Accounting for Ethnic and Racial Diversity

The Challenge of Enumeration

Patrick Simon, Victor Piché

• Publisher: Routledge
• Year: 2013
• Print ISBN: 9781138676381
• eBook ISBN: 9781317981077
• Topic: Social Sciences
• Subtopic: Sociology
• Index: Social Sciences

By the end of the 20th century, the ethnic question had resurfaced in public debate. Every country had been affected by what is commonly known as cultural pluralism, as a result of conflicts interpreted from an ethnic perspective, for instance, in the Balkans and central Africa; nationalist struggles, such as the Basque country, Quebec and Belgium; and demands for recognition and political representation by new ethnic minorities.

This resurgence or extension of the salience of ethnicity in most of the societies around the world can now be found not only in public discourse, policy making, scientific literature and popular representation, but also in the pivotal realm of statistics. This volume explores the ethnic and racial classification in official statistics as a reflection of the representations of population, and as an interpretation of social dynamics through a different lens. Spanning all continents, a wide range of international authors discuss how ethnic and racial classifications are built, their (lack of) accuracy and their contribution to the representation of ethnic and racial diversity of multicultural societies.

This book was originally published as a special issue of Ethnic and Racial Studies.

Accounting for ethnic and racial diversity: the challenge of enumeration

Patrick Simon and Victor Piché

‘The problem of the 20th century is the problem of the color-line’, claimed W. E. B. Du Bois in his famous statement in his 1903 masterpiece, The Souls of Black Folk. This prediction has been reproduced and updated for the twenty-first century in countless books and articles which open their introductions with this quote. We obviously cannot derogate from this now well established tradition. Colour, racial and/or ethnic diversities have reached an unprecedented level in the contemporary world. Migrations throughout the last centuries, from or to Europe, and then from the former colonial empires and peripheries to the former mainlands or centres, have reactivated classifications that were used before the disqualification of scientific racism. However, if the colour-line is the problem of the twenty-first century, what are colour, race, and ethnicity? This broad and fundamental question has received so many answers and disclaimers that we won’t do any justice to it in this introduction. Our purpose in this special issue is to focus on one narrow, still crucial, issue of the conundrum of ‘doing race’ (Markus and Moya 2010): the production of ethnic and racial statistics. Group boundaries, as measured in population censuses and social statistics, vary from country to country and very few definitions are truly universal or uniform: local social conventions, political traditions, legal prohibitions and political mobilization impose different group norms and definitions. These categories are not just scientific categories, invented for the sole purpose of measuring demographic changes or taking stock of ‘true and authentic’ identities. They are socially constructed, influenced by existing (and shifting) power relationships, national images and stereotypes, legal procedures and historical paths. The statistical representation of diversity is a complex process which reveals the foundations of societies and their political choices. Thus there is a gap between the apparent ethnic and racial diversity in most countries in the world and the way these societies perceive themselves and are portrayed in official and social statistics. We are categorically diverse, sometimes visibly diverse, and in some circumstances classified in categories which account for this diversity.

This special issue of the journal aims to address major epistemological questions through basic methodological questions. The articles are partly a selection of papers given at the 2007 Social Statistics and Ethnic Diversity conference: ‘Should we count, how should we count and why?’¹ and invited contributions. They cover three different though nested issues related to the broad topic of ethnic and racial statistics: Should we count? How to count? Why count?

Should we count?

Collecting ethnic and racial data is a contentious issue in a large number of countries around the world. The answer to the question ‘Should we count?’ involves two dimensions: an ethical issue and a pragmatic policy-making one. For decades, ethnic and racial classifications have been conceived and used to segregate, build hierarchies and nurture racial and ethnic stratifications and inequalities. The rationale behind collecting ethnic and racial data has dramatically changed; the purpose and meanings of data have shifted, even when the data look the same (Hacking 2005). Societies where such data have been collected for decades no longer have to justify their processes; those who do not have this tradition are facing the challenge of explaining to the general public the rationale behind such statistics. Does distinguishing and characterizing populations according to their ethnic origins constitute a risk of stigmatization or is it, on the contrary, an asset for measuring and explaining discrimination and for demanding more inclusive policies?

The use of ethnic categories is not without problems, and a growing literature discusses the issue of the social and political significance of such categories (Aspinall 2007). Socially constructed, culturally shaped, biologically determined, and genetically designed: the definitions of race and ethnicity as concepts and categories are far from being stable and shared among scientists, policy makers, public opinion and statisticians (Wimmer 2008; Brady and Kaplan 2009; Morning 2011). The linkages between political framings and the statistical categories that support them can be observed in every society (Nobles 2000). Population statistics not only aim at producing knowledge of demographic dynamics, they provide a benchmark for policies and contribute to the (re)production of national identity (Alonso and Starr 1987; Anderson 1991; Desrosières 1993). Official (and scientific) statistical categorizations both reflect and affect the structural divisions of societies, as well as mainstream social representations. As conventions, they offer arbitrary definitions of the social objects they are intended to describe, but these definitions ensue from historical, social and political processes of negotiation between public authorities and social forces. In this respect, censuses are a strategic place in which views on race and ethnicity are confronted by official statistics (Omi 1997). In this sense, censuses do more than reflect social realities; they also participate in the construction of these realities (Kertzer and Arel 2002).

The dissemination of ethnic and racial categories in public debate, policies, scientific research and statistics has sometimes been understood as one of the consequences of American imperialism. As Bourdieu and Wacquant suggest in their criticism of the hegemony of the conceptual toolkit of ‘race’, forged, they say, in the US: ‘the recent as well as unexpected discovery of the “globalization of race” results, not from a sudden convergence of forms of ethnoracial domination in the various countries, but from the quasi-universalization of the US folk-concept of “race” as a result of the worldwide export of US scholarly categories’ (Bourdieu and Wacquant 1999, p. 48). This argument had already been developed by Wacquant in his paper entitled ‘For an analytic of racial discrimination’, published in 1997. The key issue, as pointed to by some of the very critical responses to Bourdieu and Wacquant’s essay published later on,² is whether the expansion of ethnic and racial studies in the social science arena in most countries of the world is a pure artefact determined by the domination of the US academic field and institutions over the rest of the world, or whether it reflects the rise of similar social realities in multicultural countries.³

How to count?

Methodological choices to capture ethnicity or race in censuses, surveys and polls are just as important as the initial decision to count. To a certain extent, it is impossible to disentangle the ethic of data collection and the methodology of census or survey taking (Burton, Nandi and Platt 2010). In fact, there are so many ways to define ethnicity or race that finding the best and most common practices in this area, taking into account the historical, socioeconomic and political specificities of each society, is a challenge (Brady and Kaplan 2009). Recent examples of such difficulty in aligning the intention to capture ethnicity and the result obtained are to be found in the Americas, specifically in Brazil and the United States, and in the UK, where the increasing mix of population across ethnic and racial groups tends to blur the boundaries (Telles 2002; Perez and Hirschman 2009; Roth 2010; Loveman et al., this issue; Thompson, this issue). There is a growing criticism of the way official statistics do portray ethnicity and race. Beyond the issue of the legitimacy of collecting such data, the ‘realism’ and accuracy of these data to map race and ethnicity as constructed categories is under scrutiny.

Why count?

Although components of the discussion on social statistics and ethnic diversity sometimes overlap, it is possible to explore the many questions as to why it is positive to count. What are the characteristics of legal, social and economic integration? Can ethnic statistics play a role in supporting social change, in developing countries as well as in older democracies? How do we identify the existence and magnitude of discrimination, implement equality policies and assess their impacts? These questions all require innovative data sources (e.g. administrative data and longitudinal surveys), whose future orientation needs careful consideration (Wrench 2011).

Ethnic or racial data are (almost) no longer collected to preserve racist social systems, or to maintain unacceptable hierarchies among social groups, but to describe objective and subjective group realities in order to facilitate the enforcement of generally progressive social programmes (Simon 2005). Measuring the extent and nature of the diverse forms of discrimination is essential to the formulation, monitoring and evaluation of anti-discrimination policies, be they national, regional or global. Compensation for past and current discrimination, claims for group rights, recognition of minority status, and praise of cultural diversity have often led to the development of new policies, either multicultural policies or positive action measures (Grigolo, Hermanin and Moschel 2011). Indeed, among the new challenges to be faced by national governments, international organizations, and civil society, is a rising political and social demand for evidence-based policies. These measures, in order to be effective, require precise definitions of group boundaries, hence the proliferation of ethnic and racial categories in numerous population censuses, or the use of data on national origin, language spoken at home and religious affiliation.

The contributions included here are all part of the recent revisiting of mainstream assumptions about statistical categories being objective and straightforward. As mentioned above, the use of ethnic categories is not without problems, and several works discuss the issue of the social and political significance of such categories (see for instance Anderson and Fienberg 1999; Zuberi 2001; Rallu, Piché and Simon 2006). This special issue of Ethnic and Racial Studies dwells on one main idea, that there is a clear connection between social change, politics and the production of statistical categories. In this respect, social changes will have important impacts on the evolution of official categories. However, this connection is not unilateral; on the contrary, there are often competing paradigms, and policy outcomes are the result of contestation and compromise.

Census categories and competing frames of reference

While diversity is an objective fact in a large number of countries, the question remains as to how to account for it. Patrick Simon’s article argues that in the European context, the political responses to diversity have been far from homogeneous. Over and above the historical context of widespread opposition to collecting ethnic data on an official basis (almost half of the European countries reviewed do not produce ethnic data), given recent EU anti-discrimination legislation, there is as yet no consensus on the best ways to collect such data. One reason for this is related to competing models of diversity. The ‘assimilationist’ perspective claims that ethnic data should be collected and analysed to show complete assimilation, while the ‘multiculturalist’ perspective focuses on ethnic retention and self-declared ethnicity. On the other hand, the ‘anti-discrimination’ paradigm reflects a shift in the use of ethnic and racial data geared towards the monitoring and evaluation of public policies. Simon’s review of collecting ethnic statistics in Europe concludes that the current regime does not meet the standards required for effective positive action.

The Brazilian case, analysed here by Mara Loveman, Stanley Bailey and Jeronnimo Muniz, is particularly interesting because it casts competing racial classifications in a historical context and shows the empirical impact of opting for one classification over another. Historically, racial categories in Brazil used a three-fold classification system: white, black and mixed (or brown). This three-fold classification is increasingly criticized as an ideological myth of fluid racial order, as not reflecting the actual composition of the Brazilian population and as impeding the consolidation of a unified black identity. As a result, there is a momentum towards a binary approach by activists and scientists. An empirical examination of three scenarios (A =self-classification in three official census categories, B =combination of self-classified black and brown from A into a single nonwhite category, and C =self-classification in a forced binary format) shows that each classification yields very different results with respect to racial composition and inequality. For example, with scenario A, Brazil has a bare majority white population, a large mixed population and a small black population; with scenario B, Brazil becomes 50/50 white and nonwhite; and with scenario C, Brazil is a largely white country with a minority black population.

One of the most unexpected outcomes of recent competing agendas in the field of census categories is the active involvement of ethnic majorities in identity claims. Tahu Kukutai and Robert Didham discuss the significance of the new category ‘New Zealander’ which appeared in the 2006 census. What is significant about the category is that it refers to the nation rather than a particular group. The New Zealander campaign ‘Declare your pride’, and the equivalent ‘Count-me Canadian’ campaign, point to the fact that dominant groups perceive that their interests are not reflected by official practices of ethnic counting. Hence, in line with the competing claims approach, this group response in the census might serve as a denial of the importance of ethnicity and to counter the perceived threat posed by ethnic pluralism and minority group rights. It is interesting to note that this colour-blind construction, usually reserved for the exemplary French ‘republicanism’, is gaining weight in traditional immigration countries. According to Debra Thompson’s analysis (see next paragraph), this is in line with Republican Party elites favouring the ideal of having one box on all Federal forms that simply reads ‘American’.

Census politics in the United States, Great Britain and Canada with respect to the state’s role in the decision to count mixed-race categories, as presented by Debra Thompson, again clearly illustrate the connection between social changes and categorization. While the multiracial movement played a significant role, it is the three countries’ shift towards a paradigm of multiracial multiculturalism that made the difference in acknowledging mixed-race categories as a positive attribute of the multicultural nation. In brief, this is a clear example of the causal sequence between the emergence of ethnic and racial diversity, political commitments to the recognition of a multicultural society and ensuing legislative changes. The census becomes an instrument of diversity governance. But again, the connection is neither automatic nor unilateral, due to competing claims. As the article shows, there were long gaps between the adoption of the Race Relations Act in Great Britain (fifteen years) or the Employment Equity Act in Canada (ten years) and the decision to count by race. Hence, legislation is necessary but insufficient to bring about changes in data collection and production.

Hungary is another interesting case-study highlighting the effects of competing frames, as the article by Andrea Krizsan shows. Indeed, the demand for ethnic data is high in this country, but it fails to produce reliable and consistent ethnic data. According to Krizsan, three ideological frames compete for the definition of the right type of data to be collected: (i) the group self-determination perspective claims the need for ‘objective’ data to identify ‘historical’ minorities, thus excluding recent immigrant groups; (ii) the individual rights frame is more geared towards data protection; and (iii) the social inclusion model argues for ethnic data as monitoring tools. The first approach has clearly gained a dominant position in Hungary. Competing claims thus produce inconsistencies and result in the failure to adapt ethnic data to the new realities of diversity.

The South African context, presented by Tom Moultrie and Rob Darrington, remains a puzzling case. Colonial and apartheid-driven racial classifications were the result of a top-down approach, the state imposing its own racialized ideas on the enumeration process. Furthermore, census categories were fraught with great imprecision because there was a presumption that the terms employed in legislation were ‘common-sensical and self-evident’. (Aliya Saperstein below argues the same point for the United States, referring to common-sense belief that everyone knew what race was.) While it could have been expected that post-apartheid governments would have radically changed the nature of racial data, no such thing happened. Th...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Citation Information
• Notes on Contributors
• 1. Accounting for ethnic and racial diversity: the challenge of enumeration
• 2. Collecting ethnic statistics in Europe: a review
• 3. Group self-determination, individual rights, or social inclusion? Competing frames for ethnic counting in Hungary
• 4. Making (mixed-)race: census politics and the emergence of multiracial multiculturalism in the United States, Great Britain and Canada
• 5. Re-making the majority? Ethnic New Zealanders in the 2006 census
• 6. Used for ill; used for good: a century of collecting data on race in South Africa
• 7. Brazil in black and white? Race categories, the census, and the study of inequality
• 8. Capturing complexity in the United States: which aspects of race matter and when?
• Index