated 15 February 2006, the Commissioner for Human Rights observed: “Roma children are frequently placed in classes for children with special needs without an adequate psychological or pedagogical assessment, the real criteria clearly being their ethnic origin”. 44. According to the observations submitted by the International Step by Step Association, the Roma Education Fund and the European Early Childhood Research Association, countries in east central and south eastern Europe typically lacked national definitions of “disability” (related to the placement of students in special schools) and used definitions in which some form of disability was connected to the socio-cultural background of the child, thus leaving the door to discriminatory practices open. Data on children with disabilities were drawn largely from administrative sources rather than being derived from a thorough assessment of the actual characteristics of the child. Thus, divisive practices and the use of a single test were common in the 1990s. It is alleged in the observations that the assessment used to place Roma children in special schools in the Ostrava region ran contrary to effective assessment indicators that were well known by the mid 1990s, for example, those published in 1987 by the National Association for the Education of Young Children (USA). These indicators were now associated with the Global Alliance for the Education of Young Children, which included member organisations in Europe and, more particularly, the Czech Republic. Relevant indicators included: ethical principles to guide assessment practices; the use of assessment instruments for their intended purposes and in such a way as to meet professional quality criteria; assessments appropriate to the ages and other characteristics of the children being assessed; recognition of the developmental and educational significance of the subject-matter of the assessment; the use of assessment evidence to understand and improve learning; the gathering of assessment evidence from realistic settings and in situations that reflected children's actual performance; the use of multiple sources of evidence gathered over time for assessments; the constant linking of screening to follow up; limitations on the use of individually administered, norm-referenced tests; and adequate information for staff and families involved in the assessment process. Thus, the assessment of Roma children in the Ostrava region did not take into account the language and culture of the children, or their prior learning experiences, or their unfamiliarity with the demands of the testing situation. Single rather than multiple sources of evidence were used. Testing was done in a single administration, not over time. Evidence was not obtained in realistic or authentic settings where children could demonstrate their skills. Undue emphasis was placed on individually administered, standardised tests normed on other populations. According to studies cited in these observations (UNICEF, Innocenti Insight (2005); Save the Children (2000), 'Denied a future: The right to education of Roma/Gypsy and Traveller children'; D.J. Losen and G. Orfield (2002), 'Introduction: Racial inequity in special education' (Cambridge, MA: Harvard Education Press)), disproportionately placing certain groups of students in special education resulted from an array of factors, including “unconscious racial bias on the part of school authorities, large resource inequalities, an unjustifiable reliance on IQ and other evaluation tools, educators' inappropriate responses to the pressures of high-stakes testing, and power differentials between minority parents and school officials”. Thus, school placement through psychological testing often reflected racial biases in the society concerned. 45. The respondent Government observed that the unification of European norms used by psychologists was currently under way and that the State authorities had taken all reasonable steps to ensure that the psychological tests were administered by appropriately qualified experts with university degrees applying the latest professional and ethical standards in their specialised field. In addition, research conducted in 1997 by Czech experts at the request of the Ministry of Education showed that Roma children had attained in a standard test of intelligence (WISC III) only insignificantly lower results than comparable non-Roma Czech children (one point on the IQ scale). 2. Consent to placement in a special school 46. Article 7 of Decree no. 127/1997 on specialised schools made the consent of the legal guardians a condition sine qua non for the child's placement in a special school. The applicants noted that the Czech legislation did not require the consent to be in writing. Nor did information on the education provided by special schools or the consequences of the child's placement in a special school have to be provided beforehand. 47. In its report on the Czech Republic published in March 2000, ECRI observed that Roma parents often favoured the channelling of Roma children to special schools, partly to avoid abuse from non-Roma children in ordinary schools and isolation of the child from other neighbourhood Roma children, and partly owing to a relatively low level of interest in education. In its report on the Czech Republic published in June 2004, ECRI noted that when deciding whether or not to give their consent, parents of Roma children “continued to lack information concerning the long-term negative consequences of sending their children to such schools, which were often presented to parents as an opportunity for their children to receive specialised attention and be with other Roma children”. 48. According to information obtained by the FIDH from its Czech affiliate, many schools in the Czech Republic are reluctant to accept Roma children. That reluctance is explained by the reaction of the parents of non-Roma children, which, in numerous cases, has been to remove their children from integrated schools because the parents fear that the level of the school will fall following the arrival of Roma children or, quite simply, because of prejudice against the Roma. It is in that context that Roma children undergo tests designed to ascertain their capacity to follow the ordinary curriculum, following which parents of Roma children are encouraged to place their children in special schools. The parents' choice to place their children in special schools, where that is what they choose to do, is consistent with the school authorities' desire not to admit so many Roma children that their arrival might induce the parents of non-Roma children to remove their own children from the school. 3. Consequences 49. Pupils in special schools follow a special curriculum supposedly adapted to their intellectual capacity. After completing their course of compulsory education in this type of school, they may elect to continue their studies in vocational training centres or, since 18 February 2000, in other forms of secondary school (provided they are able to establish during the admissions procedure that they satisfy the entrance requirements for their chosen course). Further, Article 6 § 2 of Decree no. 127/1997 stipulated that if during the pupil's school career there was a change in the nature of his or her disability or if the specialised school was no longer adapted to the level of disability, the head teacher of the school attended by the child or pupil was required, after an interview with the pupil's guardian, to recommend the pupil's placement in another specialised school or in an ordinary school. 50. In his final report on the Human Rights Situation of the Roma, Sinti and Travellers in Europe dated 15 February 2006, the Commissioner for Human Rights noted: “Being subjected to special schools or classes often means that these children follow a curriculum inferior to those of mainstream classes, which diminishes their opportunities for further education and for finding employment in the future. The automatic placement of Roma children in classes for children with special needs is likely to increase the stigma by labelling the Roma children as less intelligent and less capable. At the same time, segregated education denies both the Roma and non-Roma children the chance to know each other and to learn to live as equal citizens. It excludes Roma children from mainstream society at the very beginning of their lives, increasing the risk of their being caught in the vicious circle of marginalisation”. 51. The Advisory Committee on the Framework Convention for the Protection of National Minorities noted in its second report on the Czech Republic, which was published on 26 October 2005, that placement in a special school “makes it more difficult for Roma children to gain access to other levels of education, thus reducing their chances of integrating in the society. Although legislation no longer prevents children from advancing from 'special' to regular secondary schools, the level of education offered by 'special' schools generally does not make it possible to cope with the requirements of secondary schools, with the result that most drop out of the system”. 52. According to the observations submitted by the International Step by Step Association, the Roma Education Fund and the European Early Childhood Research Association, the placement of children in segregated special schools was an example of a very early “tracking” of students, in this case by assigning children perceived to be of “low ability” or “low potential” to special schools from an early age. Such practices increased educational inequity as they had especially negative effects on the achievement levels of disadvantaged children (see, inter alia, the communication to the Council and the European Parliament on 'efficiency and equity in the European education and training systems' (COM/2006/0481, 8 September 2006)). The longer-term consequences of “tracking” included pupils being channelled towards less prestigious forms of education and training and pupils dropping out of school early. Tracking could thus help create a social construction of failure. 53. In their observations to the Court, the organisations Minority Rights Group International, European Network against Racism and European Roma Information Office noted that children in special schools followed a simplified curriculum that was considered appropriate for their lower level of development. Thus, in the Czech Republic, children in special schools were not expected to know the alphabet or numbers up to 10 until the third or fourth school-year, while their counterparts in regular schools acquired that knowledge in the first year. III. COUNCIL OF EUROPE SOURCES A. The Committee of Ministers Recommendation No. R (2000) 4 of the Committee of Ministers to member states on the education of Roma/Gypsy children in Europe (adopted by the Committee of Ministers on 3 February 2000 at the 696th meeting of the Ministers' Deputies) 54. The recommendation provides as follows: “The Committee of Ministers, under the terms of Article 15.b of the Statute of the Council of Europe, Considering that the aim of the Council of Europe is to achieve greater unity between its members and that this aim may be pursued, in particular, through common action in the field of education; Recognising that there is an urgent need to build new foundations for future educational strategies toward the Roma/Gypsy people in Europe, particularly in view of the high rates of illiteracy or semi-literacy among them, their high drop-out rate, the low percentage of students completing primary education and the persistence of features such as low school attendance; Noting that the problems faced by Roma/Gypsies in the field of schooling are largely the result of long-standing educational policies of the past, which led either to assimilation or to segregation of Roma/Gypsy children at school on the grounds that they were 'socially and culturally handicapped'; Considering that the disadvantaged position of Roma/Gypsies in European societies cannot be overcome unless equality of opportunity in the field of education is guaranteed for Roma/Gypsy children; Considering that the education of Roma/Gypsy children should be a priority in national policies in favour of Roma/Gypsies; Bearing in mind that policies aimed at addressing the problems faced by Roma/Gypsies in the field of education should be comprehensive, based on an acknowledgement that the issue of schooling for Roma/Gypsy children is linked with a wide range of other factors and pre-conditions, namely the economic, social and cultural aspects, and the fight against racism and discrimination; Bearing in mind that educational policies in favour of Roma/Gypsy children should be backed up by an active adult education and vocational education policy; ... Recommends that in implementing their education policies the governments of the member states: – be guided by the principles set out in the appendix to this Recommendation; – bring this Recommendation to the attention of the relevant public bodies in their respective countries through the appropriate national channels.” 55. The relevant sections of the Appendix to Recommendation No. R (2000) 4 read as follows: “Guiding principles of an education policy for Roma/Gypsy children in Europe I. Structures 1. Educational policies for Roma/Gypsy children should be accompanied by adequate resources and the flexible structures necessary to meet the diversity of the Roma/Gypsy population in Europe and which take into account the existence of Roma/Gypsy groups which lead an itinerant or semi-itinerant lifestyle. In this respect, it might be envisaged having recourse to distance education, based on new communication technologies. 2. Emphasis should be put on the need to better co-ordinate the international, national, regional and local levels in order to avoid dispersion of efforts and to promote synergies. 3. To this end member states should make the Ministries of Education sensitive to the question of education of Roma/Gypsy children. 4. In order to secure access to school for Roma/Gypsy children, pre-school education schemes should be widely developed and made accessible to them. 5. Particular attention should also be paid to the need to ensure better communication with parents, where necessary using mediators from the Roma/Gypsy community which could then lead to specific career possibilities. Special information and advice should be given to parents about the necessity of education and about the support mechanisms that municipalities can offer families. There has to be mutual understanding between parents and schools. The parents' exclusion and lack of knowledge and education (even illiteracy) also prevent children from benefiting from the education system. 6. Appropriate support structures should be set up in order to enable Roma/Gypsy children to benefit, in particular through positive action, from equal opportunities at school. 7. The member states are invited to provide the necessary means to implement the above-mentioned policies and arrangements in order to close the gap between Roma/Gypsy pupils and majority pupils. II. Curriculum and teaching material 8. Educational policies in favour of Roma/Gypsy children should be implemented in the framework of broader intercultural policies, taking into account the particular features of the Romani culture and the disadvantaged position of many Roma/Gypsies in the member states. 9. The curriculum, on the whole, and the teaching material should therefore be designed so as to take into account the cultural identity of Roma/Gypsy children. Romani history and culture should be introduced in the teaching material in order to reflect the cultural identity of Roma/Gypsy children. The participation of representatives of the Roma/Gypsy community should be encouraged in the development of teaching material on the history, culture or language of the Roma/Gypsies. 10. However, the member states should ensure that this does not lead to the establishment of separate curricula, which might lead to the setting up of separate classes. 11. The member states should also encourage the development of teaching material based on good practices in order to assist teachers in their daily work with Roma/Gypsy pupils. 12. In the countries where the Romani language is spoken, opportunities to learn in the mother tongue should be offered at school to Roma/Gypsy children. III. Recruitment and training of teachers 13. It is important that future teachers should be provided with specific knowledge and training to help them understand better their Roma/Gypsy pupils. The education of Roma/Gypsy pupils should however remain an integral part of the general educational system. 14. The Roma/Gypsy community should be involved in the designing of such curricula and should be directly involved in the delivery of information to future teachers. 15. Support should also be given to the training and recruitment of teachers from within the Roma/Gypsy community. ...” B. The Parliamentary Assembly 1. Recommendation no. 1203 (1993) on Gypsies in Europe 56. The Parliamentary Assembly made, inter alia, the following general observations: “One of the aims of the Council of Europe is to promote the emergence of a genuine European cultural identity. Europe harbours many different cultures, all of them, including the many minority cultures, enriching and contributing to the cultural diversity of Europe. A special place among the minorities is reserved for Gypsies. Living scattered all over Europe, not having a country to call their own, they are a true European minority, but one that does not fit into the definitions of national or linguistic minorities. As a non-territorial minority, Gypsies greatly contribute to the cultural diversity of Europe. In different parts of Europe they contribute in different ways, be it by language and music or by their trades and crafts. With central and east European countries now member states, the number of Gypsies living in the area of the Council of Europe has increased drastically. Intolerance of Gypsies by others has existed throughout the ages. Outbursts of racial or social hatred, however, occur more and more regularly, and the strained relations between communities have contributed to the deplorable situation in which the majority of Gypsies lives today. Respect for the rights of Gypsies, individual, fundamental and human rights and their rights as a minority is essential to improve their situation. Guarantees for equal rights, equal chances, equal treatment, and measures to improve their situation will make a revival of Gypsy language and culture possible, thus enriching the European cultural diversity. The guarantee of the enjoyment of the rights and freedoms set forth in Article 14 of the European Convention on Human Rights is important for Gypsies as it enables them to maintain their individual rights. ...” 57. As far as education is concerned, the Recommendation states: “The existing European programmes for training teachers of Gypsies should be extended; Special attention should be paid to the education of women in general and mothers together with their younger children; Talented young Gypsies should be encouraged to study and to act as intermediaries for Gypsies; ...” 2. Recommendation no. 1557 (2002): 'The legal situation of Roma in Europe' 58. This recommendation states, inter alia: “... 3. Today Roma are still subjected to discrimination, marginalisation and segregation. Discrimination is widespread in every field of public and personal life, including access to public places, education, employment, health services and housing, as well as crossing borders and access to asylum procedures. Marginalisation and the economic and social segregation of Roma are turning into ethnic discrimination, which usually affects the weakest social groups. 4. Roma form a special minority group, in so far as they have a double minority status. They are an ethnic community and most of them belong to the socially disadvantaged groups of society. ... 15. The Council of Europe can and must play an important role in improving the legal status, the level of equality and the living conditions of Roma. The Assembly calls upon the member states to complete the six general conditions, which are necessary for the improvement of the situation of Roma in Europe: ... c. to guarantee equal treatment for the Romany minority as an ethnic or national minority group in the field of education, employment, housing, health and public services. Member states should give special attention to: i. promoting equal opportunities for Roma on the labour market; ii. providing the possibility for Romany students to participate in all levels of education from kindergarten to university; iii. developing positive measures to recruit Roma in public services of direct relevance to Roma communities, such as primary and secondary schools, social welfare centres, local primary health care centres and local administration; iv. eradicating all practices of segregated schooling for Romany children, particularly that of routing Romany children to schools or classes for the mentally disabled; d. to develop and implement positive action and preferential treatment for the socially deprived strata, including Roma as a socially disadvantaged community, in the field of education, employment and housing...; e. to take specific measures and create special institutions for the protection of the Romany language, culture, traditions and identity: ... ii. to encourage Romany parents to send their children to primary school, secondary school and higher education, including college or university, and give them adequate information about the necessity of education; ... v. to recruit Roma teaching staff, particularly in areas with a large Romany population; f. to combat racism, xenophobia and intolerance and to ensure non-discriminatory treatment of Roma at local, regional, national and international levels: ... vi. to pay particular attention to the phenomenon of the discrimination against Roma, especially in the fields of education and employment; ...” C. The European Commission against Racism and Intolerance (ECRI) 1. ECRI general policy recommendation no. 3: 'Combating racism and intolerance against Roma/Gypsies' (adopted by ECRI on 6 March 1998) 59. The relevant sections of this recommendation state: “The European Commission against Racism and Intolerance: ... Recalling that combating racism, xenophobia, antisemitism and intolerance forms an integral part of the protection and promotion of human rights, that these rights are universal and indivisible, and that all human beings, without any distinction whatsoever, are entitled to these rights; Stressing that combating racism, xenophobia, antisemitism and intolerance is above all a matter of protecting the rights of vulnerable members of society; Convinced that in any action to combat racism and discrimination, emphasis should be placed on the victim and the improvement of his or her situation; Noting that Roma/Gypsies suffer throughout Europe from persisting prejudices, are victims of a racism which is deeply-rooted in society, are the target of sometimes violent demonstrations of racism and intolerance and that their fundamental rights are regularly violated or threatened; Noting also that the persisting prejudices against Roma/Gypsies lead to discrimination against them in many fields of social and economic life, and that such discrimination is a major factor in the process of social exclusion affecting many Roma/Gypsies; Convinced that the promotion of the principle of tolerance is a guarantee of the preservation of open and pluralistic societies allowing for a peaceful co-existence; recommends the following to Governments of member States: ... – to ensure that discrimination as such, as well as discriminatory practices, are combated through adequate legislation and to introduce into civil law specific provisions to this end, particularly in the fields of employment, housing and education; ... – to vigorously combat all forms of school segregation towards Roma/Gypsy children and to ensure the effective enjoyment of equal access to education; ...” 2. ECRI general policy recommendation no. 7 on national legislation to combat racism and racial discrimination (adopted by ECRI on 13 December 2002) 60. The following definitions are used for the purposes of this Recommendation: “a) 'racism' shall mean the belief that a ground such as race, colour, language, religion, nationality or national or ethnic origin justifies contempt for a person or a group of persons, or the notion of superiority of a person or a group of persons. b) 'direct racial discrimination' shall mean any differential treatment based on a ground such as race, colour, language, religion, nationality or national or ethnic origin, which has no objective and reasonable justification. Differential treatment has no objective and reasonable justification if it does not pursue a legitimate aim or if there is not a reasonable relationship of proportionality between the means employed and the aim sought to be realised. c) 'indirect racial discrimination' shall mean cases where an apparently neutral factor such as a provision, criterion or practice cannot be as easily complied with by, or disadvantages, persons belonging to a group designated by a ground such as race, colour, language, religion, nationality or national or ethnic origin, unless this factor has an objective and reasonable justification. This latter would be the case if it pursues a legitimate aim and if there is a reasonable relationship of proportionality between the means employed and the aim sought to be realised. 61. In the explanatory memorandum to this recommendation, it is noted (point 8) that the definitions of direct and indirect racial discrimination contained in paragraph 1 b) and c) of the Recommendation draw inspiration from those contained in Council Directive 2000/43/EC implementing the principle of equal treatment between persons irrespective of racial or ethnic origin and in Council Directive 2000/78/EC establishing a general framework for equal treatment in employment and occupation and on the case-law of the European Court of Human Rights. 3. The report on the Czech Republic published in September 1997 62. In the section of the report dealing with the policy aspects of education and training, ECRI stated that public opinion appeared sometimes to be rather negative towards certain groups, especially the Roma/Gypsy community, and suggested that further measures should be taken to raise public awareness of the issues of racism and intolerance and to improve tolerance towards all groups in society. It added that special measures should be taken as regards the education and training of the members of minority groups, particularly members of the Roma/Gypsy community. 4. The report on the Czech Republic published in March 2000 63. In this report, ECRI stated that the disadvantages and effective discrimination faced by members of the Roma/Gypsy community in the field of education was of particularly serious concern. It was noted that Roma/Gypsy children were vastly over-represented in special schools and that their channelling to special schools was reported to be often quasi-automatic. Roma/Gypsy parents often favoured this solution, partly to avoid abuse from non-Roma/Gypsy children in ordinary schools and isolation of the child from other neighbourhood Roma/Gypsy children, and partly owing to a relatively low level of interest in education. Most Roma/Gypsy children were consequently relegated to educational facilities designed for other purposes, offering little opportunity for skills training or educational preparation and therefore very limited opportunity for further study or employment. Participation of members of the Roma/Gypsy community in education beyond the primary school level was extremely rare. 64. ECRI therefore considered that the practice of channelling Roma/Gypsy children into special schools for the mentally-retarded should be fully examined, to ensure that any testing used was fair and that the true abilities of each child were properly evaluated. ECRI also considered that it was fundamental that Roma/Gypsy parents should be made aware of the need for their children to attend regular education. In general, ECRI considered that there was a need for closer involvement of members of the Roma/Gypsy community in matters concerning education. As a start, the authorities needed to ensure that Roma/Gypsy parents were kept fully informed of measures taken and were encouraged to participate in educational decisions affecting their children. 5. The report on the Czech Republic published in June 2004 65. With regard to the access of Roma children to education, ECRI said in this report that it was concerned that Roma children continued to be sent to special schools which, besides perpetuating their segregation from mainstream society, severely disadvantaged them for the rest of their lives. The standardised test developed by the Czech Ministry of Education for assessing a child's mental level was not mandatory and was only one of a battery of tools and methods recommended to the psychological counselling centres. As to the other element required in order to send a child to a special school – the consent of the child's legal guardian – ECRI observed that parents making such decisions continued to lack information concerning the long-term negative consequences of sending their children to such schools, which were often presented to parents as an opportunity for their children to receive specialised help and be with other Roma children. ECRI also said that it had received reports of Roma parents being turned away from ordinary schools. ECRI also noted that the Schools Act had entered into force in January 2000 and provided the opportunity for pupils from special schools to apply for admission to secondary schools. According to various sources, that remained largely a theoretical possibility as special schools did not provide children with the knowledge required to follow the secondary-school curriculum. There were no measures in place to provide additional education to pupils who had gone through the special school system to bring them to a level where they would be adequately prepared for ordinary secondary schools. ECRI had received very positive feedback concerning the success of 'zero grade courses' (preparatory classes) at pre-school level in increasing the number of Roma children who attended ordinary schools. It expressed its concern, however, over a new trend to maintain the system of segregated education in a new form – this involved special classes in mainstream schools. In that connection, a number of concerned actors were worried that the proposed new Schools Act created the possibility for even further separation of Roma through the introduction of a new category of special programmes for the “socially disadvantaged”. Lastly, ECRI noted that despite initiatives taken by the Ministry of Education (classroom assistants, training programmes for teachers, revision of the primary school curriculum), the problem of low levels of Roma participation in secondary and higher education that had been described by ECRI in its second report persisted. D. Framework Convention for the Protection of National Minorities 1. The report submitted by the Czech Republic on 1 April 1999 pursuant to Article 25 § 1 of the Framework Convention for the Protection of National Minorities 66. The report stated that the Government had adopted measures in the education sphere that were focused on providing suitable conditions especially for children from socially and culturally disadvantaged environments, in particular the Roma community, by opening preparatory classes in elementary and special schools. It was noted that “Romany children with average or above-average intellect are often placed in such schools on the basis of results of psychological tests (this happens always with the consent of the parents). These tests are conceived for the majority population and do not take Romany specifics into consideration. Work is being done on restructuring these tests”. In some special schools Roma pupils made up between 80% and 90% of the total number of pupils. 2. The report submitted by the Czech Republic on 2 July 2004 67. The Czech Republic accepted that the Roma were particularly exposed to discrimination and social exclusion and said that it was preparing to introduce comprehensive anti-discrimination tools associated with the implementation of the Council Directive implementing the principle of equal treatment. New legislation was due to be enacted in 2004 (the Act, Law no. 561/2004, was passed on 24 September 2004 and entered into force on 1 January 2005). In the field of Roma education, the report said that the State had taken various measures of affirmative action in order to radically change the present situation of Roma children. The Government regarded the practice of referring large numbers of Roma children to special schools as untenable. The need for affirmative action was due not only to the socio-cultural handicap of Roma children, but also to the nature of the whole education system and its inability to sufficiently reflect cultural differences. The proposed new Schools Act would bring changes to the special education system by transforming “special schools” into “special primary schools”, thus providing the children targeted assistance in overcoming their socio-cultural handicap. These included preparatory classes, individual study programmes for children in special schools, measures concerning pre-school education, an expanded role for assistants from the Roma community and specialised teacher-training programmes. As one of the main problems encountered by Roma pupils was their poor command of the Czech language, the Ministry of Education considered that the best solution (and the only realistic one) would be to provide preparatory classes at the pre-school stage for children from disadvantaged socio-cultural backgrounds. The report also cited a number of projects and programmes that had been implemented nationally in this sphere ('Support for Roma integration', 'Programme for Roma integration/Multicultural education reform', and 'Reintegrating Roma special school pupils in primary schools'). 3. Opinion on the Czech Republic of the Advisory Committee on the Framework Convention for the Protection of National Minorities, published on 25 January 2002 68. The Advisory Committee noted that while the special schools were designed for mentally handicapped children, it appeared that many Roma children who were not mentally handicapped were placed in these schools due to real or perceived language and cultural differences between Roma and the majority. It considered that this practice was not compatible with the Framework Convention and stressed that placing children in such schools should take place only when absolutely necessary and always on the basis of consistent, objective and comprehensive tests. 69. The special schools had led to a high level of separation of Roma pupils from others and to a low level of educational skills in the Roma community. This was recognised by the Czech authorities. Both governmental and civil society actors agreed on the need for a major reform. There was however disagreement about the precise nature of the reform to be carried out, the amount of resources to be made available and the speed with which reforms were being implemented. The Advisory Committee was of the opinion that the Czech authorities ought to develop the reform, in consultation with the persons concerned, so as to ensure equal opportunities for access to schools for Roma children and equal rights to an ordinary education, in accordance with the principles set out in Committee of Ministers Recommendation No. (2000) 4 on the education of Roma/Gypsy children in Europe. 70. The Advisory Committee noted with approval the initiatives that had been taken to establish so-called zero-classes, allowing the preparation of Roma children for basic school education, inter alia, by improving their Czech language skills, and encouraged the authorities to make these facilities more broadly available. It also considered the creation of posts of Roma pedagogical advisors in schools, a civil society initiative, to be a most positive step. The Advisory Committee encouraged the State authorities in their efforts to ensure the increase and development of such posts. A further crucial objective was to ensure a much higher number of Roma children had access to and successfully completed secondary education. 4. The Advisory Committee's opinion on the Czech Republic, published on 26 October 2005 71. In this opinion, the Advisory Committee noted that the authorities were genuinely committed to improving the educational situation of Roma children, and were trying, in various ways, to realise this aim in practice. In that connection, it noted that it was too early to determine whether the revised educational system introduced by the new Schools Act (Law no. 561/2004) would substantially change the existing situation of over-representation of Roma children in special schools or special classes. 72. The Advisory Committee noted that the authorities were paying special attention to the unjustified placement of Roma children in special schools. Tests and methods used to assess children's intellectual abilities upon school enrolment had already been revised with a view to ensuring that they were not misused to the detriment of Roma children. Special educational programmes had been launched to help Roma children overcome their problems. These included waiving fees for the last year of pre-school education, relaxing the rules on minimum class sizes, more individualised education, appointing educational assistants (mostly Roma), as well as producing methodological handbooks and guidelines for teachers working with Roma children. Preparatory pre-school classes had also been organised for Roma children, and had worked well, although on a fairly limited scale. To accommodate all the children concerned, these measures needed to be applied more widely. The Advisory Committee also took note of the special support programme for Roma access to secondary and higher education, and of the efforts that had been made to build up a network of qualified Roma teachers and educational assistants. 73. The Advisory Committee noted, however, that although constant monitoring and evaluation of the school situation of Roma children was one of the Government's priorities, the State Report said little about the extent to which they were currently integrated in schools, or the effectiveness and impact of the many measures that had been taken for them. It noted with concern that the measures had produced few improvements and that local authorities did not systematically implement the Government's school support scheme and did not always have the determination needed to act effectively in this field. 74. The Advisory Committee noted with concern that, according to non-governmental sources, a considerable number of Roma children were still being placed in special schools at a very early age, and that revision of the psychological tests used in this context had not had a marked impact. According to unofficial estimates, Roma accounted for up to 70% of pupils in these schools, and this – having regard to the percentage of Roma in the population – raised doubts concerning the tests' validity and the methodology followed. This situation was made all the more disturbing by the fact that it also made it more difficult for Roma children to gain access to other levels of education, thus reducing their chances of integrating in society. Although legislation no longer prevented children from advancing from special to ordinary secondary schools, the level of education offered by special schools generally did not make it possible to cope with the requirements of secondary schools, with the result that most dropped out of the system. Although estimates of the number of Roma children who remained outside the school system varied, those who did attend school rarely advanced beyond primary school. 75. In addition, the Advisory Committee noted that, in spite of the awareness-raising initiatives taken by the Ministry of Education, many of the Roma children who attended ordinary schools were isolated by other children and by teaching staff, or even placed in separate classes. At the same time, it was recognised that in some schools Roma children were the largest pupil-group simply because the schools concerned were located near the places where Roma resided compactly. According to other sources, material conditions in some of the schools they attended were precarious and the teaching they received was still, in most cases, insufficiently adapted to their situation. It was important to ensure that these schools, too, provided quality education. 76. According to the Advisory Committee priority had to go to placing Roma children in ordinary schools, supporting and promoting preparatory classes and also to educational assistants. Recruiting Roma teaching staff, and making all education staff aware of the specific situation of Roma children also needed to receive increased attention. An active involvement on the part of the parents, in particular with regard to the implementation of the new Schools Act, also needed to be promoted as a condition sine qua non for the overall improvement of the educational situation of the Roma. Lastly, more determined action was needed to combat isolation of Roma children in both ordinary and special schools. A clearer approach, coupled with instructions and immediate action on all levels, was needed to put an end to unjustified placement of these children in special schools designed for children with mental disabilities. Effective monitoring measures, particularly designed to eliminate undue placement of children in such schools, had to be one of the authorities' constant priorities. E. Commissioner for Human Rights Final Report by Mr Alvaro Gil-Robles on the Human-Rights Situation of the Roma, Sinti and Travellers in Europe (dated 15 February 2006) 77. In the third section of the report, which concerns discrimination in education, the Commissioner noted that the fact that a significant number of Roma children did not have access to education of a similar standard enjoyed by other children was in part a result of discriminatory practices and prejudices. In that connection, he noted that segregation in education was a common feature in many Council of Europe member States. In some countries there were segregated schools in segregated settlements, in others special classes for Roma children in ordinary schools or a clear over-representation of Roma children in classes for children with special needs. Roma children were frequently placed in classes for children with special needs without an adequate psychological or pedagogical assessment, the real criteria clearly being their ethnic origin. Being subjected to special schools or classes often meant that these children followed a curriculum inferior to those of mainstream classes, which diminished their opportunities for further education and for finding employment in the future. The automatic placement of Roma children in classes for children with special needs was likely to increase the stigma by labelling the Roma children as less intelligent and less capable. At the same time, segregated education denied both the Roma and non-Roma children the chance to know each other and to learn to live as equal citizens. It excluded Roma children from mainstream society at the very beginning of their lives, increasing the risk of their being caught in the vicious circle of marginalisation. 78. In the Czech Republic, the Commissioner was told that the young members of the Roma/Gypsy community were drastically over-represented in “special” schools and classes for children with a slight mental disability. At the same time he noted that the authorities had introduced Roma assistant teachers in ordinary classes and set up preliminary classes and that these initiatives had had promising results, though only on a small scale due to the lack of adequate resources. In particular, preparatory classes for socially disadvantaged children had been central in efforts to overcome excessive attendance of Roma children in special schools. The Czech authorities deemed that preparatory schools attached to nursery schools had been particularly successful in easing the integration of Roma children in ordinary schools. In 2004 the Czech Republic also had 332 teacher's assistants who attended the special needs of Roma pupils. 79. It was also noted that special classes or special curricula for the Roma had been introduced with good intentions, for the purposes of overcoming language barriers or remedying the lack of pre-school attendance of Roma children. Evidently, it was necessary to respond to such challenges, but segregation or systematic placement of Roma children in classes which followed a simplified or a special Romani-language curriculum while isolating them from other pupils was clearly a distorted response. Instead of segregation, significant emphasis had to be placed on measures such as pre-school and in-school educational and linguistic support as well as the provision of school assistants to work alongside teachers. In certain communities, it was crucial to raise the awareness of Roma parents, who themselves might not have had the possibility to attend school, of the necessity and benefits of adequate education for their children. 80. In conclusion, the Commissioner made a number of recommendations related to education. Where segregated education still existed in one form or another, it had to be replaced by ordinary integrated education and, where appropriate, banned through legislation. Adequate resources had to be made available for the provision of pre-school education, language training and school assistant training in order to ensure the success of desegregation efforts. Adequate assessment had to be made before children were placed in special classes, in order to ensure that the sole criterion in the placement was the objective needs of the child, not his or her ethnicity. IV. RELEVANT COMMUNITY LAW AND PRACTICE 81. The principle prohibiting discrimination or requiring equality of treatment is well established in a large body of Community law instruments based on Article 13 of the Treaty instituting the European Community. This provision enables the Council, through a unanimous decision following a proposal/recommendation by the Commission and consultation of the European Parliament, to take the measures necessary to combat discrimination based on sex, race or ethnic origin, religion or belief, disability, age or sexual orientation. 82. Thus, Article 2 § 2 of Council Directive 97/80/EC of 15 December 1997 on the burden of proof in cases of discrimination based on sex provides: “Indirect discrimination shall exist where an apparently neutral provision, criterion or practice disadvantages a substantially higher proportion of the members of one sex unless that provision, criterion or practice is appropriate and necessary and can be justified by objective factors unrelated to sex”. Article 4 § 1, which concerns the burden of proof, reads: “Member States shall take such measures as are necessary, in accordance with their national judicial systems, to ensure that, when persons who consider themselves wronged because the principle of equal treatment has not been applied to them establish, before a court or other competent authority, facts from which it may be presumed that there has been direct or indirect discrimination, it shall be for the respondent to prove that there has been no breach of the principle of equal treatment”. 83. Similarly, the aim of Council Directive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment between persons irrespective of racial or ethnic origin and of Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation is to prohibit in their respective spheres all direct or indirect discrimination based on race, ethnic origin, religion or belief, disability, age or sexual orientation. The preambles to these Directives state as follows: “The appreciation of the facts from which it may be inferred that there has been direct or indirect discrimination is a matter for national judicial or other competent bodies, in accordance with rules of national law or practice. Such rules may provide, in particular, for indirect discrimination to be established by any means including on the basis of statistical evidence” and “The rules on the burden of proof must be adapted when there is a prima facie case of discrimination and, for the principle of equal treatment to be applied effectively, the burden of proof must shift back to the respondent when evidence of such discrimination is brought”. 84. In particular, Directive 2000/43/EC provides as follows in Articles 2 (Concept of discrimination) and 8 (Burden of proof): Article 2 “1. For the purposes of this Directive, the principle of equal treatment shall mean that there shall be no direct or indirect discrimination based on racial or ethnic origin. 2. For the purposes of paragraph 1: (a) direct discrimination shall be taken to occur where one person is treated less favourably than another is, has been or would be treated in a comparable situation on grounds of racial or ethnic origin; (b) indirect discrimination shall be taken to occur where an apparently neutral provision, criterion or practice would put persons of a racial or ethnic origin at a particular disadvantage compared with other persons, unless that provision, criterion or practice is objectively justified by a legitimate aim and the means of achieving that aim are appropriate and necessary. ...” Article 8 “1. Member States shall take such measures as are necessary, in accordance with their national judicial systems, to ensure that, when persons who consider themselves wronged because the principle of equal treatment has not been applied to them establish, before a court or other competent authority, facts from which it may be presumed that there has been direct or indirect discrimination, it shall be for the respondent to prove that there has been no breach of the principle of equal treatment. 2. Paragraph 1 shall not prevent Member States from introducing rules of evidence which are more favourable to plaintiffs. 3. Paragraph 1 shall not apply to criminal procedures. ... 5. Member States need not apply paragraph 1 to proceedings in which it is for the court or competent body to investigate the facts of the case.” 85. Under the case-law of the Court of Justice of the European Communities (CJEC), discrimination, which entails the application of different rules to comparable situations or the application of the same rule to different situations, may be overt or covert and direct or indirect. 86. In its Giovanni Maria Sotgiu v. Deutsche Bundespost judgment of 12 February 1974 (Case 152-73, point 11), the CJEC stated: “... the rules regarding equality of treatment ... forbid not only overt discrimination by reason of nationality but also all covert forms of discrimination which, by the application of other criteria of differentiation, lead in fact to the same result...” 87. In its Bilka-Kaufhaus GmbH v. Karin Weber von Hartz judgment of 13 May 1986 (Case 170/84, point 31), it stated: “... Article 119 of the EEC Treaty is infringed by a department store company which excludes part-time employees from its occupational pension scheme, where that exclusion affects a far greater number of women than men, unless the undertaking shows that the exclusion is based on objectively justified factors unrelated to any discrimination on grounds of sex.” 88. In Regina v. Secretary of State for Employment, ex parte Nicole Seymour-Smith and Laura Perez (judgment of 9 February 1999, Case C-167/97, points 51, 57, 62, 65 and 77), the CJEC observed: “... the national court seeks to ascertain the legal test for establishing whether a measure adopted by a Member State has disparate effect as between men and women to such a degree as to amount to indirect discrimination... ... the Commission proposes a 'statistically significant' test, whereby statistics must form an adequate basis of comparison and the national court must ensure that they are not distorted by factors specific to the case. The existence of statistically significant evidence is enough to establish disproportionate impact and pass the onus to the author of the allegedly discriminatory measure. It is also for the national court to assess whether the statistics concerning the situation ... are valid and can be taken into account, that is to say, whether they cover enough individuals, whether they illustrate purely fortuitous or short-term phenomena, and whether, in general, they appear to be significant (see Case C-127/92 Enderby [1993] ECR I-5535, paragraph 17). ... Accordingly, ... in order to establish whether a measure adopted by a Member State has disparate effect as between men and women to such a degree as to amount to indirect discrimination for the purposes of Article 119 of the Treaty, the national court must verify whether the statistics available indicate that a considerably smaller percentage of women than men is able to fulfil the requirement imposed by that measure. If that is the case, there is indirect sex discrimination, unless that measure is justified by objective factors unrelated to any discrimination based on sex. ... ... if a considerably smaller percentage of women than men is capable of fulfilling the requirement ... imposed by the disputed rule, it is for the Member State, as the author of the allegedly discriminatory rule, to show that the said rule reflects a legitimate aim of its social policy, that that aim is unrelated to any discrimination based on sex, and that it could reasonably consider that the means chosen were suitable for attaining that aim.” 89. In its judgment of 23 October 2003 in Hilde Schönheit v. Stadt Frankfurt am Main (Case C-4/02) and Silvia Becker v. Land Hessen (Case C-5/02), the CJEC noted at points 67-69 and 71: “... it must be borne in mind that Article 119 of the Treaty and Article 141(1) and (2) EC set out the principle that men and women should receive equal pay for equal work. That principle precludes not only the application of provisions leading to direct sex discrimination, but also the application of provisions which maintain different treatment between men and women at work as a result of the application of criteria not based on sex where those differences of treatment are not attributable to objective factors unrelated to sex discrimination... It is common ground that the provisions of the BeamtVG at issue do not entail discrimination directly based on sex. It is therefore necessary to ascertain whether they can amount to indirect discrimination... To establish whether there is indirect discrimination, it is necessary to ascertain whether the provisions at issue have a more unfavourable impact on women than on men... Therefore it is necessary to determine whether the statistics available indicate that a considerably higher percentage of women than men is affected by the provisions of the BeamtVG entailing a reduction in the pensions of civil servants who have worked part-time for at least a part of their career. Such a situation would be evidence of apparent discrimination on grounds of sex unless the provisions at issue were justified by objective factors unrelated to any discrimination based on sex.” 90. In Debra Allonby v. Accrington & Rossendale College and Others, Education Lecturing Services ... and Secretary of State for Education and Employment (judgment of 13 January 2004, Case C-256/01), it stated (point 81): “... it must be held that a woman may rely on statistics to show that a clause in State legislation is contrary to Article 141(1) EC because it discriminates against female workers. ...” 91. Lastly, in Commission of the European Communities v. Republic of Austria (judgment of 7 July 2005, Case C-147/03), the CJEC observed (points 41 and 46-48): “According to settled case-law, the principle of equal treatment prohibits not only overt discrimination based on nationality but also all covert forms of discrimination which, by applying other distinguishing criteria, lead in fact to the same result (see, in particular, Case 152/73 Sotgiu [1974] ECR 153, paragraph 11; Case C-65/03 Commission v Belgium , cited above, paragraph 28, and Case C-209/03 Bidar [2005] ECR I-0000, paragraph 51). ... ... the legislation in question places holders of secondary education diplomas awarded in a Member State other than the Republic of Austria at a disadvantage, since they cannot gain access to Austrian higher education under the same conditions as holders of the equivalent Austrian diploma. Thus, although Paragraph ... applies without distinction to all students, it is liable to have a greater effect on nationals of other Member States than on Austrian nationals, and therefore the difference in treatment introduced by that provision results in indirect discrimination. Consequently, the differential treatment in question could be justified only if it were based on objective considerations independent of the nationality of the persons concerned and were proportionate to the legitimate aim of the national provisions (Case C-274/96 Bickel and Franz [1998] ECR I-7637, paragraph 27, and D'Hoop , cited above, paragraph 36).” V. RELEVANT UNITED NATIONS MATERIALS A. International Covenant on Civil and Political Rights 92. Article 26 of the Covenant provides: “All persons are equal before the law and are entitled without any discrimination to the equal protection of the law. In this respect, the law shall prohibit any discrimination and guarantee to all persons equal and effective protection against discrimination on any ground such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status.” B. United Nations Human Rights Committee 93. In points 7 and 12 of its General Observations no. 18 of 10 November 1989 on Non-Discrimination, the Committee expressed the following opinion: “... the Committee believes that the term 'discrimination' as used in the Covenant should be understood to imply any distinction, exclusion, restriction or preference which is based on any ground such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status, and which has the purpose or effect of nullifying or impairing the recognition, enjoyment or exercise by all persons, on an equal footing, of all rights and freedoms. ... when legislation is adopted by a State party, it must comply with the requirement of article 26 that its content should not be discriminatory.” 94. In point 11.7 of its Views dated 31 July 1995 on Communication no. 516/1992 concerning the Czech Republic, the Committee noted: “The Committee is of the view, however, that the intent of the legislature is not alone dispositive in determining a breach of article 26 of the Covenant. A politically motivated differentiation is unlikely to be compatible with article 26. But an act which is not politically motivated may still contravene article 26 if its effects are discriminatory.” C. International Convention on the Elimination of All Forms of Racial Discrimination 95. Article 1 of this Convention provides: “... the term 'racial discrimination' shall mean any distinction, exclusion, restriction or preference based on race, colour, descent, or national or ethnic origin which has the purpose or effect of nullifying or impairing the recognition, enjoyment or exercise, on an equal footing, of human rights and fundamental freedoms in the political, economic, social, cultural or any other field of public life. ...” D. Committee on the Elimination of Racial Discrimination 96. In its General Recommendation no. 14 of 22 March 1993 on the definition of discrimination, the Committee noted, inter alia: “A distinction is contrary to the Convention if it has either the purpose or the effect of impairing particular rights and freedoms. This is confirmed by the obligation placed upon States parties by article 2, paragraph 1 (c), to nullify any law or practice which has the effect of creating or perpetuating racial discrimination. ... In seeking to determine whether an action has an effect contrary to the Convention, [the Committee] will look to see whether that action has an unjustifiable disparate impact upon a group distinguished by race, colour, descent, or national or ethnic origin.” 97. In its General Recommendation no. 19 of 18 August 1995 on racial segregation and apartheid, the Committee observed: “... while conditions of complete or partial racial segregation may in some countries have been created by governmental policies, a condition of partial segregation may also arise as an unintended by-product of the actions of private persons. In many cities residential patterns are influenced by group differences in income, which are sometimes combined with differences of race, colour, descent and national or ethnic origin, so that inhabitants can be stigmatized and individuals suffer a form of discrimination in which racial grounds are mixed with other grounds. The Committee therefore affirms that a condition of racial segregation can also arise without any initiative or direct involvement by the public authorities. ...” 98. In its General Recommendation no. 27 of 16 August 2000 on Discrimination against Roma, the Committee made, inter alia, the following recommendation in the education sphere: “17. To support the inclusion in the school system of all children of Roma origin and to act to reduce drop-out rates, in particular among Roma girls, and, for these purposes, to cooperate actively with Roma parents, associations and local communities. 18. To prevent and avoid as much as possible the segregation of Roma students, while keeping open the possibility for bilingual or mother-tongue tuition; to this end, to endeavour to raise the quality of education in all schools and the level of achievement in schools by the minority community, to recruit school personnel from among members of Roma communities and to promote intercultural education. 19. To consider adopting measures in favour of Roma children, in cooperation with their parents, in the field of education.” 99. In its concluding observations of 30 March 1998 following its examination of the report submitted by the Czech Republic, the Committee noted, inter alia: “13. The marginalization of the Roma community in the field of education is noted with concern. Evidence that a disproportionately large number of Roma children are placed in special schools, leading to de facto racial segregation, and that they also have a considerably lower level of participation in secondary and higher education, raises doubts about whether article 5 of the Convention is being fully implemented.” E. Convention on the Rights of the Child 100. Articles 28 and 30 of this Convention provide: Article 28 “1. States Parties recognize the right of the child to education, and with a view to achieving this right progressively and on the basis of equal opportunity, they shall, in particular: (a) Make primary education compulsory and available free to all; (b) Encourage the development of different forms of secondary education, including general and vocational education, make them available and accessible to every child, and take appropriate measures such as the introduction of free education and offering financial assistance in case of need; (c) Make higher education accessible to all on the basis of capacity by every appropriate means; (d) Make educational and vocational information and guidance available and accessible to all children; (e) Take measures to encourage regular attendance at schools and the reduction of drop-out rates. 2. States Parties shall take all appropriate measures to ensure that school discipline is administered in a manner consistent with the child's human dignity and in conformity with the present Convention. 3. States Parties shall promote and encourage international cooperation in matters relating to education, in particular with a view to contributing to the elimination of ignorance and illiteracy throughout the world and facilitating access to scientific and technical knowledge and modern teaching methods. In this regard, particular account shall be taken of the needs of developing countries.” Article 30 “In those States in which ethnic, religious or linguistic minorities or persons of indigenous origin exist, a child belonging to such a minority or who is indigenous shall not be denied the right, in community with other members of his or her group, to enjoy his or her own culture, to profess and practise his or her own religion, or to use his or her own language.” F. UNESCO 101. Articles 1 to 3 of the Convention against Discrimination in Education of 14 December 1960 provide: Article 1 “1. For the purposes of this Convention, the term 'discrimination' includes any distinction, exclusion, limitation or preference which, being based on race, colour, sex, language, religion, political or other opinion, national or social origin, economic condition or birth, has the purpose or effect of nullifying or impairing equality of treatment in education and in particular: (a) Of depriving any person or group of persons of access to education of any type or at any level; (b) Of limiting any person or group of persons to education of an inferior standard; (c) Subject to the provisions of Article 2 of this Convention, of establishing or maintaining separate educational systems or institutions for persons or groups of persons; or (d) Of inflicting on any person or group of persons conditions which are incompatible with the dignity of man. ...” Article 2 “When permitted in a State, the following situations shall not be deemed to constitute discrimination, within the meaning of Article I of this Convention: (a) The establishment or maintenance of separate educational systems or institutions for pupils of the two sexes, if these systems or institutions offer equivalent access to education, provide a teaching staff with qualifications of the same standard as well as school premises and equipment of the same quality, and afford the opportunity to take the same or equivalent courses of study; (b) The establishment or maintenance, for religious or linguistic reasons, of separate educational systems or institutions offering an education which is in keeping with the wishes of the pupil's parents or legal guardians, if participation in such systems or attendance at such institutions is optional and if the education provided conforms to such standards as may be laid down or approved by the competent authorities, in particular for education of the same level; (c) The establishment or maintenance of private educational institutions, if the object of the institutions is not to secure the exclusion of any group but to provide educational facilities in addition to those provided by the public authorities, if the institutions are conducted in accordance with that object, and if the education provided conforms with such standards as may be laid down or approved by the competent authorities, in particular for education of the same level.” Article 3 “In order to eliminate and prevent discrimination within the meaning of this Convention, the States Parties thereto undertake: (a) To abrogate any statutory provisions and any administrative instructions and to discontinue any administrative practices which involve discrimination in education; (b) To ensure, by legislation where necessary, that there is no discrimination in the admission of pupils to educational institutions; ...” 102. The Declaration on Race and Racial Prejudice adopted by the UNESCO General Conference on 27 November 1978 proclaims as follows: Article 1 “1. All human beings belong to a single species and are descended from a common stock. They are born equal in dignity and rights and all form an integral part of humanity. 2. All individuals and groups have the right to be different, to consider themselves as different and to be regarded as such. However, the diversity of life styles and the right to be different may not, in any circumstances, serve as a pretext for racial prejudice; they may not justify either in law or in fact any discriminatory practice whatsoever, nor provide a ground for the policy of apartheid, which is the extreme form of racism. ...” Article 2 “... 2. Racism includes racist ideologies, prejudiced attitudes, discriminatory behaviour, structural arrangements and institutionalized practices resulting in racial inequality as well as the fallacious notion that discriminatory relations between groups are morally and scientifically justifiable; it is reflected in discriminatory provisions in legislation or regulations and discriminatory practices as well as in anti-social beliefs and acts; it hinders the development of its victims, perverts those who practise it, divides nations internally, impedes international cooperation and gives rise to political tensions between peoples; it is contrary to the fundamental principles of international law and, consequently, seriously disturbs international peace and security. 3. Racial prejudice, historically linked with inequalities in power, reinforced by economic and social differences between individuals and groups, and still seeking today to justify such inequalities, is totally without justification.” Article 3 “Any distinction, exclusion, restriction or preference based on race, colour, ethnic or national origin or religious intolerance motivated by racist considerations, which destroys or compromises the sovereign equality of States and the right of peoples to self-determination, or which limits in an arbitrary or discriminatory manner the right of every human being and group to full development is incompatible with the requirements of an international order which is just and guarantees respect for human rights; the right to full development implies equal access to the means of personal and collective advancement and fulfilment in a climate of respect for the values of civilizations and cultures, both national and world-wide.” Article 5 “1. Culture, as a product of all human beings and a common heritage of mankind, and education in its broadest sense, offer men and women increasingly effective means of adaptation, enabling them not only to affirm that they are born equal in dignity and rights, but also to recognize that they should respect the right of all groups to their own cultural identity and the development of their distinctive cultural life within the national and international contexts, it being understood that it rests with each group to decide in complete freedom on the maintenance, and, if appropriate, the adaptation or enrichment of the values which it regards as essential to its identity. 2. States, in accordance with their constitutional principles and procedures, as well as all other competent authorities and the entire teaching profession, have a responsibility to see that the educational resources of all countries are used to combat racism, more especially by ensuring that curricula and textbooks include scientific and ethical considerations concerning human unity and diversity and that no invidious distinctions are made with regard to any people; by training teachers to achieve these ends; by making the resources of the educational system available to all groups of the population without racial restriction or discrimination; and by taking appropriate steps to remedy the handicaps from which certain racial or ethnic groups suffer with regard to their level of education and standard of living and in particular to prevent such handicaps from being passed on to children. ...” Article 6 “1. The State has prime responsibility for ensuring human rights and fundamental freedoms on an entirely equal footing in dignity and rights for all individuals and all groups. 2. So far as its competence extends and in accordance with its constitutional principles and procedures, the State should take all appropriate steps, inter alia by legislation, particularly in the spheres of education, culture and communication, to prevent, prohibit and eradicate racism racist propaganda, racial segregation and apartheid and to encourage the dissemination of knowledge and the findings of appropriate research in natural and social sciences on the causes and prevention of racial prejudice and racist attitudes with due regard to the principles embodied in the Universal Declaration of Human Rights and in the International Covenant on Civil and Political Rights. 3. Since laws proscribing racial discrimination are not in themselves sufficient, it is also incumbent on States to supplement them by administrative machinery for the systematic investigation of instances of racial discrimination, by a comprehensive framework of legal remedies against acts of racial discrimination, by broadly based education and research programmes designed to combat racial prejudice and racial discrimination and by programmes of positive political, social, educational and cultural measures calculated to promote genuine mutual . respect among groups. Where circumstances warrant, special programmes should be undertaken to promote the advancement of disadvantaged groups and, in the case of nationals, to ensure their effective participation in the decision-making processes of the community.” Article 9 “1. The principle of the equality in dignity and rights of all human beings and all peoples, irrespective of race, colour and origin, is a generally accepted and recognized principle of international law. Consequently any form of racial discrimination practised by a State constitutes a violation of international law giving rise to its international responsibility. 2. Special measures must be taken to ensure equality in dignity and rights for individuals and groups wherever necessary, while ensuring that they are not such as to appear racially discriminatory. In this respect, particular attention should be paid to racial or ethnic groups which are socially or economically disadvantaged, so as to afford them, on a completely equal footing and without discrimination or restriction, the protection of the laws and regulations and the advantages of the social measures in force, in particular in regard to housing, employment and health; to respect the authenticity of their culture and values; and to facilitate their social and occupational advancement, especially through education. ...” VI. OTHER SOURCES A. European Monitoring Centre on Racism and Xenophobia (now the European Union Agency for Fundamental Rights) 103. The information on education in the Czech Republic available on the Internet site of the European Monitoring Centre includes the following: “In the Czech Republic, there are no official or non-official data on racism and discrimination in education available. The most serious problem of the Czech education system is still the segregatory placement of children from socially disadvantaged background (very often Roma) in special schools. More than half of Roma children study there. Such tendencies of the Czech education system especially at elementary schools were proved by extensive research carried out by the Institute of Sociology of the Academy of Sciences of the Czech Republic. Only very small percentage of Roma youth enter secondary schools.” 104. The Monitoring Centre's report entitled 'Roma and Travellers in Public Education', which was published in May 2006 and concerned what at the time were 25 member States of the European Union, noted inter alia that although systematic segregation of Roma children no longer existed as educational policy, segregation was practised by schools and educational authorities in a number of different, mostly indirect, ways sometimes as the unintended effect of policies and practices and sometimes as a result of residential segregation. Schools and educational authorities may, for example, segregate pupils on the basis of a perception of “their different needs” and/or as a response to behavioural issues and learning difficulties. The latter could also lead to the frequent placement of Roma pupils in special schools for mentally handicapped children, which was still a worrying phenomenon in Member States of the European Union like Hungary, Slovakia and the Czech Republic. However, steps were being taken to review testing and placement procedures taking into account the norms and behavioural patterns of the Roma children's social and cultural background. B. The House of Lords 105. In its decision of 9 December 2004 in the case of Regina v. Immigration Officer at Prague Airport and another, ex parte European Roma Rights Centre and others, the House of Lords unanimously held that British immigration officers working at Prague Airport had discriminated against Roma wishing to travel from the airport to Great Britain as they had on racial grounds treated them less favourably than other people travelling to the same destination. 106. Baroness Hale of Richmond said, inter alia: “73. ... The underlying concept in both race and sex discrimination laws is that individuals of each sex and all races are entitled to be treated equally. Thus it is just as discriminatory to treat men less favourably than women as it is to treat women less favourably than men; and it is just as discriminatory to treat whites less favourably than blacks as it is to treat blacks less favourably than whites. The ingredients of unlawful discrimination are (i) a difference in treatment between one person and another person (real or hypothetical) from a different sex or racial group; (ii) that the treatment is less favourable to one; (iii) that their relevant circumstances are the same or not materially different; and (iv) that the difference in treatment is on sex or racial grounds. However, because people rarely advertise their prejudices and may not even be aware of them, discrimination has normally to be proved by inference rather than direct evidence. Once treatment less favourable than that of a comparable person (ingredients (i), (ii) and (iii)) is shown, the court will look to the alleged discriminator for an explanation. The explanation must, of course, be unrelated to the race or sex of the complainant. If there is no, or no satisfactory explanation, it is legitimate to infer that the less favourable treatment was on racial grounds... 74. If direct discrimination of this sort is shown, that is that. Save for some very limited exceptions, there is no defence of objective justification. The whole point of the law is to require suppliers to treat each person as an individual, not as a member of a group. The individual should not be assumed to hold the characteristics which the supplier associates with the group, whether or not most members of the group do indeed have such characteristics, a process sometimes referred to as stereotyping. ... 75. The complaint in this case is of direct discrimination against the Roma. Indirect discrimination arises where an employer or supplier treats everyone in the same way, but he applies to them all a requirement or condition which members of one sex or racial group are much less likely to be able to meet than members of another: for example, a test of heavy lifting which men would be much more likely to pass than women. This is only unlawful if the requirement is one which cannot be justified independently of the sex or race of those involved... But it is the requirement or condition that may be justified, not the discrimination. This sort of justification should not be confused with the possibility that there may be an objective justification for discriminatory treatment which would otherwise fall foul of article 14 of the European Convention on Human Rights. ... 90. It is worth remembering that good equal opportunities practice may not come naturally. Many will think it contrary to common sense to approach all applicants with an equally open mind, irrespective of the very good reasons there may be to suspect some of them more than others. But that is what is required by a law which tries to ensure that individuals are not disadvantaged by the general characteristics of the group to which they belong. In 2001, when the operation with which we are concerned began, the race relations legislation had only just been extended to cover the activities of the immigration service. It would scarcely be surprising if officers acting under considerable pressure of time found it difficult to conform in all respects to procedures and expectations which employers have been struggling to get right for more than quarter of a century. 91. It is against this background that such evidence as there is of what happened on the ground at Prague Airport needs to be assessed. The officers did not make any record of the ethnic origin of the people they interviewed. The respondents cannot therefore provide us with figures of how many from each group were interviewed, for how long, and with what result. This, they suggest, makes it clear that the officers were not relying on the Authorisation: if they had been, they would only have had to record their view of the passenger's ethnicity. If correct, that would have been enough to justify refusal of leave. But what it also shows is that no formal steps were being taken to gather the information which might have helped ensure that this high-risk operation was not being conducted in a discriminatory manner. It also means that the only information available is that supplied by the claimants, and in particular the ERRC which was attempting to monitor the operation. The respondents can cast doubt on the reliability of this, but they cannot contradict it or provide more reliable information themselves. ...” C. The United States Supreme Court 107. The Supreme Court issued its decision in the case of Griggs v. Duke Power Co., 401 U.S. 424 (1971), in which it established the disparate impact test, after black employees at an electricity generating plant had brought proceedings on the grounds that their employers' practice of requiring them to hold a high school diploma or to pass an aptitude test, even for the least well-paid jobs, was discriminatory. Fewer blacks had managed to obtain the diploma or pass the standardised tests. The Supreme Court stated: “The [Civil Rights] Act [of 1964] requires the elimination of artificial, arbitrary, and unnecessary barriers to employment that operate invidiously to discriminate on the basis of race, and, if, as here, an employment practice that operates to exclude Negroes cannot be shown to be related to job performance, it is prohibited, notwithstanding the employer's lack of discriminatory intent. The Act does not preclude the use of testing or measuring procedures, but it does proscribe giving them controlling force unless they are demonstrably a reasonable measure of job performance... The Act proscribes not only overt discrimination but also practices that are fair in form, but discriminatory in operation. The touchstone is business necessity. If an employment practice which operates to exclude Negroes cannot be shown to be related to job performance, the practice is prohibited. ... ... Congress has placed on the employer the burden of showing that any given requirement must have a manifest relationship to the employment in question.” THE LAW I. SCOPE OF THE GRAND CHAMBER'S JURISDICTION 108. In their final observations, which were lodged with the Grand Chamber on 26 September 2006, the applicants repeated their contention that there had been a violation of their rights under Article 3 and Article 6 § 1 of the Convention. 109. Under the Court's case-law, the “case” referred to the Grand Chamber is the application as it has been declared admissible (see, among other authorities, Leyla Şahin v. Turkey [GC], no. 44774/98, § 128, ECHR 2005-XI; and Üner v. the Netherlands [GC], no. 46410/99, § 41, ECHR 2006‑...). The Grand Chamber notes that in its partial decision of 1 March 2005 the Chamber declared inadmissible all the applicants' complaints that did not relate to Article 14 of the Convention read in conjunction with Article 2 of Protocol No 1, including those under Articles 3 and 6 § 1 of the Convention. Accordingly, the latter complaints – assuming the applicants still wish to rely on them – are not within the scope of the case before the Grand Chamber. II. THE GOVERNMENT'S PRELIMINARY OBJECTION 110. The Court notes that in its decision on the admissibility of the application the preliminary objection made by the Government in their observations of 15 March 2004 of a failure to exhaust domestic remedies was joined to the merits of the complaint under Article 14 of the Convention read in conjunction with Article 2 of Protocol No. 1. In its judgment of 7 February 2005 (§ 31), the Chamber found that the parties' submissions on the issue of the exhaustion of domestic remedies raised questions that were closely linked to the merits of the case. It agreed with the Czech Constitutional Court that the application raised points of considerable importance and that vital interests were at stake. Accordingly, and in view of its finding that for other reasons pertaining to the merits there had been no violation, the Chamber did not consider it necessary to examine whether the applicants had satisfied that requirement in the present case. 111. It will be recalled that where a case is referred to it, the Grand Chamber may also examine issues relating to the admissibility of the application, for example where they have been joined to the merits or are otherwise relevant at the merits stage (K. and T. v. Finland [GC], no. 25702/94, § 141, ECHR 2001-VII). 112. In these circumstances, the Grand Chamber considers it necessary to determine whether the applicants have in the instant case satisfied the exhaustion of domestic remedies requirement. 113. The Government argued that the applicants had not used all available means to remedy their position. None of them had exercised their right to appeal against the decisions to place them in special schools. Six had failed to lodge a constitutional appeal. Further, of those applicants who had appealed to the Constitutional Court only five had actually contested the decisions to place them in special schools. No attempt had been made by the applicants to defend their dignity by bringing an action under the Civil Code to protect their personality rights and their parents had not referred the matter to the schools inspectorate or the Ministry of Education. 114. The applicants submitted, firstly, that there were no remedies available in the Czech Republic that were effective and adequate to deal with complaints of racial discrimination in the education sphere. More specifically, the right to lodge a constitutional appeal had been rendered ineffective by the reasoning followed by the Constitutional Court in the instant case and its refusal to attach any significance to the general practice that had been referred to by the applicants. In the applicants' submission, no criticism could therefore be made of those applicants who had chosen not to lodge such an appeal. As to why they had not lodged an administrative appeal, the applicants said that their parents had only gained access to the requisite information after the time allowed for lodging such an appeal had expired. Even the Constitutional Court had disregarded that omission. Finally, an action to protect personality rights could not be regarded as a means of challenging enforceable administrative decisions and the Government had not provided any evidence that such a remedy was effective. Further, even supposing that an effective remedy existed, the applicants submitted that it did not have to be exercised in cases in which an administrative practice, such as the system of special schools in the Czech Republic, made racism possible or encouraged it. They also drew the Court's attention to the racial hatred and numerous acts of violence directed at Roma in the Czech Republic and to the unsatisfactory nature of the penalties imposed for racist and xenophobic criminal offences. 115. The Court reiterates that the rule of exhaustion of domestic remedies referred to in Article 35 § 1 of the Convention is based on the assumption that the domestic system provides an effective remedy in respect of the alleged breach. It is for the Government claiming non-exhaustion to satisfy the Court that an effective remedy was available in theory and in practice at the relevant time; that is to say, that the remedy was accessible, capable of providing redress in respect of the applicant's complaints and offered reasonable prospects of success (V. v. the United Kingdom [GC], no. 24888/94, § 57, ECHR 1999-IX). 116. The application of the rule of exhaustion of domestic remedies must make due allowance for the fact that it is being applied in the context of machinery for the protection of human rights that the Contracting States have agreed to set up. The Court has accordingly recognised that Article 35 § 1 must be applied with some degree of flexibility and without excessive formalism. It has further recognised that the rule of exhaustion is neither absolute nor capable of being applied automatically; for the purposes of reviewing whether it has been observed, it is essential to have regard to the circumstances of the individual case. This means, in particular, that the Court must take realistic account not only of the existence of formal remedies in the legal system of the Contracting State concerned but also of the general context in which they operate, as well as the personal circumstances of the applicant. It must then examine whether, in all the circumstances of the case, the applicant did everything that could reasonably be expected of him or her to exhaust domestic remedies (İlhan v. Turkey [GC], no. 22277/93, § 59, ECHR 2000‑VII). 117. In the present case, the Government complained, firstly, that none of the applicants had sought to appeal against the decision ordering their placement in a special school or brought an action to protect their personality rights. 118. In this connection, the Court, like the applicants, notes that the Czech Constitutional Court decided to disregard that omission (see paragraph 28 above). In these circumstances, it considers that it would be unduly formalistic to require the applicants to exercise a remedy which even the highest court of the country concerned had not obliged them to use. 119. Secondly, the Government stated that of the twelve applicants who had lodged a constitutional appeal, only five had actually contested the decisions to place them in special schools, so enabling the Constitutional Court to hear their cases. 120. The Court notes that by virtue of the fact that the five applicants concerned had brought a constitutional appeal in due form, the Constitutional Court was given an opportunity to rule on all the complaints which the applicants have now referred to the Court. The Constitutional Court also found that the scope of the appeals went beyond the applicants' own personal interests so that, in that sense, its decision was of more general application. 121. Further, it can be seen from its decision of 20 October 1999 that the Constitutional Court confined itself to verifying the competent authorities' interpretation and application of the relevant statutory provisions without considering their impact, which the applicants argued was discriminatory. As regards the complaint of racial discrimination, it also stated that it was not its role to assess the overall social context. 122. In these circumstances, there is nothing to suggest that the Constitutional Court's decision would have been different had it been called upon to decide the cases of the thirteen applicants who did not lodge a constitutional appeal or challenge the decision of the head teacher of the special school. In the light of these considerations, the Court is not satisfied that, in the special circumstances of the present case, this remedy was apt to afford the applicants redress for their complaints or offered reasonable prospects of success. 123. Consequently, the Government's preliminary objection in this case must be rejected. III. ALLEGED VIOLATION OF ARTICLE 14 OF THE CONVENTION READ IN CONJUNCTION WITH ARTICLE 2 OF PROTOCOL NO. 1 124. The applicants maintained that they had been discriminated against in that because of their race or ethnic origin they had been treated less favourably than other children in a comparable situation without any objective and reasonable justification. They relied in that connection on Article 14 of the Convention, read in conjunction with Article 2 of Protocol No. 1, which provisions provide as follows: Article 14 of the Convention “The enjoyment of the rights and freedoms set forth in [the] Convention shall be secured without discrimination on any ground such as sex, race, colour, language, religion, political or other opinion, national or social origin, association with a national minority, property, birth or other status.” Article 2 of Protocol No. 1 “No person shall be denied the right to education. In the exercise of any functions which it assumes in relation to education and to teaching, the State shall respect the right of parents to ensure such education and teaching in conformity with their own religious and philosophical convictions.” A. The Chamber judgment 125. The Chamber held that there had been no violation of Article 14 of the Convention, read in conjunction with Article 2 of Protocol No 1. In its view, the Government had succeeded in establishing that the system of special schools in the Czech Republic had not been introduced solely to cater for Roma children and that considerable efforts had been made in those schools to help certain categories of pupils to acquire a basic education. In that connection, it observed that the rules governing children's placement in special schools did not refer to the pupils' ethnic origin, but pursued the legitimate aim of adapting the education system to the needs, aptitudes and disabilities of the children. 126. The Chamber noted in particular that the applicants had not succeeded in refuting the experts' findings that their learning difficulties were such as to prevent them from following the ordinary primary school curriculum. It was further noted that the applicants' parents had failed to take any action or had themselves requested their children's placement or continued placement in a special school. 127. The Chamber accepted in its judgment that it was not easy to choose an education system that reconciled the various competing interests and that there did not appear to be an ideal solution. However, while acknowledging that the statistical evidence disclosed worrying figures and that the general situation in the Czech Republic concerning the education of Roma children was by no means perfect, it considered that the concrete evidence before it did not enable it to conclude that the applicants' placement or, in some instances, continued placement, in special schools was the result of racial prejudice. B. The parties' submissions to the Grand Chamber 1. The applicants 128. The applicants submitted that the restrictive interpretation the Chamber had given to the notion of discrimination was incompatible not only with the aim of the Convention but also with the case-law of the Court and of other jurisdictions in Europe and beyond. 129. They firstly asked the Grand Chamber to correct the obscure and contradictory test the Chamber had used for deciding whether there had been discrimination. They noted that, while reaffirming the established principle that if a policy or general measure had disproportionately prejudicial effects on a group of people, the possibility of its being considered discriminatory could not be ruled out even if it was not specifically aimed or directed at that group, the Chamber had nevertheless departed from the Court's previous case-law (Thlimmenos v. Greece [GC], no. 34369/97, § 44, ECHR 2000-IV; Hoogendijk v. the Netherlands (dec.), no. 58461/00, 6 January 2005; and Nachova and Others v. Bulgaria [GC], nos. 43577/98 and 43579/98, § 157, ECHR 2005-...) by erroneously requiring the applicants to prove discriminatory intent on the part of the Czech authorities. In the applicants' submission, such a requirement was unrealistic and illogical as the question whether or not special schools were designed to segregate along ethnic lines was irrelevant since that was indisputably the effect they had in practice. The reality was that well-intentioned actors often engaged in discriminatory practices through ignorance, neglect or inertia. 130. The applicants observed in particular that in explaining why it had refused to shift the burden of proof in its Nachova and Others v. Bulgaria judgment ([GC], cited above, § 157) the Court had been careful to distinguish between racially-motivated violent crime and non-violent acts of racial discrimination in, for example, employment or the provision of services. In their submission, racial discrimination in access to education fell precisely in the latter category of discriminatory acts which could be proved in the absence of intent. More recently, the Court had ruled in the Zarb Adami v. Malta case (no. 17209/02, §§ 75 and 76, ECHR 2006-...) that a difference in treatment did not need to be set forth in legislative text in order to breach Article 14 and that a “well-established practice” or “de facto situation” could also give rise to discrimination. As, in the instant case, the applicants considered that they had indisputably succeeded in establishing the existence of a disproportionate impact, the burden of proof had to shift to the Government to prove that the applicants' ethnic origin had had no bearing on the impugned decisions and that sufficient safeguards against discrimination were in place. 131. In that connection, the applicants noted that in its General Policy Recommendation No. 7, ECRI had invited the States to prohibit both direct discrimination and indirect discrimination, with neither concept requiring proof of discriminatory intent. A clear majority of the member States of the Council of Europe had already expressly prohibited discrimination in sections of their national legislation without requiring proof of such intent and this was reflected in the judicial practice of those States. The applicants referred in this context to, inter alia, the decision of the House of Lords in the case of Regina v. Immigration Officer at Prague Airport and another, ex parte European Roma Rights Centre and others (see paragraph 105 above) and to the jurisprudence of the Court of Justice of the European Communities (CJEC). Lastly, they noted that indirect discrimination was also prohibited under international law, including the International Covenant on Civil and Political Rights and the Convention on the Elimination of Racial Discrimination. 132. Accordingly, in view of the vital importance of Article 14 protection and the need to make it effective, the applicants considered that it would be helpful for the Court to clarify the rules it applied in such situations to ensure, inter alia, that the principle of non-discrimination was interpreted and applied consistently by the two European courts. For this reason, the applicants asked the Grand Chamber to give a clear ruling that intent was not necessary to prove discrimination under Article 14, except in cases – such as, for example, of racially motivated violence – where it was already an element of the underlying offence. 133. In the instant case, the applicants did not claim that the competent authorities had at the relevant time harboured invidiously racist attitudes towards Roma, or that they had intended to discriminate against Roma, or even that they had failed to take positive measures. All the applicants needed to prove – and, in their submission, had proved – was that the authorities had subjected the applicants to differential adverse treatment in comparison with similarly situated non-Roma, without objective and reasonable justification. The question of a common European standard that had been raised by the Government was, in the applicants' view, more of a political issue and the existence or otherwise of such a standard was of no relevance as the principle of equality of treatment was a binding rule of international law. 134. Similarly, the applicants asked the Grand Chamber to provide guidance concerning the kinds of proof, including but not limited to statistical evidence, which might be relevant to a claim of a violation of Article 14. They noted that the Chamber had discounted the overwhelming statistical evidence they had adduced, without checking whether or not it was accurate, despite the fact that it had been corroborated by independent specialised intergovernmental bodies (ECRI, the Committee on the Elimination of Racial Discrimination, and the Advisory Committee on the Framework Convention for the Protection of National Minorities) and by the Government's own admission (see paragraphs 41 and 66 above). According to this data, although Roma represented only 5% of all primary school pupils at the time the application was lodged, they made up more than 50% of the population of special schools. Whereas fewer than 2% of non-Roma pupils in Ostrava were assigned to special schools, over 50% of Roma children were sent to such schools. Overall, a Roma child was more than 27 times more likely than a similarly situated non-Roma child to be assigned to a special school. 135. In the applicants' view, these figures strongly suggested that, whether through conscious design or reprehensible neglect, race or ethnicity had infected the process of school assignment to a substantial – perhaps determining – extent. The presumption that they, like other Roma children in the city of Ostrava, had been the victims of discrimination on the grounds of ethnic origin had never been rebutted. It was undisputed that as a result of their assignment to special schools the applicants had received a substantially inferior education as compared with non-Roma children and that this had effectively deprived them of the opportunity to pursue a secondary education other than in a vocational training centre. 136. In this context, they argued that both in Europe and beyond statistical data was often used in cases which, as here, concerned discriminatory effect, as sometimes it was the only means of proving indirect discrimination. Statistical data was accepted as a means of proof of discrimination by the bodies responsible for supervising the United Nations treaties and by the Court of Justice of the European Communities. Council Directive 2000/43/EC expressly provided that indirect discrimination could be established by any means “including on the basis of statistical evidence”. 137. With respect to the Convention institutions, the applicants noted that, in finding racial discrimination in the case of East African Asians v. the United Kingdom (nos. 4403/70-4530/70, Commission report of 14 December 1973, Decisions and Reports 78-B, p. 5), the Commission took into account the surrounding circumstances including statistical data on the disproportionate effect the legislation had on British citizens of Asian origin. Recently, the Court had indicated in its decision in the case of Hoogendijk v. the Netherlands (cited above) that while statistics alone were not sufficient to prove discrimination, they could – particularly where they were undisputed – amount to prima facie evidence requiring the Government to provide an objective explanation of the differential treatment. Further, in its decision in the case of Zarb Adami v. Malta (cited above), the Court had relied, inter alia, on statistical evidence of disproportionate effect. 138. The applicants added that it would be helpful for the Grand Chamber to clarify the Court's case-law by determining whether there was an objective and reasonable justification for the purposes of Article 14 for the difference in treatment in the present case and specifying the conclusions that should be drawn in the absence of a satisfactory explanation. Referring to, inter alia, the judgments in the cases of Timishev v. Russia (nos. 55762/00 and 55974/00, § 56, ECHR 2005‑...) and Moldovan v. Romania (no. 2) (nos. 41138/98 and 64320/01, § 140, ECHR 2005‑... (extracts)), they stated that where an applicant had established a difference in treatment the onus was on the respondent government to prove that it was justified. In the absence of a racially neutral explanation, it was legitimate to conclude that the difference in treatment was based on racial grounds. In the applicants' submission, neither an inadequate command of the Czech language, nor poverty nor a different socio-economic status could constitute an objective and reasonable justification in their case. They denied that the disproportionately large number of Roma children in special schools could be explained by the results of intellectual capacity tests or justified by parental consent (see also paragraphs 141 and 142 below). 139. In view of the importance of the fight against racial and ethnic discrimination that had constantly been reaffirmed by the Strasbourg institutions, the applicants considered that the Grand Chamber should state in clear terms that the States' “margin of appreciation” could not serve to justify segregation in education. The approach adopted by the Chamber, which left an unlimited margin of appreciation to the Czech State, was unjustified in view of the serious allegations of racial and ethnic discrimination in the instant case and was inconsistent with the Court's case-law. The present case warranted all the more the Court's attention in that it concerned one of the most important substantive rights, namely the right to education. 140. The applicants further argued that the Chamber had misinterpreted crucial evidence and drawn inappropriate conclusions on two decisive issues, namely parental consent and the reliability of the psychological tests. 141. There were no uniform rules at the material time governing the manner in which the tests used by the educational psychology centres were administered and the results interpreted, so that much had been left to the discretion of the psychologists and there had been considerable scope for racial prejudice and cultural insensitivity. Further, the tests which they and other Roma children had been forced to sit were scientifically flawed and educationally unsound. The documentary evidence showed that a number of the applicants had been placed in special schools for reasons other than intellectual deficiencies (such as absenteeism, bad behaviour, and even misconduct on the part of the parents). The Czech Government had themselves acknowledged the discriminatory effect of the tests (see paragraph 66 above). They had also admitted in their observations on the present case that one of the applicants had been placed in a special school despite possessing good verbal-expression skills. 142. Nor, in the applicants' submission, could the discriminatory treatment to which they had been subjected be justified by their parents' consent to their placement in the special schools. Governments were legally bound to protect the higher interest of the child and in particular the equal right of all children to education. Neither parental conduct nor parental choice could deprive them of that right. The credibility of the “consent” allegedly given by the parents of several of the applicants had been called into question by inconsistencies in the school records that raised doubts as to whether they had indeed agreed. In any event, even supposing that consent had been given by all the parents, it had no legal value as the parents concerned had never been properly informed of their right to withhold their consent, of alternatives to placement in a special school or of the risks and consequences of such a placement. The procedure was largely formal: the parents were given a pre-completed form and the results of the psychological tests, results they believed they had no right to contest. As to the alleged right subsequently to request a transfer to an ordinary school, the applicants pointed out that from their very first year at school they had received a substantially inferior education that made it impossible for them subsequently to meet the requirements of the ordinary schools. Moreover, it was unrealistic to consider the issue of consent without taking into account the history of Roma segregation in education and the absence of adequate information on the choices available to Roma parents. Referring to the view that had been expressed by the Court (in Haakansson and Sturesson v. Sweden, judgment of 21 February 1990, Series A no. 171-A, § 66) that a waiver may be lawful for certain rights but not for others and that it must not run counter to any important public interest, the applicants submitted that there could be no waiver of the child's right not to be racially discriminated against in education. 143. The instant case raised “a serious issue of general importance”, namely whether European governments were capable of coping with increasing racial and ethnic diversity and of protecting vulnerable minorities. In that connection, the most important issue was that of equality of opportunity in education as discrimination against Roma in that sphere persisted in all the member States of the Council of Europe. Putting an end to discrimination at school would enable Roma to enjoy equality of treatment generally. 144. The racial segregation of Roma children in Czech schools had not materially changed since the date the application was filed. The applicants' own futures and lack of prospects revealed the harm that their discriminatory placement in special schools had caused. Thus, in May 2006 eight of the applicants were continuing their education in a special school while a further six who had completed special school found themselves unemployed. Of the four applicants who had been allowed to attend ordinary primary school after passing the aptitude tests, two were still at school, one was unemployed and the fourth was enrolled in a vocational secondary school. The applicants considered that it was already clear that none of them would receive a general secondary school education, still less a university education. 145. Finally, the applicants pointed out that a new Schools Act had been passed in late 2004, which had purported to end the special school system. The new legislation thus acknowledged that the very existence of schools deemed “special” imposed a badge of inferiority on those assigned there. In reality, however, the new law had not brought about changes in practice as it had merely altered the criteria on which educational programmes were based. Extensive research carried out by the European Roma Rights Centre in 2005 and 2006 showed that in many cases special schools had simply been renamed “remedial schools” or “practical schools” without any substantial change in the composition of their teaching staff or the content of their curriculum. 2. The Government 146. The Government stated that the case raised complex issues concerning the social problem of the position of Roma in contemporary society. Although the Roma ostensibly enjoyed the same rights as other citizens, in reality their prospects were limited by both objective and subjective factors. There could be no improvement in their situation without the involvement and commitment of all members of the Roma community. When they attempted to eliminate these inequalities, member States were confronted with numerous political, social, economic and technical problems which could not be confined to the question of respect for fundamental rights. It was for this reason that the courts, including the European Court of Human Rights, had to exercise a degree of restraint when examining measures adopted in this field and confine themselves to deciding whether or not the competent authorities had overstepped their margin of appreciation. 147. Referring to their previous written and oral observations, the Government reiterated that race, colour or association with a national minority had not played a determining role in the applicants' education. There was no specific evidence of any difference in treatment of the applicants on the basis of those grounds. The applicants' school files showed beyond doubt that their placement in special schools was not based on their ethnic origin, but on the results of psychological tests carried out at the educational psychology centres. Since the applicants had been placed in special schools on account of their specific educational needs resulting essentially from their intellectual capacity and, since the criteria, the process by which the criteria were applied and the system of special schools were all racially neutral, as the Chamber had confirmed in its judgment, it was not possible to speak of overt or direct discrimination in the instant case. 148. The Government next turned to the applicants' argument that the instant case was one of indirect discrimination which, in some instances, could only be established with the aid of statistics. They contended that the case of Zarb Adami v. Malta (cited above), in which the Court had relied extensively on statistical evidence submitted by the parties, was not comparable to the instant case. Firstly, Zarb Adami was far less complex. Secondly, the statistical disparities found in that case between the number of men and women called to perform jury service were the result of a decision by the State, whereas the statistics relied on by the applicants in the instant case reflected first and foremost the parents' wishes for their children to attend special school, not any act or omission on the part of the State. Had the parents not expressed such a wish (by giving their consent) the children would not have been placed in a special school. Further, the statistical information that had been submitted in the instant case by the applicants was not sufficiently conclusive as the data had been furnished by the head teachers of the schools and therefore only reflected their subjective opinions. There was no official information on the ethnic origin of the pupils. The Government further considered that the statistics had no informative value without an evaluation of the socio-cultural background of the Roma, their family situation and their attitude towards education. They pointed out in that connection that the Ostrava region had one of the largest Roma populations in the Czech Republic. As to the comparative studies on countries from central and eastern Europe and beyond cited in the observations of the third-party interveners, the Government did not consider that there was any relevant link between those statistics and the substantive issues in the case to hand. In their submission, those studies tended to confirm that creating an education system optimised for Roma children was an extremely complex task. 149. Nevertheless, even assuming that the data submitted by the applicants was reliable and that the State could be considered responsible for the situation, that did not, in the Government's submission, amount to indirect discrimination that was incompatible with the Convention. The impugned measure was consistent with the principle of non-discrimination as it pursued a legitimate aim, namely the adaptation of the education process to the capacity of children with specific educational needs. It was also objectively and reasonably justified. 150. On this latter point, the Government contested the applicants' claim that the Government had not submitted any satisfactory explanation regarding the large number of Roma in special schools. While admitting that the situation of the Roma with regard to education was not ideal, the Government considered that they had demonstrated that the special schools had not been established for the Roma community and that ethnic origin had not been a criterion for deciding on placements in special schools. They reiterated that special-school placements were only possible after prior individualised pedagogical and psychological testing. The testing process was a technical tool that was the subject of continuing scientific research and for that reason could only be carried out by qualified personnel. The courts did not possess the necessary qualifications and therefore had to exercise a degree of restraint in this field. As regards the professional standards referred to in the observations of the International Step by Step Association and others, the Government emphasised that these were not legal norms possessing force of law but, at most, non-binding recommendations or indications by specialists and that the failure to apply them could not, by definition, entail international legal responsibility. 151. The files of each of the applicants contained full details of the methods that had been used and the results of the testing. These had not been challenged at the time by any of the applicants. The applicants' allegations that the psychologists had followed a subjective approach appeared to be biased and not based on any evidence. 152. The Government again conceded that there might have been rare situations where the reason for the placement in a special school was on the borderline between learning difficulties and a socio-culturally disadvantaged environment. Among the eighteen cases, this had apparently happened in one case only, that of the ninth applicant. Otherwise, the pedagogical-psychological diagnostics and the testing at the educational psychology centres had proved learning difficulties in the case of all the applicants. 153. The educational psychology centres that had administered the tests had only made recommendations concerning the type of school in which the child should be placed. The essential, decisive factor was the wishes of the parents. In the instant case, the parents had been informed that their children's placement in a special school depended on their consent and the consequences of such a decision had been explained to them. If the effect of their consent was not entirely clear, they could have appealed against the decision regarding placement and could at any time have required their child's transfer to a different type of school. If, as they now alleged, their consent was not informed, they should have sought information from the competent authorities. The Government noted in this respect that Article 2 of Protocol No. 1 to the Convention emphasised the primary role and responsibility of parents in the education of their children. The State could not intervene if there was nothing in the parents' conduct to indicate that they were unable or unwilling to decide on the most appropriate form of education for their children. Interference of that sort would contravene the principle that the State had to respect parents' wishes regarding education and teaching. In the instant case, the Government noted that apart from appealing to the Constitutional Court and lodging an application with the European Court of Human Rights, the applicants' parents had on the whole done nothing to spare their children the alleged discriminatory treatment and had played a relatively passive role in their education. 154. The Government rejected the applicants' argument that their placement in special schools had prevented them from pursuing a secondary or higher education. Whether the applicants had finished their compulsory education before or after the entry into force of the new Schools Act (Law no. 561/2004), it had been open to them to pursue their secondary education, to take additional lessons to bring them up to the appropriate level or to seek career advice. However, none of the applicants had established that they had attempted to do so (albeit unsuccessfully) or that their (alleged) difficulties were due to a more limited education as a result of their earlier placement in a special school. On the contrary, several of the applicants had decided not to pursue their studies or had abandoned them. The Government were firmly convinced that the applicants had deprived themselves of the possibility of continuing their studies through a lack of interest. Their situation, which in many cases was unfavourable, had stemmed mainly from their own lack of interest, and was not something for which the State could be held responsible. 155. The Government conceded that the national authorities had to take all reasonable steps to ensure that measures did not produce disproportionate effects or, if that was not feasible, to mitigate and compensate for such effects. However, neither the Convention nor any other international instrument contained a general definition of the State's positive obligations concerning the education of Roma pupils or, more generally, of children from national or ethnic minorities. The Government noted in this connection that when determining the State's positive obligations, the Court sometimes referred to developments in the legislation of the Contracting Parties. However, they said that no European standard or consensus currently existed regarding the criteria to be used to determine whether children should be placed in special schools or how children with special learning needs should be educated and the special school was one of the possible and acceptable solutions to the problem. 156. Moreover, the positive obligations under Article 14 of the Convention could not be construed as an obligation to take affirmative action. That had to remain an option. It was not possible to infer from Article 14 a general obligation on the part of the State actively to compensate for all the disabilities which different sections of the population suffered from. 157. In any event, since special schools had to be regarded as an alternative, but not inferior, form of education, the Government submitted that they had in the instant case adopted reasonable measures to compensate for the disabilities of the applicants, who required a special education as a result of their individual situation, and that they had not overstepped the margin of appreciation which the Convention afforded the States in the education sphere. They observed that the State had allocated twice the level of resources to special schools as to ordinary schools and that the domestic authorities had made considerable efforts to deal with the complex issue of the education of Roma children. 158. The Government went on to provide information on the applicants' current situation obtained from the files of both the school and the Ostrava Job Centre (where those applicants who were unemployed had signed on). As a preliminary, they noted that the Ostrava region was afflicted by a high rate of unemployment and that, in general, young people who had received only a primary education had difficulties in finding work. While it was possible to obtain a qualification and career counselling from the State, the active participation of the job applicant was essential. In concrete terms, two applicants were currently in their final year at primary school. Seven had begun vocational training in a secondary school in September 2006. Four had started but later abandoned their secondary-school studies, the majority through a lack of interest, and had instead signed on at the job centre. Lastly, five of the applicants had not sought to pursue their studies at secondary-school level but had registered at the job centre. Those applicants who had registered at a job centre had not co-operated with it or shown any interest in the offers of training or employment that had been made, with the result that some of them had already been struck off the job-applicants register (in some instances repeatedly). 159. Lastly, the Government rejected the applicants' claim that nothing had been changed by the introduction of the Schools Act (Law no. 561/2004). The Act unified the previously existing types of primary school and standardised the educational programmes. It did not provide for a separate, independent system of specialised schools, with the exception of schools for pupils with serious mental disorders, autism or combined mental and physical defects. Pupils with disabilities were individually integrated, wherever possible and desirable, into conventional schools. However, schools were authorised to set up separate classes with educational techniques and methods adjusted to their needs. The former “special schools” could continue to function as separate institutions, but were now “primary schools” providing education under a modified educational programme for primary education. Schools at which socially disadvantaged pupils were educated often made use of their right to establish assistant teacher's posts and preparatory classes designed to improve the children's communication skills and command of the Czech language. Teacher's assistants from the Roma community often served as a link between the school, family, and, in some instances, other experts and helped to integrate pupils into the education system. The region where the applicants lived favoured integrating Roma pupils in classes drawn from the majority population. 160. In their concluding submissions, the Government asked the Court carefully to examine the issue of the applicants' access to education in each individual case, though without losing sight of the overall context, and to hold that there had been no violation of the Convention. 3. The interveners (a) Interights and Human Rights Watch 161. Interights and Human Rights Watch stated that it was essential that Article 14 of the Convention should afford effective protection against indirect discrimination, a concept which the Court had not yet had many occasions to consider. They submitted that aspects of the Chamber's reasoning were out of step with recent developments in cases such as Timishev v. Russia (judgment cited above), Zarb Adami v. Malta (judgment cited above) and Hoogendijk v. the Netherlands (decision cited above). The Grand Chamber needed to consolidate a purposive interpretation of Article 14 and to bring the Court's jurisprudence on indirect discrimination in line with existing international standards. 162. Interights and Human Rights Watch noted that the Court itself had confirmed in Zarb Adami that discrimination was not always direct or explicit and that a policy or general measure could result in indirect discrimination. It had also accepted that intent was not required in cases of indirect discrimination (Hugh Jordan v. the United Kingdom, no. 24746/94, 4 May 2001, § 154). In their submission, it was sufficient in the case of indirect discrimination that the practice or policy resulted in a disproportionate adverse effect on a particular group. 163. As to proof of indirect discrimination, it was widely accepted in Europe and internationally and also by the Court (see Timishev, judgment cited above, § 57; and Hoogendijk, decision cited above) that the burden of proof had to shift once a prima facie case of discrimination had been established. In cases of indirect discrimination, where the applicant had demonstrated that significantly more people of a particular category were placed at a disadvantage by a given policy or practice, a presumption of discrimination arose. The burden then shifted to the State to reject the basis for the prima facie case, or to provide a justification for it. 164. It was therefore critical for the Court to engage with the type of evidence that might be produced in order to shift the burden of proof. Interights and Human Rights Watch submitted on this point that the Court's position with regard to statistical evidence, as set out in the Hugh Jordan judgment (cited above, § 154), was at variance with international and comparative practice. In European Communities Directives and international instruments, statistics were the key method of proving indirect discrimination. Where measures were neutral on their face, statistics sometimes proved the only effective means of identifying their varying impact on different segments of society. Obviously, courts had to assess the credibility, strength and relevance of the statistics to the case at hand, requiring that they be tied to the applicant's allegations in concrete ways. If, however, the Court were to maintain the position that statistics alone were not sufficient to disclose a discriminatory practice, Interights and Human Rights Watch submitted that the general social context should be taken into account, as it provided valuable insight into the extent to which the effects of the measure on the applicants were disproportionate. (b) Minority Rights Group International, the European Network against Racism and the European Roma Information Office 165. The Minority Rights Group International, the European Network against Racism and the European Roma Information Office submitted that the wrongful assignment of Roma children to special schools for the mentally disabled was the most obvious and odious form of discrimination against the Roma. Children in such special schools followed a simplified curriculum considered appropriate for their lower level of intellectual development. Thus, for example, in the Czech Republic, children in special schools were not expected to know the Czech alphabet or numbers up to 10 until the third or fourth school-year, while their counterparts in ordinary schools acquired that knowledge in the first year. 166. This practice had received considerable attention, both at the European level and within the human-rights bodies of the United Nations, which had expressed their concern in various reports as to the over-representation of Roma children in special schools, the adequacy of the tests employed and the quality of the alleged parental consent. All these bodies had found that no objective and reasonable justification could legitimise the disadvantage faced by Roma children in the field of education. The degree of consistency among the institutions and quasi-judicial bodies was persuasive in confirming the existence of widespread discrimination against Roma children. 167. The interveners added that whatever the merits of separate education for children with genuine mental disabilities, the decision to place Roma children in special schools was in the majority of cases not based on any actual mental disability but rather on language and cultural differences which were not taken into account in the testing process. In order to fulfil their obligation to secure equal treatment for Roma in the exercise of their right to education, the first requirement of States was to amend the testing process so that it was not racially prejudiced against Roma and to take positive measures in the area of language training and social-skills training. (c) International Step by Step Association, the Roma Education Fund and the European Early Childhood Education Research Association 168. The International Step by Step Association, the Roma Education Fund and the European Early Childhood Education Research Association sought to demonstrate that the assessment used to place Roma children in special schools in the Ostrava region disregarded the numerous effective and appropriate indicators that were well-known by the mid-1990s (see paragraph 44 above). In their submission, the assessment had not taken into account the language and culture of the children, their prior learning experiences or their unfamiliarity with the demands of the testing situation. Single rather than multiple sources of evidence had been used. Testing had been done in a single administration, not over time. Evidence had not been obtained in realistic or authentic settings where children could demonstrate their learning. Undue emphasis had been placed on individually administered, standardised tests normed on other populations. 169. Referring to various studies that had been carried out (see paragraph 44 above), the interveners noted that minority children and those from vulnerable families were over-represented in special education in central and eastern Europe. This resulted from an array of factors, including unconscious racial bias on the part of school authorities, large resource inequalities, unjustifiable reliance on IQ and other evaluation tools, educators' inappropriate responses to the pressures of “high-stakes” testing and power differentials between minority parents and school officials. School placement through psychological testing often reflected racial biases in the society concerned. 170. The Czech Republic was notable for its placement of children in segregated settings because of “social disadvantage”. According to a comparison of data on fifteen countries collected by the OECD in 1999 (see paragraph 18 in fine above) the Czech Republic ranked third in placing pupils with learning difficult disabilities in special school settings. Of the eight countries that provided data on the placement of pupils as a result of social factors, the Czech Republic was the only one to have recourse to special schools; the other countries almost exclusively used ordinary schools for educating such pupils. 171. Further, the practice of referring children labelled as being of low ability to special schools at an early age (educational tracking) frequently led, whether or not intentionally, to racial segregation and had particularly negative effects on the level of education of disadvantaged children. This had long-term detrimental consequences for both them and society, including premature exclusion from the education system with the resulting loss of job opportunities for those concerned. (d) Fédération internationale des ligues des droits de l'Homme (International Federation for Human Rights – FIDH) 172. The FIDH considered that the Chamber had unjustifiably placed significant weight in its judgment on the consent which the applicants' parents had allegedly given to the situation forming the subject of their complaint to the Court. It noted that under the Court's case-law there were situations in which the waiver of a right was not considered capable of exempting the State from its obligation to guarantee to every person within its jurisdiction the rights and freedoms laid down in the Convention. That applied, in particular, where the waiver conflicted with an important public interest, or was not explicit or unequivocal. Furthermore, in order to be capable of justifying a restriction of the right or freedom of the individual, the waiver of that guarantee by the person concerned had to take place in circumstances from which it could be concluded that he was fully aware of the consequences, in particular the legal consequences, of his choice. In the case of R. v. Borden ([1994] 3 RCS 145, p. 162) the Supreme Court of Canada had developed the following principle on that precise point: “[i]n order for a waiver of the right ... to be effective, the person purporting to consent must be possessed of the requisite informational foundation for a true relinquishment of the right. A right to choose requires not only the volition to prefer one option over another, but also sufficient available information to make the preference meaningful”. 173. The question therefore arose as to whether, in the light of the nature of the principle of equality of treatment, and of the link between the prohibition of racial discrimination and the wider concept of human dignity, waiver of the right to protection against discrimination ought not to be precluded altogether. In the instant case, the consent obtained from the applicants' parents was binding not solely on the applicants but on all the children of the Roma community. It was perfectly possible – indeed, in the FIDH's submission, probable – that all parents of Roma children would prefer an integrated education for their children, but that, being uncertain as regards the choice that would be made by other parents in that situation, they preferred the “security” offered by special education, which was followed by the vast majority of Roma children. In a context characterised by a history of discrimination against the Roma, the choice available to the parents of Roma children was between (a) placing their children in schools where the authorities were reluctant to admit them and where they feared being the subject of various forms of harassment and of manifestations of hostility on the part of their fellow pupils and of teachers, or (b) placing them in special schools where Roma children were in a large majority and where, consequently, they would not have to fear the manifestation of such prejudices. In reality, the applicants' parents had chosen what they saw as being the lesser of two evils, in the absence of any real possibility of receiving an integrated education which would unreservedly welcome Roma. The disproportion between the two alternatives was such that the applicants' parents had been obliged to make the choice for which the Government now sought to hold them responsible 174. For the reasons set out above, the FIDH considered that in the circumstances of the instant case, the alleged waiver by the applicants' parents of the right for their children to receive an education in normal schools could not justify exempting the Czech Republic from its obligations under the Convention. C. The Court's assessment 1. Recapitulation of the main principles 175. The Court has established in its case-law that discrimination means treating differently, without an objective and reasonable justification, persons in relevantly similar situations (Willis v. the United Kingdom, no. 36042/97, § 48, ECHR 2002‑IV; and Okpisz v. Germany, no. 59140/00, § 33, 25 October 2005). However, Article 14 does not prohibit a member State from treating groups differently in order to correct “factual inequalities” between them; indeed in certain circumstances a failure to attempt to correct inequality through different treatment may in itself give rise to a breach of the Article (“Case relating to certain aspects of the laws on the use of languages in education in Belgium” v. Belgium (Merits), judgment of 23 July 1968, Series A no. 6, § 10; Thlimmenos v. Greece [GC], no. 34369/97, § 44, ECHR 2000-IV; and Stec and Others v. the United Kingdom [GC], no. 65731/01, § 51, ECHR 2006‑...). The Court has also accepted that a general policy or measure that has disproportionately prejudicial effects on a particular group may be considered discriminatory notwithstanding that it is not specifically aimed at that group (Hugh Jordan v. the United Kingdom, no. 24746/94, § 154, 4 May 2001; and Hoogendijk v. the Netherlands (dec.), no. 58461/00, 6 January 2005), and that discrimination potentially contrary to the Convention may result from a de facto situation (Zarb Adami v. Malta, no. 17209/02, § 76, ECHR 2006‑...). 176. Discrimination on account of, inter alia, a person's ethnic origin is a form of racial discrimination. 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. It is for this reason that the authorities must use all available means to combat racism, thereby reinforcing democracy's vision of a society in which diversity is not perceived as a threat but as a source of enrichment (Nachova and Others v. Bulgaria [GC], nos. 43577/98 and 43579/98, § 145, ECHR 2005‑...; and Timishev v. Russia, nos. 55762/00 and 55974/00, § 56, ECHR 2005-...). The Court has also held that no difference in treatment which is based exclusively or to a decisive extent on a person's ethnic origin is capable of being objectively justified in a contemporary democratic society built on the principles of pluralism and respect for different cultures (Timishev, cited above, § 58). 177. As to the burden of proof in this sphere, the Court has established that once the applicant has shown a difference in treatment, it is for the Government to show that it was justified (see, among other authorities, Chassagnou and Others v. France [GC], nos. 25088/94, 28331/95 and 28443/95, §§ 91-92, ECHR 1999‑III; and Timishev, cited above, § 57). 178. As regards the question of what constitutes prima facie evidence capable of shifting the burden of proof on to the respondent State, the Court stated in Nachova and Others (cited above, § 147) that in proceedings before it there are no procedural barriers to the admissibility of evidence or pre-determined formulae for its assessment. The Court adopts the conclusions that are, in its view, supported by the free evaluation of all evidence, including such inferences as may flow from the facts and the parties' submissions. According to its established case-law, proof may follow from the coexistence of sufficiently strong, clear and concordant inferences or of similar unrebutted presumptions of fact. Moreover, the level of persuasion necessary for reaching a particular conclusion and, in this connection, the distribution of the burden of proof are intrinsically linked to the specificity of the facts, the nature of the allegation made and the Convention right at stake. 179. The Court has also recognised that Convention proceedings do not in all cases lend themselves to a rigorous application of the principle affirmanti incumbit probatio (he who alleges something must prove that allegation – Aktaş v. Turkey (extracts), no. 24351/94, § 272, ECHR 2003‑V). In certain circumstances, 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 (Salman v. Turkey [GC], no. 21986/93, § 100, ECHR 2000-VII; and Anguelova v. Bulgaria, no. 38361/97, § 111, ECHR 2002‑IV). In the case of Nachova and Others, cited above, § 157), the Court did not rule out requiring a respondent Government to disprove an arguable allegation of discrimination in certain cases, even though it considered that it would be difficult to do so in that particular case in which the allegation was that an act of violence had been motivated by racial prejudice. It noted in that connection that in the legal systems of many countries proof of the discriminatory effect of a policy, decision or practice would dispense with the need to prove intent in respect of alleged discrimination in employment or in the provision of services. 180. As to whether statistics can constitute evidence, the Court has in the past stated that statistics could not in themselves disclose a practice which could be classified as discriminatory (Hugh Jordan, cited above, § 154). However, in more recent cases on the question of discrimination, in which the applicants alleged a difference in the effect of a general measure or de facto situation (Hoogendijk, cited above; and Zarb Adami, cited above, §§ 77-78), the Court relied extensively on statistics produced by the parties to establish a difference in treatment between two groups (men and women) in similar situations. Thus, in the Hoogendijk decision the Court stated: “[W]here an applicant is able to show, on the basis of undisputed official statistics, the existence of a prima facie indication that a specific rule – although formulated in a neutral manner – in fact affects a clearly higher percentage of women than men, it is for the respondent Government to show that this is the result of objective factors unrelated to any discrimination on grounds of sex. If the onus of demonstrating that a difference in impact for men and women is not in practice discriminatory does not shift to the respondent Government, it will be in practice extremely difficult for applicants to prove indirect discrimination.” 181. Lastly, as noted in previous cases, the vulnerable position of Roma/Gypsies means that special consideration should be given to their needs and their different lifestyle both in the relevant regulatory framework and in reaching decisions in particular cases (Chapman v. the United Kingdom [GC], no. 27238/95, § 96, ECHR 2001‑I; and Connors v. the United Kingdom, no. 66746/01, § 84, 27 May 2004). In Chapman (cited above, §§ 93-94), the Court also observed that there could be said to be an emerging international consensus amongst the Contracting States of the Council of Europe recognising the special needs of minorities and an obligation to protect their security, identity and lifestyle, not only for the purpose of safeguarding the interests of the minorities themselves but to preserve a cultural diversity of value to the whole community. 2. Application of the aforementioned principles to the instant case 182. The Court notes that as a result of their turbulent history and constant uprooting the Roma have become a specific type of disadvantaged and vulnerable minority (see also the general observations in the Parliamentary Assembly's Recommendation no. 1203 (1993) on Gypsies in Europe, cited in paragraph 56 above and point 4 of its Recommendation no. 1557 (2002): 'The legal situation of Roma in Europe', cited in paragraph 58 above). As the Court has noted in previous cases, they therefore require special protection (see paragraph 181 above). As is attested by the activities of numerous European and international organisations and the recommendations of the Council of Europe bodies (see paragraphs 54-61 above), this protection also extends to the sphere of education. The present case therefore warrants particular attention, especially as when the applications were lodged with the Court the applicants were minor children for whom the right to education was of paramount importance. 183. The applicants' allegation in the present case is not that they were in a different situation from non-Roma children that called for different treatment or that the respondent State had failed to take affirmative action to correct factual inequalities or differences between them (Thlimmenos, cited above, § 44; and Stec and Others, cited above, § 51). In their submission, all that has to be established is that, without objective and reasonable justification, they were treated less favourably than non-Roma children in a comparable situation and that this amounted in their case to indirect discrimination. 184. The Court has already accepted in previous cases that a difference in treatment may take the form of disproportionately prejudicial effects of a general policy or measure which, though couched in neutral terms, discriminates against a group (Hugh Jordan, cited above, § 154; and Hoogendijk, cited above). In accordance with, for instance, Council Directives 97/80/EC and 2000/43/EC (see paragraphs 82 and 84 above) and the definition provided by ECRI (see paragraph 60 above), such a situation may amount to “indirect discrimination”, which does not necessarily require a discriminatory intent. (a) Whether a presumption of indirect discrimination arises in the instant case 185. It was common ground that the impugned difference in treatment did not result from the wording of the statutory provisions on placements in special schools in force at the material time. Accordingly, the issue in the instant case is whether the manner in which the legislation was applied in practice resulted in a disproportionate number of Roma children – including the applicants – being placed in special schools without justification, and whether such children were thereby placed at a significant disadvantage. 186. As mentioned above, the Court has noted in previous cases that applicants may have difficulty in proving discriminatory treatment (Nachova and Others, cited above, §§ 147 and 157). In order to guarantee those concerned the effective protection of their rights, less strict evidential rules should apply in cases of alleged indirect discrimination. 187. On this point, the Court observes that Council Directives 97/80/EC and 2000/43/EC stipulate that persons who consider themselves wronged because the principle of equal treatment has not been applied to them may establish, before a domestic authority, by any means, including on the basis of statistical evidence, facts from which it may be presumed that there has been discrimination (see paragraphs 82 and 83 above). The recent case-law of the Court of Justice of the European Communities (see paragraphs 88-89 above) shows that it permits claimants to rely on statistical evidence and the national courts to take such evidence into account where it is valid and significant. The Grand Chamber further notes the information furnished by the third-party interveners that the courts of many countries and the supervisory bodies of the United Nations treaties habitually accept statistics as evidence of indirect discrimination in order to facilitate the victims' task of adducing prima facie evidence. The Court also recognised the importance of official statistics in the aforementioned cases of Hoogendijk and Zarb Adami and has shown that it is prepared to accept and take into consideration various types of evidence (Nachova and Others, cited above, § 147). 188. In these circumstances, the Court considers that when it comes to assessing the impact of a measure or practice on an individual or group, statistics which appear on critical examination to be reliable and significant will be sufficient to constitute the prima facie evidence the applicant is required to produce. This does not, however, mean that indirect discrimination cannot be proved without statistical evidence. 189. Where an applicant alleging indirect discrimination thus establishes a rebuttable presumption that the effect of a measure or practice is discriminatory, the burden then shifts to the respondent State, which must show that the difference in treatment is not discriminatory (see, mutatis mutandis, Nachova and Others, cited above, § 157). Regard being had in particular to the specificity of the facts and the nature of the allegations made in this type of case (ibid., § 147), it would be extremely difficult in practice for applicants to prove indirect discrimination without such a shift in the burden of proof. 190. In the present case, the statistical data submitted by the applicants was obtained from questionnaires that were sent out to the head teachers of special and primary schools in the town of Ostrava in 1999. It indicates that at the time 56% of all pupils placed in special schools in Ostrava were Roma. Conversely, Roma represented only 2.26% of the total number of pupils attending primary school in Ostrava. Further, whereas only 1.8% of non-Roma pupils were placed in special schools, the proportion of Roma pupils in Ostrava assigned to special schools was 50.3%. According to the Government, these figures are not sufficiently conclusive as they merely reflect the subjective opinions of the head teachers. The Government also noted that no official information on the ethnic origin of the pupils existed and that the Ostrava region had one of the largest Roma populations. 191. The Grand Chamber observes that these figures are not disputed by the Government and that they have not produced any alternative statistical evidence. In view of their comment that no official information on the ethnic origin of the pupils exists, the Court accepts that the statistics submitted by the applicants may not be entirely reliable. It nevertheless considers that these figures reveal a dominant trend that has been confirmed both by the respondent State and the independent supervisory bodies which have looked into the question. 192. In their reports submitted in accordance with Article 25 § 1 of the Framework Convention for the Protection of National Minorities, the Czech authorities accepted that in 1999 Roma pupils made up between 80% and 90% of the total number of pupils in some special schools (see paragraph 66 above) and that in 2004 “large numbers” of Roma children were still being placed in special schools (see paragraph 67 above). The Advisory Committee on the Framework Convention observed in its report of 26 October 2005 that according to unofficial estimates Roma accounted for up to 70% of pupils enrolled in special schools. According to the report published by ECRI in 2000, Roma children were “vastly overrepresented” in special schools. The Committee on the Elimination of Racial Discrimination noted in its concluding observations of 30 March 1998 that a disproportionately large number of Roma children were placed in special schools (see paragraph 99 above). Lastly, according to the figures supplied by the European Monitoring Centre on Racism and Xenophobia, more than half of Roma children in the Czech Republic attended special school. 193. In the Court's view, the latter figures, which do not relate solely to the Ostrava region and therefore provide a more general picture, show that, even if the exact percentage of Roma children in special schools at the material time remains difficult to establish, their number was disproportionately high. Moreover, Roma pupils formed a majority of the pupils in special schools. Despite being couched in neutral terms, the relevant statutory provisions therefore had considerably more impact in practice on Roma children than on non-Roma children and resulted in statistically disproportionate numbers of placements of the former in special schools. 194. Where it has been shown that legislation produces such a discriminatory effect, the Grand Chamber considers that, as with cases concerning employment or the provision of services, it is not necessary in cases in the educational sphere (see, mutatis mutandis, Nachova and Others, cited above, § 157) to prove any discriminatory intent on the part of the relevant authorities (see paragraph 184 above). 195. In these circumstances, the evidence submitted by the applicants can be regarded as sufficiently reliable and significant to give rise to a strong presumption of indirect discrimination. The burden of proof must therefore shift to the Government, which must show that the difference in the impact of the legislation was the result of objective factors unrelated to ethnic origin. (b) Objective and reasonable justification 196. The Court reiterates that a difference in treatment is discriminatory if “it has no objective and reasonable justification”, that is, if it does not pursue a “legitimate aim” or if there is not a “reasonable relationship of proportionality” between the means employed and the aim sought to be realised (see, among many other authorities, Larkos v. Cyprus [GC], no. 29515/95, § 29, ECHR 1999-I; and Stec and Others, cited above, § 51). Where the difference in treatment is based on race, colour or ethnic origin, the notion of objective and reasonable justification must be interpreted as strictly as possible. 197. In the instant case, the Government sought to explain the difference in treatment between Roma children and non-Roma children by the need to adapt the education system to the capacity of children with special needs. In the Government's submission, the applicants were placed in special schools on account of their specific educational needs, essentially as a result of their low intellectual capacity measured with the aid of psychological tests in educational psychology centres. After the centres had made their recommendations regarding the type of school in which the applicants should be placed, the final decision had lain with the applicants' parents and they had consented to the placements. The argument that the applicants were placed in special schools on account of their ethnic origin was therefore unsustainable. For their part, the applicants strenuously contested the suggestion that the disproportionately high number of Roma children in special schools could be explained by the results of the intellectual capacity tests or be justified by parental consent. 198. The Court accepts that the Government's decision to retain the special-school system was motivated by the desire to find a solution for children with special educational needs. However, it shares the disquiet of the other Council of Europe institutions who have expressed concerns about the more basic curriculum followed in these schools and, in particular, the segregation the system causes. 199. The Grand Chamber observes, further, that the tests used to assess the children's learning abilities or difficulties have given rise to controversy and continue to be the subject of scientific debate and research. While accepting that it is not its role to judge the validity of such tests, various factors in the instant case nevertheless lead the Grand Chamber to conclude that the results of the tests carried out at the material time were not capable of constituting objective and reasonable justification for the purposes of Article 14 of the Convention. 200. In the first place, it was common ground that all the children who were examined sat the same tests, irrespective of their ethnic origin. The Czech authorities themselves acknowledged in 1999 that “Romany children with average or above-average intellect” were often placed in such schools on the basis of the results of psychological tests and that the tests were conceived for the majority population and did not take Roma specifics into consideration (see paragraph 66 above). As a result, they had revised the tests and methods used with a view to ensuring that they “were not misused to the detriment of Roma children” (see paragraph 72 above). In addition, various independent bodies have expressed doubts over the adequacy of the tests. Thus, the Advisory Committee on the Framework Convention for the Protection of National Minorities observed that children who were not mentally handicapped were frequently placed in these schools “[owing] to real or perceived language and cultural differences between Roma and the majority”. It also stressed the need for the tests to be “consistent, objective and comprehensive” (see paragraph 68 above). ECRI noted that the channelling of Roma children to special schools for the mentally-retarded was reportedly often “quasi-automatic” and needed to be examined to ensure that any testing used was “fair” and that the true abilities of each child were “properly evaluated” (see paragraphs 63-64 above). The Council of Europe Commissioner for Human Rights noted that Roma children were frequently placed in classes for children with special needs “without an adequate psychological or pedagogical assessment, the real criteria clearly being their ethnic origin” (see paragraph 77 above). Lastly, in the submission of some of the third-party interveners, placements following the results of the psychological tests reflected the racial prejudices of the society concerned. 201. The Court considers that, at the very least, there is a danger that the tests were biased and that the results were not analysed in the light of the particularities and special characteristics of the Roma children who sat them. In these circumstances, the tests in question cannot serve as justification for the impugned difference in treatment. 202. As regards parental consent, the Court notes the Government's submission that this was the decisive factor without which the applicants would not have been placed in special schools. In view of the fact that a difference in treatment has been established in the instant case, it follows that any such consent would signify an acceptance of the difference in treatment, even if discriminatory, in other words a waiver of the right not to be discriminated against. However, under the Court's case-law, the waiver of a right guaranteed by the Convention – in so far as such a waiver is permissible – must be established in an unequivocal manner, and be given in full knowledge of the facts, that is to say on the basis of informed consent (Pfeifer and Plankl v. Austria, judgment of 25 February 1992, Series A no. 227, §§ 37-38) and without constraint (Deweer v. Belgium, judgment of 27 February 1980, Series A no. 35, § 51). 203. In the circumstances of the present case, the Court is not satisfied that the parents of the Roma children, who were members of a disadvantaged community and often poorly educated, were capable of weighing up all the aspects of the situation and the consequences of giving their consent. The Government themselves admitted that consent in this instance had been given by means of a signature on a pre-completed form that contained no information on the available alternatives or the differences between the special-school curriculum and the curriculum followed in other schools. Nor do the domestic authorities appear to have taken any additional measures to ensure that the Roma parents received all the information they needed to make an informed decision or were aware of the consequences that giving their consent would have for their children's futures. It also appears indisputable that the Roma parents were faced with a dilemma: a choice between ordinary schools that were ill-equipped to cater for their children's social and cultural differences and in which their children risked isolation and ostracism and special schools where the majority of the pupils were Roma. 204. In view of the fundamental importance of the prohibition of racial discrimination (see Nachova and Others, cited above, § 145; and Timishev, cited above, § 56), the Grand Chamber considers that, even assuming the conditions referred to in paragraph 202 above were satisfied, no waiver of the right not to be subjected to racial discrimination can be accepted, as it would be counter to an important public interest (see, mutatis mutandis, Hermi v. Italy [GC], no. 18114/02, § 73, ECHR 2006‑...). (c) Conclusion 205. As is apparent from the documentation produced by ECRI and the report of the Commissioner for Human Rights of the Council of Europe, the Czech Republic is not alone in having encountered difficulties in providing schooling for Roma children: other European States have had similar difficulties. The Court is gratified to note that, unlike some countries, the Czech Republic has sought to tackle the problem and acknowledges that, in its attempts to achieve the social and educational integration of the disadvantaged group which the Roma form, it has had to contend with numerous difficulties as a result of, inter alia, the cultural specificities of that minority and a degree of hostility on the part of the parents of non-Roma children. As the Chamber noted in its admissibility decision in the instant case, the choice between a single school for everyone, highly specialised structures and unified structures with specialised sections is not an easy one. It entails a difficult balancing exercise between the competing interests. As to the setting and planning of the curriculum, this mainly involves questions of expediency on which it is not for the Court to rule (Valsamis v. Greece, judgment of 18 December 1996, Reports 1996-VI, § 28). 206. Nevertheless, whenever discretion capable of interfering with the enjoyment of a Convention right is conferred on national authorities, the procedural safeguards available to the individual will be especially material in determining whether the respondent State has, when fixing the regulatory framework, remained within its margin of appreciation (see Buckley v. the United Kingdom, judgment of 25 September 1996, Reports 1996‑IV, § 76; and Connors v. the United Kingdom, judgment cited above, § 83). 207. The facts of the instant case indicate that the schooling arrangements for Roma children were not attended by safeguards (see paragraph 28 above) that would ensure that, in the exercise of its margin of appreciation in the education sphere, the State took into account their special needs as members of a disadvantaged class (see, mutatis mutandis, Buckley, cited above, § 76; and Connors, cited above, § 84). Furthermore, as a result of the arrangements the applicants were placed in schools for children with mental disabilities where a more basic curriculum was followed than in ordinary schools and where they were isolated from pupils from the wider population. As a result, they received an education which compounded their difficulties and compromised their subsequent personal development instead of tackling their real problems or helping them to integrate into the ordinary schools and develop the skills that would facilitate life among the majority population. Indeed, the Government have implicitly admitted that job opportunities are more limited for pupils from special schools. 208. In these circumstances and while recognising the efforts made by the Czech authorities to ensure that Roma children receive schooling, the Court is not satisfied that the difference in treatment between Roma children and non-Roma children was objectively and reasonably justified and that there existed a reasonable relationship of proportionality between the means used and the aim pursued. In that connection, it notes with interest that the new legislation has abolished special schools and provides for children with special educational needs, including socially disadvantaged children, to be educated in ordinary schools. 209. Lastly, since it has been established that the relevant legislation as applied in practice at the material time had a disproportionately prejudicial effect on the Roma community, the Court considers that the applicants as members of that community necessarily suffered the same discriminatory treatment. Accordingly, it does not need to examine their individual cases. 210. Consequently, there has been a violation in the instant case of Article 14 of the Convention, read in conjunction with Article 2 of Protocol No. 1, as regards each of the applicants. IV. APPLICATION OF ARTICLE 41 OF THE CONVENTION 211. Article 41 of the Convention provides: “If the Court finds that there has been a violation of the Convention or the Protocols thereto, and if the internal law of the High Contracting Party concerned allows only partial reparation to be made, the Court shall, if necessary, afford just satisfaction to the injured party.” A. Damage 212. The applicants did not allege any pecuniary damage. 213. They claimed 22,000 euros (EUR) each (making a total of EUR 396,000) for the non-pecuniary damage they had sustained, including educational, psychological and emotional harm and compensation for the anxiety, frustration and humiliation they had suffered as a result of their discriminatory placement in special schools. They stressed that the effects of this violation were serious and on-going and affected all areas of their lives. 214. Further, referring to the judgments in Broniowski v. Poland ([GC], no. 31443/96, § 189, ECHR 2004‑V) and Hutten-Czapska v. Poland ([GC], no. 35014/97, §§ 235-237, ECHR 2006‑...), the applicants said that the violation of their rights “was neither prompted by an isolated incident nor attributable to the particular turn of events in [their] case, but was rather the consequence of administrative and regulatory conduct on the part of the authorities towards an identifiable class of citizens”. Accordingly, in their submission, general measures had to be taken at the national level either to remove any hindrance to the implementation of the right of the numerous persons affected by the situation or to provide equivalent redress. 215. The Government submitted, with particular regard to the psychological and educational damage, that it related to the complaints under Article 3 of the Convention and Article 2 of Protocol No. 1 taken individually, which had been declared inadmissible by the Court in its decision of 1 March 2005. In their submission, there was therefore no causal link between any violation of the Convention and the alleged non-pecuniary damage. They further contended that the sum claimed by the applicants was excessive and that any finding of a violation would constitute sufficient just satisfaction. 216. The Court reiterates, firstly, that by virtue of Article 46 of the Convention the High Contracting Parties have undertaken to abide by the final judgments of the Court in any case to which they are parties, execution being supervised by the Committee of Ministers. It follows, inter alia, that a judgment in which the Court finds a breach imposes on the respondent State a legal obligation not just to pay those concerned the sums awarded by way of just satisfaction under Article 41, but also to select, subject to supervision by the Committee of Ministers, the general and/or, if appropriate, individual measures to be adopted in their domestic legal order to put an end to the violation found by the Court and to redress so far as possible the effects. However, the respondent State remains free to choose the means by which it will discharge its legal obligation under Article 46 of the Convention, provided that such means are compatible with the conclusions set out in the Court's judgment (Broniowski, cited above, § 192; and Čonka v. Belgium, no. 51564/99, § 89, ECHR 2002‑I). The Court notes in this connection that the legislation impugned in the instant case has been repealed and that the Committee of Ministers recently made recommendations to the member States on the education of Roma/Gypsy children in Europe (see paragraphs 54 and 55 above). Consequently, it does not consider it appropriate to reserve the question. 217. The Court cannot speculate on what the outcome of the situation complained of by the applicants would have been had they not been placed in special schools. It is clear, however, that they have sustained non-pecuniary damage – in particular as a result of the humiliation and frustration caused by the indirect discrimination of which they were victims – for which the finding of a violation of the Convention does not afford sufficient redress. However, the amounts claimed by the applicants are excessive. Ruling on an equitable basis, the Court assesses the non-pecuniary damage sustained by each of the applicants at EUR 4,000. B. Costs and expenses 218. The applicants have not amended the initial claim they made before the Chamber. The costs and expenses do not, therefore, include those incurred in the proceedings before the Grand Chamber. The Court notes that the total amount claimed in the request signed by all the applicants' representatives was EUR 10,737, comprising EUR 2,550 (GBP 1,750) for the fees invoiced by Lord Lester of Herne Hill, Q.C., and EUR 8,187 for the costs incurred by Mr D. Strupek in the domestic proceedings and those before the Chamber. However, the bill of costs drawn up by Lord Lester, enclosed with the claim for just satisfaction, put his fees at GBP 11,750 (approximately EUR 17,000), including GBP 1,750 in VAT, for 45 hours of legal work. The applicants' other representatives, Mr J. Goldston and the European Centre for Roma Rights, have not sought the reimbursement of their costs. 219. The Government noted that apart from a detailed list of the legal services he had provided, Mr Strupek had not submitted any invoice to prove that the alleged costs and expenses had in fact been paid to him by the applicants. They did not comment on the discrepancy between the claim for just satisfaction as formulated by the applicants and the fee note submitted by Lord Lester. The Government further pointed out that only part of the application had been declared admissible and continued to be the subject of examination by the Court. They therefore submitted that the applicants should not be awarded more than a reasonable portion (not exceeding EUR 3,000) of the costs and expenses claimed. 220. The Court reiterates that legal costs are only recoverable to the extent that they relate to the violation that has been found (Beyeler v. Italy (just satisfaction) [GC], no. 33202/96, § 27, 28 May 2002). In the present case, this is solely the violation of Article 14 of the Convention read in conjunction with Article 2 of Protocol No. 1. The Court notes that Lord Lester has submitted details of his professional fees, which were invoiced to the European Centre for Roma Rights. Mr Strupek has produced a breakdown of the 172 hours of legal services he rendered at an hourly rate of EUR 40, to which has to be added VAT at the rate of 19%. Having regard to all the relevant factors and to Rule 60 § 2 of the Rules of Court, the Court makes a joint award to all the applicants of EUR 10,000 for costs and expenses. C. Default interest 221. The Court considers it appropriate that the default interest should be based on the marginal lending rate of the European Central Bank, to which should be added three percentage points. FOR THESE REASONS, THE COURT 1. Dismisses unanimously the Government's preliminary objection; 2. Holds by thirteen votes to four that there has been a violation of Article 14 read in conjunction with Article 2 of Protocol No. 1; 3. Holds by thirteen votes to four (a) that the respondent State is to pay the applicants, within three months, the following amounts together with any tax that may be chargeable: (i) to each of the eighteen applicants EUR 4,000 (four thousand euros) in respect of non-pecuniary damage, to be converted into the currency of the respondent State at the rate applicable on the date of payment; (ii) jointly, to all the applicants, EUR 10,000 (ten thousand euros) in respect of costs and expenses, to be converted into the currency of the respondent State at the rate applicable on the date of payment; (b) that from the expiry of the above-mentioned three months until settlement simple interest shall be payable on the above amounts at a rate equal to the marginal lending rate of the European Central Bank during the default period plus three percentage points; 4. Dismisses unanimously the remainder of the applicants' claim for just satisfaction. Done in English and in French, and delivered at a public hearing in the Human Rights Building, Strasbourg, on 13 November 2007. Michael O'Boyle Nicolas Bratza Deputy Registrar President In accordance with Article 45 § 2 of the Convention and Rule 74 § 2 of the Rules of Court, the dissenting opinions of Mr Zupančič, Mr Jungwiert, Mr Borrego Borrego and Mr Šikuta are annexed to this judgment. N.B. M.O'B. DISSENTING OPINION OF JUDGE ZUPANČIČ I join entirely in the comprehensive dissenting opinion of Judge Karel Jungwiert. I wish only to add the following. As the majority explicitly, and implicitly elsewhere in the judgment, admitted in §§ 198 and 205 – the Czech Republic is the only Contracting State which has in fact tackled the special educational troubles of Roma children. It then borders on the absurd to find the Czech Republic in violation of anti-discrimination principles. In other words, this “violation” would never have happened had the respondent State approached the problem with benign neglect. No amount of politically charged argumentation can hide the obvious fact that the Court in this case has been brought into play for ulterior purposes, which have little to do with the special education of Roma children in the Czech Republic. The future will show what specific purpose this precedent will serve. DISSENTING OPINION OF JUDGE JUNGWIERT (Translation) 1. I strongly disagree with the majority's finding in the present case of a violation of Article 14 of the Convention, read in conjunction with Article 2 of Protocol No. 1. While I am able to agree to an extent with the formulation of the relevant principles under Article 14 in the judgment, I cannot accept the manner in which the majority have applied those principles in the instant case. 2. Before specifying all the matters with which I disagree, I would like to put this judgment into a more general perspective. It represents a new development in the Court's case-law, as it set about evaluating and criticising a country's entire education system. However authoritative the precedents cited at paragraphs 175 to 181 of the judgment may be, in practice they have very little in common with the instant case other perhaps than the Roma origin of the applicants in most of the cases (for instance in Nachova and Buckley, among others). 3. In my opinion, for the principles to be applied correctly requires, firstly, a sound knowledge of the facts and also of the circumstances of the case, primarily the historical context and the situation obtaining in other European countries. As regards the historical context, the data presented in the judgment (paragraphs 14 to 16) provides information that is inaccurate, inadequate and of a very general nature. The facts as presented in the judgment do not permit the slightest comparison to be made between Roma communities in Europe with respect, inter alia, to such matters as demographic evolution or levels of school attendance. 4. I will endeavour to supply some facts and figures to make up for this lack of information. I should perhaps begin with the awful truth that, so far as the current territory of the Czech Republic is concerned, we are not talking about an “attempted” extermination of the Roma by the Nazis (see paragraph 13) but about their almost total annihilation. Of the nearly 7,000 Roma who were living in the country at the start of the war, scarcely 600 survived[2]. The situation is thus very different from that in other countries: the Czech Roma, almost all of whom were exterminated, were replaced from 1945 onwards by successive waves of new arrivals in their tens of thousands, mainly from Slovakia, Hungary and Romania. The vast majority of this new population were not only illiterate and completely uprooted, they did not speak the Czech language. The same is not true of other countries on whose territory the Roma have – in principle – been living for decades and even centuries and have attained a degree of familiarity with the environment and language. To complete and close this incursion into the historical and demographic context, I believe that a further comparison, which helps to explain the scale and complexity of the problem, would be useful. An estimation of the numbers of Roma living in certain European countries has given the following minimum and maximum figures (which of course remain approximate): Germany 110,000 – 140,000 for a population of 80,000,000 France 300,000 – 400,000 for a population of 60,000,000 Italy 90,000 – 120,000 for a population of 60,000,000 United Kingdom 100,000 – 150,000 for a population of 60,000,000 Poland 35,000 – 45,000 for a population of 38,000,000 Portugal 40,000 – 50,000 for a population of 10,000,000 Belgium 25,000 – 35,000 for a population of 10,000,000 Czech Republic 200,000 – 250,000 for a population of 10,000,000[3]^, [4] These figures provide an indication of the scale of the problem facing the Czech Republic in the education field. 5. An important question that needs to be asked is what is the position in Europe and what standards or minimum requirements have to be met? The question of the schooling and education of Roma children has for almost 30 years been the subject of analysis and, on the initiative of the Council of Europe, proposals by the European Commission and other institutions. The judgment contains more than 25 pages (paragraphs 54 to 107) of citations from Council of Europe texts, Community law and practice, UN materials and other sources. However, the majority of the recommendations, reports and other documents it cites are relatively vague, largely theoretical and, most important of all, were published after the period with which the instant case is concerned (1996-1999 – see paragraph 19 of the judgment). I should therefore like to quote the author mentioned above, whose opinion I agree with. In his book Roma in Europe, J.-P. Liégeois stresses: “We must avoid over-use of vague terms ('emancipation', 'autonomy', 'integration', 'inclusion', etc.) which mask reality, put things in abstract terms and have no functional value ... ... officials often formulate complex questions and demand immediate answers, but such an approach leads only to empty promises or knee-jerk responses that assuage the electorate, or the liberal conscience, in the short term.”[5] In this connection, the sole resolution on the subject that is concrete and accurate – a major founding text of perhaps historic value – is the Resolution of the Council and the Ministers of Education meeting within the Council of 22 May 1989 on school provision for gypsy and traveller children[6]. 6. Regrettably and to my great surprise, this crucial document is not among the sources cited in the Grand Chamber's judgment. I should therefore like to quote some of the passages from this resolution: “THE COUNCIL AND THE MINISTERS FOR EDUCATION, MEETING WITHIN THE COUNCIL, ... Considering that the present situation is disturbing in general, and in particular with regard to schooling, that only 30 to 40 % of gypsy or traveller children attend school with any regularity, that half of them have never been to school [emphasis added], that a very small percentage attend secondary school and beyond, that the level of educational skills, especially reading and writing, bears little relationship to the presumed length of schooling, and that the illiteracy rate among adults is frequently over 50 % and in some places 80 % or more, Considering that over 500 000 children are involved and that this number must constantly be revised upwards on account of the high proportion of young people in gypsy and traveller communities, half of whom are under 16 years of age, Considering that schooling, in particular by providing the means of adapting to a changing environment and achieving personal and professional autonomy, is a key factor in the cultural, social and economic future of gypsy and traveller communities, that parents are aware of this fact and their desire for schooling for their children is increasing, ...” 7. How astonishing! In the twelve countries that formed the European Union in 1989 it is acknowledged that between 250,000 and 300,000 children had never attended school. It is an inescapable fact that the trend since then has tended to confirm this diagnosis. There is nothing to suggest an improvement in the situation in this sphere, especially with the enlargement of the European Union. The population of the Roma community is estimated (by the same source) at 400,000 in Slovakia, 600,000 in Hungary, 750,000 in Bulgaria and 2,100,000 in Romania. In total, there are more than 4,000,000 Roma children in Europe, more than 2,000,000 of whom will, in all probability, never attend school in their lifetimes. 8. I am determined to bring this terrible and largely concealed truth out into the open, as I consider it shameful that such a situation should exist in Europe in the 21st century. What has caused this alarming silence? 9. Statistical data on the former Czechoslovakia indicates that in 1960 some 30% of Roma had never attended school. This figure has fallen and was only 10% in 1970. A numerical comparison of the Czech Republic data on the number of children born and the number attending school shows school attendance levels attaining almost 100% twenty years later[7]. 10. Nevertheless, in this sorry state of affairs, some people consider it necessary to focus criticism on the Czech Republic, one of the few countries in Europe where virtually all children, including Roma children, attend school. Further, for the school year 1989-1990 there were 7,957 teachers for 58,889 pupils and for the school year 1992-1993 8,325 teachers for 48,394 pupils[8], that is to say one teacher for every seven pupils. 11. For years, European States have produced an often strange mix of achievements and projects which combine successes with failures. The problem concerns the education systems of many countries, not just the special schools[9]. The Czech Republic has chosen to develop a system that was introduced back in the 1920s (see paragraph 15 of the judgment), and to improve it while providing the following procedural safeguards for placements in special schools (paragraphs 20 and 21) : – parental consent, – recommendations of the educational psychology centres, – a right of appeal, – an opportunity to transfer back to an ordinary primary school from a special school. In a way, the Czech Republic has thereby established an education system that is inegalitarian. However, this inegalitarianism has a positive aim: to get children to attend school in order to have a chance to succeed through positive discrimination in favour of a disadvantaged population. Despite this, the majority feel compelled to say that it is not satisfied that the difference in treatment between Roma children and non-Roma children pursued a legitimate aim of adapting the education system to the needs of the former and that there existed a reasonable relationship of proportionality between the means used and the aim pursued (see paragraph 208 of the judgment). No one has conveyed the following opinion better than Arthur Schopenhauer, who was the first to express it: “This peculiar satisfaction in words contributes more than anything else to the perpetuation of errors. For, relying on the words and phrases received from his predecessors, each one confidently passes over obscurities and problems...”[10] 12. I fully accept that while much has been done to help certain categories of pupil acquire a basic knowledge, the situation regarding the education of Roma children in the Czech Republic is far from ideal and leaves room for improvement. Nevertheless, a closer examination of the situation leads me to ask but one question: which country in Europe has done more, or indeed as much, in this sphere? To require more, to require an immediate and infallible solution, is to my mind asking too much, perhaps even the impossible, at least as far as the relevant period, which began just a few years after the fall of the Communist regime, is concerned. 13. I consider it important both in the analyses and in all the assessments and conclusions for a distinction to be drawn between what is desirable and what one might term realistic, possible or simply feasible. This rule should also apply to the sphere of law generally and in the instant case in concreto. According to the applicants, no measures were taken to enable Roma children to overcome their cultural and linguistic disadvantages in the tests (see paragraph 40). However, this is but another excellent illustration of their lack of realism. It is, in my view, illusory to think that a situation that has obtained for decades, even centuries, can be changed from one day to the next by a few statutory provisions. Unless the idea is to dispense with the tests altogether or to make them an irrelevance. 14. Nor should it be forgotten that every school system entails not only education but also a process of assessment, differentiation, competition and selection. This fact of life is currently the subject of a wide debate on the reform of the French education system. The President of the French Republic has in a letter of 4 September 2007 to the teaching professions introduced the notion of a selection procedure for entry to lower and higher secondary education: “No one should go into the first form unless he has shown that he is able to follow lower secondary-school education. No one should enter the fifth form unless he has demonstrated his ability to follow an upper secondary-school education.” 15. I find the conclusions reached by the majority (see paragraphs 205 to 210 of the judgment) somewhat contradictory. They note that difficulties exist in the education of Roma children not just in the Czech Republic but in other European States as well. To describe the total absence of a school education for half of Roma children (see points 6 and 7 above) in a number of States as “difficulties” is an extraordinary euphemism. To explain this illogical approach, the majority note with satisfaction that, unlike some countries, the Czech Republic has chosen to tackle the problem (see paragraph 205 of the judgment). The implication is that it is probably preferable and less risky to do nothing and to leave things as they are elsewhere, in other words to make no effort to confront the problems with which a large section of the Roma community is faced. 16. In my view, such abstract, theoretical reasoning renders the majority's conclusions wholly unacceptable. DISSENTING OPINION OF JUDGE BORREGO BORREGO (Translation) 1. I am somewhat saddened by the judgment in the present case. 2. In 2002 Judge Bonello said that he found it “particularly disturbing that the Court, in over fifty years of pertinacious judicial scrutiny, has not, to date, found one single instance of violation of the right [guaranteed by] ... Article 2 or ... Article 3 induced by the race ... of the victim” (Anguelova v. Bulgaria, judgment of 13 June 2002, no. 38361/97, dissenting opinion). While I agree with Judge Bonello's criticism that the absence, five years ago, of a single case of racial discrimination concerning the core Convention rights was disturbing, the judgment in the present case has now got the Court off to a flying start. The Grand Chamber has in this judgment behaved like a Formula One car, hurtling at high speed into the new and difficult terrain of education and, in so doing, has inevitably strayed far from the line normally followed by the Court. 3. In my opinion, the Second Section's judgment of 17 February 2006 in the present case was sound and wise and a good example of the Court's case-law. Regrettably, I cannot say the same of the Grand Chamber judgment. (The Chamber judgment is 17 pages long, the Grand Chamber's, 78 pages, which all goes to show that the length of a judgment is no measure of its sagacity). I will focus on two points only. 4. The approach: After noting the concerns of various organisations about the realities of the Roma's situation, the Chamber stated: “The Court points out, however, that its role is different from that of the aforementioned bodies and that, like the Czech Constitutional Court, it is not its task to assess the overall social context. Its sole task in the instant case is to examine the individual applications...” (at paragraph 45). 5. Yet the Grand Chamber does the exact opposite. In contradiction with the role which all judicial bodies assume, the entire judgment is devoted to assessing the overall social context – from the first page (“historical background”) to the last paragraph, including a review of the “Council of Europe sources” (14 pages), “Community law and practice” (5 pages), United Nations materials (7 pages) and “other sources” (3 pages, which, curiously, with the exception of the reference to the European Monitoring Centre, are taken exclusively from the Anglo-American system, that is, the House of Lords and the United States Supreme Court). Thus, to cite but one example, the Court states at the start of paragraph 182: ”The Court notes that as a result of their turbulent history and constant uprooting the Roma have become a specific type of disadvantaged and vulnerable minority”. Is it the Court's role to be doing this? 6. Following this same line, which to my mind is not one appropriate for a court, the Grand Chamber stated in paragraph 209 after finding a discriminatory difference in treatment between Roma and non-Roma children: “... since it has been established that the relevant legislation ... had a disproportionately prejudicial effect on the Roma community, the Court considers that the applicants as members of that community necessarily suffered the same discriminatory treatment. Accordingly, it does not need to examine their individual cases”. 7. This, then, is the Court's new role: to become a second ECRI (European Commission against Racism and Intolerance) and dispense with an examination of the individual applications, for example the situation of applicants nos. 9, 10, 11, 16 and 17, in complete contrast to the procedure followed by the Chamber in paragraphs 49 and 50 of its judgment. 8. At the hearing on 17 January 2007 the representatives (from London and New York) of the applicant children (from Ostrava) confined themselves in their oral submissions to an account of the discrimination which they say the Roma are subjected to in Europe. 9. None of the applicant children or the parents of those applicants who were still minors were present at the hearing. The individual circumstances of the applicants and their parents were forgotten. Since Rule 36 § 4 of the Rules of Court states that representatives act on behalf of the applicants, I put a very simple question to the two British and American representatives – had they met the minor applicants and/or their parents? And had they been to Ostrava? I did not receive an answer. 10. I still have the same impression: the hearing room of the Grand Chamber had become an ivory tower, divorced from the life and problems of the minor applicants and their parents, a place where those in attendance could display their superiority over the absentees. 11. The Roma parents and the education of their children: On the subject of the children's education, the Chamber judgment states: “[T]he Court notes that it was the parents' responsibility, as part of their natural duty to ensure that their children receive an education...” (at paragraph 51). After an analysis of the facts the Chamber went on to hold that there had been no violation of Article 14, read in conjunction with Article 2 of Protocol No. 1. 12. I consider the stance taken by the Grand Chamber with respect to the parents of the minor applicants to be extremely preoccupying and, since it concerned all the Roma parents, one that is quite frankly, unacceptable. It represents a major deviation from the norm and reflects a sentiment of superiority that ought to be inconceivable in a court of human rights and strikes at the human dignity of the Roma parents. 13. The Grand Chamber begins by calling into question the capacity of Roma parents to perform their parental duty. The judgment states: “The Court is not satisfied that the parents of the Roma children, who were members of a disadvantaged community and often poorly educated, were capable of weighing up all the aspects of the situation and the consequences of giving their consent” (at paragraph 203). Such assertions are unduly harsh, superfluous and, above all, unwarranted. 14. The Grand Chamber then proceeds to compound its negative appraisal of the Roma parents: “ The Grand Chamber considers that, even assuming the conditions referred to in paragraph 201 above were satisfied, no waiver of the right not to be subjected to racial discrimination can be accepted, as it would be counter to an important public interest...” (paragraph 204). I find this particularly disquieting. The Grand Chamber asserts that all parents of Roma children, “even assuming” them to be capable of giving informed consent, are unable to choose their children's school. Such a view can lead to the awful experiences with which we are only too familiar of children being “abducted” from their parents when the latter belong to a particular social group because certain “well-intentioned” people feel constrained to impose their conception of life on all. An example of the sad human tradition of fighting racism through racism. 16. How cynical: the parents of the applicant minors are not qualified to bring up their children, even though they are qualified to sign an authority in favour of British and North American representatives whom they do not even know! 17. Clearly, I agree with the dissenting opinions expressed by my colleagues, whose views I wholly subscribe to. 18. Any departure by the European Court from its judicial role will lead it into a state of confusion and that can only have negative consequences for Europe. The deviation from the norm implicit in this judgment is substantial and the fact that all Roma parents are deemed unfit to educate their children is, in my view, insulting. I therefore take my place alongside the victims of that insult and declare: “Jsem český Rom” (I am a Czech Roma). DISSENTING OPINION OF JUDGE ŠIKUTA To my great regret, I cannot share the opinion of the majority, which has found that in the instant case there has been a violation of Article 14 of the Convention read in conjunction with Article 2 of Protocol No. 1. I wish to briefly explain my main reasons for not concurring. I do agree that, in general terms, the situation of Roma in Central and Eastern Europe is very complex, not easy and simple, and requires efforts from all the key players involved, in particular the Governments. This situation, however, has developed over hundreds of years and been influenced by various historical, political, economic, cultural and other factors. Governments have to play a proactive role in this process and are obliged therefore to adopt relevant measures and projects, with a view to reaching a satisfactory situation. The Roma issue should be seen from that perspective, as a living and continuously evolving issue. The Court's case-law[11] clearly establishes that a difference in treatment of “persons in otherwise similar situations” does not constitute discrimination contrary to Article 14 where it has an objective and reasonable justification; that is, where it can be shown that it pursues “a legitimate aim” or there is “a reasonable relationship of proportionality” between the means employed and the aim sought to be realised. The validity of the justification must be assessed by reference to the aim and effects of the measures under consideration, regard being had to the principles that apply in democratic societies. In assessing whether and to what extent differences in “otherwise similar situations” justify different treatment, the Court has allowed the Contracting States a certain margin of appreciation[12]. The fact that the Government chose to fulfil the task of providing all children with compulsory education through the establishment of special schools was fully within the scope of their margin of appreciation. The special schools were introduced for children with special learning difficulties and special learning needs as a way of fulfilling the Government's task of securing to all children a basic education, which was fully compulsory. The introduction of special schools should be seen as another step in the above-mentioned process, whose ultimate aim was to reach a satisfactory, or at least an improved, educational situation. The introduction of special schooling, though not a perfect solution, should be seen as positive action on the part of the State to help children with special educational needs to overcome their different level of preparedness to attend an ordinary school and to follow the ordinary curriculum. It can therefore be seen that, in general, there existed objective and reasonable justification for treating children placed in special schools differently from those placed in ordinary schools, on the basis of objective results in the psychological tests, administered by qualified professionals, who were able to select suitable methods. I do agree that the treatment of the children attending ordinary schools on the one hand and of those attending special schools on the other was different. But, at the same time, both types of school, ordinary and special, were accessible and also de facto attended, at the material time, by both categories of children – Roma and non-Roma. The only decisive criterion, therefore, for determining which child would be recommended to which type of school was the outcome of the psychological test, a test designed by experts, qualified professionals, whose professionalism none of the parties disputed. The difference in treatment of the children attending either type of school (ordinary or special) was simply determined by the different level of intellectual capacity of the children concerned and by their different level of preparedness and readiness to successfully follow all the requirements imposed by the existing school system represented by the ordinary schools. Therefore, isolated statistical evidence, especially when from a particular region of the country, does not by itself enable one to conclude that the placement of the applicants in special schools was the result of racial prejudice, because, by way of example, special schools were attended by both Roma and, at the same time, non-Roma children. Statistics are not by themselves sufficient to disclose a practice which could be classified as discriminatory (Hugh Jordan v. the United Kingdom, no. 24746/94, § 154). The fact that ordinary schools were attended by Roma children as well proves only that there existed other selection criteria than race or ethnic origin. Also, the fact that some of the applicants were transferred to ordinary schools proves that the situation was not irreversible. It should also be noted that the parents of the children placed in the special schools agreed to their placement and some of them actually asked the competent authorities to place their children there. Such positive action on the part of the applicants' parents only serves to show that they were sufficiently and adequately informed about the existence of such schools and about their role in the schooling system. I have no doubt that, in general, a professional will be more competent to take a decision on the education of a minor child than its parents. Be that as it may, had there been any doubt that a decision of the parents to place their children in a special school was not “in the best interest of the child”, the Child Care Department of the Ostrava Welfare Office, which had the power and duty to bring such cases to the Juvenile Court to assess the best interest of the child, could have intervened. But that was not the case, as neither the Welfare Office, nor the applicants' parents, turned to the Juvenile Court, which was competent to deal with this issue. Having said all this, I have come to the conclusion that the difference in treatment was between children attending ordinary schools on the one hand and children attending special schools on the other, regardless of whether they were of Roma or non-Roma origin. Such difference in treatment had an objective and reasonable justification and pursued a legitimate aim – providing all children with compulsory education. However, I have also come to the conclusion that there was no difference in treatment between children attending the same special school, which children (Roma and non-Roma) are to be considered as “persons in otherwise similar situations”. I found no legal or factual ground in the instant case for the conclusion that Roma children attending special school were treated less favourably than non-Roma children attending the same special school. It is not acceptable to conclude that only Roma children attending special schools were discriminated against in comparison to non-Roma children (or all children) attending ordinary schools, since these two groups of children are not “persons in [an] otherwise similar situation”. It is also not acceptable to conclude this because both “groups” had the same conditions of access and attended both types of school: non-Roma children were attending special schools and, at the same time, Roma children were attending ordinary schools solely on the basis of the results achieved by passing the psychological test, which test was the same for all children regardless of their race. Based on the above, I do not share the opinion that the applicants, because of their membership of the Roma community, were subjected to discriminatory treatment by their placement in special schools. A N N E X LIST OF THE APPLICANTS 1. Ms D.H. is a Czech national of Roma origin who was born in 1989 and lives in Ostrava-Přívoz; 2. Ms S.H. is a Czech national of Roma origin who was born in 1991 and lives in Ostrava-Přívoz; 3. Mr L.B. is a Czech national of Roma origin who was born in 1985 and lives in Ostrava-Fifejdy; 4. Mr M.P. is a Czech national of Roma origin who was born in 1991 and lives in Ostrava-Přívoz; 5. Mr J.M. is a Czech national of Roma origin who was born in 1988 and lives in Ostrava-Radvanice; 6. Ms N.P. is a Czech national of Roma origin who was born in 1989 and lives in Ostrava; 7. Ms D.B. is a Czech national of Roma origin who was born in 1988 and lives in Ostrava-Heřmanice; 8. Ms A.B. is a Czech national of Roma origin who was born in 1989 and lives in Ostrava-Heřmanice; 9. Mr R.S. is a Czech national of Roma origin who was born in 1985 and lives in Ostrava-Kunčičky; 10. Ms K.R. is a Czech national of Roma origin who was born in 1989 and lives in Ostrava-Mariánské Hory; 11. Ms Z.V. is a Czech national of Roma origin who was born in 1990 and lives in Ostrava-Hrušov; 12. Ms H.K. is a Czech national of Roma origin who was born in 1990 and lives in Ostrava-Vítkovice; 13. Mr P.D. is a Czech national of Roma origin who was born in 1991 and lives in Ostrava; 14. Ms M.P. is a Czech national of Roma origin who was born in 1990 and lives in Ostrava-Hrušov; 15. Ms D.M. is a Czech national of Roma origin who was born in 1991 and lives in Ostrava-Hrušov; 16. Ms M.B. is a Czech national of Roma origin who was born in 1991 and lives in Ostrava 1; 17. Ms K.D. is a Czech national of Roma origin who was born in 1991 and lives in Ostrava-Hrušov; 18. Ms V.Š. is a Czech national of Roma origin who was born in 1990 and lives in Ostrava-Vítkovice. ------------------------------- 1. P. Evans (2006), ‘Educating students with special needs: A comparison of inclusion practices in OECD countries’, Education Canada 44 (1): 32-35. [2]. A. Frazer (M. Mikulášková), The Gypsies (Cikáni), Prague 2002, p. 275. 1. J.-P. Liégeois, Roma in Europe, to be published by Council of Europe Publishing. 2. Nevertheless, in a census taken of the population of the Czech Republic on 3 March 1991, only 32,903 people claimed to be members of the Roma (Statistical Yearbook of the Czech Republic 1993 ,Prague 1993, p. 142). 1. Op. cit. (text subject to editorial revision). 2. Official Journal of the European Communities C 153 of 21/06/1989, pp. 3 and 4. 1. Statistical Yearbook of the Czech Republic 1993, Prague 1993, pp. 88 and 302 2. Statistical Yearbook of the Czech Republic 1993, Prague 1993, p. 307. 3. In the public debate currently underway in France, it has been noted that “40% of pupils entering the first form do not have a basic education. At the end of the fourth form, 150,000 young people leave the system without mastering any subject (Editorial in the ‘Figaro’, 4 September 2007). The same newspaper related in an article on 7 September 2007 that “according to the Education Board, 40% of primary-school pupils – 300,000 children in all – leave each year with severe failings or in great dificulty”. 1. A. Schopenhauer, The World as Will and Representation (Volume II), this tranlsation by EFJ Payne, Dover, New York 1966, p. 145. 1. E.g. Willis v. the United Kingdom, no. 36042/97, § 48, ECHR 2002-IV 2. Gaygusuz v. Austria, judgment of 16 September 1996, Reports of Judgments and Decisions 1996-IV, par. 42