The Dark Side of Numbers: The Role of Population Data Systems in Human Rights Abuses

Social Research, Summer, 2001 by William Seltzer, Margo Anderson

Potential Safeguards: An Introduction

A variety of safeguards are available that may help to deter the use of population data systems in assisting in the planning or carrying out of major human rights abuses. While in most circumstances few of these safeguards are absolute, they each can help to discourage a contemplated misuse by raising the cost of such a misuse, either in financial or political terms. Moreover, even if one or more safeguards successfully discourages the use of a population data system in assisting in a human rights abuse, the underlying human rights abuse may still take place. However, without the assistance of the data sought, the efficiency of the perpetrators is likely to be reduced. As a result, lives may be saved and the extent and duration of other harms reduced. (On the other hand, most, but not all, safeguards have the unfortunate side effect of also reducing the analytical usefulness of the resulting data.) Five different safeguards may be distinguished.

Substantive Safeguards

The ultimate safeguard is not to gather or save data that permits associating an individual with a potentially vulnerable group. This safeguard, while often perceived as reducing the analytical or policy usefulness of the involved data system, has been deliberately employed in several countries that had histories of misuses associated with major abuses. For example, shortly after World War II, the Dutch authorities removed the item on religion from both the census and the population registration system (Berger, 1998). In France, "it is forbidden to stock in computer memory. .. any personal data relating directly or indirectly to the racial origins, political, philosophical and religious opinions, and trade union membership of individuals" (Article 31 of the Law of 6 January 1978, quoted by Leridon, 1999: 189).

Two major objections to this approach are often advanced. First, it is not always possible to know in advance which variables will turn out to be of relevance to potential perpetrators of human rights abuses. Although the topics that have usually been used to define target populations in national data systems come from a well-defined list (race, ethnicity, religion, country of birth or ancestry, mother tongue), in special circumstances victims have sometimes been defined in very broad terms (for example, urban residents, those with advanced education, or even the literate population). Second, and of more practical relevance, the elimination of variables may impose a considerable price in terms of a reduction of the descriptive and analytical usefulness of the data system. For example, in recent years, both in the United States and the United Kingdom, advocates for some but not all potentially vulnerable groups have been among the strongest supporters of continued or more detailed coverage. Such detailed identification is seen as important element in providing the statistical basis for various programs enacted to redress past discrimination as well as recognizing the groups' presence within the overall body politic. Since such data have also been used to abuse vulnerable target populations either in the official statistical reports of individual countries or in studies by other users, the trade-offs involved deserve to be further examined.