Strasbourg, 2223 May 2000
MG-S-ROM (2000) 13

Roma and Statistics

Table of contents

Introduction 5

I. The collection of data of an ethnic nature :

Why collect data on Roma/Gypsies ?
Quantitative and qualitative evaluation of the existing data 7
Criteria, methods and control of the collection and use of data of an ethnic nature 9
Ethnic statistics and criminality: bias and taboos 11
Are there any alternatives to the collection of data of an ethnic nature? 12

II. Data protection in international law 13

Convention 108 of the Council of Europe for the protection of individuals with regard to automatic processing of personal data 14
The basic principles contained in Convention 108 and Recommendation (97) 18 (on the protection of personal date collected and processed for statistical purposes) pertinent to the question of the collection of data on an ethnic basis 16
Control of the application of Convention 108 17

III. Population censuses 18

The 1994 census in « the former Yugoslav Republic of Macedonia » 18

Debate on census operations: how to promote the involvement and participation of Roma/Gypsies in censuses? 19

Debate on the participation of Roma/Gypsies in society 21

Follow-up 23

I. List of participants 25
II. Principal relevant texts (data protection, see web site on data protection); basic texts for the EU and UN. 31



1. Following discussions in the Specialist Group on Roma/Gypsies concerning the problems and questions connected with the collection of data of an ethnic nature in the course of the past two years, the Council of Europe, in cooperation with the Project on Ethnic Relations (USA), organised a two-day Round Table entitled "Roma and statistics" (22-23 May 2000).

2. The participants (some 30) were Roma/Gypsy representatives from different European countries and diverse political persuasions, statisticians, demographers and lawyers specialised in data protection (see the list of participants in Appendix I).

3. The debate on the collection of data on an ethnic basis is now more than ever on the agenda of various organisations active in the field of Roma/Gypsies (OSCE/ODIHR, various NGOs such as European Roma Rights Centre, Open Society Institute, Roma/Gypsy organisations, etc.). As regards the Roma/Gypsies themselves, their opinions are divided and often very strongly held.

4. Schematically, there are two diametrically opposed positions. The first is that of a good proportion of the Roma/Gypsy representatives and populations who are opposed to any form of collection of data on an ethnic basis arguing, that in the past these data have always been used against them, the most terrible example being the Holocaust during the second world war, which was made possible by the existence of lists identifying the Roma/Gypsy population. In addition, it has to be borne in mind that the police services in many member States make frequent use of crime statistics compiled on an ethnic basis, which reinforce the prejudices and stereotypes of the majority population. These statistics are often collected in a dubious fashion and in breach of most national legislations, which prohibit this type of data collection.

5. Opposing this is the position of other Roma leaders, as well as the organisations working on these questions, who believe that quantitative and qualitative data are necessary if we want to be able to: 1) develop and implement policies aimed at improving the situation of Roma/Gypsies in fields such as employment, education, housing, etc., 2) evaluate the policies and projects implemented and 3) combat discrimination and, in particular, bring proof of discrimination before the judicial authorities.

6. The two opposing positions presented in paragraphs 4 and 5 constitute in schematic fashion the two aspects of the dilemma faced by those who are active in the field of Roma/Gypsies. The debate held in Strasbourg on 22 and 23 May – which was the first held specifically on this question - nevertheless showed that the terms of the dilemma are more complex than they appear at first sight and that therefore the solutions required cannot be clear-cut or simplistic. The Strasbourg meeting launched the debate on this issue, but it is clear that more work and reflection should be devoted to this subject, both in different contexts (among Roma/Gypsy leaders, among policymakers, at national and international level, etc.) and concerning the different specific problems arising from the collection of data on an ethnic basis.

7. This report summarises the different points discussed at the Strasbourg meeting in the following order: 1) The collection of data of an ethnic nature and the different ways in which it is perceived, 2) Data protection in international law, 3) Population censuses.


8. In this first section we seek to reflect as faithfully as possible the first part of the debate, but arranged thematically rather than chronologically, in order to facilitate the reader’s understanding of the questions raised by the collection of ethnic data.

Why collect data on Roma/Gypsies? Quantitative and qualitative evaluation of the existing data

9. MM. Kassof and Mirga pointed out that at present we lacked certain data concerning Roma, both quantitative and above all qualitative, and this lack of data made it difficult to frame policies aimed at improving the position of Roma in various fields of everyday life, including their political representation. They added that a great deal of data actually existed, but that these data were often of poor quality and that the aim should therefore be to collect data of better quality. They acknowledged that ethnic data were often wrongfully used to stigmatise a particular group, but said that the need to collect data in the fields of health, housing, employment, etc., should be recognised as legitimate.

10. Other participants (Mr Gheorghe and Mr Kawczynski in particular) considered on the one hand that there were at present sufficient statistical data on Roma/Gypsies and on the other hand that the collection of this type of data was never innocent but always politically motivated. Mr Gheorghe added that the processes of collecting data on Roma/Gypsies were at present much more advanced and widespread than the processes aimed at implementing policies in favour of Roma, and that there was a lack of political actors capable of using the data already existing. He therefore considered that there was no need to collect more data.

11. Mr Gheorghe then explained that the opposition of many Roma/Gypsies to the collection of ethnic data was not only connected with the use made by the police services of data on "Roma criminality", but also with the fact that other government departments also collected data (for example on migratory phenomena) and that in addition there were exchanges of data between States, in particular those belonging to the Schengen group. He also said that the apprehension felt by many Roma/Gypsies with respect to data collection was probably connected with the fact that there were many who feared that it was aimed at limiting the freedom of movement of Roma/Gypsies and hence exercising tighter control over them.

12. Mr Kawzcynski also stressed the difficulty of exercising control over the use that was made of the data collected. In his opinion it was in fact not possible to say yes or no to the collection of ethnic data. It was more a matter in each case of knowing who was collecting the data, for what purpose, who prepared the questions for collecting the data and what types of data were collected. He thus highlighted the complexity and the multiplicity of the problems arising. Like Mr Gheorghe, he considered that the collection of data of an ethnic nature was much more a matter of politics than of statistics.

13. Mrs Verspaget raised the question of whether more figures were really an essential precondition for improving the situation of Roma/Gypsies. She was of the opinion that it was more useful to concentrate on the necessary legislative changes, on the adoption and implementation of appropriate policies, and that if there was to be data collection, this should be in connection with the implementation of policies in favour of Roma only, and with the participation of Roma organisations and representatives.

14. Taking the opposite view, Mr Goldston stressed the fact that it was necessary to have a better knowledge of the present availability of data in different countries and the existing legal norms in this field, and to broaden the debate on the collection of ethnic data. He was nevertheless in favour of the collection of this type of data, arguing that such data were absolutely necessary to anybody who wished to bring proof of discrimination and submit to the judicial authorities cases of discrimination against Roma. He developed another argument in favour of the collection of ethnic data which was also advocated by the representatives of Project on Ethnic Relations: the interests of Roma were often poorly evaluated and poorly represented because of the absence of numerical data in various fields of everyday life. He also pointed out that the difference between the collection of individual data and the collection of collective data should be borne in mind.

15. Mr Holomek explained that the great majority of Roma in the Czech Republic refused the collection of data of an ethnic nature for fear that they would be used against them. He nevertheless considered that the fact as a result it was necessary to work with estimates only gave rise to many problems. He gave some examples: it had never really been possible to know the exact number of Roma who had suffered as a result of the 1993 law on nationality in the Czech Republic and hence to precisely estimate its consequences; at present it was not known exactly how many Roma children were in special schools for the mentally handicapped; the 1990 census gave the figure of 33,000 Roma in the Czech Republic, whereas the actual number was much higher, and this influenced the way in which Roma matters were considered and the policies that were adopted. He nevertheless stated that even though more statistical data were necessary, it was essential that the use made of these data should be strictly controlled, which was not the case at present. Mrs Battaini-Dragoni added that more statistical data concerning Roma/Gypsies might perhaps make it possible to better identify the obstacles encountered by them regarding access to basic rights and public services.

16. Mr Gamez stepped in to introduce certain refinements into the debate. He explained the differences between population censuses (which gave a picture of a country at a point in time), administrative data (collected by institutions such as the police, not subject to the general rules established by the United Nations and collected for the specific purposes of an institution) and official statistics (collected for the use of the society as a whole, and not the property of a government). In the opinion of most participants, the type of data collection that caused the most concern was administrative data. The paradox often lay in the fact that to the extent that Roma tended not to identify themselves as such in the censuses, governments felt justified in collecting administrative data in order to compensate for the resulting lack of information from censuses (Mr Gheorghe).

Criteria, methods and control of the collection and use of data of an ethnic nature

17. Mr Gheorghe said the basic problem with the collection of data on Roma/Gypsies was connected with the fact that the identification was most often based on skin colour, because except in the case where self-identification was used, it was difficult to determine the criteria for identifying Roma/Gypsies. He therefore felt that identification could serve to institutionalise discrimination against Roma/Gypsies on the basis of the colour of their skin and, in fact, to accentuate the "ethnicisation" of relations between communities in central and eastern Europe. He concluded by stressing the urgent need to draw up good practice guidelines for the attention of those who collected data.

18. Mr Totev for his part said that other indicators, such as socio-economic status, education, etc., could be used instead of identification merely according to the colour of the skin. He added that it was also necessary to consider whether ethnicity could be regarded a statistical parameter and wondered whether the questions concerning the reliability of the data were not connected with the fact that these data often no longer reflected the present position. Mrs Kalibova explained that in censuses, data on the ethnicity of the individual were in fact collected only in central and Eastern Europe and this was due to the old practice of registration according to nationality. She wondered whether there could be any identification criteria to take into account other than self-definition.

19. Mr Gheorghe pointed to the fact that the pressure for collecting more data on Roma was in fact largely generated by the international organisations, who wanted governments to base themselves on recent studies and data when introducing measures for improving the situation of Roma. He cited in particular the action of the European Commission, which, through the PHARE programme for the countries in the accession phase, led governments to carry out detailed studies with precise figures on the situation of Roma. He wondered whether such data collection was lawful from the standpoint of the international norms. He concluded that these same organisations should also take care to ensure that the international standards concerning the protection of the rights of the individual in the collection of data were respected, should draw up codes of conduct for the States concerned and exercise control over the process. He suggested that this subject should be on the agenda of the next informal trilateral meeting between the Council of Europe, the OSCE and the European Union.

20. The participants in favour data collection stressed the need for the increasingly active involvement of Roma in the collection of data concerning them, as well as the tightening of the control procedures.

Ethnic statistics and criminality: bias and taboos

21. All participants denounced the collection, generally illicit, of administrative data on Roma criminality, which only went to reinforce the stereotypes and often did not reflect reality. This type of data was at the heart of the dilemma posed by the theme of the collection of ethnic data. Mrs Plaks described for example the absurd position of certain political leaders who claimed not to have any data concerning Roma because the collection of data on an ethnic basis was unconstitutional, but who were able to give very precise figures concerning the supposed criminality of Roma. Mr Gheorghe added that it seemed to him that Roma/Gypsies were the only minority group for which crime figures were generally collected, which showed both a lack of coherence and the desire to stigmatise.

22. He then presented another aspect this question: on the one hand, the reflection on the collection of ethnic data should now also include statistics and other data on race-motivated crimes committed against Roma; on the other hand, the question of "Roma criminality", supposed or real, was a taboo subject for the majority of Roma leaders. Mr Gheorghe wondered whether admitting its existence might be the price that Roma leaders had to pay to build confidence between Roma and the majority society.

23. Mrs Verspaget reacted to this last point by describing an experiment that had been conducted in the Netherlands which consisted of promoting joint efforts by the police, teachers and parents to combat the juvenile delinquency among migrant groups. She explained that even though the programme had had very convincing results, the lessons and statistics drawn from it which had been made public had only reinforced the stereotype view that immigrants were criminals.

Are there any alternatives to the collection of data of an ethnic nature?

24. Mr Holomek considered that the improvement of the situation of Roma had to come about through the emancipation of Roma and their affirmation as a national minority and that therefore statistical data concerning this minority were necessary, but the protection of these data had to be guaranteed. He recalled that this question was very closely connected with the problem of the representation of Roma at all levels (local, national, international) and that it was vital that Roma themselves should decide whether it was a good thing to collect data on an ethnic basis.

25. Mr Mirga summed up the way in which he perceived the problem: according to him, Roma did not want statistics on crime; furthermore, they did not want the collection other types of data or censuses either. He concluded that whatever the situation, their answer was no, and so he wondered what perception they had of their place in a democratic society. He also thought that there should not be this constant harking back to the memory of the holocaust to justify the refusal. On the contrary, he was of opinion that in order to be competitive Roma should have the courage to declare themselves as such and allow the majority society to know them better. He considered that Roma should also better demonstrate the discrimination and social exclusion they were the victims of by means of statistics and use the political argument of their sheer numbers in order to be better able to push their claims (as other minorities did). He also cited the example of certain Roma leaders who proclaimed their identity in order to be able to play a role in elections.

26. The debate also touched on the virtual absence of any collection of data of an ethnic nature in the European Union countries. The question, which then arose, was whether this absence of data had any consequences for the definition, coherence and implementation of policies. Mr Goldston, for his part, considered that the absence in several EU member States of any statistics of an ethnic nature was not necessarily a good example to follow, and that this made it very difficult to prove discrimination before a court of justice. He stressed the fact that the solution to the problem of the collection of data of an ethnic nature involved the development of effective control mechanisms and the education of those who handled this type of data.

27. Mr Gheorghe, was of the opinion that the absence of ethnic data did not prevent western European governments from taking action and cited in this connection the EU Resolution of the Council and the Ministers of Education meeting within the Council of 22 May 1989 on school provision for Gypsy and Traveller children.

28. Mrs Kalibova emphasises the fact that if we wish to have a coherent and pan-European approach, data should also be collected in Western European countries

29. He also proposed that projects demonstrating good practice in data collection should implemented at local level, in partnership with local Roma organisations, in order to see whether it was possible to collect ethnic data for useful purposes and to avoid any misappropriation of the data. Mrs Verspaget supported this idea and repeated that if there was to be data collection, it should be closely connected with the implementation of actions aimed at improving the situation of Roma.


30. Mrs Wiederkehr, Deputy Director of Legal Affairs, gave participants detailed information concerning data protection in the international texts, and in particular the Council of Europe Conventions and other pertinent texts.
We summarise her contribution here and include the references to the pertinent texts (Appendix II). We recommend the reader to consult the Directorate of Legal Affairs web site on data protection:

Convention 108 of the Council of Europe Convention for the protection of individuals with regard to automatic processing of personal data (1981)

31. There is a great diversity of legal instruments concerned with data protection. This protection finds its source in two articles of the European Convention for the Protection of Human Rights and Fundamental Freedoms (ECHR): Article 8 (Right to respect for private and family life) and Article 10 (Freedom of expression). However, it was recognised that the protection provided by the ECHR was not sufficient. The member States of the Council of Europe therefore decided to elaborate a specific instrument, which was Convention 108 of the Council of Europe Convention for the protection of individuals with regard to automatic processing of personal data, adopted in 1981, which was the first legal instrument on personal data protection of universal application. It has so far been ratified by 20 member States, while several others are in the process of ratification. This Convention implies the existence of national legislation on data protection and has thus stimulated an intense process of elaboration of legal instruments concerned with data protection in the States parties to the Convention.
Following the entry into force of Convention 108, a Consultative Committee was set up to permit experts in the field to meet regularly and to proceed to exchanges of experience in order to resolve the problems arising at national level.
The text of the Convention was also to be completed by guidelines on the implementation of the basic principles contained in it. A series of recommendations accompanied by filled out explanatory reports was therefore adopted concerning, for example, data protection and social security, the police, statistics, etc. (see the list of texts adopted in Appendix II). These latter instruments are not legally binding, but are very complete and detailed as to the practical implementation of the various principles contained in Convention 108.

32. In 1995, the European Union adopted directive 95/46/EC on the protection of individuals with regard to the processing of personal data and on the free movement of such data, based on the principles set out in Convention 108 and very detailed. The result is that European Union member States of the Union have had to adapt their national legislation to comply with the directive. Furthermore, this directive forms part of the acquis communautaire that candidate States have to adopt and implement.
In preparation for accession to the EU, central and eastern European countries such as Slovenia and Hungary have ratified Convention 108. Other countries of the region are preparing to do so too. The Council of Europe is offering them its legal expertise to help them adapt their legislation accordingly.

33. The European Court of Human Rights, in its judgement in the case of Rotaru v. Romania of 5 April 2000, confirmed the fact that Convention 108 is an essential instrument for the protection of private life and that data protection is an integral part of Article 8 of the ECHR.

34. Lastly, the Consultative Committee on Convention 108 recently adopted, in June 2000, a draft additional protocol regarding supervisory authorities and transborder data flows. This draft should shortly be submitted to the Committee of Ministers with a view to its adoption and opening for signature. It establishes an obligation for the Parties to set up supervisory authorities in the field of data protection, exercising their functions in complete independence. In addition, transborder flows of personal data to a state or organisation which is not subject to the jurisdiction of a Party to the Convention should in principle be prohibited unless that State or organisation ensures an adequate level of protection for the intended data transfer.

35. It can therefore be seen that there is a complex range of legal instruments available. Besides the texts mentioned above, the United Nations have elaborated a number of instruments and Eurostat, the European Union statistics agency, is also very active in this field.

36. The basic principles contained in Convention 108 and Recommendation (97) 18 (on the protection of personal data collected and processed for statistical purposes) pertinent to the question of the collection of data on an ethnic basis:

- Article 51
- Article 6 on “sensitive” data2
- Principle 1 of Recommendation (97) 18.

Article 6 stipulates that sensitive data, including data on racial origin, are subject to special rules and may be collected only where domestic law provides appropriate safeguards (protection of private life). This means in particular the need to have the consent of the person concerned by the collection of data (explicit, written, free consent given in full knowledge of the facts).

The collection of sensitive data on unidentified persons normally gives rise to no real problems, but where it is a matter of data that make it possible to identify the person, protection measures are again required: data collection should in this case be exceptional and domestic law must provide specific guarantees concerning the rapid dissociation of the sensitive data from those permitting identification.

Control of the application of Convention 108:

37. Responsibility for the control of the proper application of principles contained in the Convention lies essentially with the person subject to the collection of data, through the right to access to data of a personnel nature and the right to rectification when there is collection of sensitive data. It results from this that information and awareness raising work is required to make sure that people know their rights in this respect.

38. There are nevertheless supervisory authorities for the application of the Convention: in the first place the European Court of Human Rights, which can be invoked in domestic law, and the Council of Europe Human Rights Commissioner, who can deal with cases that cannot be referred direct to the Court.
In addition, thought is being given to a consolidated legal instrument which would make it possible to have a single reference instrument encompassing the various existing texts on data protection.

39. Lastly, the draft additional Protocol to Convention 108, through the creation of supervisory authorities for data protection, should make it possible to better inform people of their rights and duties concerning the collection and exchange of personal data and it will be possible for individuals to take their complaints to these authorities.


40. Mr Millich, principal Administrator in the Population Division, informed the participants of the work done in recent years by the European Population Committee (CDPO), in particular on population censuses and the demography of national minorities3.

We summarise his contribution below.

41. First, it must be borne in mind that 50% of the Council of Europe member States do not collect data of an ethnic nature because their constitutions prohibit it. Furthermore, where data of an ethnic nature are collected, the figures obtained for Roma/Gypsies are generally not correct. This is due to several factors, but mainly to the fact that national minorities generally do not have any faith in the collection of data concerning them, the use to which it may be put (which results in under-declarations).
However, there are safeguards in census practices that should ensure the respect of international data protection standards.

The 1994 census in "the former Yugoslav Republic of Macedonia"

42. The Council of Europe had the opportunity to put some of these principles into practice on the occasion of the population census in "the former Yugoslav Republic of Macedonia" in 1994, conducted under the control and with the assistance of the international community, the aim of which was to obtain reliable data acceptable to all the communities present. (It is to be reminded that one of the national minorities had boycotted the 1991 census.

43. Bearing in mind the experience of the 1994 census, it can be recognised that, in order to give confidence to all groups in the population, the coming censuses should fulfil the following conditions:
- Elaboration of the census law in conformity with, among others, the international standards on this subject and ensuring that the data collected would not be used for any purposes other than those provided for by the census;
- Setting up of census committees, made up of representatives of the minority groups at national and local level; representatives of the minorities therefore had to be appropriately trained;
- The census form had to be printed in a language that could be understood by the respondent; the census interviewer also had to be able to speak the respondent’s language;
- The first page of the census form, which contained personal data permitting individuals to be identified, had to be destroyed as soon as the reliability checks for the census process had been completed (checks for possible double-counting, etc.);
- Representatives of the minorities had to be present during the counting and the analysis of the data.

44. Mr Millich concluded by recommending that the Roma/Gypsy communities should accept to actively participate in the next census campaign, on the one hand by asking to be represented in all the national and local bodies in charge of the preparation and supervision of the censuses and on the other hand by encouraging their members to answer to questionnaires. It is also thanks to this awareness and civil participation that Roma/Gypsies may pursue their emancipation process in the civil society. He recalls the the Council of Europe is at the disposal of the member States, should the latter wish to improve the confidence of the population in the censuses.

Debate on census operations: how to promote the involvement and participation of Roma/Gypsies in censuses?

45. Censuses provide a precise picture of a country’s situation at a given point in time. The fact is that Roma often appear under-represented because they often fail to declare themselves as Roma. The unreliable results of past censuses show that Roma were opposed to the idea of being counted and in fact, as pointed out by Mrs Kalibova, the results of censuses generally reflect the political position of the minorities in the society.

46. Mr Gheorghe stated that, contrary to what had been asserted by several participants, Roma had participated in the latest censuses in central and eastern Europe and that the number of Roma declaring themselves as such had increased, in particular in Bulgaria, Romania and Hungary (unlike the Czech Republic and Slovakia). Mrs Kalibova was able to introduce these data on the Roma in her publication “The demographic characteristics of national minorities in certain European states”. According to him there were therefore usable data available and he concluded on this point by stressing the fact that the variations in the Roma population census figures probably depended much more on the way in which the censuses were organised (and the involvement of Roma) than on the attitude of Roma vis-à-vis the collection of data.

47. The majority of the participants agreed on the fact that population censuses represented an acceptable form of the collection of data with an ethnic component, insofar as the identification of ethnic origin was made on the basis of voluntary declaration and it was a form of data collection that concerned the whole population and did not single out Roma/Gypsies.

48. In addition, all stressed that the success of such operations depended mainly on the full and complete involvement of Roma/Gypsies in the entire process (preparation and supervision committees, analysis, control mechanisms) and the transparency of the operations. This required on the one hand, that Roma organisations and leaders should undertake to mobilise the communities to participate (information, preparation) and, on the other hand, that the Roma leaders should receive training concerning the aims and methods of population censuses.

49. Replying to a question from a participant, Mr Courbage stated that censuses nevertheless did not include persons in illegal situations (which was the case for certain Roma in particular refugees and migrants) unless there were regularisation operations, and that in addition surveys complementary to the censuses could be necessary in order to obtain a truer picture of the situation of marginalized groups such as Roma/Gypsies. According to him, relying solely on the results of censuses could lead to a distorted picture of the reality and exaggeration of certain characteristics and, for example, given the low figures for the Roma population generally obtained, give a very exaggerated idea of the crime rate in the Roma population.

50. Lastly, the question was raised of what it was still possible to do with respect to the preparation of the series of censuses to be conducted in 2001. Mr Rosza explained that in Hungary representatives of the Gypsy self-government had been involved in the preparation of the census forms. In other countries it appeared that the preparation had already been completed and it was too late to try to better involve the Roma leaders. It was nevertheless stressed that mobilisation campaigns could be organised for the Roma and that, above all, discussions between Roma leaders should be organised and encouraged on this subject, in order to have a clear position before the census campaigns of 2001.

Debate on the participation of Roma/Gypsies in society:

51. The debate on participation in population censuses led to a more general debate on the participation of Roma in society.

52. Mr Kawczynski and Mrs Verspaget were of the opinion that too much attention should not be concentrated of the forthcoming censuses because, according to them, they would not give decisive results. Mr Kawcynski thought that Roma should stop offering their services (for example concerning participation in censuses) to the majority society and its decision-makers, who were now asking a great deal of the Roma whereas they had always been kept at a distance in the past. He wondered about the fact that many were now trying to obtain data on Roma but were not, according to him, interested in Roma as people. He also stated that he did not want to play the role of guarantor vis-à-vis the Roma communities and assure them that the data collected by a government would be collected legitimately and for useful purposes, in order to encourage Roma to participate.

53. Mr Holomek for his part was of the opinion that even if there was doubt as to the results it was necessary to be involved in order to put an end to the non-participation of Roma in society, which often led to the emigration of Roma. For him, the only viable solution for improving the position of Roma was participation in the sedentary, non-gypsy society. In this connection, Mr Gheorghe pointed out that the report prepared by Y. Matras in 1996-97 for the European Committee on Migration (CDMG) of the Council of Europe4 had already identified as a factor leading to the migration of Roma the lack of confidence in and non-participation in the institutions of the majority society. Mr Goldston stressed the fact that states could win Roma confidence only if their procedures were made absolutely clear, transparent and close to the citizen.

54. Mr Millich intervened in the debate to point out that censuses were one of the ways of contributing to the emancipation of Roma and that in any event they could not harm their cause. He believed that participation should be encouraged, and that as time went on the data collected would become more and more precise and reliable, and suggested that the Roma leaders should discuss in the first place what they wanted the position of Roma to be in 15 or 20 years' time. He added in fact that once a consensus was found on this question, other questions such as those concerned with data collection would be easier to resolve. He finished by encouraging once more the Roma leaders to ask for the participation of Roma in the census control committees and for the training and recruitment of Roma census interviewers.

55. Mr Gheorghe continued by proposing that independent census observation committees should be set up, on the lines of the election observation committees.


56. The participants all felt it was necessary to continue to discuss this question and to go more deeply into the various aspects raised at the Strasbourg meeting.

57. Mr Gheorghe, for the OSCE/ODIHR, recalled that the OSCE High Commissioner on National Minorities had requested that a seminar on this same subject should be organised soon by the OSCE. He supported the idea that the different parties concerned by the question of collection of data of an ethnic nature should discuss this theme in depth, and in particular the Roma leaders among themselves. He added that this question would of course be at the centre of the debate to be held in October 2000 on Roma and their recent migrations, organised by the OSCE/ODIHR in Warsaw. He suggested that other meetings should be devoted to alternative options to census data such as the collection of administrative data, those collected by NGOs, research institutes, etc. He concluded by suggesting that the Council of Europe Specialist Group on Roma/Gypsies (MG-S-ROM) should ask its Romanian representative, Mr Zamfir, to inform it about his studies of 1993 and 1998 on Roma in Romania, carried out on the basis of samples selected from the electoral rolls.

58. Mr Kassof and Mrs Plaks also considered that broader discussions should be held, in particular with the representatives of the ministries of justice. They, as well as Mr Mirga, were clearly in favour of the collection of data of various types concerning Roma, as a way of drawing Roma out of their marginal situation. The Project on Ethnic Relations would prepare its own report after the meeting.

59. Mrs Verspaget proposed that the MG-S-ROM should start to draw up standards for the collection of data concerning Roma/Gypsies, including in particular the followings principles: consent of the person concerned, self-definition, involvement of Roma, control, objectives connected with the implementation of policies in favour of Roma, etc.

60. The Secretariat added that the report that it would prepare after the meeting (this report) would be broadly disseminated, in particular among member States. He recalled that the Council of Europe was available to member States to assist in the organisation of censuses and in the field of data protection.

61. In response to the request made by Mr Kawzcynski to finance round tables bringing together Roma leaders to discuss data collection, the Secretariat undertook to finance such actions. The first meeting of this type, co-financed by the Council of Europe and the Roma National Congress, will be held in Hamburg on September 2000.

62. Finally, the question of data collection on Roma/Gypsies, encouraged by various international organisations, in particular in the framework of the preparation programmes for accession to the European Union, will be on the agenda of the next informal meeting of the international organisations involved in Roma/Gypsy issues (Budapest, November 2000).



Representatives of Romani Organisations
/Représentants d’organisations romani :

Mr. Costel BERCUS, Coordinator, Romani CRISS
Str. Buzesti 19 Sector 1 PO BOX 22-68, Bucharest 1, ROMANIA
Tel./Fax : + 401 231 41 44 ; Mobile : 094 815 460,;

Mr. Karel HOLOMEK, Director, Central and East European Center for Roma
Jugoslavska 17, 613 00 Brno, CZECH REPUBLIC
Tel. : +420 5 45 21 14 11/21 57 45 ;Fax : + 420 5 57 17 98 or 420 5 42 21 02 65 ;
E-mail : or

Mr. Rudko KAWCZINSKI, President, Roma National Congress
Central European Office, Simon Von Utrecht Str 85, 20359 Hamburg, GERMANY
Tel : + 49 40 319 42 49 - Fax : + 49 40 31 04 75 ;- E-mail:

Ms. Klara ORGOVANOVA, Programme Director, Open Society Fund–INFORAMA Bajkalska 25 Bratislava, SLOVAK REPUBLIC
Tel : + 421 754 414 730 ; Fax : +421 754 418 867 ; E-mail :

Mr. Emil SCUKA, apologised/excusé
President of International Romani Union (IRU)
Podstrasnickou vinici 6, 100 00 Praha 10, CZECH REPUBLIC
Tel ./ Fax : + 420 321 24 661 ; Mobile : +420 602 201 509

Mr. Nikolay KIRILOV, Roma Lom Foundation
71 Pirot Street Lom, BULGARIA
Tel /Fax : +359 971 28 751 - E-mail :

Ms. Carmen SANTIAGO REYES, apologised/excusée
Member of Council of Europe Specialist Group on Roma/Gypsies
C/Ronda de los Tejares N° 16-2, 4° dcha, 14001 Cordoba, SPAIN
Tel : + 34 957 47 01 93 - Fax : +34 957 48 39 03


Mr. Jordan TOTEV , Chief Expert, National Statistical Institute
2 P. Volov Str, 1504 Sofia, BULGARIA
Tel : + 359 2 9857 2861 – Fax : + 359 2 9857 2488 or 359 2 9857 2000
E-mail :

Mr Gabor ROZSA, Deputy Director Population Census/Hungarian Central Statistical Office/Chef adjoint du Dpt des recensements de la population, Bureau central de la statistique de Hongrie, HUNGARY
Tel.: 361 34 56 802 ; Fax: 361 34 56 803 ; e-mail:

Mr Gabriel GAMEZ, Deputy Director, Department of international relations and co-operation, Swiss Federal Statistical Office , SWITZERLAND
Tel: + 41-32-7136769; Fax: + 41-32-7136002; EMail:

Mrs. Carol MOTTET, senior officer and head of the co-operation program with Central and Eastern Europe, Department of international relations and co-operation, Swiss Federal Statistical Office , SWITZERLAND
Tel: + 41-32-7136009; Fax: + 41-32-7136002; EMail:


Mr. Gerard G. de BOER, Permanent Representation of the Netherlands to the Council of Europe /Représentation Permanente des Pays-Bas au Conseil de l’Europe - 3 place S. Brant - 67000 Strasbourg, FRANCE
Tel. : +33 3 88 36 20 48 ;Fax : +33 3 88 36 70 10 E-mail :

Mr. Nicolae GHEORGHE, Adviser on Roma and Sinti Issues, OSCE / Office for Democratic Institutions and Human Rights
Al. Ujazdowskie 19, 00-557 Warsaw, POLAND
Tel :+48 22 520 06 00/ 41 43 ;Fax :+ 48 22 520 0605 ;

Mr. James A. GOLDSTON, Deputy Director, Open Society Institute (OSI)
Oktober 6 utca 12, 1051 Budapest, HUNGARY
Tel : + 361 327 3100 - Fax : + 361 327 3101 : E-mail : 

M. Youssef COURBAGE ; Directeur de Recherches, Institut National d'Etudes Démographiques, 133 bld Davout 75020 Paris, FRANCE
Tél.: +33 1 56 06 21 47 .Fax: +33 1 56 06 21 93 E-mail:

Ms. Kveta KALIBOVA, Department of Demography and Geodemography, Charles University, Faculty of Science, Albertov 6, 128 43 Prague2, CZECH REPUBLIC
Tel : + 420 2 2195 2163 ; Fax : +420 2 249 20 657 ; E-mail :

United Nations High Commissioner for Refugees (UNHCR)
Mme Pascale MOREAU, Senior Liaison Officer to the European Institutions for the UNHCR/Administrateur Principal de Liaison auprès des Institutions Européennes à Strasbourg pour le Haut Commissariat des Nations Unies pour les Réfugiés, CEPI, Bureau 34, 67000 Strasbourg, FRANCE
Tel.: +33 3 88 41 28 69; Fax: +33 3 88 41 39 79; E-mail: or


Ms. Josephine VERSPAGET, Chair of the Specialist Group on Roma/Gypsies Nieuwe Boschstr., 4811 CX BREDA, NETHERLANDS
Tel.: +31 76 52 12 108 , Fax: +31 76 51 41 198 ; E-mail:

Mme Marie-Odile WIEDERKEHR, Deputy Director, Legal Affairs, Directorate General II
Tél.: +33 3 88 41 22 89; Fax: + 33 3 88 41 37 39; E-mail:

Mr. Franco MILLICH, Population Division, DG III Social Cohesion
Tel. +33 3 88 41 23 31 ; Fax : +33 3 88 41 27 31, E-mail :


Mr. Allen KASSOF, President, Project on Ethnic Relations, 15 Chambers Str. Princeton, NJ 08542, USA
Tel. : +1 609 683 5666 ; Fax : +1 609 683 5888 ; E-mail :

Ms. Livia PLAKS, Executive Director, Project on Ethnic Relations, 15 Chambers Str. Princeton, NJ 08542, USA
Tel. : +1 609 683 5666 ; Fax : +1 609 683 5888 ; E-mail :

Mr. Andrzej MIRGA, Chair (PERRAC ; PER Roma Advisory Council), Co-Chair Specialist Group on Roma/Gypsies (COE), Project on Ethnic Relations, 15 Chambers Str. Princeton, NJ 08542, USA
Tel. : +1 609 683 5666 ; Fax : +1 609 683 5888 ; E-mail :

Ms. Ann Marie GROCHOLSKI, Program Officer, Assistant to the Executive Director, PER, Project of Ethnic Relations, 15 Chambers Street, Princeton, NJ 08542-3718, USA
Tel : +1 609 683 5666 – Fax : + 1 609 683 5888


Mme Gabriella BATTAINI-DRAGONI, Director of Social Affairs and Health, Directrice des Affaires Sociales et de la Santé
Tel.: +33 3 88 41 21 55 ; Fax: +33 3 88 41 37 78

Mr. John MURRAY, Coordinator of Activities on Roma/Gypsies/Coordonnateur des Activités concernant les Roms/Tsiganes
Tel.: +33 3 88 41 21 67; Fax: +33 3 88 41 27 31; E-mail:

Ms. Maria OCHOA-LLIDO, Secretary of the MG-S-ROM /
Secrétaire du MG-S-ROM
Tel.: +33 3 88 41 21 79; Fax: +33 3 88 41 27 18; E-mail:

Ms. Françoise KEMPF, Administrator/Administratrice, Roma/Gypsy activities/
Activités roms/tsiganes
Tel.: +33 3 88 41 38 54; Fax: +33 3 88 41 27 31; E-mail: franç

Mrs Gilda FARRELL, Administrative Officer, Directorate of Social Affairs and Health, Direction des Affaires Sociales et de la Santé, Social Cohesion Specialised Unit
Tél.: +33 3 88 41 30 57 ; Fax: +33 3 88 41 ; E-mail:

Mr Simon TONELLI, Administrative Officer, Directorate of Social Affairs and Health, Direction des Affaires Sociales et de la Santé
Tél: +33 3 88 41 21 62 ; Fax: +33 3 88 41 27 18 ; e-mail:

Mr Robert DRAKE, Conseiller Technique Unité Emploi/ Technical Advisor Employment Unit
Tel: +33 3 99 41 48 20; Fax: +33 3 88 27 18 ; E-mail:

Mme Brigitte THOMAS, Assistant/Assistante, Roma/Gypsy Activities/Activités roms/tsiganes
Tel.: +33 3 88 41 21 66; Fax: +33 3 88 41 27 31; E-mail:

Ms. Elzbieta KUZMA, Trainee/Stagiaire,
Tel.: +33 3 90 21 47 47; Fax: +33 3 88 41 27 31; E-mail:



1. Council of Europe

You will find the principle relevant texts on data protection (Convention 108, treaties, recommendations, realisations) on web site on data protection

Data protection Unit

Web site

Mail address:
Data Protection Unit
Directorate General I (Legal Affairs)
Council of Europe
Avenue de l’Europe
F-67075 Strasbourg Cedex

Tel.: +33 3 88 41 2551
Fax: +33 3 88 41 27 64

Convention 108
Convention for the protection of individuals with regard to automatic processing of personal data (STE N°108, 28 January 1981. Entrée en vigueur le 1.10.1985).

2. Other international organisations texts :

European Union

Directive cadre 95/46/CE
To consult on the following address :

United Nations

1990 Guidelines

adopted by the General Assembly on 14 December 1990

The procedures for implementing regulations concerning computerized personal data files are left to the initiative of each State subject to the following orientations:

A. Principles concerning the minimum guarantees that should be provided in national legislations

Information about persons should not be collected or processed in unfair or unlawful ways, nor should it be used for ends contrary to the purposes and principles of the Charter of the United Nations.

Persons responsible for the compilation of files or those responsible for keeping them have an obligation to conduct regular checks on the accuracy and relevance of the data recorded and to ensure that they are kept as complete as possible in order to avoid errors of omission and that they are kept up to date regularly or when the information contained in a file is used, as long as they are being processed.

The purpose which a file is to serve and its utilization in terms of that purpose should be specified, legitimate and, when it is established, receive a certain amount of publicity or be brought to the attention of the person concerned, in order to make it possible subsequently to ensure that:
(a) All the personal data collected and recorded remain relevant and adequate to the purposes so specified;
(b) None of the said personal data is used or disclosed, except with the consent of the person concerned, for purposes incompatible with those specified;
(c) The period for which the personal data are kept does not exceed that which would enable the achievement of the purpose so specified.

Everyone who offers proof of identity has the right to know whether information concerning him is being processed and to obtain it in an intelligible form, without undue delay or expense, and to have appropriate rectifications or erasures made in the case of unlawful, unnecessary or inaccurate entries and, when it is being communicated, addressees. Provision should be made for a remedy, if need be with the supervisory authority specified in principle 8 below. The cost of any rectification shall be borne by the person responsible for the file. It is desirable that the provisions of this principle should apply to everyone, irrespective of nationality or place of residence.

Subject to cases of exceptions restrictively envisaged under principle 6, data likely to give rise to unlawful or arbitrary discrimination, including information on racial or ethnic origin, colour, sex life, political opinions, religious, philosophical and other beliefs as well as membership of an association or trade union, should not be compiled.

Departures from principles 1 to 4 may be authorized only if they are necessary to protect national security, public order, public health or morality, as well as, inter alia, the rights and freedoms of others, especially persons being persecuted (humanitarian clause) provided that such departures are expressly specified in a law or equivalent regulation promulgated in accordance with the internal legal system which expressly states their limits and sets forth appropriate safeguards.
Exceptions to principle 5 relating to the prohibition of discrimination, in addition to being subject to the same safeguards as those prescribed for exceptions to principles 1 and 4, may be authorized only within the limits prescribed by the International Bill of Human Rights and the other relevant instruments in the field of protection of human rights and the prevention of discrimination.

Appropriate measures should be taken to protect the files against both natural dangers, such as accidental loss or destruction and human dangers, such as unauthorized access, fraudulent misuse of data or contamination by computer viruses.

The law of every country shall designate the authority which, in accordance with its domestic legal system, is to be responsible for supervising observance of the principles set forth above. This authority shall offer guarantees of impartiality, independence vis-a-vis persons or agencies responsible for processing and establishing data, and technical competence. In the event of violation of the provisions of the national law implementing the aforementioned principles, criminal or other penalties should be envisaged together with the appropriate individual remedies.

When the legislation of two or more countries concerned by a transborder data flow offers comparable safeguards for the protection of privacy, information should be able to circulate as freely as inside each of the territories concerned. If there are no reciprocal safeguards, limitations on such circulation may not be imposed unduly and only in so far as the protection of privacy demands.

The present principles should be made applicable, in the first instance, to all public and private computerized files as well as, by means of optional extension and subject to appropriate adjustments, to manual files. Special provision, also optional, might be made to extend all or part of the principles to files on legal persons particularly when they contain some information on individuals.

B. Application of the guidelines to personal data files kept by governmental international organizations
The present guidelines should apply to personal data files kept by governmental international organizations, subject to any adjustments required to take account of any differences that might exist between files for internal purposes such as those that concern personnel management and files for external purposes concerning third parties having relations with the organization.
Each organization should designate the authority statutorily competent to supervise the observance of these guidelines.
Humanitarian clause: a derogation from these principles may be specifically provided for when the purpose of the file is the protection of human rights and fundamental freedoms of the individual concerned or humanitarian assistance.
A similar derogation should be provided in national legislation for governmental international organizations whose headquarters agreement does not preclude the implementation of the said national legislation as well as for non-governmental international organizations to which this law is applicable.

1 . Article 5 - Quality of data
Personal data undergoing automatic processing shall be:
a. obtained and processed fairly and lawfully;
b. stored for specified and legitimate purposes and not used in a way incompatible with those purposes;
c. adequate, relevant and not excessive in relation to the purposes for which they are stored;
d. accurate and, where necessary, kept up to date;
e. preserved in a form which permits identification of the data subjects for no longer than is required for the purpose for which those data are stored.

2 . Article 6 - Special categories of data
Personal data revealing racial origin, political opinions or religious or other beliefs, as well as personal data concerning health or sexual life, may not be processed automatically unless domestic law provides appropriate safeguards. The same shall apply to personal data relating to criminal convictions.

3 See in particular the publication: The demographic characteristics of national minorities in certain European states - Volume 2 (Population Studies No. 31), by Werner Haug and Youssef Courbage, Council of Europe Publishing, 2000.

4 Doc. MG-S-ROM (2000) 5: Problems connected with the international mobility of Roma in Europe and the recent emigration of Roma from the Czech Republic and the Slovak Republic, Yaron Matras, (1996 et 1998), adopted by the CDMG at its 41st meeting, 1-3 December 1999.