SECOND SECTION

CASE OF HORVÁTH AND KISS v. HUNGARY

( Application no. 11146/11 )

JUDGMENT

STRASBOURG

29 January 2013

FINAL

29/04/2013

This judgment has become final under Article 44 § 2 of the Convention. It may be subject to editorial revision.

In the case of Horváth and Kiss v. Hungary ,

The European Court of Human Rights ( Second Section ), sitting as a Chamber composed of:

Guido Raimondi , President, Danutė Jočienė , Peer Lorenzen , András Sajó , Işıl Karakaş , Nebojša Vučinić , Helen Keller , judges, and Stanley Naismith , Section Registrar ,

Having deliberated in private on 18 December 2012 ,

Delivers the following judgment, which was adopted on that date:

PROCEDURE

1 . The case originated in an application (no. 11146/11) against the Republic of Hungary lodged with the Court under Article 34 of the Convention for the Protection of Human Rights and Fundamental Freedoms (“the Convention”) by two Hungarian nat ionals, Mr István Horváth and Mr András Kiss (“the applicants”), on 11 February 2011 .

2 . The applicants were represented by Mrs L. Farkas, a lawyer practising in Budapest , and the European Roma Rights Centre , a non-governmental organisation with its seat in Budapest . The Hungarian Government (“the Government”) were represented by Mr Z. Tallódi , Agent, Ministry of Public Administration and Justice .

3 . The applicants alleged under Article 2 of Protocol No. 1 read in conjunction with Article 14 of the Convention that their education in a remedial school had amounted to direct and/or indirect discrimination in the enjoyment of their right to educati on, on the basis of their Roma origin, in that their schooling assessments had been paper-based and culturally biased, their parents could not exercise their participatory rights, they had been placed in schools designed for the mentally disabled whose curriculum had been limited, and they had been stigmatised in consequence.

4 . On 4 January 2012 the application was communicated to the Gove rnment. It was also decided to rule on the admissibility and merits of the application at the same time (Article 29 § 1).

THE FACTS

I. THE CIRCUMSTANCES OF THE CASE

5 . The applicants were born in 1994 and 1992 respectively and live in Nyíregyháza .

A . General background

6 . The appli ca nts are two young Roma men, who were diagnosed as having mental disabilities. As a result of these diagnoses, the applicants were educated at the Göllesz Viktor Remedial Primary and Vocational School, a remedial school (“special educational program me ” or “special” school) in the city of Nyíregyháza , created for children with mental disabilities.

7 . The proportion of Roma students at the Göllesz Viktor Remedial Primary and Vocational School was 40 to 50% in the last ten years. Statistical data indicate that in 2007 Roma represented 8.7% of the total number of pupils attending primary school in Nyíregyháza. In 1993, the last year when ethnic data were officially collected in public education in Hungary, at least 42% of the children in special ed ucational program me were of Roma origin according to official estimates , though they represented only 8.22% of the total student body.

8 . According to statistical data in the Statistical Yearbook of Edu cation, in 2007/2008 only 0.4– 0.6% of students with special needs had the opportunity to participate in integrated mainstream secondary education providing the Baccalaureate. Although one of the second applicant ’ s classmates was admitted to a secondary vocational school offering the Baccalaureate, ne ither of the applicant s was enrolled in a Baccalaureate program me , which limit ed their access to higher education and employment. The first applicant was unable to follow a course to become a dance teacher, the career of his father ; instead , he received special vocational training to become a baker. The second applicant continued his studies in a mainstream secondary vocational school which did not offer the Baccalaureate, and was unable to pursue his ambition to become a car mechanic.

B . Societal c ontext

9 . Scholarly literature suggests that the systemic misdiagnosis of Roma children as mentally disabled has been a tool to segregate Rom a children from non-Roma children in the Hungarian public school system since at least the 1970s.

10 . The national Gypsy research in 1971 made it clear that a major obstacle to the education of Gypsy children was the existence of remedial (special) schools. In 1974/1975 , 11.7% of Gypsy children attended special schools and classes. Due to the steady increase in Gypsy enrolment , by 1985/1986 their proportion had reached 17.5%, whereas only 2% of majority Hungarian students studied in special schools and classes. Eight grades finished in special education amounted to six grades in a normal school. Between 1972 and 1975 , almost 50% of the lower grade special school students in Budapest were re - tested. The most significant result of the Budapest review was that if the borderline between sound and disabled mental abilities were set at IQ 70, the figure recommended by the World Health Organisation ( WHO ) , then only 49.3% of students participating in special education qualified as mentally disabled, whereas 50.7% qualified as normal, of wh om 12% had average intellect and 38.7% w ere borderline case s , that is, on the brink of mental retardation. However, only 7% were qualified as having average mental abilities through a complex evaluation. The complex evaluation qualified children whose test results suggested otherwise as intellectually disabled . In order to come to this conclusion, the category of familial intellectual disability was introduced , a notion distinct from pathological mental disability.

11 . According to the Hungarian authorities, in 2004, 5.3% of primary school children were mentally disabled in Hungary , whereas this ratio stood at 2.5% in the European Union. In the last decade the rate of mentally disabled children has been continuously increasing in Hungary , especially in the ‘ mild mental disability ’ and ‘ other disability ’ categories. Children with disadvantag ed background, especially Roma ones , are significantly over-represented amongst children with a disability.

12 . The shortcomings of the diagnost ic system were acknowledged by S tate authorities when in 2003 the Ministry of Education launched a programme entitled “Out of the Back Bench” with the stated aim of reviewing children and, after re-diagnosis, channelling those back to mainstream school who had been misdiagnosed. Through the program me , 2,100 children were reassessed and 11% of the re-diagnosed children were channelled back to normal school. In Szabolcs-Szatm á r-Bereg County , where the applicants are from, this rate was 16%.

13 . Part of the reason for the fact that so many children wer e considered disabled was that the definition of special educational needs in Act no. LXXIX of 1993 on Public Education (“the PEA”) and the definition of mental disability prior to 1 September 2003 (see paragraph 63 below) went beyond mental disability and included educational challenge, dyslexia and behavioural problems.

14 . In 2007, the National Expert and Rehabilitation Committee (NERC) explained that an IQ between 70 and 85 represented a borderline intellect. A child in this range of IQ could have serious and persistent learning impairment. The expert evaluating each case had to assess what factors tilt ed the balance towards mental disability or sound mental ability. For example, weak abilities of abstraction or associative learning could indicate menta l disability even above IQ 70. “Borderline intellect” was not on its own considered as mental retardation or a cause for placement in special school.

15 . In 2004 the Mini ster of Education requested the expert p anels to stop transferring children with scores above IQ 70 to special schools. That year, a new protocol and new standardised proceedings were adopted, calling for the disadvantaged situation of the child to be taken into account. If a child spoke the language of an ethnic minority, for instance, he or she could not be examined using verbal tests in Hungarian . Still, inequalities persisted. The gr eatest difference between Roma and non-Roma children occurred in a performance test, the so-called “Mosaic Te st ” . One explanation for this is that Roma children have less experience with toys and games where units from bits or pictures from pieces (e.g. toy cubes with different pictures on each side, or puzzles, etc.) had to be assembled.

C . Mr Horváth ’ s assessments

16 . Mr Horváth started elementary education in the Göllesz Viktor Remedial Primary and Vocational School on the basis of the recommendation of the Expert and Rehabilitation Panel of Szabolcs-Szatmár-Bereg County (“the Expert Panel”). His examination was requested on 19 April 2001 by the nursery he was attending at that time. The nursery claimed that his mental and social abilities were lower than normal for his age, which showed in his sense of logic, drafting skills and communication. He spent very little time in the nursery, as he was sick most of the time. This, although a common cause for bad performance in tests, was not taken into account when his results were assessed.

17 . The examination requested by the nursery was performed on 17 May 2001. In addition to the observation of his behaviour, his abilities (verbal, counting, cognitive, attention/concentration, visuo-motor coordination) and his performance, the following IQ tests were done : “Budapest Binet Test” – IQ 64; “Coloured Rav en Test” – IQ 83; “Goodenough ‘ draw-a-p erson ’ Test ” – D Q 67. The Expert Panel did not elaborate in its opinion on the causes of the disparate results.

18 . In its opinion, the Expert Panel diagnosed Mr Horváth with “mild mental disability”, of which the origin was declared unknown. The diagnosis stated that Mr Horváth was “two and a half years behind normal”, together with an immature central nervous system. Therefore , he was channelled to remedial school. As opposed to the WHO value of IQ 70, expert panels in Hungary applied , according to the Ministry of National Resources, IQ 86 as a border value between sound intellectual ability and mild mental disability.

19 . Mr Horváth ’ s parents had been told by the Expert Panel even before the examination took place that he was going to be placed in a remedial school and they had been asked to sign the expert opinion before the examination took place.

20 . On 3 December 2002 the E xpert Panel re-examined Mr Horvá th. It found that there was no development in his abilities, and reported that he was still suffering from mild mental disability.

21 . On 28 April 2005 the Expert Panel again examined Mr Horváth. According to this examination, his “ Raven Test ” result was IQ 61. Therefore the Expert Panel declared that his status had not changed and upheld its previous opinion.

22 . On 20 March 2007 another examination took place. This time, Mr Horváth ’ s “ Raven Test ” value was IQ 71. The Expert Panel noted that he had better knowledge than this test score reflected, had good results at school in 2006 and 2007, was integrated in his school system and able to study individually, had no impediment in speech and only needed some reassurance. In addition, it noted that he was active in classes, hard-working and complied with all the requirements of the curriculum. Noting that Mr Horváth studied in a remedial school, the Expert Panel again diagnosed him with mild mental disability and special educational needs. Therefore it upheld his placement in remedial school.

23 . Mr Horváth ’ s parents were not invited to participate in the diagnostic assessments. His father signed only the opinion of 17 May 2001. It is unclear if the parents were provided with information about the procedure and their respective rights, including a right to appeal, or if a copy of the opinion was given to them. His father accompanied Mr Horváth to the first examination but was not allowed to attend the examination itself. The parents were told the result but no explanation about the consequences was given .

24 . On 26 September and 2 October 2008 Mr Horváth was re-examined by the NERC as ordered by the first instance court (see paragraph 38 below). This opinion stated that the applicant had “mild mental disability” although the causes of the disability could not be established.

D . Mr Kiss ’ s assessments

25 . After spending seven months in nursery, Mr Kiss started elementary education in September 1999 in a mainstream school, Primary School No. 13 located in a Roma settlement of Nyíregyháza. In its decision of 4 January 1999, the local pedagogical advisory service concluded that he had learning difficulties “deriving from his disadvantaged social and cultural background” and advised him to be educated under a special programme but in a mainstream school. On 14 December 1999 the school requested an expert diagnosis based on his results in the first quarter of the school year, claiming that he had poor results, was often tired, his attention was volatile and his vocabulary poor. His IQ then measured 73.

26 . On 15 May 2000 the Expert Panel diagnosed Mr Kiss with “mild mental disability”. According to the “ Budapest Binet Test ” , his IQ was 63, and he scored IQ 83 in the “ Raven Test ” . Relying on the results, the Expert Panel arranged for Mr Kiss to be placed at a school for children with mild mental disabilities. As rehabilitation, the Expert Panel proposed that his concentration and analytical-synthetical ability should be developed. The Panel ’ s opinion did not contain any explanation for the discrepancies between Mr Kiss ’ s IQ results in the various tests.

27 . Mr Kiss ’ s parents objected to the placement of their child in the remedial school and insisted that he should be educated in a mainstream school, but in vain. They were not informed of their right to appeal against the Panel ’ s decision. Mr Kiss was then placed in Göllesz Viktor Remedial Primary and Vocational School.

28 . During his studies, Mr Kiss won numerous competitions, including a poetry reading contest and sports competitions, and he was an A student until 7th grade. However, his teacher told him that he could not continue his studies to become a car mechanic as he intended to, because as a remedial school pupil, he could only choose between training courses offered by a special vocational school.

29 . The Expert Panel subsequently re-assessed Mr Kiss twice, on 14 December 2002 and 27 April 2005. On the latter occasion the Expert Panel noted that, despite the fact that he had achieved good results at school, his analytical thinking was underdeveloped. His IQ based on the “ Raven Test ” scored 71, yet the Expert Panel stated that he needed to be educated further at the remedial school.

30 . During the court procedure in the case (see below) , the first-instance court ordered that Mr Kiss be examined by the NERC . According to the expert opinion of 20 November 2008, his mental capacity was normal, he was not mentally disabled and his SQ (social quotient) score was 90, which excluded mental disability. However, he had significant deficiencies with regard to acquired knowledge and had a learning impairment. As with the first applicant, the NERC found that the Expert Panel ’ s decision should have noted that socio-cultural factors had played a significant role in the shaping of their status from an early age, but in fact these factors and M r Kiss ’ s disadvantaged situation were not taken into account.

The NERC concluded that both applicants were provided with education adequate to their abilities.

E. Review of the applicants ’ intellectual ability by independent experts

31 . In August 2005 both applicants participated in a summer camp where the testing of 61 children with ‘ special educational needs ’ took place. The testing was carried out by independent experts.

32 . Both applicants were assessed with various tests. With regard to Mr Horváth, the experts noted that his “ Raven Test ” (IQ 83) was under the average, but did not correspond to the “mentally disabled” score; therefore , he was not mentally disabled. His “Bender B Test” referred to immature nervous system potentially causing behavioural problems and problems in studying but he was not considered mentally disabled or unfit for an integrated mainstream class.

33 . Mr Kiss ’ s “ Raven Test ” score was IQ 90, his “MAVGYI-R Test” score was IQ 79, and his verbal intelligence was 91. According to the assessment, he suffered from immaturity of the nervous system and dyslexia. The experts noted that he was sound of mind and could be educated in a school with a normal curriculum. They suggested immediate intervention by the authorities in order to place him into a mainstream school and to provide him with appropriate education. The experts also suggested a thorough pedagogical examination and the development of a subsequent individual learning plan with pedagogical and psychological help. They noted that he had to catch up with his studies in order to reduce the deficiencies he had as a result of studying under a lower curriculum.

34 . The experts noted that the diagnostic methods applied sho uld be reviewed, and that Roma children could have performed better in the tests if those had not been designed for children belonging to the ethnic majority. They stressed that the “ Raven Test ” measured intelligence only in a narrow margin and therefore provided less data with regard to intelligence. The experts further recommended that the “ MAVGYI-R ” child intelligence test should be reviewed and updated as it was outmoded and because oral tests were culturally biased and poorly compatible with the present lifestyle and knowledge of children. The experts also noted that the intelligence tests had a close correlation with school qualification; therefore education in a remedial class might significantly influence the results of an intelligence test of a 13/14-year-old child.

The NERC found the independent experts ’ conclusions open to doubt.

F . First-instance court proceedings

35 . On 13 November 2006 the applicants filed a claim for damages with the Szabolcs-Szatmár-Bereg County Regional Court , requesting the court to establish a violation of the principle of equal treatment amounting to a violation of their personal ity rights under section 76 of the Civil Code and section 77(3) of the PEA . The action was directed against the Expert Panel, the Szabolcs-Szatmár-Bereg County Cou ncil and the Göllesz Viktor Remedial Primary and Vocational School.

36 . The applicants claimed that the Expert Panel had discriminated against them and misdiagnosed them as being “mildly mentally disabled” on the basis of their ethnicity, social and economic background, and had subsequently ordered them to be educated in a special school, although they had normal abilities. They asserted that the expert panels were free to choose the tests applied by them , and it was well-known among experts that some tests were culturally biased and le d to misdiagnosis of disadvant aged children, especially Roma ones . This systemic error originated in the flawed diagnostic system itself, which did not take into account the social or cultural background of Roma children, was as such culturally biased, and therefore l ed to the misdiagnosis of Roma children. They claimed that it was the responsibility of the experts who were required by the law to be experienced in the field of mental disabilities and thus obliged to know the symptoms of such disabilities to ensure that only children with real mental disability were educated in special/disabled/special educational needs classes. In addition, and in violation of the respective rules of procedure, the applicants ’ parents had not been informed of the Panel ’ s procedure or its consequences or of their rights to participate in the proceedings and to appeal against the decision s in question , so their constitutional right to a remedy was violated.

37 . The applicants further asserted that the County Council had failed effectively to control the Expert Panel. They also claimed that the teachers working at the Remedial School should have not ic ed that they were of normal abilities.

38 . The Regional Court ordered the applicants to be examined by the NERC .

39 . On 27 May 2009 the Regional Court found that the aggregate of the respondents ’ handlings of the applicants ’ education had amounted to a violat ion of the ir rights to equal treatment and education and therefore ordered them , jointly and severally, to pay 1,000,000 Hungarian forints (HUF) in damages to each applicant.

The court explained that it was called on to investigate whether the respondents had complied with the Constitution and the PEA , that is, ensured the applicants ’ civil rights without any discrimination, promoted the realisation of equality before the law with positive measures aiming to eliminate their inequalities of opportunity, and provided them with education in accordance with their abilities. It reasoned that – while the statutory definition of “special needs” had been amended several times in the relevant period – the relevant regulations clearly stipulated that the expert panels should individualise each case, decide on possible special needs in each case according to the needs and circumstances of the individual child, identify the reasons underlying any special needs, and establish specific support services which a child needed according to the extent of disability.

40 . The court held that this kind of individualisation was lacking with regard to the applicants ’ diagnoses and that the Expert Panel had failed to identify those specific professional services that would help the applicants in their education. It had failed to establish during the applicants ’ examination and re-examination the reasons for which they were in need of special education, and whether they needed that as a result of their behaviour or of organic or non-organic reasons.

41 . The court emphasised that the principle of equal treatment required that the Expert Panel decide whether children reaching school age might study in school s with a standard curriculum or in remedial schools with a special one. At the same time, the court noted that , in the present case, the operation of the Expert Panel was stalled due to ongoing restructuring and the low number of professional and other staff. Therefore , the Expert Panel could not perform its duty of continuous control examinations.

42 . Moreover, in the court ’ s view, t he County Council had failed to ensure effective control over the Expert Panel and therefore failed to note that the P anel had not informed the parents appropriately. In addition, the County Council had not ensured that the expert decisions were individualised according to the law.

Therefore , the respondents had violated the applicants ’ right to equal treatment.

G . Appeal procedure

43 . The Expert Panel did not appeal and so the above decision became final and enforceable with regard to it .

On appeal by the Remedial School and the County Council, on 5 November 2009 the Debrecen Court of Appeal reversed the first - instance judgment and dismissed the applicants ’ claims against those respondents .

44 . The Court of Appeal accepted the Remedial School ’ s defence, namely that it had done no more than enrol the applicants according to the Expert Panel ’ s decision. It held that it was for the County Council to ensure effective control over the lawful operation of the Remedial School and the Expert Panel. An omission in this regard might establish the County Council ’ s liability, in particular because the parents ’ participatory rights had not been respected.

45 . The Court of Appeal further noted that, in order to prevent the misdiagnosis and con sequent segregation of Roma children into remedial schools, there was a need, unfulfilled as yet, for the development of a new diagnostic testing system which should take into account the cultural, linguistic and social background of children. However, it held that the lack of appropriate diagnostic tools and the subsequent placement of the applicants into remedial schools did not have any connection to their ethnic origin, and therefore found no discrimination against the applicants, concluding that their personal ity rights had not been violated. In its view, the applicants had not suffered any damage as a result of the unlawful conduct of the respondents, since, according to the court-appointed experts ’ opinion, they had been educated in accordance with their mental abilities. That opinion effectively confirmed the Expert Panel ’ s decisions.

The Court of Appeal ’ s judgment further contains the following passage:

“Examining the – not at all comprehensive – amendments [of the PEA and the decrees on its implementation which occurred after 1 January 2007], it can be established on the one hand that those amendments were predominantly and evidently occasioned by the progress of related science, the researches and the results of surveys, and on the other hand that the following of legislative developments in this period was almost an impossible task for those applying the law.”

H . Review proceedings

46 . The applicants subsequently submitted a petition for review to the Supreme Court. They argued that there was no national professional standard established with regard to the diagnostic system in Hungary . The well-known systemic errors of the diagnostic system, together with the disregard of the socially, culturally and linguistically disadvantaged background, had resulted in a dispropor tionately high number of Roma children diagnosed as having “mild mental disability”.

47 . The applicants requested the Supreme Court to establish, as an analogy with the case of D.H. and Others v. the Czech Republic ([GC], no. 57325/00, ECHR 2007 ‑ IV) , the misdiagnosis of Roma children, that is , that the channelling of Roma children with normal mental abilities into remedial schools constituted discrimination. Such misdiagnosis represented direct – or alternatively indirect – discrimination, based on the ethnic, social and economic background of the applicants.

48 . The applicants further claimed that the Court of Appeal had wrongly concluded that there was no connection between the lack of appropriate diagnostic tools and the ethnic origin of the applicants. The fact that the tests themselves had no indication of ethnicity did not preclude that they forced a dispropo rtionately high number of Roma children into a disadvantaged position in comparison with majority children. This practice amounted to a violation of section 9 (indirect discrimination) of Act no. CXXV of 2003 on Equal Treatment and the Promotion of Equal Opportunities (“the ETA”) . In addition, the fact that the experts had disregarded the specific social, cultural and language components when assessing the test results had led to direct discrimination in breach of section 8 of the ETA .

49 . The applicants also asserted that the respondents had not acted with due diligence in the circumstances, when – aware of the systemic error of the diagnostic system – they had failed to act according to international standards. In addition, Mr Kiss had been placed in a remedial school despite the explicit objection of the parents.

50 . The Supreme Court reviewed the second-instance judgment and found it partly unfounded. It stated as follows:

“Considering the relevant provisions of the [ETA] and the [PEA] ... the Supreme Court has to decide whether the respondents discriminated against the plaintiffs on the basis of their ethnic, social, economic and cultural background, which resulted in the deprivation of their rights to be educated in accordance with their abilities and therefore their rights to equal treatment, and subsequently whether their personal ity rights have been violated.”

51 . The Supreme Court upheld the second-instance judgment with regard to the finding that the conduct of the Remedial School and the County Council had not violated the applicants ’ right to equal treatment, either in terms of dir ect or indirect discrimination.

52 . The Supreme Court further noted:

“The systemic errors of the diagnostic system leading to misdiagnosis – regardless of its impact on the applicants – could not establish the respondents ’ liability ... The creation of an appropriate professional protocol which considers the special d isadvantaged situation of Roma children and alleviates the systemic errors of the diagnostic system is the duty of the State.”

53 . The Supreme Court noted, however, that:

“[T]he failure of the State to create such a professional protocol and [an eventual] violation of the applicants ’ human rights as a result of these systemic errors exceed the competence of the Supreme Court ... the applicants may seek to have a violation of their human rights established before the European Court of Human Rights. Therefore the Supreme Court has not decided on the merit of this issue.”

54 . The Supreme Court further examined whether the respondents ’ liability could be established under the general rules of tort liability regardless of the fact that it had not established a violation of the applicants ’ personal ity rights. It found no such liability in respect of the Remedial School . However, it observed that the Expert Panel ’ s handling of the parental rights had violated the relevant law (Ministerial Decree no. 14/1994 . (VI.24.) MKM ). The County Council was found liable for this on account of its failure to supervise the legality, or to organise the supervision of the legality , of the functioning of the Expert Panel, as well as to put an end to the unlawful practice. The prejudice to the applicants was caused by their deprivation of the right to a remedy provided for by law and thereby of the theoretical chance of obtaining a more favourable assessment of their learning abilities.

The Supreme Court consequently upheld the first-instance judgment with regard to the payment of HUF 1,000,000 in damages to each applicant by the Expert Panel, out of which sum the County Council was obliged to pay HUF 300,000, on account of its deficient control.

This decision was served on 11 August 2010.

II. RELEVANT DOMESTIC LAW

A. Elements of domestic law submitted by the Government

55 . T he work of the expert and rehabilitation committees examining learning abilities was, at the material time, regulated by Ministerial Decree n o. 14/1994 . (VI. 24.) MKM. This Decree dealt with procedural issues, regulated the operation of expert committe es , secured the complexity of the expert and rehabilitation committee examinations , and required that the committees ’ recommendations be based on a complex assessment of the results of medical, pedagogical and psychological examinations. As to the methods of examination to be used, a protocol was outlined in a manual entitled “ Transfer Examinations” ( “ the Manual ” ), the publication of which was commissioned by the Ministry of Education in the 1980s.

56 . The Manual states with emphasis that performance disorder s may have two causes: the lack of knowledge or the lack of ability. It specifies the diagnostic signs indicating that the lack of knowledge is not caused by ability disorder as follows: where the lack of knowledge is explained by previous poor developmental conditions and poor socio-cultural environment; where the task can be simplified so as to suit the child ’ s level of knowledge and at that level no performance disorder can be observed; where during the examination the manner of making use by the child of the help provided by the examining teacher and the child ’ s capability to be oriented and taught indicate that his abilities are developable; and where the child ’ s social maturity, general knowledge and performance in life situations indicate that his abilities are intact.

57 . Consequently, in examining a child ’ s task- solving performance, the interdependence of four factors shall always be examined, namely previous educational effects, the child ’ s scope of knowledge, the child ’ s abilities and his age-related maturity.

58 . The Manual further contains the following guidelines:

“W here a child from a socio-culturally retarded environment is being examined, tests free of cultural elements should be used. Certain tasks of a given test may be transformed in order to adjust them – at the same level of difficulty – to the child ’ s scope of knowledge ...

W hen a socially disadvantaged child is being examined, special attention must be paid to his capability to learn i n the examination situation ... ”

59 . The Manual also draws experts ’ attention to the desirable procedures to be followe d in examining a child of Roma ethnicity as follows:

“The fact that a child does not know the language of school instruction or that his command of language does not attain the level of mother tongue would, in itself, constitute a serious disadvantage even if the child had no school integration problems resulting from social and/or cultural problems. Therefore, the special education or psychological examination of children coming from a disadvantageous social situation and underdeveloped linguistic environment should be carried out with special care. From a delay in speech development no conclusions concerning the child ’ s mental maturity should be drawn. In such cases the child ’ s practical intelligence should be assessed, or his cognitive abilities should be examined through non-verbal tasks.”

60 . This protocol was reviewed and updated between 2004 and 2008 and a new M anual was published. In 2010 a new Min i sterial D ecree ( no. 4/2010. (I.19 . ) OKM ) was issued for the regulation of the work of the pedagogical expert services. Th is Decree prescribes a uniform procedural order for expert and rehabilitation committees , and specifies the professional requirements to be met in carrying out the examinations, based on wh ich expert opinions are drafted ; moreover, i n addition to the remedies formerly introduced, it provides for the involvement of an independent equal opportunity expert , if appropriate .

B. Elements of domestic law submitted by the applicants

61 . Before the ETA entered into force in 2004, discrimination based on ethnic origin had been prohibited by the Constitution, the Civil Code and the PEA . On the enactment of the ETA, the PEA was amended to provide that the requirement of equal treatment shall apply to all p articipants in public education and permeate all segme nts and procedures of the same.

62 . R elevant provisions of the PEA are as follows:

Section 4

“ (7) Those co-operating in the organisation, control and operation of public education and in the performance of the tasks of public educati on shall take account of the children ’ s interest, which is placed above everything else, when making decisions and taking measures.

The children ’ s interests which are placed above everything else are the following in particular: ...

b) that they should be given every kind of assistance to evolve their abilities and talents, to develop their personalities and to update their knowledge continually as prescribed by this Act; ... ”

Section 10

“ (3) Children and pupils have the following rights:

a) they shall receive education and teaching according to their abilities, interest and faculties, continue their studies according to their abilities and participate in primary art education in order that their talent should be recognised and developed; ...

f) they shall receive particular care – special nurture or care with the purpose of rehabilitation – according to their conditions and personal endowments, they shall appeal to the institution of pedagogical assistance service, irrespective of their age; ... ”

63 . The PEA further gives the definition of special educational needs (“SEN ”).

Between 1 September 1996 and 1 September 2003 , it prov ided as follows :

Section 121

“ (18) (later 20): [ The term of ] o ther disability [concerns] those children/ pupils who, on the basis of the opinion of the expert and rehabilitation committee :

a) struggle with pervasive development disorder (for example, autism) , or

b) struggle with disorders in school performance ... because of other psychic disorders ... as a consequence of which are lastingly impeded in development and learning (for example, dyslexia ... ); ... ”

64 . By 1 September 2003 the PEA was amended; and the term SEN was introduced instead of ‘ other disability ’ :

Section 121

“ ( 29 ) [C] hildren/ pupils with [SEN] are those who, on the basis of the opinion of the expert and rehabilitation committee :

a) suffer from physical, sensory, mental, speech deficiency or autism, or multiple disabilities in case of the joint occurrence thereof , or

b) are lastingly and substantially impeded in development and learning because of psychic disorders (for example, dyslexia ... ); ... ”

65 . As of 1 September 2007, s ection 121 of the PEA reads as relevant :

“(29) [C] hildren/ pupils with special educational needs are those who, on the basis of the opinion of the expert and rehabilitation committee :

a) suffer from physical, sensory, mental, speech deficiency or autism, or multiple disabilities in case of the joint occurrence thereof , and struggle with lasting and serious disorders in the cognitive functions or behavioural development, attributable to organic causes, or

b) struggle with long-term and serious disorders in the cognitive functions or behavioural development, not attributable to organic causes. ”

66 . As demonstrated above , as of 1996, the PEA differentiated between two categories of disability, namely the category of mentally disabled children and the one of those who suffered from adaptive, learning or behavioural difficulties.

As of 2003, the term SEN was introduced and the category of mentally disabled children was defined as SEN( a) whereas the one of those who suffered from adaptive, learning or behavioural difficulties was defined as SEN( b).

In 2007, the law redefined the se categories and since then has differentiate d between the two categories according to the origin of special needs: organic disabilit ies correspond to SEN ( a) whereas special needs with non-organic causes correspond to SEN ( b) . If the disability is attributable to organic causes , the child is declared by the rehabilitation committee of experts as having mild mental disability and will be educated in a specialised institution with specialised teachers. If the special needs do not originate in organic causes th e n the child can be educated in an integrated way, that is, in normal mainstream schools but with the support of special education teachers. Nevertheless , the PEA also allow ed ‘ SEN(b) children ’ to be educated in special schools or classes, under a special curriculum ; i n o rder to change this practice, a subsequent amendment was introduced to the effect that only those mentally disabled children should be placed in segregated special schools whose disability derived from organic causes.

However, i n 2008, a new amendment reinstalled the previous provision of educating SEN children, again allowing children who were not mentally disabled and had no organic disability to be educated in segregated special schools.

67 . As of 1 September 2007 the PEA introduced a provision for pupils suffer ing from adaptive, learning or behavioural difficulties , who can be educated in an integrated way:

Section 30

“(7) If a child / pupil struggles with adaptive, learning or behavioural difficulties ... or the chronic and serious derangement of cognitive functions or of development of behaviour ascribable to organic reasons, he or she is entitled to developmental education. ...

(8) The question whether a child/pupil struggles with adaptive, learning or behavioural difficulties or has special educational needs shall be decided by t he rehabilitation committee of expe rts at the request of the educational counselling service. ”

68 . As of 2003, the PEA also regulates the necessary conditions for educating children with special educational needs:

Section 121

“( 28 ) The necessary conditions for the education and teaching of children with special educational needs are as follows : employment of conductive therapists and therapeutic teachers according to the separate kindergarten education or school education and teaching of children/pupils and the type and severity of the special educational need; application of a special curriculum, textbooks or any other special aids necessary for education and teaching; engagement of therapeutic teachers with qualifications in a special field necessary for private tuition, integrated kindergarten education, school educatio n and teaching, developmental preparation and activities specified by the competent committee of experts; a sp ecial curriculum, textbooks and special therapeutic and technical tools necessary for the activities; provision of the professional services specified by the rehabilitation com mittee of experts for children students; ... ”

69 . Under the PEA , the term “ special curricul um” means that ‘ SEN children ’ may be exempt from certain subjects fully or partially, according to the opinion of the expert and rehabilitation committee or the pedagogical advisory committee .

70 . Lastly, t he PEA also defines the different ca tegories of secondary education and provides that , in order to educate children with special educational needs, secondary school s shall operate as special vocational school. Such school s shall educate those pupils who , as a result of their disabilities, cannot be educated in mainstream school .

C. Natio nal Social Inclusion Strategy (Extreme Poverty, Child P overty, the Roma) (2011–2020)

71 . This document, published by the Ministry of Public Administration and Justice (State Secretariat for Social Inclusion) in December 2011, contains the following passages:

“II.2. Providing an inclusive school environment, reinforcing the ability of education to compensate for social disadvantages

The development of an inclusive school environment that supports integrated education and provides education that breaks the inheritance of segregation and disadvantages as well as the development of services assisting inclusion play a primary role in the reduction of the educational failures of disadvantaged children, including Roma children.

As empha sised in the national strategy “ Making Things Better for Our Children” (2007), « in an educational system creating opportunities, children, regardless of whether they come from poor, under-educated families, live in segregated living conditions, are disabled, migrants or blessed with outstanding talent, must receive education suited to their abilities and talents throughout their lifetime, without their education being influenced or affected by prejudices, stereotypes, biased expectations or discrimination. Therefore, this must be the most important priority of Hungary ’ s educational policy. »

In the interest of reducing the extent of educational exclusion, we must reduce the selectivity of the educational system. Institutions must have effective tools against discrimination and need major methodological support for promoting the integration of pupils encumbered with socio-cultural disadvantages; this is also the way to reduce the out-migration of non-Roma pupils from certain schools. The development and application of an inclusive school model is a fundamental criterion concerning the regulation, management and coordination of public education that is also key in methodological developments as well as in the renewal of teacher training and the determination of the content of cooperation between institutions.

In the interest of ensuring that, likewise, children should not be unnecessarily declared disabled, we must provide for the enforcement of procedures determined in the relevant rule of law and professional criteria concerning the examinations serving as the basis for the subsequent expert opinion by providing professional assistance on an ongoing basis and with independent and effective inspections. In the spirit of prevention and in the interest of ensuring the timely and professional development of children, we must create standard procedures, professional contents and requirements also in the areas of early childhood development, educational consulting and speech therapy. The range of tests, examination methods and means used in the course of the testing and examination of children must be continuously extended. We must pay particular attention to avoiding declaring children disabled unnecessarily in the case of disadvantaged children transferred into long-term foster care and the Roma and must ensure that the tests, methods and procedures employed for the determination of the child ’ s actual abilities should be able to separate any deficiencies that may arise from environmental disadvantages.”

III. RELEVANT INTERNATIONAL TEXTS

A. Council of Europe sources

72 . Recommendation n o. 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) provides as follows:

“The Committee of Ministers, under the terms of Article 15.b of the Statute of the Council of Europe,

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 ...

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.”

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

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. ”

73 . The Opinion on Hungary of the Advisory Committee on the Framework Convention for the Protection of National Minorities, adopted on 22 September 2000 (CM(2000)165)) , contains the following passage:

“41. The Advisory Committee is deeply concerned about the well documented cases of improper treatment of Roma children in the field of education, notably through putting them in “special schools”, which are reserved ostensibly for mentally disabled children. The Advisory Committee stresses that placing children in such special schools should take place only when it is absolutely necessary on the basis of consistent, objective and comprehensive tests, which avoid the pitfalls of culturally biased testing. It considers it a positive step that the existence of and the need to address this unacceptable phenomenon has been recognised by the Ministry of Education. The Advisory Committee considers that the current situation is not compatible with Article 12(3) of the Framework Convention and must be remedied. ”

74 . The F ollow-up Report on Hungary (2002-2005) of the Council of Europe Commissioner for Human Rights (29 March 2006 ) ( CommDH(2006)11) c ontains the following passages:

“29. The Ministry of Education estimates that 95% of children of school age are registered school attenders. Alongside the normal schooling programme, there is special educational provision for children regarded as requiring special attention on account of handicap. While the maximum size of ordinary classes is 25 children, the special classes have a maximum of 13 so as to ensure quality instruction. The per-pupil grant which central government makes to local authorities is doubled for children in the special classes.

30. Around 20% of Roma children continue to be assigned to special classes as against only 2% of Hungarian children. It should be noted that dyslexia is regarded as a serious difficulty requiring placement in a special class and that social marginality has sometimes also been treated as a handicap. As a result, whereas the proportion of handicapped children in Europe is 2.5%, it is 5.5% in Hungary on account of inappropriate or abusive placements of this kind.

31. A protection mechanism has recently been introduced which requires parental consent for a child to be placed in a special class. In addition, the child must be tested without delay to assess its abilities. During the visit it was explained to the delegation that the files of 2,000 children regarded as handicapped had been thoroughly checked to make sure that placement in a special class was strictly necessary and to put right any abusive placements which authorities had made for financial or segregation reasons. Of the 2,000 children concerned, 10% had been returned to ordinary schooling after the check – evidence that close supervisi on of placements must continue.”

75 . The Report on Hungary of the European Commission against Racism and Intolerance (ECRI) (fourth monitoring cycle) , a dopted on 20 June 2008 and p ublished on 24 February 2009 , contains the following passages:

“ 81. [O f ] the three levels of disabilities into which children in special schools may fall (“very serious” (requiring residential care), “medium-severe” or “mild ” disability ), the vast majority of children assessed as having a “mild disability” could, in the view of many NGOs, be integrated relatively easily in the ordinary school system: many children are misdiagnosed due to a failure to take due account of cultural differences or of the impact of socio-economic disadvantage on the child ’ s development, and others suffer from only very minor learning disabilities that do not warrant the child ’ s removal from the mainstream system. ECRI repeatedly heard that investments in teacher training should primarily be directed towards ensuring that teachers in the mainstream school system are equipped to deal with diverse, integrated classes, rather than towards perpetuating a system from which children, once streamed into it, are unlikely to break out, and which overwhelmingly results in low levels of educational achievement and a high risk of unemployment. Some actors have suggested that – bearing in mind that the best way of ensuring that children do not wrongly become trapped in special schools is to ensure that they are never sent down that track in the first place – the category of children with mild disabilities should simply be deleted from the Education Act and all children with mild disabilities integrated in the mainstream school system.

82. ECRI notes that the efforts made to date to combat the disproportionate representation of Roma children in special schools for children with mental disabilities, though they have had some positive effects, cannot be said to have had a major impact in practice so far. It stresses that, in parallel to assisting wrongly diagnosed children already in the special school system to return to the mainstream system, putting an end to this form of segregation also implies ensuring that children are not wrongly streamed into special schools. ”

B. Other international texts

76 . For other relevant international texts, see D.H. a nd Others v. the Czech Republic [GC], cited above , §§ 81 to 107; and Oršuš and Others v. Croatia [GC], no. 15766/03, §§ 87 to 97, ECHR 2010 .

THE LAW

I. ALLEGED VIOLA TION OF ARTICLE 2 OF PROTOCOL No . 1 READ IN CONJUNCTION WITH ARTICLE 14 OF THE CONVENTION

77 . The applicant s argued that their education in a remedial school represented ethnic discrimination in the enjoyment of their right to education, in breach of Article 2 of Protocol No. 1 read in conjunction with Article 14 of the Convention.

Article 2 of Protocol No. 1 provides:

“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.”

Article 14 of the Convention provides:

“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.”

78 . The Government contested that argument.

A. Admissibility

1. The parties ’ submissions

a. Victim status

i. The Government

79 . The Government argued that the applicants could no longer claim to be victims of a violation of their rights within the meaning of Article 34 of the Convention given that the Regional Court had found in respect of the Expert Panel that the applicants ’ right to equal treatment and education had been violated by the Expert Panel ’ s failure to individualise their diagnoses or to specify the cause and nature of their special educational needs. Each of the applicants had been awarded HUF 1,000,000 as non-pecuniary damages. Moreover, the Supreme Court had found that the County Council was liable for its failure to supervise the legality of the functioning of the Expert Panel which had conducted a gravely unlawful practice by failing to observe the legal guarantees concerning the parents ’ rights to be present, be informed, consent or seek a remedy. The prejudice suffered on account of the applicants ’ deprivation of the right to a remedy provided for by law and thereby of the theoretical chance of obtaining a more favourable assessment of their learning abilities had been compensated by non-pecuniary damages.

ii. The applicants

80 . The applicants contested the Government ’ s assertion that these judgments fully and effectively remedied the violation of their rights. The damages provided in regard to the omissions of the County Council and the Expert Panel did not respond to their claim of structural direct/indirect discrimination, i.e. the flawed system of diagnosis in Hungary , or to their claim of misdiagnosis and inadequate education. It was also established by the Regional Court that the damage caused derived from the convergence of the actions of each of the respondents . Because of the appellate process, it was only with regard to the Expert Panel that the judgment had become final. However, the applicants asserted that a final judgment in respect to an authority last in line of culpability, i.e. the Expert Panel, could not effectively remedy the violation of their rights to equal treatment in education. Given that respondents ’ actions had been inseparable, the Expert Panel alone could not have changed the structure under which the applicants had been misdiagnosed. Therefore, they continued to be victims of a violation of their rights under the Convention.

b. Exhaustion of domestic remedies

i. The Government

81 . Concerning the applicants ’ claim that the assessment of their learning abilities had not been made with culturally unbiased tests which amounted to a general claim of a systemic error, the Government submitted that in this respect the applicants had failed to exhaust domestic remedies in accordance with Article 35 § 1 of the Convention. Such claims should have been raised by the applicants in proceedings instituted against the ministry responsible for education. The availability of this remedy was undisputable and there was record of successful such actions. Moreover, as to the issue of segregation, the Government submitted that this issue had not been raised before the competent domestic authorities; in particular, the question of the County Council ’ s liability for the eventual discriminatory effect of its education policy had been not addressed by the applicants in the domestic proceedings although the local authorities were better placed to determine the adequacy of an education policy to the needs of the children concerned. It was true that the applicants had initially filed an action against the County Council on account of its alleged failure to provide them with an education adequate to their abilities, however, they had withdrawn that action on 26 February 2007 and 9 March 2007, respectively.

ii. The applicants

82 . The applicants contested the Government ’ s position, claiming that they had submitted their claim before the domestic courts against respondents who were – each to a different extent as part of a system – all responsible for their misdiagnoses. They claimed that the ministry responsible for education oversaw the whole education sector, while at the local level it was the county councils which maintained, supervised and controlled the expert panels assessing children. In Hungary , certain S tate duties were transferred to local public authorities due to decentralisation of the public administration.

c . S ix-month time-limit

i. The Government

83 . The Government were of the opinion that the application was also inadmissible for the applicants ’ failure to observe the six-month time-limit laid down in Article 35 § 1 of the Convention. On the issue of whether the applicants ’ education was channelled into special education on the basis of assessments made with culturally biased or unbiased tests and methods, the Regional Court ’ s judgment of 27 May 2009 had been the final domestic decision. This judgment became final in regard to the Expert Panel on 2 July 2009 . The applicants, however, had not submit ted their application until 11 February 2011, that is, more than six months later .

ii. The applicants

84 . In order to find redress for the violation of their rights, the applicants stressed that they had needed to exhaust all effective domestic remedies available to them against all respondents who bore joint liability for the alleged breaches. Therefore the six- month time-limit r a n from the receipt of the Supreme Court judgment on 11 August 2010 . Indeed, the Government did not claim that the review by the Supreme Court had not been an effective remedy.

2. The Court ’ s assessment

85 . The Court finds that the above objections are interrelated and must be examined together . In so far as the applicants ’ claim of d iscrimination and/or misdiagnosi s is concerned, the Court observes that the Supreme Court did not sustain the applicants ’ claim of discrimination and breach of equal treatment. In particular, it confirmed the position of the lower courts regarding the respondents ’ joint liability, finding that, in the adjudication of the claims against the appealing parties, it was appropriate to evaluate the conduct of the School and the County Council in relation to the unlawful acts of the Expert Panel, as established by the Regional Court, even if the latter ’ s judgment had become final in the absence of appeal in regard t o the Expert Panel. In view of this finding of joint liability, the Court will consider the alleged violations as deriving from the joint acts of the School, the County Council and the Expert Panel. However, the applicants obtained redress only in regard to the Expert Panel ’ s handling (see paragraph s 43 to 54 above) , and none in regard to their claims of discrimination. In these circumstances , the Court is satisfied that the applicants have retained their victim status for the purposes of Article 34 of the Convention.

86 . Moreover, the Court observes that the applicants pursued claims of discrimination and u nequal treatment before all domestic j udicial instances, including the Supreme Court , which however held in essence (see paragraph 53 above) that the applicants ’ claim of systemic error amounting to a violation of their Convention rights could not, in the circumstances, be redressed by means of the national law. The Court is therefore satisfied that – in respect of the alleged discrimination in the enjoyment of their right to education – the applicants have taken all the requisite steps to exhaust domestic remedies that can be reasonably expected in the circumstances .

87 . Concerning the applicants ’ claim about the unsuitability of the tes t battery applied in their case , the Court notes that the applicants could have brought an action against the education authorities under this head. However, they did not do so . This aspect of the case cannot therefore be examined on the merits for non-exhaustion of domestic remedies (see also Horváth and Vadászi v. Hungary (dec.), no. 2351/06, 9 November 2010).

88 . It follows from the above considerations that, to the extent that the applicants have exhausted domestic remedies, the six - month time-limit ran from the service of the Supreme Court ’ s judgment on 11 August 2010 and has thus been respected.

89 . Furthermore, t he Court notes that the application is not manifestly ill-founded within the meaning of Article 35 § 3 (a) of the Convention. It further notes that it is not inadmissible on any other grounds. It must t herefore be declared admissible, apart from the applicants ’ claim about the unsuitability of the test battery applied in their cases (see paragraph 87 above) , which must be rejected for non-exhaustion of domestic remedies, pursuant to Article 35 §§ 1 and 4.

B. Merits

1. The parties ’ arguments

a. The applicants

90 . According to the applicants, the improper shunting of Roma children into special schools constituted indirect discrimination, and was impermissible under Article 2 of Protocol No. 1. Under domestic law, indirect discrimination occur red where an apparently neutral provision, criterion or practice would put persons of a specific racial or ethnic origin at a particular disadvantage compared with other persons, unless that provision, criterion or practice was objectively justified by a legitimate aim and the means of achieving that aim were appropriate and necessary.

91 . The applicants submitted that Roma were uniquely burdened by the current syst em; no other protected group had been shown to have suffered wrongful placement in special schools based on the diagnostic system. S ocial deprivation was in great part linked to the concept of familial disability. This notion had been formulated during the first big wave of re-diagnosis of Roma children transferred to special schools in the 1970s. According to contemporary research , familial disability could not amount to any type or form of mental disability, as it was in essence based on the social deprivation and the non-mainstream, minority cultural background of Roma families and children. The definition of mental disability as comprising social deprivation and/or having a minority culture amount ed to bias and prejudice.

92 . In addition, the tests used for placement had been culturally biased and knowledge-based, putting Roma children at a particular disadvantage. None of the applicants had been observed in their home, and their ethnicity had not been taken into account when assessing the results. Consequently, their socio-cultural disadvantaged background resulting from their ethnicity had not been taken into consideration .

93 . The applicants further fault ed the examination process for its not being sufficiently individualis ed. After the first assessment , based on which the applicants had been transferred to a special school, the applicants had in fact not been re-examined. The “ review ” had been paper-based, their diagnose s had never been individualised , and their parents ’ rights had not been respected. These failures had been established by the domestic courts. Indeed, it had been a violation to assign them to special schools when their tests had indicated IQ scores higher than WTO standards for mental disability. For the applicants, the issue was why the Government had allowed expert p anels across the country and in Nyíregyháza in particular to diagnose mild mental disability contrary to WHO standards. Given that the WHO standards had been applicable at the time, the development of science and the changing terminology could not serve as a reasonable j ustification for the misdiagnose s of the applicants and the deprivation of their right to access adequate education. Until 2007, special schools had not only educate d mentally disabled children, but also educated children with special education needs, including educational challenge and poor socio-economic background. Due to an amendment in 2007, the PEA had prescribed that all children who had been sent to special schools because of “psychological disorders” or “learning difficulties” had to be re - tested in order to establish whether the disorder was the result of organic reasons ; if not, those children had to be tran sferred back to normal schools.

b. The Government

94 . The Government denied that the applicants had been treated less favourably than non-Roma children in a comparable situation. Moreover, inasmuch as their treatment in education had been different from that of non-Roma (and other Roma) children of the same age, it had had an objective and reasonable justification. Moreover , they had not been treated differently from non-Roma children with similar socio-cultural disadvantages.

95 . The Government were of the opinion that tests and standards tailored to the Roma population would have no sensible meaning from the point of view of assessing a child ’ s ability to cope with the mainstream education system – which was the purpose of the assessment of learning abilities of children and of the psychometric tests applied in the process. They refer red to NERC ’ s expert opinion of 28 June 2007 , which stated that the culture-bias of the “Budapest Binet T est ” was less apparent in younger ages ( three to six years of age) because it measured primarily basic practical knowledge. When this test was applied , its cultural bias could be compensated by a pedagogical examination aimed at exploring practical knowledge. Moreover, this one had not been the only test applied ; and the applicants had been tested with a complex method. The diagnose s that the applicants needed special education had not been based on a single test; they had not even been exclusively based on the results of various tests obtained in a single examination session.

96 . Moreover, the results of standardis ing the recently developed “ WISC-IV Child Intelligence Test ” show ed that there we re no ethnically determined differences b etween the test scores of Roma and non-Roma children. Therefore, in light of foreign experience gained in this field, it had been decided in the standardisation process not to lay down separate norms specifically appl icable to Roma children but to use other means to ensure the fair assessment of all children in the course of the application of standardised tests. Relying on expert opinions , the Government claim ed that socio-cultural background had been decisive for the mental development of the child , and when the actual level of a child ’ s mental development (IQ) had been measured, the result had necessarily been influenced by the same socio-cultural effects that had shaped the child ’ s mental development. In sum, the above results of the standardisation prove d that IQ tests did not measure any difference between Roma and non-Roma culture or any cul tural differences between Roma and non-Roma children. What they did measure was the effect of cultural deprivation or insufficient cultural stimuli in early childhood on the mental development of children, irrespective of their ethnic origin. Dispropor tionate representation of Roma children in special education was explained by their disproportionate representation in the group deprived of the beneficial effects of modernisation on the mental development of children. These factors concern ed areas of social development which fell outside the scope of the right to education or any of the rights enshrined in the Convention.

97 . The Government were further of the opinion that the testing (or assessment) of the applicants ’ abilities had been sufficiently ind ividualised even if their diagnose s had not been so, as it had been established and redressed by the Regional Court ’ s final judgment against the Expert Panel.

98 . Moreover, the Government agreed that the ensuing possibility of errors of assessment resulting from eventual personal biases or professional mistakes being committed must be counterbalanced by appropriate safeguards. Such procedural safeguards, including the parents ’ rights to be present, be inf ormed, consent or seek remedy, we re provided for by Hungarian law. The fact that these safeguards had not been respected in the applicants ’ case was not disputed: it had been established by the Supreme Court which had found that the Expert Panel had conducted a gravely unlawful practice in this respect and that the County Council had also been liable for this on account of its failure to supervise the legality of the functioning of the Expert Panel, as well as to put an end to the unlawful practice.

99 . The assessment by the Expert Panel had not been carried out for medical purposes but with a view to determining whether the applicants could successfully be educated in a mainstream school. Therefore, contrary to the applicants ’ opinion, it could not be regarded as misdiagnosis if a diagnosis of learning disability , in terms of special education , did not coincide with a medical diagnosis of mild mental retardation as defined by the WHO.

100 . Therefore , it had not been unreasonable for the Supreme Court to examine the applicants ’ diagnose s, contrary to the medical approach proposed by them, from the point of view of their right to an education adequate to their abilities and to find that, from this aspect , the Expert Panel ’ s original diagnose s es tablishing that the applicants had needed education under a special curriculum had been confirmed by the forensic experts ’ opinion , even in the second applicant ’ s case.

2. The Court ’ s assessment

a. General principles

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. 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 (see D.H. and Others , cited above, §§ 175-176).

103 . Furthermore, the Court reiterates that the word “respect” in Article 2 of Protocol No. 1 means more than “acknowledge” or “take into account”; in addition to a primarily negative undertaking, it implies some positive obligation on the part of the State ( see Campbell and Cosans v. the United Kingdom , 25 February 1982, § 37, Series A no. 48 ). Nevertheless, the requirements of the notion of “respect”, which appears also in Article 8 of the Convention, vary considerably from case to case, given the diversity of the practices followed and the situations obtaining in the Contracting States. As a result, the Contracting States enjoy a wide margin of appreciation in determining the steps to be taken to ensure compliance with the Convention with due regard to the needs and resources of t he community and of individuals (see Lautsi and Others v. Italy [GC], no. 30814/06, § 61, ECHR–2011 (extracts); Leyla Åžahin v. Turkey [GC], no. 44774/98, § 135, ECHR 2005 ‑ XI; Case “relating to certain aspects of the laws on the use of languages in education in Belgium” (merits), 23 July 1968, pp. 30-31, § 3 , Series A no. 6).

104 . In the context of the right to education of members of groups which suffered past discrimination in education with continuing effects, structural deficiencies call for the implementation of positive measures in order, inter alia , to assist the applicants with any difficulties they encountered in following the school curriculum. These obligations are particularly stringent where there is an actual history of direct discrimination. Therefore, some additional steps are needed in order to address these problems, such as active and structured involvement on the part of the relevant social services (see Oršuš and Others , cited above, § 177) .

The Court would note in this context Recommendation no. R(2000)4 of the Committee of Ministers (see paragraph 72 above) according to which a ppropriate 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.

105 . Furthermore, t he 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. S uch a situation may amount to “indirect discrimination”, which does not necessarily require a discriminatory intent (see, amongst other authorities , D.H. and Others , cited above, § 184).

A general policy or measure which is apparently neutral but has disproportionately prejudicial effects on persons or groups of persons who, as for instance in the present case, are identifiable on the basis of an ethnic criterion, may be considered discriminatory notwithstanding that it is not specifically aimed at that group, unless that measure is objectively justified by a legitimate aim and the means of achieving that aim are appropriate, necessary and proportionate (see OrÅ¡uÅ¡ and Others , cited above, § 150 ) . Furthermore, discrimination potentially contrary to the Convention may result from a de facto situation (see Zarb Adami v. Malta , no. 17209/02, § 76, ECHR 2006 ‑ VIII ).

106 . Where it has been shown that legislation produces such indirect discriminatory effect, the Court would add that, as with cases concerning employment or the provision of services (see, mutatis mutandis , Nachova and Others v. Bulgaria [GC], nos. 43577/98 and 43579/98, § 157, ECHR 2005 ‑ VII) , it is not necessary , in cases in the educational sphere , to prove any discriminatory intent on the p art of the relevant authorities (see D.H. and Others , cited above, § 194).

107 . 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 (see D.H. and Others , cited above, § 188).

108 . W here an applicant alleging indirect discrimination establishes a rebuttable presumption that the effect of a measur e or practice is discriminatory, the burden of proof shifts to the respondent State . T he latter must show that the difference in treatment is not discriminatory (see, mutatis mutandis , Nachova and Others , loc. cit.). Regard being had in particular to the specificity of the facts and the nature of the allegations made in this type of case (see, mutatis mutandis , Nachova and Others , cited above, § 147), it would be extremely difficult in practice for applicants to prove indirect discrimination without such a shift in the burden of proof.

b. Application of those principles to the present case

109 . The Court notes that the applicants in the present case made complaints under Article 2 of Protocol No. 1 read in conjunction with Article 14 of the Convention, claiming that the fact that they had been assigned to a remedial school for children with special educational needs during their primary education violated their right to receive an education and their right to be free from discrimination. In their submission, all that has to be established is that, without objective and reasonable justification, they were assigned to a school where, because of the limited curriculum, they were treated less favourably than non-Roma children in a comparable situation and that this amounted in their case to indirect discrimination (compare with the above-mentioned D.H. and Others judgment , § 183 ) .

110 . The Court notes that Roma children have been overrepresented among the pupils at the Göllesz Viktor Remedial Primary and Vocational School (see paragraph 7 above) and that Roma appear to have been overrepresented in the past in remedial schools due to the systematic misdiagnosis of mental disability (see paragraph 10 above). The underlying figures not having been disputed by the Government – wh o have not produced any alternative statistical evidence – the Court considers that these figures reveal a dominant trend. It must thus be observed that a general policy or measure exerted a dispro portionately prejudicial effect on the Roma, a particularly vulnerable group. For the Court, t his disproportionate effect is noticeable even if the policy or the testing in question may have similar effect on other socially disadvantaged groups as well . T he Court cannot accept the applicants ’ argument that the different treatment as such resulted from a de facto situation that affected only the Roma . However, it is uncontested – and the Court sees no reason to hold otherwise – that the different, and potentially disadvantageous, treatment applied much more often in the case of Roma than for others. The Government could not offer a reasonable justification of such disparity, except that they referred , in general terms, to the high occurrence of disadvantageous social background among the Roma ( see paragraph 96 above ) .

111 . Although the policy and the testing in question have not been argued to aim specifically at that group, for the Court there is consequently a prima facie case of indirect discrimination. It thus falls on the Government to prove that in the case of applicants the difference in treatmen t had no disproportionately prejudicial effects due to a general policy or measure th at is couched in neutral terms, and that therefore the difference in treatment was not discriminatory.

112 . 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, OrÅ¡uÅ¡ and Others , cited above, § 196; Larkos v. Cyprus [GC], no. 29515/95, § 29, ECHR 1999-I; Stec and Others v. the United Kingdom [GC], no. 65731/01, § 51, ECHR 2006 ‑ VI ). The Court stresses that w here 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.

113 . The Court notes the Government ’ s submissions (see paragraph 94 above) according to which the impugned treatment is neutral ( that is, based on objective criteria) and results in the different treatment of different people, and moreover the education program me in its existing form is beneficial to pupils with different abilities. The Court accepts that the Government ’ s position to retain the system of special school s/classes has been 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 which the system causes (see paragraph s 73 to 75 above) – even if in the present case the applicants were not placed in ethnically segregated classes .

114 . The Court notes that the Hungarian authorities took a number of measures to avoid misdiagnose s in the placement of children. Nevertheless, the Council of Europe Commissioner for Human Rights found in 2006 that 20% of Roma children continued to be assigned to special classes, as compared with only 2% of majority children (see paragraph 74 above) . Moreover, the ECRI Report published in 2009 ( see para graph 75 above ) indicated a high number of misplaced Roma pupils . For the Court, t hese facts raise serious concerns about the adequacy of these measures at the material time.

115 . The Court notes that the misplacement of Roma children in special schools has a long history across Europe .

Regarding the Czech Republic, the Advisory Committee on the Framework Convention for the Protection of National Minorities pointed out that children who were not mentally handicapped were frequently and quasi-automatically placed in Czech remedial schools “[owing] to real or perceived language and cultural differences between Roma and the majority” (see D.H. and Others , cited above, § 68).

In Hungary , the concept of “familial disability” ( see paragraph s 10 and 91 above ) resulted in comparable practices. The ECRI R eport published in 2009 notes that the vast majority of children with mild learning disabilities could easily be int egrated into mainstream schools; and many are misdiagnosed because of socio-economic disadvantage or cultural differences. T hese children are unlikely to break out of this system of inferior education, resulting in their low er educational achievement and poorer prospects of employment. The R eport also noted that efforts to combat the high proportion of R oma children in special schools – both by assisting wrongly diagnosed children and preventing misdiagnosis in the first place – have not yet had a major impact (see paragraph 75 above).

116 . In such circumstances – and in light of the recognised bias in past placement procedures ( see paragraph 115 above) – the Court considers that the State has specific positive obligations to avoid the perpetuation of past discrimination or discriminative practices disguised in allegedly neutral tests.

117 . While in the present case the Court is not called on to examine the alleged structural problems of bias ed testing, the related complaint being inadmissible ( see para graph 87 above) , it is nevertheless incumbent on the S tate to demonstrate that the tests and their application were capable o f determin ing fairly and objectively the school aptitude and mental capacity of the applicants.

118 . The Court observes that the Hungarian authorities set the borderline value of mental disability at IQ 86, significantly higher th an the WHO guideline of IQ 70 (see paragraph 18 above). The Expert Panel found disparate measurements of Mr Horváth ’ s IQ between IQ 61 and 83. Mr Kiss had an IQ of 63 according to the “ Budapest Binet Test” and an IQ of 83 accord ing to the “Raven T est ” . However, when taking the latter test at a summer camp (see paragraph 31 above) , Mr Horvá th scored IQ 83 and Mr Kiss IQ 90.

The Court cannot take a position as to the acceptability of IQ scores as the sole indicators of school aptitude but finds it troubling that the national authorities significantly departed from the WHO standards.

119 . The Court observes, further, that the tests used to assess the applicants ’ learning abilities or difficulties have given rise to controversy and continue to be the subject of scientific debate and research. The Court is aware that it is not its role to judge the validity of such tests, or to identify the state-of-the- art, least culturally biased test of educational aptitude . It is only called on to ascertain whether good faith efforts were made to achieve non-discriminatory testing . Nevertheless , various factors in the instant case lead the Court to conclude that the results of the tests carried out in regard to applicants did not provide the necessary safeguards against misdiagnosis that would follow from the positive obligations incumbent on the State in a situation where there is a history of discrimination against ethnic minority children.

120 . In the first place, the Court notes that it was common ground between the parties that all the children who were examined sat the same tests, irrespective of their ethnic origin.

T he Government acknowledged that at least part of the test battery applied ( namely, the “ Budapest Binet T est ” ) was culturally biased (see paragraph 95 above) .

Moreover, certain tests used in the case of the applicants were found to be obsolete by independent experts (see paragraph 34 above) .

121 . In these circumstances, t he Court considers that, at the very least, there is a danger that the tests were culturally biased. For the Court, t he issue is therefore to ascertain to what extent special safeguards were applied that would have allowed the authorities to take into consideration , in the placement and regular biannual review process, the particularities and special characteristics of the Roma applicants who sat them , in view of the high risk of discriminatory misdiagnosis and misplacement.

122 . The Court relies in this regard on the facts established by the Regional Court which were not contradicted on appeal (see paragraph s 39 to 42 above) . This court found that the Expert Panel had failed to individualise the applicants ’ diagnoses or to specify the cause and nature of their special educational needs and therefore violated the applicants ’ rights to equal opportunity. Moreover, the social services administering the placement had been subject to constant reorganis ation. In this regard , the c ourt had found that the conditions necessary for the functioning of the Expert Panel had not been provided. Consequently, the Expert Panel and the County Council could not provide the necessary guarantees against misplacement which was historically more likely to affect Roma. Moreover, after a careful analysis of the applicable law , the Court of Appeal and the Supreme Court concluded that , as of 2003 , children with special educational needs had included students with psychological developmental trouble s (learning disabilities). It was not clear whether the applicants had mental (or learning ) disabilities that could not have been taken into consideration within the normal education system by providing additional opportunities to catch up with the normal curriculum. Those courts found that, because of th e changes in legislation, related to changing concepts on integrated education, there was lack of legal certainty from 1 January until 1 September 2007 (see paragraph 45 in fine above)

123 . In the face of these findings, it is difficult for the Court to conceive t hat there was adequate protection in place safeguarding the applicants ’ proper placement. Therefore, the tests in question, irrespective of their allegedly biased nature, cannot be considered to serve a s sufficient justification for the impugned treatment.

124 . As regards the question of parental consent, the Court accepts the Government ’ s submission that in this regard the violation of the applicants ’ rights to education was recognis ed and adequate remedies were provided in the domestic procedure (see paragraph 79 above) . However, in the case of Mr Kiss, the absence of parental participation and the parents ’ express objection to the placement can be seen as having contributed to the discrimination.

125 . The Court notes that the identification of the appropriate educational program me for the mentally disabled and students with a learning disability, especially in the case of Roma children, as well as 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. The Court notes in the Hungarian context that the 2003 programme (see paragraph 12 above) and the 2011 National Inclusion Strategy (see paragraph 71 above) advocate an integrated approach in this respect .

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 (see Valsamis v. Greece , 18 December 1996, § 28, Reports of Judgments and Decisions 1996-VI).

126 . 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 , 25 September 1996, § 76, Reports 1996– IV, and Connors v. the United Kingdom , no. 66746/01, § 83 , 27 May 2004).

127 . The facts of the instant case indicate that the schooling arrangements for Roma applicants with allegedly mild mental disability or learning disability were not attended by adequate safeguards 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 consequence , they received an education which did not offer the necessary guarantees stemming from the positive obligations of the State to undo a history of racial segregation in special schools. The education provided might have compounded their difficulties and compromised their subsequent personal development instead of helping them to integrate into the ordinary schools and develop the skills that would facilitate life among the majority population.

In that connection, the Court notes with interest that the new legislation intends to move out students with learning disabilities from special schools and provides for children with special educational needs, including socially disadvantaged children, to be educated in ordinary schools enabling the diminution of the statistical overrepresentation of Roma in the special school population. This integration process requires the use of state - of - the - art testing.

However, in the present case the Court is not called on to examine the adequacy of education testing as such in Hungary .

“[I] f a restriction on fundamental rights applies to a particularly vulnerable group in society, who have suffered considerable discrimination in the past, such as the mentally disabled, then the State ’ s margin of appreciation is substantially narrower and it must have very weighty reasons for the restrictions in question ... .[T]he treatment as a single class of those with intellectual or mental disabilities is a questionable classification, and the curtailment of their rights must be subject to strict scrutiny.” (paragraphs 42 and 44).

129 . Consequently, there has been a violation in the instant case of Article 14 of the Convention taken in conjunction with Article 2 of Protocol No. 1 in respect of each of the applicants.

II . APPLICATION OF ARTICLE 41 OF THE CONVENTION

130 . 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

131 . The applicant s made no damages claims .

B. Costs and expenses

132 . The applicant s claimed , jointly, 6,000 euros (EUR) for the costs and expenses incurred before the Court. This claim corresponds to 100 hours of legal work billable by their lawyer at an hourly rate of EUR 60.

133 . The Government contested this claim.

134 . According to the Court ’ s case-law, an applicant is entitled to the reimbursement of costs and expenses only in so far as it has been shown that these have been actually and necessarily incurred and are reasonable as to quantum. In the present case, regard being had to the documents in its possession and the above criteria , the Court considers it reasonable to award the sum of EUR 4,5 00 jointly to the applicants, who were represented by a lawyer and a non-governmental organisation , covering costs under all heads.

C. Default interest

135 . The Court considers it appropriate that the default interest rate 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 UNANIMOUSLY

1. Declares the complaint concerning the alleged unsuitability of the test battery applied in the applicants ’ case inadmissible and the remainder of the application admissible ;

2 . Holds that there has been a violation of Article 2 of Protocol No. 1 read in conjunction with Article 14 of the Convention;

3 . Holds

(a) that the respondent State is to pay the applicant s , jointly, within three months from the date on which the judgment becomes final in accordance with Article 44 § 2 of the Convention, EUR 4,5 00 ( four thousand five hundred euros), to be converted into the currency of the respondent State at the rate applicable at the date of settlement , plus any tax that may be chargeable to the applicants, in respect of costs and expenses;

(b) that from the expiry of the above-mentioned three months until settlement simple interest shall be payable on the above amount at a rate equal to the marginal lending rate of the European Central Bank during the default period plus three percentage points;

4 . Dismisses the remainder of the applicant s ’ claim for just satisfaction.

Done in English, and notified in writing on 29 January 2013 , pursuant to Rule 77 §§ 2 and 3 of the Rules of Court.

             Stanley Naismith Guido Raimondi Registrar President