some current political issue usually unrelated to the goal of improving the socio-economic status of minorities. belong to a particular minority. ‘The individual’s subjective choice is inseparably linked to objective criteria relevant to the person’s identity’.47 Obstacles to effective monitoring Equally problematic is external identification. The State may not impose an identity on individuals so it is not acceptable to use the perception of the interviewer as the sole means of identifying different individuals’ membership of a group. Practically, this method would also be subject to the prejudices of the interviewer and therefore likely to be inaccurate. Resolving these ambiguities about self-identification requires confidence and trust building efforts by the government and minority NGOs. The principle of self-identification There is widespread agreement that data on ethnicity or religion are necessary for the design and implementation of effective policies to combat discrimination. At the same time, under international law, no one can be compelled to reveal certain kinds of sensitive information, including data on ethnic origin and religion. This standard is sometimes misinterpreted as prohibiting any collection of data on ethnicity. In fact, international law supports the principle of selfidentification, leaving the individual to choose with which ethnic, religious or linguistic group(s), if any, to identify. Further, the Committee on the Elimination of Racial Discrimination (CERD) stated in General Recommendation VIII that the way individuals be identified as belonging to ethnic groups will ‘if no justification exists to the contrary, be based upon self-identification by the individual concerned’. Although the principle of self-identification is useful for resolving legal-ethical dilemmas concerning the collection of data on ethnicity in general, this principle alone is not sufficient to ensure meaningful data on minority groups. In the case of many minorities, a deep-seated resistance to declare their ethnic or religious identity is rooted in lived experiences of abuse of personal data. On the other hand, where programmes are established for particular groups, such as a programme to assist members of minorities to obtain jobs, individuals who do not meet any of the objective criteria for membership of a particular ethnic group (culture, ethnicity, religion, language), may attempt to self-identify with that group in order to benefit from the programme; however, there is no right to arbitrarily choose to Fear, stigmatization and confounded identities Fear of the consequences of ethnic data collection is pervasive. The fears of minorities and governments differ. Government fears include concerns that data showing large inequalities between groups will cause conflict or exacerbate historical conflicts between groups. Minorities’ fear may include distrust towards government claims that data intended for beneficial use instead will create more discrimination and stigmatization. Another aspect of under-reporting of minorities is related to the multiple identities minorities might have. Experience shows that ethnicity is often confounded with civic, confessional, and linguistic identities. Underestimation and overestimation Taken together, the various pitfalls associated with measuring the size of ethnic populations combine to yield considerable discrepancies between official and unofficial numbers, with the official figures often considerably lower than the number of persons who identify themselves as ethnic minority in daily life. Among the problems associated with underestimating the size of a given country’s minority population are overestimation on socially sensitive indicators such as birth-rate, unemployment, and criminality. Council of Europe, Framework Convention for the Protection of National Minorities (FCNM), Article 3.1 and Explanatory Report, H(1995)010, paragraph 35. 47 130 M A R G I N A L I S E D M I N O R I T I E S I N D E V E LO P M E N T P R O G R A M M I N g