Everlasting Countdowns : Race, Ethnicity and National Censuses in Latin American States 1443841498, 9781443841498

Citation preview

Everlasting Countdowns

Everlasting Countdowns: Race, Ethnicity and National Censuses in Latin American States

Edited by

Luis Fernando Angosto Ferrández and Sabine Kradolfer

Everlasting Countdowns: Race, Ethnicity and National Censuses in Latin American States, Edited by Luis Fernando Angosto Ferrández and Sabine Kradolfer This book first published 2012 Cambridge Scholars Publishing 12 Back Chapman Street, Newcastle upon Tyne, NE6 2XX, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2012 by Luis Fernando Angosto Ferrández and Sabine Kradolfer and contributors All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-4438-4149-8, ISBN (13): 978-1-4438-4149-8

TABLE OF CONTENTS

List of Figures............................................................................................ vii List of Tables............................................................................................ viii Acknowledgments ....................................................................................... x Chapter One: Race, Ethnicity and National Censuses in Latin American States: Comparative Perspectives.......................................... 1 Luis Fernando Angosto Ferrández and Sabine Kradolfer Chapter Two: Are there Still “Indians” in Argentina? Indigenous Peoples and the 2001 and 2010 Population Censuses.......................... 41 Pilar Barrientos Chapter Three: Bolivia: Indigenous Identities and Collective Subjects in the Andes........................................................................... 69 Pablo A. Regalsky Chapter Four: Who Counts Indigenous People, How are they Counted and What For? Census Policies and the Construction of Indigeneity in Colombia .................................................................. 94 Gloria Patricia Lopera Mesa Chapter Five: The Convergence of the “Indigenous” and “Ladino” Categories in the Guatemalan Census of 2002................................... 128 Gemma Celigueta Comerma Chapter Six: The Social and Political Construction of Racial and Ethnic Categories in National Censuses of Panama, 1911-2010 .......................................................................................... 155 Mònica Martínez Mauri

vi

Table of Contents

Chapter Seven: From Pre-Modern “Indians” to Contemporary “Indigenous People”: Race and Ethnicity in Peruvian Censuses 1827-2007 .......................................................................... 185 David Sulmont and Néstor Valdivia Chapter Eight: National Censuses and Indigeneity in Venezuela............ 221 Luis Fernando Angosto Ferrández Chapter Nine: Ethnic/Racial Statistics: Brazil and an Overview of the Americas .................................................................................. 264 José Luis Petruccelli Chapter Ten: Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting ............................................................ 304 Fabiana Del Popolo and Susana Schkolnik Contributors............................................................................................. 335

LIST OF FIGURES

Fig. 2-1. Argentina: Census Questionnaire. 2001 National Population, Household and Housing Census. Question No. 2. “Self-recognition” as a descendant or member of an indigenous people ........................... 48 Fig. 2-2. Questionnaire, Supplementary Survey of Indigenous Peoples of the Argentine Republic (2004-2005). Set of questions on the land tenure system..................................................................... 57 Fig. 2-3. Argentina: Questionnaire, 2010 National Population, Household and Housing Census. Question No. 5. “Identification of membership of or descent from an indigenous people”............................................ 64 Fig. 3-1. Bolivia: highlands and lowlands ................................................. 77 Fig. 4-1. Colombia: indigenous macro-regions according to ONIC’s organisational structure ........................................................................ 96 Fig. 4-2. Indigenous territories in Colombia and population density ........ 97 Fig. 5-1. Guatemala: percentage of indigenous population by department .................................................................................... 129 Fig. 6-1. Indigenous peoples of Panama (%) 2010.................................. 163 Fig. 7-1. Peru 1876-2007: total population (in millions) by census year and estimates of indigenous population ............................................. 198 Fig. 7-2. Peru ENAHO 2009: % of head of households identified as indigenous, by type of indicator (self-identification or maternal language), and age groups.................................................................. 204

LIST OF TABLES

Table 2-1. Argentina: population according to whether they belong to and/or are first generation descendants of the Kolla people. Jujuy and Salta. 2004-2005.................................................................. 62 Table 3-1. Bolivia: population self-identified as indigenous in the 2001 Census.................................................................................................. 71 Table 3-2. Native speaker population in Bolivia’s Andean departments... 76 Table 3-3. Bolivia: distribution of agricultural land in 1950 (before the Agrarian Reform) .............................................................. 86 Table 3-4. Bolivia: land property distribution according to the 1984 Agricultural Census ............................................................................. 87 Table 3-5. Bolivia: land distribution according to type of land title (Agrarian Reform 1953-1993) ............................................................. 87 Table 4-1. Colombia: population census of the Viceroyalty of New Granada, 1778.................................................................................... 100 Table 4-2. Colombia: population censuses in the nineteenth century...... 104 Table 4-3. Population censuses in Colombia and identification criteria for ethnic groups (twentieth and twenty-first centuries) .................... 106 Table 5-1. Changes in the relative size of the indigenous, non-indigenous, ladino, Maya, Xinka and Garifuna populations of Guatemala........... 129 Table 5-2. Guatemala: total population, ethnic group, ethnic belonging and language in which the respondent learned to speak, by department, according to the 2002 census ............................................................. 130 Table 6-1. Number of indigenous and Afro-descendants in the Republic of Panama .......................................................................................... 163 Table 6-2. Panama’s indigenous comarcas: Legislation, size and population 2000-2010 .......................................................................................... 165 Table 6-3. Change in Panama’s indigenous population according to censuses, 1911-2010 ...................................................................... 172 Table 7-1. Peru 1876 - 2007: Ethnic and racial indicators used in national censuses and estimations of indigenous population ........................... 202 Table 7-2. Peru 1876-2007: Dimensions and questions used in censuses and surveys to measure ethnicity and race ......................................... 209 Table 8-1. The indigenous population in Venezuelan national censuses... 223 Table 8-2. Venezuela: 2001 census results for denominations not included in the list of indigenous peoples backed by the LOPCI ..................... 249

Everlasting Countdowns

ix

Table 8-3. Indigenous peoples registered in Venezuela according to different official census and other sources..................................... 255 Table 9-1. Resident population by colour/race, Brazilian censuses 1872-2000 .......................................................................................... 272 Table 9-2. Distribution of resident population by colour/race, Brazilian censuses 1872-2000 ........................................................................... 273 Table 9-3. Brazil: breakdown for the spontaneous question on colour/race in the 1998 employment surveys and 1976 household surveys ......... 276 Table 9-4. Comparative table of ethnic/racial classification questions.... 292 Table 10-1. Latin America: identification criteria for Afro-descendant and indigenous populations in the 1980-2000 censuses..................... 316 Table 10-2. Latin America: questions associated with the criterion of self-identification in the censuses of 2000..................................... 319

ACKNOWLEDGMENTS

We should like to thank all the contributors who agreed to accompany us in this project. Their expertise and close co-operation are the fundamental pillars upon which this volume has been constructed. However, the views expressed in the introductory chapter and any errors that it may contain are our sole responsibility. We also want to acknowledge the work of the translators who brought into English some chapters initially written in Spanish: Marion Marshrons translated the chapters by Barrientos, Celigueta and Petrucelli, and Brett Todd translated Del Popolo and Schkolnik’s. Brett also worked on the proofreading of some of the chapters, which were also proofread by Richard Nice; both of them provided very valuable idiomatic advice. The translations and proofreading were made possible thanks to funding from the Office of Equal Opportunities of the University of Lausanne (Switzerland) and from the School of Social and Political Sciences of the University of Sydney (Australia).

CHAPTER ONE RACE, ETHNICITY AND NATIONAL CENSUSES IN LATIN AMERICAN STATES: COMPARATIVE PERSPECTIVES LUIS FERNANDO ANGOSTO FERRÁNDEZ AND SABINE KRADOLFER

Politics, not demographics, is at the core of this book on censuses. This caveat will hardly surprise social scientists, always on their guard before clichés about the politically neutral functioning of statistics bureaus, but may however be reassuring for the general reader who opens this type of book all too ready to drop it as soon as the first numbers are within sight. No statistical or demographic expertise is required to engage debate around social categories in censuses, and one of the goals pursued by this work is in fact grounded on that basic premise: we should like to increase the existing degree of public awareness, scrutiny and discussion around national population counts. Contributors to this volume are removing the fig-leaves from censuses by historicising and contextualising a type of statistical practice that has become essential for the functioning (and understanding) of the contemporary state. Through censuses, states both obtain information from and give shape to “national” populations upon which government will be exercised. Nowadays these populations are characteristically diverse, and racial and ethnic identification has become one of the means through which censuses officially classify that diversity in discrete units. The contentious and complex history (and present) of this type of categorisation is a telling testimony to the political load embedded in census-making, and also an excellent platform to tackle ongoing debates on statecraft, cultural diversity and democracy in contemporary Latin America. By focusing on the contested and contingent realm of census-

2

Chapter One

making, we aim to contribute to a more fertile understanding of these contemporary debates. Ethnicity and race have re-emerged as pivotal social categories in this continent. There is a growing influence of social movements, political parties and civil society organisations that, with differing strategies and outcomes, appeal to these categories as sources of identification, mobilisation and rights claims. In parallel, some governments are drawing on narratives of ethnicity, and particularly on notions of indigeneity, in order to reshape national identities and create new legitimacies for postneoliberal political models. This amalgam of social processes “from below” and “from above” powerfully influences today’s Latin American politics and, though slippery in analytical terms, it feeds into the realm of census-making. A decade ago, seventeen out of nineteen Latin American states were incorporating categories to identify ethnic or racial belonging in their first census round of the century (Del Popolo et al., 2009; see Petruccelli in this volume for a list of census questions and race and ethnic categories in that round). A new census round is now taking place and, yet again, one of the aspects that attracts most expert and general attention concerns the inclusion and operational shaping of those categories. Both socialistleaning and liberal-leaning governments are currently including ethnic and/or racial categories in their national censuses, in a factual demonstration that the tendency towards developing a degree of sensitivity for diversity in population counts can be regarded as a consolidating ideological trend in the continent. This fact presents great interest for comparative studies, and this in turn relates to another of the volume’s goals: we intend to provide information and grounded discussion for future decision-making in this transcendent field of contemporary politics. Race and ethnicity have a consolidated presence in Latin American censuses for reasons that, to a considerable extent, are associated with current global politics, but which are nonetheless strongly conditioned by the historical and sociological peculiarities of the region. In the broad picture, national censuses are recognised world-wide as transcendent (and laborious and costly) administrative events. They are essential instruments of state governance, a means to “govern by numbers” (Desrosières, 2008b). They provide data for the definition of public agendas and for the subsequent design and implementation of government policy. But, in addition, they are spaces of contested social representations and considered essential in the struggles for group recognition. Censuses thus relate to the whole continuum of politics that oscillates between redistribution and “recognition”1. Against this international background,

Race, Ethnicity and National Censuses in Latin American States

3

and with the ideology of multiculturalism so closely intertwined with both the re-legitimisation of Latin American democracies and in general with neoliberal governance (Assies et al., 2000; Hale, 2002; Van Cott, 2000), it is unsurprising that at present these national counts pay distinctive attention to identity groups identified through ethnic or racial criteria. In addition to these global conditionings, regional idiosyncrasies certainly modulate interests in and approaches to census-making, and Latin American actors have elaborated their own set of arguments in the justification of the use of race and ethnic categories. Latin America harbours the world’s most unequal wealth distribution and, given the central role censuses can play in portraying and tackling that issue, these national counts have understandably become the target of increasing political and technical scrutiny. “Ethnicity” and “race” are social categories that have been related to unequal distribution of both wealth and power and to the different degrees to which a person (as a member of a particular identity group) can enjoy citizen rights (Byrne, 1999; Lovell and Wood, 1998; Valle Silva, 2000). In addition, the demands and expectations projected upon censuses in this continent are distinctively strong for other political reasons: powerful emergent narratives of democracy and citizenship are addressing social groups as much as individuals, and human rights debates are as focused on the particularities of cultural diversity as on universal concerns. Two pursuits thus become intertwined in cultural-diversity sensitive population counts: first, since they produce numerical data for social indicators in areas such as education, health or income, censuses are expected to contribute to tackling social inequality through a more effective orientation of the state’s “politics of redistribution”. On the other hand they are considered to be tactical for a positive re-valorisation of already existing collective identities which have previously been subordinated and pervaded by a negative “minority” status within national constructs. Officially recognised as politically and economically significant, and in harmony with the substantially different but on this point convergent ideologies of multiculturalism and plurinationalism that have been shaping Latin American constitutionalism in the past two decades, race and ethnic categories are gaining presence in administrative records. Census designs, their implementation programmes and the ideological conceptualisations that articulate them have accordingly been in a continuous process of remodelling. Studying these changes is a way to identify ideological affinities and cleavages in national and regional politics of governmentality, for censuses are administrative tasks in which the state is obliged to transform ideology into practice. In this practice states project

4

Chapter One

dominant views on what constitutes the “national body” and, to a large extent, condition the policies that can be developed in regard to that body. Critiques of the inclusion of racial and ethnic categories in Latin American censuses are rare, in general terms, or weakly articulated–Brazil being the most salient exception. On the contrary, there is a predominant concern with improving the quality of censuses by creating mechanisms that portray the population’s cultural diversity with more reliability and contrast it with other social diversities (such as economic ones). However, the development of case studies in the continent is very uneven, with countries such as Brazil, and to a lesser extent Colombia and Chile, concentrating a large share of the existing analytical efforts. We hope that the chapters included in this volume, which provide insights into state-ofthe-art census research in the continent, will stimulate further debates in countries where this research is still only incipient.

1. So what is political about censuses? Ideology pervades the census approach to race and ethnic identities. Ideology determines in the first place whether or not ethnic and racial categories will be accepted as legitimate and realistic divisions of the “national body” that the census represents. Then, if those categories are accepted, ideology, articulated in this stage with technical considerations, continues to influence the national population image through the specific shaping given to those categories in operational definitions that are far from being universal: they do vary and can do so substantially. All this ideology-pervaded process impacts on society in practical terms, both through the ways in which it pre-informs public policy and through a practical act of administrative recognition that implies the legitimisation, consolidation or even the creation of social categories that may condition citizen interaction. Census bureaus tend to present themselves as “fact-finder” technocratic bodies, thus setting boundaries to external political influences and aiming to maintain stable institutional environments. However, research on censuses has been consistently showing how rusty that armour of bureaucratic neutrality is, and a “loss of innocence” story has gained ground, emphatically stressing the political nature of these statistical undertakings. Though technical expertise at the bureaus remains centrally accredited, it is widely recognised that census-making is deeply enmeshed in broader political processes that shape them as governmental tools. In the USA, where there is by now a rich tradition of debate around racial and ethnic census categories that is rooted in civil right activism and more

Race, Ethnicity and National Censuses in Latin American States

5

recently in the wave of debates around the ideology of multiculturalism and its practical articulations, research produced on the occasion of the first census round of the new century was extraordinarily emphatic and convergent on this point. In her book on race and census, Melissa Nobles remarked that her general goal was “to confront the presumed transparency and political neutrality of racial categorisation on censuses”, and reminded the reader that “census bureaus are not innocent bystanders in the arena of politics; census data are never merely demographic data” (2000: x). Kertzer and Arel similarly argued that “the formulation of census questions and categories is inextricably embroiled in politics” (2002: 18) and, against the portrayals of censuses as neutral scientific enterprises, they suggestively depicted these counts as “a political battleground” where the existence of a given imagined group can be scientifically legitimised (ibid.: 20-21). Peter Skerry pithily stated that “the census is inherently–and properly–political in nature” (Skerry, 2000: 7) and presented his book on the topic with the subtitle “race, group identity and the evasion of politics”, in a belligerent allusion to the still lingering bureaucratic reluctance to openly accept political debate on census design and implementation. Nancy Krieger stressed the closely-knit relation between ideology and practical effects as mediated by identity categories in censuses with the following comments on USA experiences: “Change racial/ethnic categories […] and you change denominators for rates of birth, disease, disability, and death. Change rates, and you change assessments of need, understandings of social inequalities in health, and claims for resources. Change racial/ethnic categories, and you change our view of ourselves in relation to what even the US federal government now recognises, explicitly, as the ‘social-political construct’ of race/ethnicity” (2000: 1687)2. These geographically localised views are far from being exceptional among social scientists working in other very different social and political milieus. Existing work on censuses in or on Latin America goes along these lines. In his study of the use of statistics as ideological nation-making instruments in Ecuador, A. Kim Clark was aware that for many people “the very use of numbers seems to preclude the contaminating subjectivity of opinions”. However, far from subscribing to such views, he contended that “while statistics seem to present objective data through the simple counting of already existing facts, the very categories used in their collection reveal conceptions of society and personhood” (1998: 185). Other scholarly research may not have been so explicit about this point, but the generalised critical stance demonstrates that social scientists working on the region share the premise that there is nothing apolitical in

6

Chapter One

census undertakings as a whole (Harris et al., 1993; Magno et al., 2004; Sulmont, 2011; Telles, 2002; Telles and Lim, 1998; Urrea, 2011; Valdivia, 2011). The works produced by analysts linked to regional bureaus of statistics in the region or to supranational agencies specialised in population studies advocate the use of ethnic and racial categories as instruments to combat structural marginalisation, which is another way to acknowledge the political quality of the census enterprise (CELADE/CEPAL, 2009; CEPAL, 2006; Del Popolo, 2008; Jhon Antón et al., 2009; Petruccelli, 2007; Scholnik and Del Popolo, 2005). In relation to the political nature of censuses, it has been remarked that these counts, rather than merely adapting to the social world, play a central role in configuring it. In this regard Bourdieu’s ideas on the symbolic might of the modern state have had a strong influence, as well as Foucauldian notions of governmentality (Brubaker et al., 2004: 33-34; Desrosières 2008a and 2008b). It is nowadays generally accepted among scholars that censuses, as statecraft tools, project an illusory image of societies as well-bounded entities constituted by an addition of discrete cultural, racial or ethnic sub-units (Anderson, 1991), but also that this is related to modern “governmental technologies”. Foucault’s insights (2008) into the emergence of statistics in the eighteenth century as closely associated with such technology have been recovered by authors such as Desrosières (2008a and 2008b) in his analysis of the tensions between descriptive and normative dimensions in statistics as a characteristic of these undertakings. Censuses legitimise certain social identities, but they can also contribute to generating those identities. For some analysts the constitutive potential of censuses is so absolute that they have been said to “nominate into existence” (Goldberg, 1997) and also to “make up people” (Ian Hacking, cited by Brubaker et al., 2004). Kertzer and Arel worked on the premise that the “census does much more than simply reflect social reality: rather, it plays a key role in the construction of that reality” (2002: 2). Nobles suggested that “racial enumeration itself creates and advances concepts of race” (2000: xi) and, furthermore, that the “census bureaus are political actors that help to make race a political reality and do not simply count by it” (ibid.: 22). Skerry, providing concrete examples of the influence of census in shaping social categorisations, pointed out how a term such as “Hispanic”, so consolidated in USA nowadays, was not that common before its being adopted in the USA by the Bureau of the Census in the 1970s (2000: 31). In smaller scale studies, other authors have also

Race, Ethnicity and National Censuses in Latin American States

7

acknowledged “the constitutive power of the census with respect to race” (Mezey, 2003: 1703). All this research situates bureaus of statistics as targets of political scrutiny and, to some extent, it is not surprising that they have tried to defend their administrative neutrality. Besides the search for stable institutional environments, the existing reluctance within these bureaus to accept the premise that censuses are “political” relates to potential collateral damage that understandably concerns functionaries: the fact that they can be associated with wilful and fraudulent data production and management. However, when social scientists recognise the political nature of censuses they do not (necessarily) contend that census results are being “cooked”. “Political” is not (in principle) “unethical”. It is not even conterminous with “fabricated” in the arena of census categorisation, since what is being contended by scholars and activists is precisely that in the practice of counting social identities there is no such a thing as an exact or naturally objective method, and therefore that in this aspect of censusmaking there is nothing that is not “fabricated” (Kertzer and Arel, 2002: 21-22). Unfortunately, certain bureaucracies (and also certain governments) cultivate a (misleading) narrative of distrust of the term “political”, contributing to normalise a hegemonic discourse, triumphant in the heyday of neoliberalism, that presents the orientation of public administration as a de-ideologised practice that merely requires attention to unquestionable (market-determined) facts. This does not benefit civic engagement with the public sphere, and certainly does not facilitate the increase of legitimacy in census processes. There is an ample scope for political decision in the process that conditions the inclusion, definition and operationalisation of social identities in national censuses. We contend that the only way to avoid lack of legitimacy and problematic outcomes in the shaping and management of these social categories in censuses is precisely by making that decision process as open to informed public debate, participation and scrutiny as possible.

2. The chapters This volume includes chapters on ethnic and racial census categorisation in Argentina, Bolivia, Brazil, Colombia, Guatemala, Panama, Peru and Venezuela. There is a strong focus on the examination of state-fostered constructions of indigeneity as mediated by national censuses, though the Argentinian, Brazilian and Panamanian case studies address ethnic and racial categorisation at large. In addition, two chapters

8

Chapter One

of the volume explicitly develop a comparative perspective that tackles the types of questions used in different South, Central and North American and Caribbean national censuses to classify population according to ethnic and racial criteria. Like most volumes with a continental scope, ours deals with the dilemmas of representation and with the challenge of identifying regional trends with a limited sample of national studies. The diversity of analytical perspectives to be found here not only reflects the varied disciplinary backgrounds of contributors, but also testifies to the complexities associated with census categorisation and the range of political stances that censuses generate as soon as they are critically scrutinised. We nevertheless consider that the combination of South and Central American case studies, the range of theoretical and methodological perspectives developed by contributors and the comparative efforts explicitly developed in some of the chapters play to the volume’s advantage. As a whole, it provides insights into regional trends as well as a large amount of empirical data that not only enriches our general introductory reflections but will be a valuable input into any forthcoming comparative analysis. All the chapters present some common grounds. The authors have examined the political aspects of census-making, which is an essential step forward for bringing them closer to public debate. Some degree of diachronic analysis has been adopted in the examination of national censuses in all case studies, and notions of “change” in state approaches to indigeneity in particular and to identity classification in general become apparent. This is probably the simplest and strongest testimony to the constructed character of state sanctioned ethnic and racial categories: in censuses they are defined, used or obliterated in accord with malleable conceptions of nationality, democracy and justice that depend on hegemonic ideologies and the goals that states set for themselves at particular historical periods. A brief introduction to the chapters follows. Pilar Barrientos provides a historical overview of indigenous categorisation in Argentinian censuses and discusses the practical implications of current working dynamics within the national bureau of statistics. The link between census-making and nation-building are made salient by a critical examination of the historically contentious relationship between the Argentinean state and the indigenous population, and then by analysing the terms under which Afro-descendancy has been recently (2010) conceptualised as an ethnic category. From a standpoint informed by personal experience in the bureau of statistics, Barrientos highlights how its decision-making processes can be conditioned by an underlying tacit hierarchisation of knowledge types. The “technical” one ranks at the

Race, Ethnicity and National Censuses in Latin American States

9

very top, and for some inside technicians “bureaucratic procedures take precedence over any other logic”. As a result, despite the existence of rhetoric and positive regulations that promote the participation of civil society representatives (such as indigenous ones) in census design, actors who participate but are not bearers of “technical knowledge” can be factually excluded from decision-making. Besides the value of her reflexive authorship in this context, with this work Barrientos generates avenues for the study of power relations ingrained in institutional place and practice, something about which science historians have been long writing without many echoes in the social sciences. The chapter by Pablo Regalsky revolves around one of the major issues associated with contemporary census politics: the individual subjectification of identities implied by the consolidation of self-ascription as the central criterion for ethnic identification. The author approaches censuses from a relational understanding of ethnicity within state power relations. The national census, rather than merely providing a snapshot of demographic conditions, becomes in this view an indication of the existing power balance between indigenous peoples and the state. This analysis of national censuses is developed within a broad political frame that takes into account the global conjunctures of contemporary capitalism and the ongoing hegemony of neoliberal political narratives. Against that background, Regalsky explores alternative methods for indigenous identification in Andean Bolivia. He points out objective grounds upon which that identification can be undertaken by resorting to existing forms of recognising community belonging and membership in the region: indigenous notions of descent and the mechanisms through which holding rights over communal lands are granted are central in that process. This proposal sharply highlights the contrast between the use of census paradigms that promote individual self-identification (and therefore count “individual bodies”) and the possibilities to be found in paradigms based on “collective bodies” and communal ideologies. The author contends that the critical analysis of this contrast is particularly relevant in a period of ongoing debates about the territorial and political configurations that plurinational Bolivia requires. Gloria Lopera Mesa examines the relation between national censuses and the construction of indigeneity in Colombia by questioning who counts indigenous peoples, how are they counted and what for. The author’s densely documented historical overview generates analytical threads that link the racialised colonial order established by the conquest with contemporary models of differentiated citizenship sustained by notions of cultural difference. At present, census categorisation in

10

Chapter One

Colombia faces what the author calls “the paradox of self-recognition”: self-identification is the central criterion for ethnic identification in the national census, yet its results continue to be treated with suspicion from state agencies that in their institutional praxis retain the power of defining indigeneity through other mechanisms. Lopera additionally examines the appearance of a new ethnic/legal category, that of the “indigenous (or Afro-Colombian) victim of the armed conflict”. The author argues that this category, a combination of ethnic identification and penal typification, is showing an unexpected potential for the displacement of other categories of “mere indigeneity” as a condition to access basic state-regulated benefits. This category presents the double challenge of identifying who is a “victim” and setting boundaries to an indigenous or Afro-Colombian subjectivity, and Lopera suggestively draws attention to the role that the judiciary might play in the process of identification as the ultimate sphere in which existing notions of indigeneity can be legitimised or rejected. Gemma Celigueta Comerma analyses the appearance of new forms of indigenous identification in the Guatemalan census round 2002. In a country with a long history of categorical dichotomisation between “indigenous” and ladino population, over 100,000 people simultaneously self-identified as both indígenas and ladinos in 2002. The author remarks on the puzzling fact that these amalgamated indígenas-ladinos identifications precisely sprang up in a period in which the prevailing ideology of multiculturalism had been stimulating differentiated ethnic ascriptions, rather than cohesive hybrid categories (such as those previously promoted through Latin American paradigms of national integration). With the support of her own ethnographic work in Quetzaltenango and in critical dialogue with recent work on indigeneity and mestizaje, Celigueta suggests that these indígenas-ladinos figures incarnate understandings of indigeneity that depart from notions of “purity and authenticity”. Albeit calling for further ethnographic studies in order to bring new insights into this phenomenon, the author points to the influence that current redefinitions of indigeneity at international level might be exerting over contemporary representations and conceptualisations of indigeneity in Guatemala. The chapter by Mònica Martínez Mauri explores how racial and ethnic categories have been constructed in Panama with the support of national censuses and the state’s biopolitical management of a culturally diverse population. She highlights the relation between the official population categories and the ways in which the national territory has been conceptualised, largely along the lines of a centre-periphery dichotomisation

Race, Ethnicity and National Censuses in Latin American States

11

that overlaps with another, separating “civilised” and “uncivilised” spheres. Martínez shows how the relations between centre and periphery and between urban and rural were not only reflected in but also supported by census categories used to identify population inhabiting territories differently related to national power. In this context, racial ideology is shown to have been central in the shaping of census categories, with notions of race determining categorisation until the 1950s. Even when “mixed” categories appeared in censuses they were conceptualised as resulting from the combination of “pure” ones, thus reinforcing racial thinking. The author also emphasises the strong influence of international organisations in the definition of indigeneity and discusses the links between contemporary census categorisation and the Panamanian version of the neoliberal model. She suggests that the fact that indigenous organisations are accepting policies of differentiation consistent with the scripts fostered by guiding institutions within this model relates to the existence of an assistance-based local development frame that offers advantages to the members of differentially targeted cultural groups. David Sulmont and Néstor Valdivia tackle the case of Peru, a country where ethnic self-identification has not yet been introduced in national censuses. The authors provide a historical overview of these censuses and identify four distinguishable phases. Racial categorisation prevailed until the 1940s, when censuses shifted to register indicators of what dominant anthropological schools identified as “indigenous” cultural traits. By the 1970s, ideological changes in the state’s approach to ethnicity meant that it would not be directly registered in censuses. However, the persistence of different forms of discrimination continued to spur debates about how to officially identify and portray the perceived relation between ethnic and racial differentiations and social disadvantage, and since 2001 those concerns have been reflected in a variety of research projects and practical statistical experiences that are critically explored by the authors. With a focus on applied research, Sulmont and Valdivia partake in the efforts to improve the reliability of ethnic and racial identification in the next national census round (to be held in 2017). They distinguish sets of goals in the advocacy of ethnic and racial categories in official statistics, showing how census methodologies vary in accord with the goals that are to be pursued. They emphasise the importance of sound methodological and conceptual foundations for census undertakings. Though the contribution of technical expertise is acknowledged and the investment of adequate resources is of course called for, the authors rightly underscore the crucial importance of wide public engagement in the shaping of

12

Chapter One

censuses, whose results always have wide-ranging and transcendent practical implications. Luis Fernando Angosto Ferrández examines the relation between indigeneity and census-making in Venezuela, where official data on the relative size of the indigenous population has always projected its “minority” status. With a particular focus on the analysis of censuses from the 1980s onwards, the author shows how different historical and political conjunctures have influenced census designs and results. Operational definitions, questions and methodologies used in national censuses to identify and characterise the indigenous population are analysed in relation to particular political milieus. Angosto contends that fluctuations in the size of the indigenous population reflected in official statistics relate more directly to political than to demographic factors. The analytical comparison of the number of indigenous peoples registered in different census rounds supports that argument and clearly illustrates the ways in which contingent political milieus and bureaucratic decision influence census results. Regarding contemporary census affairs, Angosto also identifies noticeable divergences within state organs in the approach to “indigeneity” and in the understanding of concepts such as “indigenous community”, and shows how those divergences are currently reflected in institutional praxis. The realm of the census becomes one of the arenas where those divergences are negotiated, but the author reveals the ways in which that realm is more broadly framed and influenced by contextual political forces. New discursive narratives of national identity that intensively draw on an imaginary of indigeneity and governmental plans of national development (articulated through the so-called “new geometry of power”) are pointed out as particularly influential in that regard. José Luis Petruccelli situates his analysis of Brazilian censuses within a continental comparative frame. He shares the view that, besides providing data for public policy, censuses legitimise a particular social perception of the groups that constitute the society. In the examination of Brazilian censuses from the late nineteenth century onwards, he shows how, besides describing and classifying the population through physical traits, race categories have contributed to naturalise culture. Petruccelli tackles in applied terms the debate around the number and definition of the social categories that ought to be used to provide an appropriate representation of the Brazilian population in national censuses. This has been a politically loaded and paradigmatic debate, given the prominence of the racial democracy ideology in the country and the often alleged difficulties in the establishment of discrete classifications in the Brazilian “racial continuum”. Petruccelli nevertheless shows that in fact a

Race, Ethnicity and National Censuses in Latin American States

13

small number of categories concentrate most responses of racial selfidentification when surveys with national scope have been conducted, and argues that this should constitute the basis for informed debates around censuses. The author contends that the removal of racial classifications from Brazilian censuses, a proposal consonant with the views of those who characterise the country as a racial democracy, would hinder the struggle against still existing forms of race-based marginalisation in the country. On another front, Petruccelli makes a remarkable compilation of data regarding the use of ethnic and racial census indicators in American countries. This compilation, along with the one provided in the following chapter by Fabiana Del Popolo and Susana Schkolnik, offers an excellent comparative platform for anyone interested in contemporary censusmaking in Latin America and beyond. Fabiana Del Popolo and Susana Schkolnik share the premise that indigenous peoples and Afro-descendants in Latin America are disadvantaged populations as regards the enjoyment of human rights. They set out to identify ways to gather information that can be used to effectively tackle those disadvantages, making an overview of statistical experiences in the continent. Out of this comparative exercise, the authors elaborate conceptual tools that could orientate the inclusion of ethnic and racial categories in forthcoming national censuses. They remark the centrality of these counts in processes of official ethnic and racial identification. However, Del Popolo and Schkolnik consider that, beyond national censuses, it is also essential to extend ethnic and racial identification to all official sources of data, and close their chapter with a list of recommendations for that purpose. Given the centrality of the authors’ own institution, the CEPAL, in the channelling of continental census trends, this work provides insights into some of the central guidelines that Latin American bureaus of statistics will be using in the next few years.

3. Historical framings The establishment of parameters for the historical comparison of racial and ethnic categorisation in Latin American censuses is a complex task with which we will only partially deal in this volume. Instead of embarking on a systematic comparison of census methodologies and results, we have focused on the identification of common sociological patterns, salient ideological currents and statecraft models that have conditioned the ways in which race and ethnic census categories have been shaped and used by Latin American governments.

14

Chapter One

3.1. Liberal projects and state modernisation in nineteenth century Latin America The early stages of modern Latin American censuses are closely intertwined with the processes of state development and modernisation that spread over much of the continent in the latter part of the nineteenth century. In the second half of the century the continent witnessed what somewhat celebratorily has been called “the heyday of liberal reform” (Bushnell and Macaulay, 1994: 193-221). In effect, that reform had little room for liberal democracy and even less for the participation of subaltern classes in the orientation of government, with political dominance often expressed through figures who concentrated and even personalised state power. However, it has to be conceded that, though frequently at the expense of violently suppressed internal conflict and with political rule sustained by quite impermeable (if heterogeneous) elites and/or parties of notables, the last decades of the century and the early 1900s established the grounds for new avenues of economic growth. Albeit principally benefiting those elites (Burns, 1980: 10-15), that economic growth was accompanied by a gradual development of the state apparatus. Particularly from the 1870s onwards, with relative political stability and ongoing projects of modernisation under some variant of the “order and development” motto, governments sought mechanisms that would facilitate the consolidation of state institutionality and governmentality over the national population. Significantly, bureaus of statistics were created in several countries during this period, and many of the first “modern” national censuses were conducted in or around the 1870s: Argentina (1869), Brazil (1872), Guatemala (1870, with its Bureau of Statistics created in 1879), Peru (1876) and Venezuela (1873) are examples of this trend to be found amongst the national case studies included in this volume3. This was also the period of new intellectual influences in the continent. Positivism had gained terrain in intellectual circles4 and, albeit with distinctive national characteristics, at large it became an “ideology more than a philosophy [of science]”; it was used for political purposes, finding expression in governmental institutions (Ardao, 1963). In addition, notions of social Darwinism had become a “meta-language” that provided elites with frameworks for scientific and political analysis of Latin American histories and futures (Stepan, 1991: 41). Theories of science and social conclusions were thus intertwined in a system of interpretation which provided grounding for state approaches to population diversity through the accommodation of evolutionist views that equated “white race” and “progress”. Whilst race in this period was often conceived in terms of

Race, Ethnicity and National Censuses in Latin American States

15

cultural (hierarchised) difference rather than merely based on biology, it determined the fate of a social group: inferior cultures would need to be “civilised”. Censuses maintained their usual importance for tax collection and were used for the establishment of quotas of representation in the consolidating parliamentary regimes, but in this period they also became the foundations of development projects and population management in general. They were instrumental in the pursuit of the sought-after “modernised” national image. Accordingly, censuses marked boundaries between “civilised” and “uncivilised” populations, the former being sustained by largely racialised notions of national identity in which “white” (or “creolised”), “cultured” and “urban” profiles represented the peak of evolution and articulated the interests of the dominant classes. In this context, the indigenous population that had not been “reduced”, effectively integrated into the national economy and subjected to direct state control (that is, the population that maintained more autonomy) would fall under the “uncivilised” category, and it was thus reflected in censuses.

3.2. National projects and the ideologies of mestizaje and “whitening” Although nation-state building processes have had distinctive features in different Latin American countries, most of them faced similar challenges with regards to the construction of their sought-after “unitary” population. After nineteenth-century independences, the colonial order was theoretically contested through liberal ideals that projected images of a citizenry based on individual subjects equal before the law. Governments were charged with the responsibility of overcoming or removing the cultural and juridical distinctiveness that subaltern groups were ascribed in the previous order. Cultural difference was causally narrowed down to lack of educational enlightenment, and legal difference, particularly the existence of collective rights, was targeted as antithetic to the emancipatory values of political liberalism and modern individualism. The juridical distinctiveness that indigeneity preserved in much of the continent under colonial law was one of the first obstacles to be erased. However, the young and unstable Republics fell short in the realisation of their ideals. Weak state apparatuses and inconsistencies in the policies that were to split indigenous collective bodies into individual citizens and proprietors converged with the resistance deployed by some of those collectives. Furthermore, as slavery was formally abolished during the second half of the century, the Republics had to face additional challenges in the

16

Chapter One

constitution of a national population that, though theoretically equal, continued to be perceived as racially and culturally diversified. There would be additional strong opposition from elites that loathed the potential consequences of eliminating legal differences and the implications of blurring conceptual and cultural ones. The increasing “vulgar” miscegenation (Cañizares-Esguerra, 2007: 32) was seen as a threat to their ideals of a (hierarchised) nation based on Hispanic heritage. However, as outlined above, Latin American evolutionist science and political philosophy of this period tended to explain race hierarchies through cultural rather than purely biological reasons. Whilst white superiority was tantamount to cultural superiority, racial mixture would not necessarily be conceptualised as leading to degeneration and corruption, as was the assumption for hegemonic political and scientific discourse in the north of America (cf. Baker, 1998). Parts of the continent had already witnessed an ideologically approved miscegenation of Spanish and Amerindian elites that facilitated colonial rule at early stages. On the other hand, nationalist movements throughout the nineteenth century had made a considerable theoretical investment in presenting the political value of miscegenation and mestizo identities as a guarantee for and label of American autonomies. With variants of this socio-political background in different parts of the continent, a basic ingredient for late nineteenth and twentieth-century nationalism would be the concern with national identity and racial mixture (Miller, 2004; Wade, 1997: 84-87). The general goal of presenting a culturally homogenised and unitary population came through distinctive national projects, but it is possible to identify within them two dominant, if often inter-related, trends: the ideologies of mestizaje and the ideologies of “whitening”. In order to consolidate the unitary character of the national population base, and depending on where the cause of the problem of diversity was placed, two avenues were followed. When the problem was categorised as “cultural”, it was to be tackled through educational programmes and the search for amelioration in conditions of work, health, etc. If it was labelled biological, it would be addressed through programmes of eugenics (that were expected to result in generalised “whitening” of the population), through a management of migration policies that favoured the arrival of certain types of (“white”) migrants and forbade the arrival of (“nonwhite”) others5, or else by isolation of Blacks and Indians in enclaves (cf. Wright, 1990: 76). Notwithstanding the recognition of the hybrid cultural and biological origins of the Latin American nations, both the ideologies of mestizaje (and also those of “whitening”) were based on the idea that

Race, Ethnicity and National Censuses in Latin American States

17

the “indigenous” and “black” groups that had not already disappeared were likely to disappear soon into the new mestizo identity of the nation. When the projects of assimilation and cultural and economic merging into the new national conglomerate proved a blatant failure, the indigenous population was left outside the boundary of the project of nationality (such as in Guatemala), with the underlying assumption that it was essentially atavistic and could never form part of the new nation. These ideologies of mestizaje and “whitening” were reflected in census programmes that would register racial categories as associated with extreme alterity from the referential “mestizo” or “white” spheres. The “indigenous” population was defined through its lack of “civilisation” and its “isolation”. As soon as the Spanish language was spoken, overt signs of indigeneity had been eliminated or modified to a mestizo or creole standard and/or there was urban residence, the assumption was that the indigenous ingredient had merged into the mestizo nation. Beyond cultural matters, these conceptions implied that the integrated population took part in national production. With the expansion of the capitalist frontier and urbanisation, the numbers in groups categorised as “indigenous” would tend to decrease. A good illustration of the ideology that pervaded this period is to be found in this chapter in this volume by Sulmont and Valdivia: in Peru, the increase of “white” and mestizo population that was projected by the 1940 census was interpreted by state agents as an expression of the tendency towards the “formation of the specific type of national race: the Mestizo, in which racial crossbreeding is synthesised, with a predominance of the ethnic characteristics of whites and Indians”.

3.3. From indigenismo to participative anthropology 3.3.1. The influence of institutional indigenismo The emergence of indigenismo as a distinctive state ideology orienting policy for indigenous peoples is one of the most influential milestones in the shaping of approaches to population management in the continent. Institutionalised with the creation of the Interamerican Indigenist Institute (INII) after the First Interamerican Indigenist Congress of 1940 (Pátzcuaro, Mexico), it had the International Convention on the Interamerican Indigenist Congress and the Interamerican Indigenist Institute (Convención Internacional Relativa a los Congresos Indigenistas Interamericanos y al Instituto Indigenista Interamericano) as its legal foundations. An institution with such Panamerican scope could only materialise after a historical period of gestation in which its necessity was defined and

18

Chapter One

discursively recognised (Giraudo, 2006). The legitimising discourse had been gaining strength for a decade, revolving around the necessity of coordinating state approaches to the so-called “indigenous issue” (la cuestión indígena) in the continent. The First American Conference of Labor (Primera Conferencia Americana del Trabajo [Santiago, 1936]) and the Second General Assembly of the Panamerican Institute of Geography and History (Segunda Asamblea General del Instituto Panamericano de Geografía e Historia [Washington, 1937]) had converged on that point: indigenous issues should be addressed in meetings of experts, supported by scientific research and specialised institutions. Two years before Pátzcuaro, the Eighth Panamerican Conference (Lima, Peru) openly recommended the creation of an indianist institute and urged American governments “to develop policy aiming to completely integrate [the indigenous population] in the respective national milieus”, since this population, as the descendant of the first ones in American lands, deserved “a preferential right to protection from public authorities in order to overcome the deficiencies of their physical and intellectual development”6 (Giraudo 2006: 6-7). The problem was in those terms associated with physical and cultural dimensions, but the first one would be abandoned soon after. A resolute departure from the explanation of difference and marginalisation through biological criteria would be one of the characteristics of state approaches to cultural diversity (and to indigeneity in particular) after the Second World War and after emergent anthropological theories that focused on cultural dynamics started to become functional for the Institute (see next section). Race seemed to be a category in “extinction” in that post-War world context. At an international level, those scientific epistemes converged with antiracist movements that went from antifascism to the struggles in the sphere of civil rights in countries like the USA, but also included the anti-colonial wave and the anti-apartheid movement. This was all framed by the Cold War and the competition of two super-powers for global supremacy. Although the nominal obliteration of race would gradually take over censuses, this did not imply that racial dominance disappeared. The socalled “post-racial” era could be interpreted as a transition from a system of racial dominance to one of racial hegemony (Taguieff, 2001; Winan 2004). The influence of indigenismo, and particularly of its Mexican current, was strong in the continent. It shaped research agendas and analytical perspectives in countries with most diverse sociological and political histories7, giving legitimacy to social orders and reproducing the state’s “right to rule” (Saldivar, 2011: 69-70). But, notwithstanding its

Race, Ethnicity and National Censuses in Latin American States

19

governmental elements, indigenista praxis advanced ideas of cultural relativism. In the causal explanation of inequalities between the groups that constituted society, it placed emphasis on the sphere of culture. Censuses in the continent would be consistent with that orientation and in this period one can identify a clear shift towards this type of approach to indigeneity as defined through certain cultural traits considered to be “traditional”. Chapters in this volume provide significant illustrations of this process. In Guatemala, the 1950 census, regarded as the first professionalised census programme in the country, had the imprint of the INII and the Interamerican Statistical Association (see Celigueta in this volume). The census abandoned the definitions of the ladino and the indígena groups in racial terms and started to build upon idioms of ethnicity, defining group difference in terms of culture and compiling information about a whole list of cultural traits (from language and type of dwelling to clothing and food). In Panama census re-orientation had similar characteristics, and the INII’s emphasis on integrating the indigenous population was felt. The 1950 census abandoned racial criteria too; indígenas were defined as those “who live primordially separated from the country’s socio-economic structure and at times even under tribal organisation” (see Martínez in this volume). In Venezuela, the influence of indigenismo was also noticed, and indeed fostered by the father of Venezuelan anthropology, Acosta Saignes (see Angosto in this volume). In the 1950s the indigenous population which was not integrated in “civil life” was labelled “inaccessible”–a step away from “uncivilised”. In Colombia, indigenismo was influential from the 1950s onwards (see Lopera in this volume), in a twofold way: on the one hand, national censuses introduced modifications in the criteria used to identify the indigenous population which, without totally eliminating phenotypical characterisation, emphasised cultural traits and residence; on the other hand, those criteria started to be contested by peoples who selfidentified as “indigenous” but who did not match the “anthropological” criteria used in censuses to set boundaries to indigenous identities. 3.3.2. Anthropology, academia and the shaping of indigeneity through censuses The historic relationship between anthropology and colonialism has been widely commented and documented. Along with other scientific disciplines, anthropology has also been associated with nation-state building processes, contributing to legitimise social orders sustained by racial hierarchies (Baker, 1998; Haller, 1971; Stocking, 1968) or facilitating

20

Chapter One

knowledge about indigenous populations in order to facilitate national integration (Hewitt de Alcántara, 1984; Saldivar, 2011). Against this background there has also been a considerable academic production addressing the history of “national anthropologies” in relation to social structures of power (Beckett, 2010; Lins and Escobar, 2006; Lomnitz, 2001). In Latin America there have been studies of the influence of local and global institutions and political contingencies in the shaping and legitimisation of academic work in countries such as Ecuador (Martínez Novo, 2007). Nevertheless, there seems to be sufficient scattered evidence to anticipate a fruitful harvest for those who conduct further research into the complex relation between political milestones in the continent and the twists and turns of national anthropologies. The relation between indigenismo and anthropological views of human diversity, or that between the 1970s international context and the emergence of currents of applied and politically engaged anthropology, are certainly interesting spheres of research. A largely unexplored aspect of these histories is the role of anthropologists in the shaping of public policy or administration. As regards the criteria that have been used to count population in ethnic and racial terms in the Latin American continent, this influence appears to be considerable. Census-making is a field in which the relations between the political and scientific spheres need to be finely tuned. Census categories require scientific legitimisation and, particularly with regard to the shaping of ethnic and racial categories, the influence exerted by anthropologists as scientific authorities at certain periods of census history is very noticeable. Anthropologists have had a continuing presence as qualified experts on indigenous issues since the discipline began to be institutionalised in the continent and particularly since the INII helped in raising their professional profile as experts that could contribute to good government and to the consolidation of national projects. Dominant views from the discipline have helped to “normalise” conceptions of indigeneity and ethnicity through politically loaded administrative processes. In this volume, that influential weight of anthropology as a discipline is made salient in several chapters. The shift from “race” to “ethnicity” in the nominal and conceptual framing of indigeneity well illustrates that influence. Celigueta emphasises in her chapter the influence of dominant anthropological theories in the definition of indigeneity in Guatemala through the selection of certain cultural traits as its diacritica. The role of anthropologists was central in the shift from the conceptualisation of ladinos and indígenas as racial groups to their conceptualisation as ethnic

Race, Ethnicity and National Censuses in Latin American States

21

groups, as well as in the production of organic ideas about acculturation that were consistent with policies of integration of indigenous peoples into nationality. Once ladinos and indígenas were categorised according to “cultural” rather than “racial” differences, there were in theory no insurmountable barriers for the ladinización of the indígenas through cultural processes of acculturation in which education would play a fundamental role. When cultural traits such as language, clothing or rural residence are established to define indigeneity, it follows that, if these traits disappear, so does indigeneity. The case of Peru presents some parallels. The anthropological perspectives on indigenous peoples from the 1940s to 1960s were determinant in the shaping of the 1961 census results and in the abandonment of racial categories to classify the population (see Sulmont and Valdivia in this volume). In Venezuela, the direct influence of anthropologists peaks in the 1970s, when the lack of national census data on the indigenous population led the state agency in charge of indigenous issues to resort to an academic department for the elaboration of a census (see Angosto in this volume). The criteria then established to identify indigeneity in the country still resonate and are reflected in contemporary census practice. In spite of the general move from externally imposed objective criteria of identification to self-identification, a basic distinction remains in Venezuela that separates the “general” from the “indigenous” census: the latter is only applied in pre-established areas of “traditional occupation”, whereas through the former individuals can selfascribe to the indigenous categories in any part of the country. In Colombia, the intervention of anthropologists in the debates around the definition of indigeneity was noticeable at least from the 1950s (see Lopera in this volume). Reichel-Dolmatoff questioned the existing criteria for the identification of indigenous people in the census of 1951, arguing that many people who were indigenous “from an anthropological viewpoint” had been obliterated from the indigenous census. That anthropological viewpoint was in fact based on certain cultural and phenotype or diacritica, but according to Reichel-Dolmatoff those traits should not include aboriginal language, since some (otherwise objectively identifiable) indigenous people spoke Spanish whilst maintaining the other diacritica. 3.3.3. New forms of political activism, avant-garde indigenismo and international networking The 1970s were years of global structural adjustments and turmoil. Financial capitalism would become the spurring force of new forms of

22

Chapter One

globalisation and the welfare state would be gradually displaced as the political institutional formula of social democracy. After some years of germination, identity politics gained terrain during that decade as the frontline in contestation of power imbalances. In this context, the politics of indigeneity in Latin America went through a period of redefinition while being exposed to two intertwined foci of transformation. On the one hand, the principles of institutional indigenismo, characterised by a topdown approach to designing and implementing policy for indigenous peoples, started to be strongly contested by indigenous actors and sympathetic activists. On the other hand, there was a gradual move towards a trans-continental internationalisation of indigenous struggles and networks and the creation of new permanent supranational platforms that addressed the “indigenous issue” and aimed at the coordination of state policy for indigenous peoples (Briones and Kradolfer, 2010; Brysk, 2000: 86-105). There were varied expressions and articulations of these trends in different countries, but they all pointed to a gradual (if contentious) recognition of indigenous peoples as political actors. In the 1970s emerged an “avant-garde indigenismo”. It placed emphasis on ideas of indigenous agency and autonomy within national politics, principles which were to some extent outlined in the renowned Barbados Declaration (1971). Some anthropologists, a wing of the Catholic Church influenced by liberation theology, and other political activists participated as “allies” in this process. The seeds of an indigenous social movement germinated during this period, and indigenous organisations sprang out throughout the continent. The emphasis on agency and autonomy was shaped through cultural claims and new forms of representation which established the foundations of what has been critically labelled “ethnicism”: a form of channelling claims that would have diverted indigenous peoples from more radical ones and which was rapidly adapted and adopted by regional governments (Díaz Polanco, 1978; 1997). The fact that this all came in a period of hegemonic shift towards financial capitalism has not been as widely examined and discussed as the contemporary relations between the ideology of multiculturalism and the heyday of neoliberalism (cf. Goldberg, 1994; Hale, 2002; Hobsbaum, 1996; Zizek, 1997), but it is the source of contextual political forces that cannot be disregarded. At any rate, all these transformations found expression in the design and implementation of censuses, as becomes apparent by examining the case studies included in this volume. In Colombia, an “indigenous renaissance” took place in the 1970s. Indigenous organisations, as in many other parts of the continent, emerged

Race, Ethnicity and National Censuses in Latin American States

23

during this period and started to re-claim territory and demand degrees of cultural and political autonomy (see Lopera in this volume). From 1973 a specific questionnaire would be used for indigenous censuses. These censuses, the author suggests, contributed to the “re-indigenisation phenomena” that would be reinforced with the shift to multicultural constitutionalism in the 1990s. From this period onwards, “state institutions in charge of dealing with the territorial claims of the indigenous movement have played a major role as agents in charge of counting and, thus, building indigeneity” (ibid.). In Venezuela, the 1970s harboured the first indigenous organisations at regional and, albeit fleetingly, national level. The imprint of the avant-garde indigenismo marked that process, supported by a range of non-indigenous activists. In the 1970s the national census did not produce data identifying the indigenous population (see Angosto in this volume), but precisely that gap facilitated the germination of new criteria to identify this population. The lack of relevant data led the state agency in charge of indigenous issues to commission a count of this population from an academic department of the Central University of Venezuela, and this department established the criteria that associated indigeneity with certain traits and territories of traditional occupation. In 1982, the new census round included a specific questionnaire for the indigenous population, which was to be identified basically according to criteria established in the 1970s. In Panama, the census in that decade reflected a shift towards a territorialisation of indigeneity with the introduction of the concept of “indigenous populated site”, which connoted the possibility of levels of “autonomy”. It is however to be noted that this was also the period of General Torrijos as Head of State and a new constitution that was hotly debated on the issue of indigenous lands and the possibilities for its collective ownership (see Martínez in this volume). In Peru, the situation was slightly different due to the influence of orthodox Marxist ideology in the shaping of political discourse that introduced the category of “peasant” as a substitute for indígena (see Sulmont and Valdivia in this volume). The revolutionary government that would remain in power until 1980 undertook an agrarian reform as part of a national project of development in which the population previously categorised as “indigenous”, now conceptualised as “peasants”, were to become an essential productive and political force. The “indigenous issue” had to be resolved through economic development and by strengthening a peasantry as a self-conscious class with revolutionary potential. Indigenous identification disappeared from censuses.

24

Chapter One

3.3.4. International organisations and the shift towards selfidentification As regards the front of internationalisation, from the late 1970s indigenous peoples gained increasing importance within UN agencies (Schulte-Tenckhoff, 1997). In 1971, the UN Economic and Social Council (ECOSOC) created the Sub-Commission on Prevention of Discrimination and Protection of Minorities. An ambitious study on discrimination against indigenous peoples was undertaken, coordinated by José Martínez Cobo and published in five volumes between 1981 and 1984 (Martínez Cobo, 1986). In 1982, the UN created a Working Group on Indigenous Populations (WGIP) with the goals of addressing issues concerning the promotion and protection of the fundamental rights of native peoples and developing normative frames consonant with those rights. In 1985, the WGIP began drafting a Declaration on the Rights of Indigenous Peoples (UNDRIP) which would only be finally approved by the UN General Assembly on 13 September 2007, after 22 years of negotiations8. The Permanent Forum on Indigenous Issues was created in 2000, meeting for the first time two years later in New York. Other acts of recognition took place during this period in various UN bodies (Kradolfer 2011). Also in this context, the concept of “social inclusion” became central to United Nations discourse on equality and has been fostered in events such as the Social Summit (1996) and the United Nations Conference Against Racism (2001). As indigenous peoples have been amongst the most marginalised populations, the design and formulation of development projects and programmes addressing them in the spectrum of the Millennium Development Goals of the United Nations highlighted the lack of information and disaggregated data. The Workshop on Data Collection and Disaggregation for Indigenous Peoples, held in New York in January 2004, advanced in its 10th recommendation that “the rights-based approach to development requires the development of a conceptual framework for rights-based indicators, that are relevant to indigenous and tribal peoples” (Economic and Social Council, 2004). In parallel, other supranational institutions such as the Inter-American Development Bank shifted towards targeting ethnic and racial excluded groups in its projects and policies, and the World Bank moved to a more qualitative approach to analysing poverty, making notions of “culture” central to the discourse of economic growth (Magno et al., 2004: 332). Despite all these processes revolving around indigenous issues, a definition of “indigenous peoples” has always been a contentious point and indigenous actors have been reluctant to be bound by one because of its restrictive potential and because of the difficulty of applying it trans-

Race, Ethnicity and National Censuses in Latin American States

25

continentally (Méndez and Martín, 2006: 348). Self-identification became in this context a central element in the definition of indigeneity, and on the legal front the International Labour Organization (ILO) Convention 169 for Indigenous and Tribal Peoples (1989) constituted a pivotal reference in this regard. Article 2 of this convention stipulated that “Self-identification as indigenous or tribal shall be regarded as a fundamental criterion for determining the groups to which the provisions of this Convention apply”. This criterion of identification would be consistently reflected in Latin American censuses from the 1990 round onwards, and indeed it has become the referent for all types of ethnic or racial census identification with very few exceptions (such as Cuba and Peru). This shift was of course spurred by UN agencies too. The UN had produced since 1946 four population programmes aimed at setting common standards for national censuses while improving their quality. There was therefore an expectation that “census-taking methods, and, by extension, census bureaus, can transcend particular political and economic environments” (Nobles, 2000: 15). The introduction of self-identification has become an example of such universalising methods. The participation of indigenous actors in the design and implementation of administrative measures that may affect them gained momentum with this process of internationalisation. The promotion of that participation found explicit articulation in the UNDRIP, whose Article 19 stipulates that “States shall consult and cooperate in good faith with the indigenous peoples concerned through their own representative institutions in order to obtain their free, prior and informed consent before adopting and implementing legislative or administrative measures that may affect them”. The presence of indigenous delegations has increased in all UN agencies, as well as in other international organisation (e.g. UNESCO, WHO, WIPO, etc.), as well as in other important international fora (as, for a recent example, the Rio+20 United Nations Conference on Sustainable Development held in June 2012).

4. Contemporary challenges 4.1. Self-identification Self-identification has indeed become a “regime in principle” (Skerry, 2000: 47) as regards ethnic and racial classification in national censuses, but of course it is not a panacea that precludes controversies, contradictions, dissatisfactions and challenges.

26

Chapter One

Its political foundations and practical implications are generating new sets of questions for its advocates and detractors alike. Despite its takenfor-granted connections with group recognition and its links with the advancement of collective rights, the principle of self-identification is paradoxically very consistent with individualistic political values and processes of subjectification of “the social” that are characteristic of the neoliberal era. Censuses are reinforcing the subjective character of social identities in different ways. Self-identification as currently articulated ultimately transforms mechanisms of group affiliation into individualised acts of subjective psychological choice. Additionally, it does not attach permanent identifications or legal status to people and this projects a potentially evanescent and “fleeting” character upon social identities. These dilemmas are at the core of contemporary census-making and have practical implications. Self-identification overcomes undignifying census practices, avoids reification of identities and is expected to provide more realistic portraits of the cultural diversity in a country. However, whilst officially unquestioned as a census criterion, self-identification is in fact not treated as a “reliable” source of data for the implementation of government policy addressing the practical articulations of differentiated rights or affirmative action. In parallel to self-identification, states establish their own objective criteria to discern, for instance, who and who is not indigenous when it comes to affirmative action, or they resort to ethnic or racial civil organisations to act as identity “gatekeepers”. Venezuelan censuses have established since the 1990s a factual distinction between a type of indigeneity that can be claimed through selfidentification and a type of indigeneity that is defined through officially established objective criteria (see Angosto in this volume). The latter is associated with particular areas of “traditional [indigenous] occupation”, areas which in turn were identified according to “anthropological” criteria four decades ago. Special questionnaires (and census programmes) are designed to obtain information from this category of (indigenous) population. The other category of indigenous population, the one registered through acts of self-ascription, can in principle be located anywhere within the national territory (the question of indigenous identification through self-ascription is included in the general census). The enjoyment of some of the distinctive rights that the constitution of 1999 grants to indigenous peoples is only made possible for the indigenous population that is preidentified through the questionnaires applied in pre-selected areas of traditional indigenous occupation. Similarly, the Colombian state has

Race, Ethnicity and National Censuses in Latin American States

27

introduced “additional criteria, other than self-identification that factually distinguish between the ‘truly indigenous’” and those who would be so only at a different level (see Lopera in this volume). The discourse of selfidentification collapses before the existence of those objective filters; in practice, self-identification becomes in this situation a marker of a type of indigeneity that is separated from notions of “traditionality” or “authenticity”. But the co-existence of objective criteria (as demarcated by cultural or geographical-territorial diacritica) and subjective criteria (as articulated through self-identification) in the social processes of definition of contemporary indigeneity is not only to be found in state practice and administrative procedure. In fact, and paradoxically, it is echoed in the praxis of advocates of self-identification. This contradiction is expressed when these advocates claim that in a census in which self-identification is already used for ethnic or race categories there has been an “undercount”: as a matter of fact those claims imply a (surprisingly overlooked) understanding of ethnic or race categories through some sort of objectivised criteria. For one could not talk about “undercounts” (or for that matter “overcounts”) unless those claims were grounded on the assumed existence of objective criteria to recognise ethnic or racial identities. Alternative methods and political foundations from which to undertake indigenous censuses are discussed in this volume. Regalsky provides an example of how anthropological knowledge can be applied to this task and how the locus of power in the identification process can be transferred onto indigenous groups. This type of proposal poses new challenges to prevalent values of census-making and relates to broader discussion of sovereignty, democracy and state monopolies, but is certainly a stimulating way to question long-sustained paradigms in a period in which new forms of organising national co-habitation are being debated in Latin America. As managed within the state apparatus, objective criteria of ethnic and racial classifications have generally been volatile instruments of governmentality, but the introduction of self-identification in censuses does not overcome that use–nor does it imply, as is sometimes assumed, that the state abandons the use of objective criteria to ground governmental practice related to groups defined through social identities.

4.2. The political grammar of social identities The relation between conceptions of social identities and their operational materialisation in censuses is mediated by semantics. That

28

Chapter One

mediation, which is articulated within the bureaus of statistics, is instrumental in the distillation of political struggles over the determination of key-concept meanings. For this reason and for other more practical socio-linguistic issues, the choice of terminology and wording is far from being secondary for census programmes and can decisively condition the results of social identity counts. Technicians and researchers generally agree that census results can be significantly affected by “social milieu and collection methods” and by questions such as “wording, format, and placement [and the] timing of a survey” (Skerry, 2000: 59). But that agreement does not always translate into the opening of census design methods to more public participation and debate in order to maximise the political legitimacy of these undertakings. In Latin America, discussions around the impact of the semantics of ethnic and race categorisations in census results are not new. The choice of appropriate terminology for questionnaires has been a long term concern, particularly since the shift towards self-identification. For instance, by the early 1990s, authors such as Harris et al. (1993) were using experimental methodology to show how the substitution of census categories could impact results. The choice of vernacular terms and the avoidance of rigid dichotomous categories in censuses was advocated on that occasion: according to those authors, when a vernacular term such as moreno was used in Brazilian censuses, the proportion of respondents self-identifying as being of mixed colour-race would be considerably larger than when the term pardo was used. In this volume we find some interesting contemporary examples of the semantics of categorisation; let us consider some of them. The use of the term “people” (pueblo) has been central to the processes of indigenous political re-emergence since the 1990s, yet as a politically loaded term it had (and still has) to overcome a number of challenges springing from both within and without the indigenous movement. At an early stage in those processes it was only accepted with reluctance, or directly rejected, by political forces and lobbies which presented it as a threat to national sovereignty and the seed for secession claims. Thus, for instance, its inclusion in Latin American constitutions as the substantive that would identify indigenous collectives was often firmly resisted. This very same reason was behind its substitution in censuses by other terms such as “ethnic group” or even “cultures”. After the approval of the ILO Convention 169 (1989), and more recently that of the UNDRIP (2007), the use of the term “indigenous peoples” was explicitly detached from potentially secessionist claims and became more accepted; but official acceptance in many countries has only come after a long process that is

Race, Ethnicity and National Censuses in Latin American States

29

still ongoing. Let us look at examples of how these processes have impacted census-making in the continent. In the Chile of 1992, still in the early transition to democracy after the fall of Pinochet’s dictatorial regime, the inclusion of a question to identify the indigenous population in the national census was being discussed, but the general recognition of indigenous peoples was still debated and opposed by conservative political groups. The identification of indigenous population in the census was finally agreed upon and it would result from acts of self-ascription; but, given the opposition generated by the use of the term “people”, this was discarded and the census question was finally posed in the following terms (Gundermann et al., 2005): “If you are Chilean, do you consider yourself as belonging to any of the following cultures? 1. Mapuche. 2. Aymara. 3. Rapanui. 4. None of the previous”. 10.33% of the country’s population was identified as indigenous under those terms. In 2002, with a different political climate in the country and with legal and administrative reforms as a background, the census question established to identify the indigenous population accommodated the use of the term “people” and was posed in the following terms: “Do you belong to any of the following peoples? 1. Alacalufe (Kawashkar). 2. Atacameño. 3. Aymara. 4. Coya. 5. Mapuche. 6. Quechua. 7. Rapanui. 8. Yámana (Yagán). 0. None of the previous”. According to the results, the indigenous population had come down to 4.58% of the national one. In the analysis of these results, researchers have discussed the potential impact of several factors related to the “data collection environment”, but amongst them the substitution of the term “people” by “culture” has been emphasised: the sense of belonging to a “people” is perceived as more restrictive (Gundermann et al., 2005: 99-101). This is an interesting example of the semantic politics around the term “people”, but we can find others more broadly related to socio-linguistics. Let us look at the case of Venezuela. In the census round of 2001, the question posed in the general questionnaire to enable self-ascription to the indigenous category was: “Do you belong to an indigenous people?” and, when the answer was affirmative, the respondent would be asked to indicate which was his/her people of ascription (with no pre-established options); however, in the previous round (1990/92), the question had been: “To which indigenous group do you belong?”. The change between one and the other responded to the political climate in the country and to the explicit recognition that indigenous peoples had gained with the 1999 Constitution, and unexpected census results ensued. Many “indigenous” respondents in 2001 were not

30

Chapter One

familiar with the new use of the term “people” in the identification of their collective ethno-political identities. In Spanish, this term [pueblo] has another meaning besides that of “people”: it can also mean “village” or “town”. The latter was in fact the meaning inferred by many of those indigenous respondents, who named their place of residence when asked about the people [pueblo] they belonged to. This created problems for the members of the National Institute of Statistics, who, when processing results had to assign “peoplehood” to “townhood” whenever that was feasible according to their pre-existing knowledge of the territorial units where indigenous peoples are located in the country (see Angosto in this volume). In the following census round (2011) changes were introduced to the question, in order to avoid that “people” problem: “Do you belong to an indigenous people or ethnic group?”9 More examples can be drawn, for instance, from the semantics of the term “indigenous”, which, depending on national and regional contexts, may connote different combinations of positive and negative meaning. Whilst in certain social sectors of countries such as Peru it is still pervaded by discriminatory meanings that may for many people preclude its acceptance as a source of self-identification, in some other countries such as Venezuela it is growing as a category of public ascription and being invested with positive meanings: indigeneity is one of the central symbolic sources that inspires the process of national identity re-foundation spurred by the Bolivarian governments of the past decade. Other significant examples of this political grammar of social identities have been examined in other parts of the continent. In Canada, the questionnaire of the 1996 national census listed the term “Canadian” as an ethnic origin category for the first time in national history. It appeared in 1991 as a statistically significant self-identification despite not being listed as one of the existing “example” categories: 3.3% of the population identified as “Canadian” in the category “Other-specify”. This came after a political campaign launched through a regional newspaper that aimed at overcoming particularistic ethnic identifications and promoted that nationality as a way to avoid the so-called “Balkanisation” of the country (Ketzer and Arel, 2002: 16). The results in 1996, after the inclusion of the “Canadian” category in the national census, were a massive surprise: 24.1% of the population self-identified as such. Analysts argued that, amongst other factors, the astonishing results could be considerably related to the fact that, for Québécois respondents, the term “Canadian” might have been associated with “Canadien”, in turn a term historically used to identify French Catholics (ibid.).

Race, Ethnicity and National Censuses in Latin American States

31

If to the identifiable influence of the grammar of social identities in census results related to ethnic and race identification we add the acknowledged influence of the so-called “example effects”, which names the demonstrated tendency of census respondents to reproduce in responses what has been included in questions (Skerry, 2000: 60), we find yet more strong arguments to support the depiction of the work in bureaus of statistics as a politically loaded process.

4.4. State monopolies and the census Despite divergences over many other issues, much of the social theory on the state, from liberal to Marxist to functionalist to post-structuralist perspectives, converges in the presentation of this apparatus of institutionalised power monopolies as a well coordinated apparatus cemented by the rule of law, class ideology, bureaucratic logic, the governmental will to discipline and control or any other goal. In Latin America, contemporary critical political philosophy from both the poles of revolution and reform criticises the foundations or the leanings of states as homogeneous (if complex) institutional structures with few nuances; either it is to be extinguished or it has to be taken over, transformed and used as an instrument for social re-ordering. In either case, it is implied that there is a closely-knit, well-integrated and well coordinated apparatus with a coherent set of bureaucratic behaviours and actions behind it. However, though the coercive, uniformising and impositional might of the state on certain fronts cannot be denied, it is difficult to sustain that in politically divided (and sometimes polarised) societies the state organs orchestrate their functioning without dissonant notes–even if arguably they follow class-bounded general orientations. The field of census-making generates in this regard thought-provoking dynamics. In this volume, Sulmont and Valdivia point out different conceptions of race and ethnicity within Peruvian state organs as potential obstacles for the improvement of the quality of the national censuses, and Angosto examines in detail concrete examples of substantial differences between a government ministerial branch and the bureau of statistics in their approaches to defining and counting indigenous population. These divergences could be simplistically interpreted in the light of long-held views concerning the weak institutionalisation of public powers in Latin American states, too often portrayed as prone to work inefficiently, uncoordinatedly and in the pursuit of particularised or corporative interests. In our opinion, there are other forces at work. The state in Latin America, in the midst of institutional renewal and rapid social change, is

32

Chapter One

traversed by multiple political forces. In this milieu there is scope for negotiation and manoeuvre within state agencies and certainly within bureaus of statistics which, besides responding to supranational orientation and guidance (such as the one originating in UN agencies), are supposed to respond to national interests. We believe that this is another positive reason to increase the level of public engagement with census-making, but we also warn against the dangers of transforming statistics bureaus into targets of petty interests and partisan discontents. The history of censuses is sufficiently complicated in itself, with its origins so closely associated with tax collection and forced recruitment into non-redistributive states, that the records of people fleeing from enumerators as one would run for one’s life (Clark, 1998: 195) cannot but be received with sympathetic understanding. Census programmes do not need extra strain on their reputation in a historical period of state re-construction and social reform in Latin America. But we nonetheless contend that more and better informed political debate around social categories in census-making is necessary and democratically healthy. Let it be clear that this is radically different from using this principle as the basis for orchestrated attacks that it may receive as spurred by party interests, as unfortunately still occurs on occasion.

5. The future that lies in contemporary censuses At present there is a broad consensus around the necessity of including ethnic and racial categories in Latin American censuses, and there are many reasons for that. Those social identities matter to many people at a symbolic level, and at a material level research has shown correlations between ethnic or racial ascriptions and economic and political hierarchies in the continent. In addition, the debates on democracy and human rights nowadays address collectives as much as individuals, and Latin America is one of the main foci of generation of such debates. At a trans-continental level, neoliberalism and global forms of governance have accommodated some degree of cultural sensitivity within their ideological dogmas. All these factors converge in explanations as to why there is such widespread backing for the inclusion of ethnic and racial categories in contemporary Latin American censuses, despite this being a subversion of deeply rooted principles of cultural and racial amalgamation as the pillars for national construction in the region throughout most of the twentieth century. But the fact that there is this consensus about the need for more realistic representations of the existing population diversity in a given country should not necessarily lead to uncritical acceptance of any

Race, Ethnicity and National Censuses in Latin American States

33

category presented as “ethnic” or “racial” in character. This is where public debate and participation in census design, along with increased levels of qualitative and quantitative research, should play a role. On the other hand, it is far from clear that these social categories should be used in every single sphere of official registration. Qualified voices advocate the inclusion of ethnic identification in all sources of official data, but this may be fraught with dangers. Firstly because of the well-known forces of normalisation of (potentially discriminatory) difference that are fostered by officially legitimised social categorisation. Secondly, because it is not clear what material or symbolic benefits derive (either for society as a whole or for those being thus categorised) from that way of proceeding in certain fields such as, for example, medicine. In the field of health studies there is increasing concern about the embedded racial pre-conceptions that might be affecting medical practitioners and researchers in certain areas, with potentially negative consequences for patients and society at large (Cooper et al., 2003; Foster and Sharp, 2002; Kierans and Cooper, 2011). In spite of the discourse of inclusion and all the emphasis that is placed on the participation of representatives of collectives that may be affected by census categorisations, existing research on Latin America has not yet focused on the concrete influences that civil society and grass-roots organisations exert over censuses. Elsewhere it has been argued that a shift in the “locus of power” has taken place, bringing decisive power to those groups (Kertzer and Arel, 2002: 27-8). But it is still unclear how far that might have effectively happened in Latin America. Social movements and grass-roots organisations have undeniably been exerting massive influence in the political arena through activism which, among other cleavages, identifies around ethnic and social identities, and bureaus of statistics have been gradually opening up to the participation of representatives of organisations associated with those movements. However, the examination of existing research suggests that international directives and global ideological influences have been more important so far in shaping census trends in the continent. It could be argued that, as with many of the initial multicultural reforms of the 1990s, the introduction of race and ethnic categories in some countries was more of a top-down adjustment to general crises of governmental and state legitimacy than a direct reflection of bottom-up mobilisation and political strength. On the other hand, even when there is nowadays evidence that states are becoming, at least formally, more responsive to participatory claims in public administration, there is always difficulty in knowing the real extent to which a participatory process becomes a process of power sharing. Barriento’s chapter in this volume provides a good example of the way in which

34

Chapter One

bureaucratic procedure and covert hierarchisation of types of knowledge may result in the practical neutralisation of the inputs of, for instance, indigenous representatives in the census design process. The contents of social identities are always open to contestation. Debates within the social sciences illustrate this point, with experts still finding it difficult to define and establish a widely accepted and analytically sound differentiation between processes of identification based on “race” and “ethnicity”, for instance. It is not surprising that in censuses the conceptual confusion between these categories is still unresolved, and also that many respondents present what in analytical terms some denominate “inconsistencies” in response to questions related to these categories. Popular conceptions of what constitutes a race or an ethnic group vary just as scientific ones do, and it is difficult to identify them as prevalent without considering in-group differences in terms of social class, nationality, age or gender. Even if census programmes place more emphasis on developing conceptual conciseness in the area of measuring social identities in a way that attunes them as much as possible to those popular conceptions, and even if those concepts were to become clearly accessible and explained before census programmes, “inconsistencies” will probably continue to be an intrinsic part of counting people according to racial or ethnic self-identifications. In this regard, the ethnographic study of everyday categorisations and classification practices can hardly keep up with an ever changing “official” modification of census categories over the past few decades (or at least with the continuous fluctuation of results in response to the same type of categorisations). Ethnography-based studies focus on interaction in local or micro-spheres from which categorisations and classifications can be elicited and, though they provide a depth and richness of knowledge that cannot be matched by macro-interactionist approaches, they have generally been regarded as works of cultural critique rather than as sources of proposals for census categorisations, for instance. Ethnographically grounded anthropological work in particular, and qualitative research in general, could indeed make more explicit efforts to reach the sphere of political decision in census-making; but this would only become a fruitful exercise if bureaus of statistics became more open to informed discussion and invest more decisively in civic engagement and grounded public debate in order to inform their praxis.

Race, Ethnicity and National Censuses in Latin American States

35

Bibliographical references Anderson, Benedict. 1991 [1983]. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso. Ardao, Arturo. 1963. Assimilation and Transformation of Positivism in Latin America. Journal of the History of Ideas, 24(4): 515-522. Assies, Willem et al. 2000. The Challenge of Diversity: Indigenous Peoples and Reform of the State in Latin America. Amsterdam: Thela Thesis. Baker, Lee D. 1998. From Savage to Negro: Anthropology and the Construction of Race, 1896-1954. Berkeley: University of California Press. Beckett, Jeremy. 2010. National Anthropologies and their Problems. In Jon Altman and Melinda Hinkson (eds) Culture Crisis: Anthropology and Politics in Aboriginal Australia. Sydney: University of New South Wales Press, pp. 32-44. Briones, Claudia and Sabine Kradolfer. 2010. Dilemas y paradojas de la internacionalización de los movimientos indígenas en América Latina: una introducción. Bulletin de la Société suisse des Américanistes, 70 (Inter-políticas indígenas y antropológicas): 11-19. Brubaker, Rogers, Mara Loveman and Peter Stamatov. 2004. Ethnicity as Cognition. Theory and Society, 33: 31-64. Brysk, Allison. 2000. From Tribal Village to Global Village: Indian Rights and International Relations in Latin America. Stanford: Stanford University Press. Burns, E. Bradford. 1980. The Poverty of Progress: Latin America in the Nineteenth Century. Berkeley: University of California Press. Bushnell, David and Neil Macaulay. 1994 [1988]. The Emergence of Latin America in the Nineteenth century. Oxford: Oxford University Press. Byrne, David. 1999. Social Exclusion. Buckingham: Open University Press. Cañizares-Esguerra, Jorge. 2007. Creole Colonial Spanish America. In Charles Stuart (ed.) Creolization: History, Ethnography, Theory. Walnut Creek, CA: Left Coast Press, pp. 26-45. CELADE/CEPAL–Centro Latinoamericano y Caribeño de Demografía/ Comisión Económica para América Latina y el Caribe. 2009. Censos 2010 y la inclusión del enfoque étnico: hacia una construcción participativa con pueblos indígenas y afrodescendientes de América Latina. Santiago de Chile: Naciones Unidas (CEPAL). CEPAL- Comisión Económica para América Latina y el Caribe. 2006. Pueblos indígenas y afrodescendientes en América Latina y el Caribe:

36

Chapter One

información sociodemográfica para políticas y programas. Santiago de Chile: Naciones Unidas (CEPAL). Charters, Claire and Rodolfo Stavenhagen (eds). 2009. Making the Declaration Work: The United Nations Declaration on the Rights of Indigenous Peoples. Copenhagen: IWGIA. Clark, A. Kim. 1998. Race, “Culture”, and Mestizaje: the Statistical Construction of the Ecuadorian Nation, 1930-1950. Journal of Historical Sociology, 11(2): 185-211. Cooper, Richard, Jay Kaufman and Rik Ward. 2003. Race and Genomics. The New England Journal of Medicine, 348(12): 1166-1170. Del Popolo, Fabiana. 2008. Los pueblos indígenas y afrodescendientes en las fuentes de datos: experiencias en América Latina. Santiago: CEPAL. Del Popolo, Fabiana, Ana María Oyarce, Susana Schkolnik and Fernanda Velasco. 2009. Censos 2010 y la inclusión del enfoque étnico: hacia una construcción participativa con pueblos indígenas y afrodescendientes de América Latina. Santiago de Chile: Naciones Unidas. Desrosières, Alain. 2008a. L’argument statistique I. Pour une sociologie historique de la quantification. Paris: Presses de l’Ecole des Mines. —. (2008b). L’argument statistique II. Gouverner par les nombres. Paris: Presses de l’Ecole des Mines. Díaz Polanco, Héctor. 1978. Indigenismo, populismo y marxismo. Nueva antropología, III (9): 7-32. —. 1997. Indigenous Peoples in Latin America: the Quest for SelfDetermination. Boulder: Westview Press. Economic and Social Council. 2004. Workshop on Data Collection and Disaggregation for Indigenous Peoples New York, 19 to 21 January 2004. New York: Permanent Forum on Indigenous Issues. [online: http://nas.ucdavis.edu/varese/DataforIndPeople.html, accessed 30 June 2012] Foster, Morris W. and Richard R. Sharp. 2002. Race, Ethnicity, and Genomics: Social Classifications as Proxies of Biological Heterogeneity. Genome Research, 12: 844-850. Foucault, Michel. 2008. The Birth of Biopolitics. Lectures at the Collège de France, 1978-1979. Basingstoke: Palgrave Fraser, Nancy. 2000. Rethinking Recognition: Overcoming Displacement and Reification in Cultural Politics. New Left Review, 3: 107-120. Fraser, Nancy and Axel Honneth. 2003. Redistribution or Recognition: a Political-Philosophical Exchange. London: Verso.

Race, Ethnicity and National Censuses in Latin American States

37

Giraudo, Laura. 2006. “No hay propiamente todavía instituto”: los inicios del Instituto Indigenista Interamericano (abril 1940 – marzo 1942). América Indígena, LXII (2): 6-32. Goldberg, David T. 1994. Multiculturalism: a Critical Reader. Oxford: Blackwell. Goldberg, David T. (ed.). 1997. Racial Subjects: Writing on Race in America. New York: Routledge. Gundermann, Hans, Jorge Vergara and Rolf Foerster. 2005. Contar a los indígenas en Chile. Autoadscripción étnica en la experiencia censal de 1992 y 2002. Estudios Atacameños, 30: 91-115. Hale, Charles. 2002. Does Multiculturalism Menace? Governance, Cultural Rights and the Politics of Identity in Guatemala. Journal of Latin American Studies, 34: 485-524. Haller, John S. 1971. Outcasts from Evolution: Scientific Attitudes of Racial Inferiority, 1859-1900. Urbana: University of Illinois Press. Harris, Marvin, Josildeth Gomes, Joseph Lang and Bryan Byrne. 1993. Who are the Whites? Imposed Census Categories and the Racial Demography of Brazil. Social Forces, 72(2): 451-462. Hewitt de Alcántara, Cynthia. 1984. Anthropological Perspectives on Rural Mexico. London: Routledge. Hobsbaum, Eric. 1996. Identity Politics and the Left. New Left Review, 217: 38-47. Jhon Antón et al. 2009. Afrodescendientes en América Latina y el Caribe: del reconocimiento estadístico a la realización de derechos. Santiago: Naciones Unidas (CEPAL). Kertzer, David and Arel, Dominique. 2002. Census and Identity: the Politics of Race, Ethnicity and Language in National Censuses. Cambridge: Cambridge University Press. Kierans, Ciara and Jessie Cooper. 2011. Organ Donation, Genetics, Race and Culture: The Making of a Medical Problem. Anthropology Today, 27(6): 11-14. Kradolfer, Sabine. 2011. The Transnationalisation of Indigenous Peoples’ Movements and the Emergence of New Indigenous Elites. International Social Science Journal, 202: 377-388. Krieger, Nancy. 2000. Editorial: Counting Accountably: Implications of the New Approaches to Classifying Race/Ethnicity in the 2000 Census. American Journal of Public Health, 90(11): 1687-1689. Lins, Gustavo and Arturo Escobar. 2006. World Anthropologies: Disciplinary Transformations within Systems of Power. London: Berg. Lomnitz, Claudio. 2001. Deep Mexico, Silent Mexico: an Anthropology of Nationalism. Minneapolis: University of Minnesota Press.

38

Chapter One

Lovell, Peggy and Charles Wood. 1998. Skin Color, Racial Identity, and Life Chances in Brazil. Latin American Perspectives, 25(3): 90-109. Magno, José A., Charles H. Wood, Flávia Cristina and Drummond Andrade. 2004. Estimating the Stability of Census-Based Racial/Ethnic Classifications: the Case of Brazil. Population Studies, 58(3): 331-343. Martínez Cobo, José. 1986. Study of the Problem of Discrimination against Indigenous Populations. Geneva and New York: UNO. http://www.un.org/esa/socdev/unpfii/en/spdaip.html, accessed 1 June 2012 Martínez Novo, Carmen. 2007. Antropología indigenista en el Ecuador desde la década de 1970: compromisos politicos, religiosos y tecnocráticos. Revista Colombiana de Antropología, 43: 335-366. Méndez, Marta and David Martín. 2006. Introducción al proyecto de Declaración de las Naciones Unidas sobre los Derechos de los Pueblos Indígenas. AIBR, Revista de Antropología Iberoamericana, 1(2): 331361. Méndez-Reyes, Johan and Lino Morá-Beltrán. 2009. Las ideas positivistas y evolucionistas en la obra de Laureano Vallenilla Lanz. Revista de Filosofía, 61: 41-64. Mezey, Naomi. 2003. Erasure and Recognition: the Census, Race and the National Imagination. Northwestern University Law Review, 97(4): 1701-1768. Miller, Marilyn G. 2004. Rise and Fall of the Cosmic Race: the Cult of Mestizaje in Latin America. Austin: University of Texas Press. Nobles, Melissa. 2000. Shades of Citizenship: Race and the Census in Modern Politics. Stanford: Stanford University Press. Petruccelli, José Luis. 2007. A cor denominada. Estudos sobre a classificação racial. Rio de Janeiro: DP&A/LPP-UERJ. Primera Declaración de Barbados: por la liberación del indígena. 1971. Barbados. Saldivar, Emilio. 2011. Everyday Practices of Indigenismo: an Ethnography of Anthropology and the State in Mexico. The Journal of Latin American and Caribbean Anthropology, 16(1): 67-89. Schulte-Tenckhoff, Isabelle. 1997. La question des peuples autochtones. Bruxelles, Paris: Bruylant et Librairie Générale de Droit et de Jurisprudence. Skerry, Peter. 2000. Counting on the census: race, group identity, and the evasion of politics. Washington: Brookings Institution Press. Schkolnik, Susana and Fabiana Del Popolo. 2005. Los censos y los pueblos indígenas en América Latina: una metodología regional. Santiago de Chile: CEPAL/CELADE.

Race, Ethnicity and National Censuses in Latin American States

39

Simon, Patrick. 2005. The Measurement of Racial Discrimination: the Policy Use of Statistics. International Journal of Social Science, 183: 9-25. Skidmore, Thomas E. 1995 [1974]. Black into White: Race and Nationality in Brazilian Thought. Durham: Duke University Press. Stepan, Nancy L. 1991. The Hour of Eugenics: Race, Gender and Nation in Latin America. Ithaca: Cornell University Press. Stocking, George W. 1968. Race, Culture and Evolution: Essays in the History of Anthropology. New York: The Free Press. Sulmont, David. 2011. Race, Ethnicity and Politics in Three Peruvian Localities: An Analysis of the 2005 CRISE Perceptions Survey in Peru. Latin American and Caribbean Ethnic Studies, 6(1):47-78. Taguieff, Pierre-André. 2001. The Force of Prejudice: on Racism and its Doubles. Minneapolis: University of Minnesota Press. Taylor, Charles. 1994. The Politics of Recognition. In Amy Gutman (ed.) Multiculturalism. Princeton: Princeton University Press. Telles, Edward. 2002. Racial Ambiguity among the Brazilian Population. Ethnic and racial studies, 25(3): 415-441. Telles, Edward and Nelson Lim. 1998. Does It Matter Who Answers the Race Question? Racial Classification and Income Inequality in Brazil. Demography, 35(4): 465-474. Urrea, Fernando. 2011. Contar y ser contados. El Censo 2005 y las minorías étnicas. In Margarita Chaves (ed.) La multiculturalidad estatalizada. Bogotá: Instituto Colombiano de Antropología e Historia ICANH. Valdivia, Néstor. 2011. El uso de categorías étnico/raciales en censos y encuestas en el Perú: balance y aportes para una discusión. Lima: GRADE. Valle Silva, Nelson do. 2000. A Research Note on the Cost of Not Being White in Brazil. Studies in Comparative International Development, 35(2): 18-28. Van Cott, Donna Lee. 2000. The Friendly Liquidation of the Past: the Politics of Diversity in Latin America. Pittsburgh: University of Pittsburgh Press. Wade, Peter. 1997. Race and Ethnicity in Latin America. London: Pluto Press. Winant, Howard. 2004. The New Politics of Race: Globalism, Difference, Justice. Minneapolis: University of Minnesota Press. Wright, Winthorp R. 1990. Café con leche: Race, Class and National Image in Venezuela. Austin: Texas: University of Texas Press.

40

Chapter One

Žižek, Slavoj. 1997. Multiculturalism, or the Cultural Logic of Multinational Capitalism. New Left Review, 225: 28-51.

Notes 1

For a classic conceptualisation of this term associated with the liberal theory of multiculturalism and differentiated citizenship, see Taylor (1994). For a discussion of politics of recognition as associated with a status model in which “misrepresentations” of groups are translated into tangible forms of institutional subordination that are not merely cultural, see Fraser (2000); for a dialogue on the possible articulations of recognition and redistribution politics, see Fraser and Honneth (2003). 2 For a comparative study of the political imperatives of statistics used in antidiscrimination policy in the USA, the UK, Canada, and Australia, see Simon (2005). 3 The situation in Colombia and Panama differed, the latter being still in the 1870s a department of Colombia. It was in 1911 that the first modern census in Panama as an independent nation was conducted. Before positivism-influenced censuses, Colombia and Bolivia had carried out nineteenth century national censuses in 1825-1827 and 1831. 4 For an overview of the influence of positivism in Brazil, see Skidmore (1995: 1014); for the case of Venezuela, see Méndez-Reyes and Morán-Beltrán (2009). 5 The Argentinean Constitution (1994) maintains vestiges of this orientation (with antecedent in the 1853 Constitution): in the country in which Juan B. Alberdi coined the famous dictum that “to govern is to populate”, article 25 compels the federal Government to promote “European immigration”. For an illustration of restrictions on the immigration of Blacks and Chinese and the favouring of immigrants of European ancestry in Brazil, see Skidmore (1995: 192-200). 6 All the translations are by the authors. 7 For comments on the influence of Mexican indigenismo and anthropology in Ecuador, see Martínez Novo (2007: 340, 345-346). 8 However, the impact of this document in various countries (and especially in Latin America) predates its acceptance by the Assembly General, for elements of the draft declaration influenced legislation since the 1990s (see Charters and Stavenhagen, 2009). 9 The wording of the question was grammatically confusing and should have rather been “¿Pertenece a algún pueblo o etnia indígena?” (instead of the existing “¿Pertenece a algún pueblo indígena o etnia?”). Asking if one belongs to an indigenous people or to an ethnie can be misinterpreted: without an adjective that indicates otherwise, an ethnie could also include all those which are not considered “indigenous” ethnies.

CHAPTER TWO ARE THERE STILL “INDIANS” IN ARGENTINA? INDIGENOUS PEOPLES AND THE 2001 AND 2010 POPULATION CENSUSES PILAR BARRIENTOS

1. Introduction This chapter seeks to highlight certain aspects of the process for including the so-called “indigenous variable” in the 2001 Censo Nacional de Población, Hogares y Vivienda (National Population, Household and Housing Census) and its Encuesta Complementaria de Pueblos Indígenas 2004-2005–ECPI (Supplementary Survey of Indigenous Peoples 20042005), an operation which provided more detailed data on households who stated that at least one of their members belonged to, or was a descendant of, an indigenous people. I shall first give a brief overview of the history of ethnic monitoring in Argentina, and of how the legislation that required the question to be included on the census questionnaire came into being: What criteria were taken into account? I shall then focus on the methodologies used to implement the ECPI and the criteria used to select some of its variables; I shall then go on to comment on how the data were presented and the implications this had for their use. Lastly, I will look at the inclusion of questions on ethnicity and people of African descent in the last population census (2010). In this chapter I have taken an anthropological approach, giving preference to qualitative processes, and have sought to analyse the areas of agreement/disagreement between state organisations and indigenous organisations and the complexity of developing intercultural mechanisms for bringing the two together. Having worked on different stages of implementation of the ECPI as a member of the training team from the Instituto Nacional de Estadística y Censos–INDEC (National Statistics

42

Chapter Two

and Census Institute), I am familiar with those areas1. The post enabled me to sit as a participating observer on several different bodies, attending both meetings to plan the survey operation and those at which the variables to be included in the questionnaire were decided. What took place at such meetings prompted me to look for conceptual tools from within the field of anthropology with which to analyse what happens when “difference” is encountered in the state sphere. The question that provides the title for this chapter was put to me by an official from the very body that was charged with counting all those who affirmed that they were members or descendants of an indigenous people.

2. Great Argentine fictions “Doesn’t Argentina have anything to do with Indians?” (Viñas, 1983: 18)2

In Imagined Communities, Anderson argues that censuses are presented as a fiction in which “everyone is in it [and] everyone has one–and only one–extremely clear place” (2006: 166). The national population, household and housing censuses conducted in Argentina are no exception. In the tradition of national censuses, which were begun in 1869 by Domingo F. Sarmiento–a statesman who advocated a positivist approach to public administration–large-scale statistical surveys have come to be seen as a patriotic act and are the largest civilian operation carried out in peacetime (INDEC, 1991: 1). Their implementation involves thousands of human resources and taking part in them, both as a census taker and as a respondent to the questionnaire, assumes the status of a “public obligation”, a civic duty and a duty incumbent on all inhabitants. Censuses thus create the illusion, proclaimed from the advertisements announcing them, that each and every one of the country’s inhabitants has to be counted and located territorially. Pursuing this idea, they can be likened to Shumway’s notion of “guiding fictions” which “cannot be proven, and indeed are often fabrications as artificial as literary fictions. Yet they are necessary to give individuals a sense of nation, peoplehood, collective identity, and national purpose” (1993: 13). Over and above the information it provides, a census contributes to the idea, in the sense used by Anderson (2006: 33), of “imagined linkage” that stems from calendrical coincidence and the associated publicity disseminated in the media. On that day all of us living in Argentina will be counted and thereby become “part of the supracommunity”, the nation.

Are There Still “Indians” in Argentina?

43

In the case of Argentina, the so-called Primer Censo Nacional de Población (First National Population Census) in 1869 appeared in the context of building a liberal nation. It was a tool not only for generating policies that intervened in people’s lives but also for constructing a hegemonic discourse about Argentine identity. That census was the starting point for categorising and classifying the population according to certain variables. And although there were some notable omissions in the count (especially in the Chaco and Patagonia, where resistance from the indigenous inhabitants prevented its implementation), its results were deemed to give a true and accurate picture of reality. The process of selecting variables in statistical operations has an ideological correlate. The variables included in questionnaires, as well as those omitted, set the parameters for describing the Argentine nation. Amadassi and Massé, quoting Otero, point out that, “if the Desert/Patagonia Campaign of 1879 and the Chaco Campaign of 1884 succeeded at the military level in eliminating the indigenous peoples in those territories which up to then had yet to be incorporated into the Argentine nation-state, the corresponding national population censuses at the end of the nineteenth and beginning of the twentieth centuries expressed through the census process what arms expressed in the conquest of territory” (2005: 3). By using a scientific discourse, demography legitimised the “awareness [of the apparent absence of indigenous peoples] possible on the part of the military and political movements which meant that the capture of a bountiful land was guaranteed” (González in Viñas, 1983: 5). Specific actions of different kinds converged in an effort to find homogeneity and uniformity so that a fictional sociological discourse that reinforced the idea of a European–or North American-style white nation was created–together with, in the words of Viñas (1983), a discourse of silence. This silence has been maintained throughout almost every national population census, in keeping with the development of a literary and journalistic discourse that confirmed the absence of indigenous people from within the frontiers of the civilised nation, thereby appearing to reinforce Renan’s view that “forgetting, I would even go so far as to say historical error, is a crucial factor in the creation of a nation” (1999: 56).

3. A cry in the silence When the 500th anniversary of the arrival of Christopher Columbus in the Americas was being commemorated, what had lain dormant in Latin America gradually came to the surface. The indigenous presence, as well

44

Chapter Two

as the genocide that had been perpetrated, became visible. This constituted “indigenous emergence” (Bengoa, 2009: 7), in other words, the presence of new identities and ethnic expressions, the assertion of demands and claims by indigenous peoples who had found an opportunity to make themselves heard through the use of different strategies. One such strategy was to get indigenous issues onto the state agenda and at the right moment: namely, the convening of the Constituent Assembly in 1994 for the purpose of amending the Constitution. Section 75, paragraph 17 of the new Argentine Constitution specifically recognises and grants force of law to the pre-existence of indigenous peoples by empowering Congress: To recognise the ethnic and cultural pre-existence of indigenous peoples of Argentina. To guarantee respect for their identity and their right to bilingual and intercultural education; to recognise the legal capacity of their communities, and the community possession and ownership of the lands they traditionally occupy; and to regulate the granting of other lands adequate and sufficient for human development; none of them shall be sold, transmitted or subject to liens or attachments. To guarantee their participation in issues related to their natural resources and in other interests affecting them. The provinces may jointly exercise these powers.

Later on, in 1998, and for the purposes of “quantifying the population that recognises itself as being indigenous or the descendants of indigenous people” (INDEC, 2006: 4), Congress passed Aboriginal Census Law 24,956 which ordered indigenous monitoring to be included in the next national population census on the basis of self-recognition. Article 1–Self-identification of identity and membership of aboriginal communities shall be incorporated into the National Population and Housing Census for the year 2000 by expanding the modules established within it.

In order to do so and to comply with the requirements of the law in question, INDEC launched a “process of consultation with governmental and non-governmental bodies involved with the issue and with organisations of indigenous peoples, to be used as the basis for drafting a comprehensive methodological proposal, namely one that amalgamates different measuring tools” (INDEC, 2006: 6). In this way a completely unprecedented mechanism was developed. On the one hand, an “indigenous variable” was to be included in the national demographic census itself and, on the other, there was to be a supplementary survey to the census, in the design and development of

Are There Still “Indians” in Argentina?

45

which the interested actors themselves, in other words, representatives of the different indigenous peoples, had to participate. Therefore, even though INDEC was invoking a state requirement, the mechanism was in fact presented as being a response to a clear demand, an expression of the struggle being waged by organisations, within the context of the abovementioned “indigenous emergence”, as a Latin American movement seeking to raise the profile of peoples who, though silenced, had been in existence for years. It is interesting to look at the tensions between the ability of the indigenous peoples to convey their demands and the steps taken by the state, which put resources at their disposal for the purposes of “enforcing the law” within the framework of a “politically correct” rhetoric. What is striking is the gulf between that rhetoric and the specific actions that were taken. For example, little effort was made to disseminate the data relating to indigenous people to society as a whole. It amounted to illusory strategies in which the indigenous participation required under the 1994 constitutional amendment (Section 75, paragraph 17) was only superficially implemented and a kind of “theatre of operations” recreated, possibly becoming, in what may be a risky hypothesis, another fiction that largely but not entirely guides the relationship between the indigenous peoples and the state.

4. And so, how can it be done? Organisational aspects Ensuring that an ethnic monitoring question was included in the 2001 Census and its supplementary survey presented an unprecedented difficulty. It was not a question of putting together a team of demography experts who would draft the question and test out “how it worked”. The genuine and actual participation of the indigenous peoples themselves had to be taken into account, in accordance with both constitutional law and the international recommendation made in Article 6 of ILO Convention 169 (1989), which Argentina signed in 1992 (and ratified in 2000). Article 6 1. In applying the provisions of this Convention, governments shall: (a) consult the peoples concerned, through appropriate procedures and in particular through their representative institutions, whenever consideration is being given to legislative or administrative measures which may affect them directly; (b) establish means by which these peoples can freely participate, to at least the same extent as other sectors of the population, at all levels of decision-making in elective institutions and administrative and other bodies responsible for policies and programmes which concern them;

46

Chapter Two (c) establish means for the full development of these peoples’ own institutions and initiatives, and in appropriate cases provide the resources necessary for this purpose.

How to go about this task? Where to start? What was the motivation for this meeting with Others? Would they know how to do this work? Among this evolving set of standard state-controlled questions used in both the census and the ECPI questionnaires, there were also unwritten questions: questions about the Other; about the Others; about other indigenous people and other white people. It is worth emphasising that this meeting I am talking about was not taking place at an ahistorical moment or in a virtual space. It was taking place in a specific place at a specific time. And was part of a process of dialogue between representatives of indigenous organisations and representatives of the state, the aim of which was to enable a complicated job to be done. I should clarify that, in the context of this article, the two interlocutors are not seen as homogeneous wholes, two solid opposing blocks. Although for these purposes I shall refer to two interlocutors, each of them is seen as speaking with many different voices, with different interests, transmitting a variety of messages from one or other side of the argument. In turn, the discourse of each interlocutor was not fixed but kept changing as the tasks to be done to implement the survey were carried out. In this dialogue, while it was clear that the state exercised hegemonic power, it did not have a power monopoly. It could be said that there were webs of power. For example, the power held by those indigenous organisations that managed to secure a seat at the negotiating table where questions to be included in both the census and the ECPI were designed. These powers are bound up with ownership processes that have been constructed on the basis of those organisations and with a series of institutionalised practices that have their own history and which, in order to be made legitimate, ought, in any event, to be framed in law. Though the demand for the indigenous variable to be included had come from organised indigenous peoples, the state redefined it and set the legal and technical framework in which had to be implemented. For their part, the indigenous organisations maintained an openly confrontational discourse with the state while at the same time accepting these state arrangements which entailed their having a relationship with it– a relationship of demand, negotiation and opposition to certain practices. Among the indigenous representatives, there was no single discourse; even though there may have been many points of agreement, the discourse varied, even among the representatives of a single people. One of the clearest examples was the differences found among the Mapuche people,

Are There Still “Indians” in Argentina?

47

where, on the one hand, representatives of several organisations, including Mapuche Newentuayiñ, Organización de Jóvenes Mapuche We Kimvn, We Kvyeh and Grupo Kojog, complained in a public statement that “this survey acts as a filter so that the number of indigenous people considered in the statistics is further reduced” (2001) and, on the other, two of their representatives were regional coordinators in the survey structure and actively involved in designing the questionnaire. I have described this panorama in order to give an idea of how complex it was to think in these terms. It was a space in which tensions and disputes arose not only in “western” political terms but also within a discourse laden with spirituality that had a community-oriented world view, something which was disconcerting for state officials who, though careful to involve “the Indians”, in reality doubted that they could participate effectively from the technical point of view.

5. How ethnicity was addressed in the 2001 census “By hypostatising a data result, a specific outcome was turned into something that was inevitably determined to the point where it became fiercely mechanistic; and by projecting the social aspects of rivalry onto Darwinian factors, the traditional was seen as passive, immobile, a thing, matter.” (Viñas, 1983: 79)

Within the basic methodological definitions used in statistical research, the word “variable” refers to “a property, characteristic or attribute that may be found in certain subjects or which may be found in different degrees or forms. They are qualifying concepts that allow individuals to be placed in categories or classes and are capable of being identified or measured” (Briones, 1987: 34). In common parlance, it means something that is unstable, inconstant, mutable, changeable. In order to conduct social statistical research, variables are determined, specific attributes of the population to be studied are highlighted and the subjects are classified in line with such variables. As mentioned earlier, in 2001 an “indigenous variable” was incorporated into the census. This variable was determined on the basis of at least one member of a household recognising that they were a member or descendant of an indigenous people, and was presented as follows in the census questionnaire:

48

Chapter Two

Figure 2-1. Census Questionnaire. 2001 National Population, Household and Housing Census. Question No. 2. “Self-recognition” as a descendant or member of an indigenous people”.

Source: INDEC 2001a.

If the answer was yes, the interviewer had to ask the respondent which people they belonged to or were a descendant of and wait for their reply. The template listed 17 peoples. The 17 peoples who in 1998 had registered their legal status with the Instituto Nacional de Asuntos Indígenas–INAI (National Institute for Indigenous Affairs), the body responsible for implementing the procedure for granting legal status to indigenous communities, whose registration in the Registro Nacional de Comunidades Indígenas–Re.Na.C.I. (National Register of Indigenous Communities) implies formal recognition of their ethnic and cultural pre-existence, in accordance with the provisions of Resolution N° 4,811/96 of the former Secretaría de Desarrollo Social (Ministry of Social Development). Other variables which are hard to monitor operated on this first question about indigenous otherness–subjective variables, representations which are shaped by the hegemonic discourse, the one that is maintained from the very earliest school days and imbues perceptions. Thus, when I was accompanying census takers in urban areas, I observed the question being omitted, and the “No” category being ticked, when the census takers–teachers, civil servants and members of the Armed Forces–decided that the interviewee did not have the “look” of an indigenous person, in other words, did not have the phenotypic features that essentialise indigenous people in Argentina. In other cases, the question was openly prompted: “No one in this household recognises that they are a member or descendant of an indigenous people, do they?”. Despite the efforts made in the course of training the census takers, when the importance of correctly formulating the question was emphasised, the prevailing social representations of indigenous peoples carried more weight than the intentions of the trainers. All the same, the question was answered

Are There Still “Indians” in Argentina?

49

extensively in the cities, resulting in data that suggests that 50% of recognised indigenous people live in urban areas. The criterion of self-recognition favoured by international organisations calls for self-identification. The criterion of descent refers back to “our ancestors” and has a hint of essentialism. The so-called “indigenous variable” seeks to monitor ethnicity through identity, assuming that the two are equivalent and synonymous. It stems from a “supra-ethnic” category (Bartolomé, 1987: 19), that of “indigenous peoples”, which classifies the options in advance and at the same time specifies them. The 17 peoples are presented as alternatives on the basis of a bureaucratic criterion: as mentioned earlier, in order for a person to be recognised as a member of an “indigenous” community, that community has to have legal status. And so, in Argentina, the state ascribes itself the right to recognise who qualifies as indigenous and who does not on the basis of an administrative procedure–the registration of communities in the Re.Na.C.I. The design of these state-run statistical operations shows the tension that exists between this ascribing of identity on the basis of external recognition and self-identification. These categories emerged as a result of bringing together the information provided by the communities who had submitted documents and gone through the necessary procedures. In other words, those considered as such were those who had satisfied the conditions laid down in Article 2 of resolution 4,811/96 of the country’s former Ministry of Social Development which lists the information that an indigenous community needs to supply in order to be recognised by INAI: name and geographical location, a profile confirming its ethnic-cultural and historical origin, including the provision of documentation, a description of its patterns of organisation, a list of members, including their degree of kinship, and the mechanisms it uses to include or exclude members. As far as the 2001 Census is concerned, it should be noted that the data relating to the presence within a household of one or more persons who recognised that they were descendants or members of an indigenous people have not been published. However, as stated in official publications, this was the quantitative data that was used to design the statistical sample with which the ECPI would be implemented: In 2004 and 2005 INDEC conducted the ECPI in order to quantify and depict the population that recognises itself as belonging to and/or descendants of indigenous peoples. The framework used was those households in which the 2001 National Population, Household and Housing Census recorded at least one person as belonging to and/or a descendant of indigenous peoples.

50

Chapter Two

6. From the census to the supplementary survey The ECPI required a particular methodological approach. It was a lengthier operation than the census, consisting of two years of field work– across the whole country–in which the effective participation of “the Indians” had to be ensured. Evidence of INDEC’s position can be found in its methodological documents: Given the lack of experience in the subject area, efforts were made to come up with a mechanism that could include indigenous organisations. One of the most striking features of the ECPI is the design of its institutional architecture. There were two reasons why particular care was taken with it. The first was lack of experience in the subject area, which made it crucial to add a specifically sectoral view to the ample experience of INDEC in social and population statistics. In practice, therefore, the ECPI was the product of joint work between INDEC and INAI. The second reason is the aforementioned constitutional framework which specifically determines that the National Constitution shall ensure the participation of indigenous peoples in the management of interests affecting them. The carrying out of statistical measuring that makes it possible to report on their numbers and living conditions and which also becomes an important potential data source for formulating public policy is undoubtedly in the interests of those peoples. The institutional architecture designed in order to implement the ECPI therefore took their participation into account.3

INDEC ensured that indigenous peoples participated in the ECPI by involving individuals belonging to such peoples in different aspects of the survey, especially the conceptual design of the questionnaire, the training of interviewers, awareness-raising and the actual implementation of the survey. At national level, a structure was set up consisting of six Coordinadores Regionales Indígenas–CRI (Regional Indigenous Coordinators), one for each of the six regions in which, at INAI’s suggestion, the field work was organised (INDEC, 2006: 8). Using the census results that had yielded a positive answer to the question referred to above, a representative sample (consisting of 57,000 cases) was chosen from throughout the country, with which various topics of interest to indigenous organisations and others of interest to the population studies INDEC normally carries out were to be examined in more detail. The ECPI was carried out by means of a direct interview in which a series of questions was put to each member of the household. There were two levels of organisation:

Are There Still “Indians” in Argentina?

51

1) At national level, there was a coordinating group made up of technical teams from INDEC and INAI and the 6 CRIs that were administratively answerable to INDEC. 2) At provincial level, the survey was carried out in accordance with national guidelines, using a team made up, on the one hand, of provincial referents, motivators, aides, interviewers and instructors, who were members of indigenous peoples, and, on the other, coordinators, supervisors, receptionists and provincial state officials, who were “white”. In the preparatory stages, at national level joint work was done with the CRIs (representing the Mapuche, Toba, Kolla and Rankulche peoples) on designing the questionnaire and drafting training materials. INDEC proposed the training model, about which the CRIs were consulted. Consensus was reached on the use of a “pair teaching” mechanism in which instructors from the indigenous peoples would run the courses jointly with an instructor from INDEC. The former would explain to the respondents how to fill in the questions on the specific variables relating to indigenous topics and the INDEC instructors would deal with the part containing the questions related to the comparative variables used in the 2001 Census. The CRIs were present throughout this process as consultants and supervisors. However, the division of labour was clear. It is interesting to point out that these discussions exhibited the dyadic relationship highlighted by Miguel Bartolomé (1997) between “people of custom”–the indigenous organisers who were assigned the topics relating to ancestral cultural practices–and “people of reason”–the professionals from INDEC who were responsible for matters relating to statistical consistency.

7. Discussion of the survey’s thematic content The ECPI was a statistical investigation conducted within the framework of official statistics, in other words, data that is collected and organised by and from the state. Even if its technical nature redeems it, the fact that it is “statistical” and “official” puts it on a political plane. In the words of Foucault (1999), it is a form of biopolitics, a style of government that regulates populations through biopower. Some of the variables considered in the survey were: - Self-recognition and indigenous ancestry - Knowledge and use of the indigenous language - Work - Education

52

Chapter Two

-

Migration Fertility System of land tenure Housing conditions Specific indigenous topics: Problems affecting them, use of ancestral cultural patterns, communal work. The variables listed by Foucault as essential for studying populations (fertility, work, migration, housing conditions and education) were the ones INDEC and INAI deemed unmodifiable. They were non-negotiable. These decisions were taken on the basis of technical arguments which from that perspective had their merits. Those variables could not be changed; the decision to formulate them in one way and not another was based on the need for comparability that stemmed from taking the population census as the framework for the sample. The other variables (traditional medicine practices, communal ownership, among others) that the indigenous representatives wanted to be investigated could be discussed. For me the carrying out of the ECPI, with all its emphasis on the technical aspects–and despite it–, was, over and beyond the methodological arguments already mentioned, a political act and one that was important. During the discussions on the survey there was constant evidence of power-related negotiations, hence its political nature. From the outset the “classification/characterisation” of the teams as, on the one hand, technicians and, on the other, indigenous people placed the latter in “a secondary position” (Pacheco de Oliveira, 2007: 89). While it is true that they were consulted about the topics to be covered in the survey, the fact that they were not recognised as having the technical know-how required by a statistical operation divested them of any “scientific” authority. Although the legal requirement concerning participation appeared to be fulfilled, it was never more than partially met. It should be noted that one of the CRIs, the Kolla representative, was a leading member of the Coordinadora de Estudiantes Indígenas (Indigenous Students’ Coordinating Body) and only one module short of completing a degree in sociology at the University of Buenos Aires. The team of social demographers from INDEC took no account of her academic background in order to rethink the formulation of the questions or to shape the variables and the methodological design of the survey. She was only consulted in order to clarify points relating to the “indigenous topics”, about ceremonials, customs, etc., but not to examine how to apply the ideas that emerged from addressing those topics–as if it was not possible to combine both types of knowledge.

Are There Still “Indians” in Argentina?

53

The technical arguments were mainly monopolised by the INDEC team–although (as mentioned earlier) I believe that technical decisions have an inevitable ideological substrate. The political arguments were mainly monopolised by the CRIs. They constantly brought indigenous demands (around land tenure, communal work, child labour) to the discussion table. In the face of these differences, the technical discourse was the one authorised to settle disputes. The social demographers often said, “They don’t understand, allow us to explain to them that things have to be measured as they were in the census”. On several occasions, the indigenous leaders pointed out that “the census asked it that way, but it’s wrong”. Thus, each variable, each dimension and each category was a source of tension between the demands of the representatives of the indigenous peoples and the methodological needs of the members of the conceptual design team. In the following section I shall describe some of the most representative examples of this.

8. “Indians”, land, children Below are some situations that illustrate some of the tensions there were between the indigenous organisations and the state officials when it came to discussing three particular variables.

8.1. Variable: Identity. Dimension: Self-recognition as a member of an indigenous people Going back to the survey questionnaire, let us look again at the question: “Do you recognise that you are a member of an indigenous people?”. If the answer is yes, the respondent is then asked “Which people do you belong to?” And here, the interviewer–in all cases an indigenous person–must note and codify the response. The table of codes appears in the questionnaire itself. It consists of the 17 peoples mentioned above, together with seven others (Atacama, Ava guaraní, Comechingón, Charrúa, Lule, Tonocoté, Vilela) added both because the organisations representing those peoples requested it as well as because they were present in significant numbers according to the 2001 Census data. Is identity measurable? If it is considered to be something “fluctuating, historical and variable, the realisation of which is partly due to the changing structure of inter-ethnic systems” (Bartolomé, 1987: 9), are statistics the best tool for reporting on that dynamic? Taking this path of analysis may result in the political nature of the survey being disguised. The above-mentioned law 24,956/98 ordering an Aboriginal Census to be

54

Chapter Two

conducted was the result of a demand from indigenous organisations. As Carola Goldberg writes: These days, the resurgence of ethnic identities has become a tool for enabling indigenous peoples to exert political pressure on the dominant society to the extent that it allows positive images of the groups to be constructed. Ethnic awareness, which has often seemed to be confined to merely reclaiming past customs and practices, is now acting as a political tool for these groups (2007: 6).

The ECPI “is one more arena in which they can fight for their rights to be recognised and extended” (Pacheco de Oliveira, 2007: 198). And so, over and above whether it is technically viable to measure identity, it can be strategically useful to indigenous organisations for such a survey to be conducted. However, accepting this methodology as a step towards achieving recognition and visibility for indigenous peoples means accepting a reduction in the complexity of ethnic diversity. Statistics need to be classified. They have their origins in a biopower, in a need to implement a biopolicy on human populations. In addition, they are accompanied by a global governmentalisation of society, in the Foucauldian sense, meaning societies in which power is decentralised and the members play an active role in their own self-government (1999). And so the indigenous people themselves are acting in the measurement of their own interests. That makes sense. Guillaume Boccara sees the current state as “liberal multicultural”, a hegemonic power that tends to “hold ethnic groups responsible when things turn out badly or get out of its hands; to blame these same ethnic groups which it itself helped mould into their essentialised and reified forms and whose memories it helped to shape through colonial violence” (2006, 189) It essentialises them through the use of, among other things, one of the state’s most powerful ideological apparatuses, the education system, by including content in the curriculum such as “the extinction of the indigenous Argentine population” as a result of successive “Desert Campaigns” (which went on from 1820 until 1888 in Patagonia) or the “Gran Chaco Campaigns” (military incursions that took place between 1870 and 1917), and describing what were in fact campaigns of extermination as civilising. It reifies them, for example, and in a similar vein, by means of a school discourse that has objectified indigenous peoples as “savages”, “barbarians”, “backward” and part of the past. The actors involved have accumulated a great deal of historical experience as far as relating to “Others” is concerned. Given this background, I think it is valuable to see the tension between the state’s

Are There Still “Indians” in Argentina?

55

governmentalisation processes, on the one hand, and the strategies adopted by indigenous sectors, on the other, as experiences that foster the construction of those sectors as political subjects. The indigenous variable was unevenly applied throughout the ECPI’s implementation. It was included under “people” but not under “community”. When referring to “people”, the criterion of self-recognition was used while, as already explained, “community” referred specifically to the state’s recognition of some indigenous organisations, using the indicators mentioned, one of them being their geographical location. In the case of people who recognise themselves as indigenous and live in cities (which, according to the data from the ECPI itself, is the situation of about 50% of the population), there is tension between these concepts. For the Argentine state, a community is recognised as such if it lives as a group within a shared territory. The situation of some communities such as, for example, those belonging to the Kolla people, which–as a result of internal migration towards urban centres–are scattered across cities but call themselves communities, challenges this view and requires an approach that goes beyond the scope of this article. Indigenous identity may thus depart from its association with territory, as understood by the state, and be understood as a type of grouping that extends beyond its spatial reference and which has to do with cultural and linguistic links, feelings of belonging and bonds of solidarity. This raises a new problem, still under discussion: will indigenous people who recognise themselves as such but are not recognised by the state because they do not belong to a community that conforms to the parameters set by the authorities be considered indigenous?

8.2. Variable: Land tenure system “No Potosí, no mining for Argentina but land and more land for its agricultural destiny” (Viñas, 1983: 87)

The claiming back of land is one of indigenous peoples’ pivotal demands. A correlate of the massacres perpetrated since colonial times was the expulsion of the inhabitants from their territories and the appropriation of the latter for economic purposes. Since the constitutional reform of 1994, which recognises the ethnic and cultural existence of indigenous peoples prior to the constitution of the Argentine nation-state– referring to their presence as pre-Colombian peoples–, there has been talk of the “right to … community possession and ownership of the lands they traditionally occupy” (Section 75, paragraph 17, of the National Constitution).

56

Chapter Two

Over the past 20 years their claims have been united around the notion of “territory” and it is one of the main ideas around which the different stances taken by indigenous organisations converge. The use of the words “territories” and “ancestral” has political significance. Preparation of the ECPI questionnaire opened up space for a discussion of the variables to take place. This was one of the most controversial issues. Below is an extract from the minutes of the meeting at which this debate began: The purpose of the meeting was to reach agreement about the questionnaire to be used in the pilot study to be carried out in November in the Province of Salta and the City of Buenos Aires. The different parts of the prototype questionnaire were run through and questions and suggestions were shared. Land ownership: As presented, Question 23: Is the land your own? Answer options: Yes, No, Don’t know. The indigenous representatives said that it did not cover the “ancestral” situation, since they considered the territories to belong to the peoples even if legally they did not. The thematic team countered with a proposal to open the question with a) Who does this land belong to? b) Then specify whether the land is … your own … leased, etc. This proposal was discussed and it was suggested that there should be a direct question: Is tenure ancestral? Because for indigenous peoples there are no state-owned lands, only indigenous territories (INDEC, ECPI Minutes, 17/10/2003).

The Mapuche representatives insisted: We are talking about territory, not land, territory is broader, it encompasses the air, the ground, the water, the subsoil (INDEC, ECPI Minutes, 17/10/2003).

Nevertheless, and again citing the census (which by then had almost acquired the status of technical god), the word “land” (tierra) was kept. The next issue was community ownership of land. For the state officials present, operationalising this concept in the questionnaire was extremely difficult. Following arduous discussions, it was possible to arrive at a question that had no correlate in the census: “Do you consider the land on which you live to be your own?” The answer options were “Yes” or “No”, with no room for explanations. Also, to the question “In whose name is the property title?” the option “The community” was added. The questions included in the census on this issue had remained identical to the ones contained in the questionnaire used in 2001. In this way they were asked the question in two ways, with both points of view

Are There Still “Indians” in Argentina?

57

being taken into account. In the end, over and above the issues raised in the transcript, the questionnaire adopted the following approach for dealing with this topic: Figure 2-2. Questionnaire, Supplementary Survey of Indigenous Peoples of the Argentine Republic (2004-2005). Set of questions on the land tenure system.

Source: INDEC, 2004a.

The word “land” was not changed despite the arguments put forward by the CRIs and the provisions of Convention 169 which states that, “The use of the term ‘lands’ [tierras] (…) shall include the concept of

58

Chapter Two

territories, which covers the total environment of the areas which the peoples concerned occupy or otherwise use” (Convention 169, ILO, Art. 13.2, 1989). The category “ancestral” was omitted while, in the end, “state-owned” (fiscal) was kept in. Several arguments were put forward but the strongest was a technical one based on the need to compare the results with those obtained in the 2001 Census. The technical aspect was, once again, uncontested and uncontestable and concealed the fact that within the proposed participation power was unequally distributed. To date–31 October 2011–the information relating to “Characteristics of households”: “Table 10.9. Households in which the land is recognised as being owned in accordance with the requirements for legal land ownership” and “Table 10.10. Households that live on land that neither they nor the indigenous community own under the land tenure system according to land ownership” has not been made available on the INDEC website4. I believe that this variable is crucial. And in a way it shows how the state reserves the right to decide whether certain subjects can be debated. It shows “the rock-hard immovability of its classification system” (Gorosito Kramer, 2007: 6). Allowing inclusion of the category of ancestral territory would, then, have provided a space in which–even using state logic–laws and official data could be articulated–and, with them, the possibility to make claims based on information generated by the state itself. And so why did the coordinators of the indigenous organisations agree to this category not being included? Can the force of bureaucratic logic be stronger than political negotiation? Beyond the technical argument which, as has been pointed out, was the standard counterargument made in response to the indigenous representatives and a point of contention, the imposition of timetables, budgets that needed to be implemented and other urgent matters meant that, in many cases, the questionnaires presented were the only ones viable because “there was no time left to discuss concepts, they have to go off for printing now or we won’t make it”. Furthermore, from my 20 years’ experience of “state affairs”, the technicians themselves are convinced that bureaucratic procedures take precedence over any other logic. This attitude is part of the habitus of the professional working within the state sphere, in other words those things which, though acquired, become incorporated in a lasting way through the structuring of practices and naturalisation of actions (Bourdieu, 1988). And so, although the apparently solid wall of the state has cracks through which it is possible to raise grievances and obtain results, there is also a logic that limits demands and which could be deemed equally as “ancestral” as the territories the indigenous people are claiming for themselves: the inertia of state bureaucracy.

Are There Still “Indians” in Argentina?

59

8.3. Variable: Fertility There was a question in the questionnaire on women’s fertility: “How many children do you have who were born alive?” “In my community, you are going to have problems,” said Rosalía, a Kolla coordinator. “We all consider everyone’s children to be our own. You are not going to be able to count them in the way you want. We would ask how many children you have given birth to. The number you get will be wrong” The formulation of the question was retained because that is what the international recommendations on fertility say.

8.4. Behind the scenes The scenes described above show some of the tensions that arose with regard to the design of the survey questionnaire, in particular, between the so-called “sociodemographic topics” and the “indigenous topics”. In documents such as the Plan Guía para el Instructor de la ECPI (Guide for ECPI Instructors) (INDEC 2004c), the allocation of tasks involved in addressing each topic in the training courses for interviewers was also clearly delimited. In the course of negotiations, the technicians reserved the teaching of operational statistical concepts for themselves and said of the indigenous instructors that “[t]hey are very weak in these subjects and we can’t guarantee good quality data”, while the indigenous instructors preferred to go ahead with the “indigenous variables” because of the possibility that the whites would interfere in their specific topics: “You know nothing about this topic, we will be able to explain it better to our brothers”.

9. The ECPI: an intercultural relationship? As soon as I started work on this survey (I have worked with INDEC on over 20 different national surveys and the four national censuses that have been conducted over the past decade), I felt that it was a different kind of process. INDEC also designs surveys for other institutional counterparts (including the Federal Public Revenue Administration, the Ministry of Labour, the Ministry of Education and local governments). In the joint work processes, there are discussions, things go to and fro, amendments are made to agreements and questionnaires, the questions are reformulated, etc. In the case of the ECPI the situation was different. Although there were representations about “indigenous matters” and “the whites” and there was accumulated historical experience, there were no

60

Chapter Two

previous–concrete, genuine–parameters for sitting down to talk to indigenous representatives. In previous papers (Barrientos, 2007) I have argued that this situation is a meeting in which otherness is present, making it possible for the anthropological question to be asked. In Argentina, specifically attempting interculturality as a new way of building relationships is becoming extremely complicated–especially when that word is incorporated into state rhetoric but does not further transform power relationships in the way Catherine Walsh proposes: While interculturality is founded on the need to build relationships between groups, as well as between different practices, logics and types of knowledge, with a desire to confront and transform the power relations (including the structures and institutions of society) that have naturalised social asymmetries, multi- or pluri-culturality simply starts from the ethnocultural plurality of society and the right to difference. (2002: 1)

From that perspective the entire ECPI process was lacking in “intercultural” features and in so saying I am not seeking to deny that substantial progress has been made as far as the history of ethnic monitoring in Argentina is concerned. However, it was not possible for the code that Argentinian state rhetoric uses when dealing with indigenous affairs to be changed, resulting in “a great deal of spin and little delivery”. I also want to make my position clear about the fact that this operation was political in nature. Seeing it as a political act entails recognising its “dimension of negotiation in which the actors try to maximise their interests within a certain historical context and a specific correlation of forces” (Pacheco de Oliveira, 2007: 58). The advertising slogan for the ECPI–which was poorly disseminated–was: “One more step towards recognition of the rights of indigenous peoples”. In order to be a subject of rights, it is necessary to be seen. In my opinion, the ECPI was a step forward but only in that it rendered indigenous peoples more visible; for their rights to be recognised, there is still a long way to go. The census question addressed to each household about whether or not they belong to an indigenous people enables new questions to be posed about the subjects that make up the Argentine nation and gives them the chance to be named within the Argentina of today, to form part of the present. Just as hegemonic discourses had the persistence to repeat themselves as often as necessary but failed to achieve absolute domination, I believe that the inclusion of the indigenous variable, over and beyond the results it produces, is gaining strength as a counterhegemonic strategy. And in turn it is calling attention to ethnic diversity, thereby challenging the guiding fiction of white uniformity.

Are There Still “Indians” in Argentina?

61

10. On how the official data were presented “The state concentrates, treats, and redistributes information and, most of all, effects a theoretical unification. Taking the vantage point of the Whole, of society in its totality, the state claims responsibility for all operations of totalisation (especially thanks to census-taking and statistics or national accounting) and of objectivisation, through cartography.” (Bourdieu, 1998: 45)

According to the ECPI results, 600,329 individuals recognised that they were members and/or first generation descendants of indigenous peoples. The results also show the range of different peoples and the rural and urban areas in which they have settled. The data presented both on the INDEC website and in the paper publication (which is only provisional) was prepared according to the following criteria: The ECPI provides results for the whole of the population who recognise that they are members and/or first generation descendants of indigenous peoples (the indigenous population) and the data from those households can be consulted in: National Results. It also provides specific estimates for a wide range of peoples; this information can be consulted in: Results by People. In addition, for each of those peoples, information relating to national level and the Sampled Regions is given. The regional samples enable information to be obtained for each people according to geographical regions set up on an ad hoc basis. These sampled regions were determined by combining two criteria: 1) the traditional location of indigenous peoples within our country’s current territory, according to the specialist literature on the subject; 2) current place of settlement, which was determined by combining the information on the geographical distribution of households who answered in the affirmative to the question in the 2001 Census about whether they belonged to or were descendants of an indigenous people and data on indigenous communities supplied by INAI, NGOs and recognised researchers on the subject. For this reason, each one of the indigenous peoples selected was studied in the regions necessary to ensure their appropriate analysis. In some cases, these regions coincided with a provincial boundary while in others they covered two or more provinces. Some peoples were studied in several of the sampled regions and others in just one. Annexes 1 and 2 show the sampled regions determined for each people, organised by people and by province.5

62

Chapter Two

Table 2-1. Population according to whether they belong to and/or are first generation descendants of the Kolla people. Jujuy and Salta. 2004-2005 Members of and/or first generation descendants of the Kolla people Table 1.1 Population according to whether they belong to and/or are first generation descendants of the Kolla people. Jujuy and Salta. 2004-2005 Recognise that they belong to and/or are first generation 53,106 descendants of the Kolla people Recognise that they are Kolla 46,044 Recognise that they are Kolla and descendants of the 44,375 Kolla people through their father and/or mother Recognise that they are Kolla but are not first generation 1,669 descendants of the Kolla people Do not recognise that they are Kolla(1) but are first 7,062 generation descendants of the Kolla people Do not recognise that they are Kolla but are descendants 3,318 of the Kolla people through their father and mother Do not recognise that they are Kolla but are first generation descendants of the Kolla people and another 810 indigenous people(2) Do not recognise that they are Kolla but are first generation descendants of the Kolla people and another 2,934 non-indigenous people (1) This population group does not recognise that it specifically belongs to the Kolla people or any other indigenous people. (2) The population who are descendants of the Kolla people and another indigenous people will also be counted in the tables corresponding to the other indigenous peoples of which they are descendants. (..) Information estimated from a sample with coefficient of variation (CV) over 25%. Note: Final data. The total for each indigenous people corresponds to the number of people who recognised that they were members and/or first generation descendants of that people. Source: INDEC, 2004-2005b6.

Are There Still “Indians” in Argentina?

63

This way of displaying the data makes it impossible for people with no knowledge of statistics to have clear access to them. To whom is this information addressed? Which users? What is a “sampled region”? What previous knowledge is required to interpret this data? How can it be used to implement public policies if it cannot be clearly separated out by jurisdiction? How can indigenous organisations have access to this data in order to support their demands? The example above, taken from the INDEC website, that illustrates the situation. In this case the information is grouped under two Argentinian provinces (Salta and Jujuy), making it impossible to disaggregate the data by province. The grounds on which it was not possible to “show” the data in any other way were argued from the perspective of what was possible for “statistical procedures”. On consulting the staff responsible for the Statistical Information System in operation in INDEC, I was told that “very few people ask for that information, generally students or researchers... no, people from indigenous organisations, no”. Meticulous preservation of the statistical system and international standards relating to publication procedures were prioritised over the need for the data to be accessible and available. From what has been seen, new questions arise: What involvement did the representatives of indigenous peoples have in deciding on this way of reporting back the survey data to society? Were there demands for this situation to be reversed? Why did the organisations not use these arguments to back up their demands? Could this data have been used to implement subsequent laws such as Emergency Law 26,160/06 on the possession and ownership of land traditionally owned by the country’s original indigenous communities? I believe that the fact that the data from the survey was published in this way has turned it into an information record that is almost as invisible as the indigenous peoples themselves were for centuries. And this forms part of a specific logic, as Ana Gorosito Kramer states: “It is not the categories it uses but its logic which extends its influence over us, setting the limits of what belongs to us and what belongs to others, meaning simultaneously indicators of potential loyalty and betrayal, in other words, the enemy within” (2007: 6).

11. Ethnicity and Afrodescendant populations in the 2010 census The inclusion of ethnicity and Afrodescendant variables in the 2010 National Population, Household and Housing Census was also controversial. It was considered feasible for indigenous peoples to be included within

64

Chapter Two

poverty indicators without it being necessary to identify them as such. In the end these variables were included in the census questionnaire because of demands from the sectors involved and other national bodies. In this case, the formulation of the question on ethnicity was slightly amended. Figure 2-3. Questionnaire, 2010 National Population, Household and Housing Census. Question No. 5. “Identification of membership of or descent from an indigenous people”.

Source: INDEC. 2010 National Population, Household and Housing Census. Expanded Census Questionnaire.

It still called for self-recognition, albeit using a more colloquial formulation. Compared to the 2001 Census, more effort was put into publicising the issue. Specific media work was carried out to stress the importance of answering this question. Some of the advertisements were broadcast in indigenous languages (Quechua and Huarpe) and narrated by the interested parties themselves. The slogan used was: “Everyone has the right to identity. Let’s announce our presence” [“La identidad es un derecho de todos. Digamos presente”]. At the time of writing the data has still not been published. Similarly, for the first time a question relating to Afrodescendant populations was included in an Argentine population census. In this case a more extensive awareness campaign was carried out and joint work was done with the Instituto Nacional contra la Discriminación, la Xenofobia y el Racismo (National Institute against Discrimination, Xenophobia and Racism). It is not the purpose of this article to examine this topic in more depth but it is particularly interesting to see how these questions were incorporated into the round of 2010/2011 censuses conducted in Latin America. They should be incorporated into the state agenda and will certainly provoke political debate between the state and the groups undergoing these identity-related processes.

Are There Still “Indians” in Argentina?

65

12. Some ideas for further reflection I believe it is necessary to develop some ideas about the processes through which indigenous groups can own these state mechanisms so that they are able to reactivate their role as protagonists and recreate the political means they need to realise their demands. I have taken the idea from Ana Ramos, who says in this regard: In order to participate as legitimate adversaries in conflict negotiation, indigenous actors have described their practices in terms that render them acceptable; however, matters which have been kept outside of politics still, in short, form the mainstay of alliances and disputes in public arenas. (…) Historical and current recognition of the political as a typical mode of relationality does not mean disregarding differences, heterogeneity or even conflicts in the ways of moving in social spaces. (2008: 5)

Beyond casting a critical eye over the operations described here, I wish to stress that this initiative was unprecedented, both because indigenous peoples were included as workers in the conduct of the surveys and because of the results that were obtained and the importance that “difference” acquired. In other words, new meaning was given to indigenous people’s ability to be active subjects. Despite this, I still wonder: can one talk of participation when the state has sole authority for the management of funds and their allocation? Is it possible for surveys on topics demanded by indigenous peoples to be carried out when from the outset the ECPI was a technical prisoner of the census on account of its supplementary nature? Why did its coordinators refuse to include crucial variables that are important to indigenous peoples? What decision-making mechanisms were used to ensure indigenous participation in the ECPI? How did the indigenous peoples own this experience? How were they involved in the affirmation of ethnic identity? In these national statistical operations, the issue of identity has once again been presented by the state as a question of managing otherness. In this regard, it is reasonable to ask: “How can the problem of identity be addressed if both sides gain some visibility and end up recognising each other as partly the same but partly different and heteronomous?” (Gorosito Kramer, 2007: 7). It would thus be possible “[t]o discover […] that ‘the Other’ is never outside or beyond us” (Bahbha, 2003: 216). The questions asked in the questionnaires transcended the rigid possibilities offered by the statistical mechanism and, in a country in which popular imagination believes that “all the ‘Indians’ and ‘blacks’ were killed”, provided an

66

Chapter Two

opportunity to show that they are present and that they have not renounced their right to identity.

Bibliographical references Amadassi, Enrique and Gladys Massé. 2005. Censos y estudios en profundidad: caso argentino. Paper presented at the international seminar Pueblos indígenas y afrodescendientes de América Latina y el Caribe. Relevancia y pertinencia de la información sociodemográfica para políticas y programas. Santiago de Chile: CEPAL, 27-29 April. Anderson, Benedict. 2006 [1983]. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso. Barrientos, Pilar. 2007. La pregunta antropológica y la Encuesta Complementaria de Pueblos Indígenas. Paper presented at the VII RAM – Reunión de antropología del Mercosur, GT 8: violencia estatal e indigenismo. Porto Alegre: Universidad Federal de Porto Alegre, 23-27 July. Bartolomé, Miguel. 1987. Afirmación estatal y negación nacional. El caso de las minorías nacionales en América Latina. Suplemento Antropológico, XXII (2): 7-43. —. 1997. Gente de costumbre y gente de razón. Mexico: Siglo XXI. Bengoa, José. 2009. ¿Una segunda etapa de la emergencia indígena en América Latina? Cuadernos de Antropología Social, 29: 7-22. Bahbha, Homi. 2003. Narrando la nación. In Fernández Bravo (ed.) Literatura y frontera. Procesos de territorialización en las culturas argentina y chilena del siglo XIX. Buenos Aires: Sudamericana. —. 2003. El entremedio de la cultura. In Stuart Hall and Paul du Gay (eds) Cuestiones de identidad cultural. Buenos Aires: Amorrortu. Boccara, Guillaume. 2006. Dossier hegemonías, clasificaciones etnopolíticas y protagonismo indígena, siglos XVII-XXI. Anuario IEHS, 21: 53-189. Bourdieu, Pierre. 1998. Practical Reason: on the Theory of Action. Stanford: Stanford University Press. Briones, Guillermo. 1987. Metodología de la investigación cuantitativa en ciencias sociales. Módulo de investigación social. ICFES: Bogotá. Foucault, Michel. 1999. Estética, ética y hermenéutica. Barcelona: Paidós. Goldberg, Carola. 2007. Reflexiones en torno a la medición de la etnicidad. Paper presented to the Asociación de Estudios de Población de la Argentina. Córdoba: Huerta Grande, 1-2 November. Gorosito Kramer, Ana. 2007. Identidades bajo sospecha. Notes from a seminar on Culture, Nation and Identity, given by Professor Ana María Gorosito Kramer, in the context of the Master’s degree course on

Are There Still “Indians” in Argentina?

67

Social Anthropology at the Universidad Católica. Asunción (Paraguay), September-October. Pacheco de Oliveira, Joao. 2007. Hacia una antropología del indigenismo. Río de Janeiro: Contracapa. Ramos, Ana. 2008. “La lucha”: retrospectivas, contextos y prácticas políticas mapuches y tehuelches. Paper presented at the IX Congreso Argentino de Antropología Social Fronteras de la antropología. Posadas (Argentina), 5-8 August. Renan, Ernest. 1999. ¿Qué es una nación? In Fernández Bravo (ed.) Literatura y frontera. Procesos de territorialización en las culturas argentina y chilena del siglo XIX. Buenos Aires: Sudamericana. Shumway, Nicolás. 1993. La invención de la nación. Historia de una idea. Buenos Aires: Emecé. Viñas, David. 1983. Indios, ejército y frontera. Buenos Aires: Siglo XXI. Walsh Catherine. 2002. Interculturalidad, reformas constitucionales y pluralismo jurídico. Boletín Rimay. Nº 36 [On line, consulted on 30 August 2011]. http://icci.nativeweb.org/boletin/36/walsh.html

Documents consulted ILO–International Labour Organization. 1989. Convention Nº 169 on Indigenous and Tribal Peoples in Independent Countries. Geneva. [On line, consulted on 29 October 2011] http://www.ilo.org/ilolex/spanish/convdisp1.htm INDEC–Instituto Nacional de Estadística y Censos. 1991. Manual del Censista. Censo Nacional de Población, Hogares y Viviendas 1991. Buenos Aires: INDEC. —. 2001a. Cuestionario Censal de Población, Hogares y Viviendas 2001. Buenos Aires: INDEC. —. 2001b. Manual del Censista. Censo Nacional de Población, Hogares y Viviendas 2001. Buenos Aires: INDEC. —. 2004a. Cuestionario Encuesta Complementaria de Pueblos Indígenas 2004-2005. Buenos Aires: INDEC. —. 2004b. Manual para el Encuestador ECPI. Buenos Aires: INDEC. —. 2004c. Plan Guía para el Instructor de la ECPI. Buenos Aires: INDEC. —. 2004-2005a. Encuesta Complementaria de Pueblos Indígenas de la República Argentina. Resultados provisionales. Buenos Aires: INDEC. [On line, consulted on 31 October 2011] http://www.indec.gov.ar/webcenso/ECPI

68

Chapter Two

—. 2004-2005b. Documentos internos. Relatorías Encuesta Complementaria de Pueblos Indígenas. Buenos Aires: INDEC. —. 2006. Encuesta Complementaria de Pueblos Indígenas. Resultados abreviados. Serie Estudios N° 4. Buenos Aires: INDEC. —. 2010a. Manual del Censista. Censo Nacional de Población, Hogares y Viviendas 2010. Buenos Aires: INDEC. —. 2010b. Cuestionario Censal de Población, Hogares y Viviendas 2010. Buenos Aires: INDEC. Ley Nacional 23.302/85. Asuntos Indígenas. Boletín Oficial. Buenos Aires. 12 November 1985. —. 24.965/98. Censo Aborigen. Boletín Oficial. Buenos Aires. 28 May 1998. —. 26.160/06. Comunidades indígenas. Emergencia en materia de posesión y propiedad de tierras. Boletín Oficial. Buenos Aires. 29 November 2006. Resolution N° 4,811/96. [On line, consulted on 20 October 2011] www.mecon.gov.ar/onp/html/presutexto/ley2011/.../D11E118.doc Organización Mapuche Newentuayiñ, Organización de Jóvenes Mapuche We Kimvn, We Kvyeh, Grupo Kojog. 2001. Comunicado público. Frente al Censo 2001. [On line, consulted on 23 October 2011] http://www.mapuche.info/mapu/puel011116.html

Notes 1

The views expressed in this article are the sole responsibility of the author and may not reflect the point of view of the organisations that were involved in implementing the ECPI. 2 All literal quotations from Spanish sources have been translated into English by the translator Marion Mashrons. 3 http://www.indec.gov.ar/webcenso/ECPI/ECPI-Caracteristicas y resultados.pdf [consulted on 31 October 2011]. 4 http://www.indec.gov.ar/webcenso/ECPI/index_ecpi.asp [consulted on 31 October 2011]. 5 http://www.indec.gov.ar/webcenso/ECPI/ECPI-Caracteristicas y resultados.pdf [consulted on 31 October 2011]. 6 http://www.indec.gov.ar/webcenso/ECPI/pueblos/ampliada_index.asp?mode=06 [consulted on 31 October 2011].

CHAPTER THREE BOLIVIA: INDIGENOUS IDENTITIES AND COLLECTIVE SUBJECTS IN THE ANDES PABLO A. REGALSKY

There is a continuous re-definition of what is and who are “indigenous”. Andean communities are, nowadays, collective subjects that define their members as indigenous people. Of course, this is not the only way to understand “indigenous,” but, by examining the reproduction of these collective subjects, a better understanding of the current debate around Bolivia’s state territorial reconfiguration as a “pluri-national state” may be achieved. The category of indigeneity as produced by the nation-state purports that identity is located in the individual body. This rationale is implicit in the national census methodology. Thus the census is an instrument, along with other institutions such as the school, for the invisibilisation of collective subjects and consolidation of the individualisation processes. Here I rather focus on the way identity is produced by the collective subject, within which indigenous individuals are socialised. The results of the official census are compared with the outcome of this latter criterion, which is based on the link between collective subjects and indigenous territorial rights. Even if these different criteria may lead to differences in the numbering of indigenous people as I show below in this chapter, that is not the main purpose of my discussion. The concern of my argument is wider. There is a need to take into account the process of emergence of collective subjects which is very much related to the new forms of state organisation and how this dynamics is taking place under the government of Evo Morales in Bolivia (Keating, 2001). The state controls its territory and population through classificatory systems (organised through census and taxing) and with the help of social and political intermediaries (Migdal, 1988; Scott, 1998). The sociopolitical relations of intermediation along with the established meanings of

70

Chapter Three

classificatory categories are always in transformation. Long periods of stability, social subordination and/or integration usually alternate with relatively short waves of disruption, when forms of social autonomy emerge on the basis of shifting cycles in relations of force. The transformation of those relations of force at different levels means that the scale and magnitude acquired by indigenous collective subjects is variable; it is not fixed in time and space. Adopting this dynamic approach moves away from the binary/dichotomous idea of ethnic identities taken as “strategy” or as an intrinsic/essential quality of people. This chapter begins by showing the results of self-identification by individuals as shown by the 2001 Census and how those results represent a shift in the relations of force between the state and the indigenous organisations, the latter being umbrellas for the communities which exercise territorial jurisdictional powers parallel to those of the state. I describe how Andean communal organisations behave like collective subjects, while managing their own territories, and how communal control over the rights of access to land constitutes a basis for the existence of those indigenous collective subjects. I discuss the need to define identity in connection with self-governing territorial organisations which constitute the core of indigeneity not only for Andean communities but for all indigenous people in Bolivia.

1. Who is indigenous in Bolivia? Several censuses have attempted to define the number of indigenes in Bolivia. The colonial empadronamientos, first established in 1591, had the goal of registering the indigenous tributarios for the collection of the personal tax. Since Bolivia established itself as a Republic there have been ten censuses–in 1831, 1835, 1845, 1854, 1882, 1900, 1950, 1976, 1992 and 20011. The criteria used to include or not and to define the category “Indian” have varied widely in the last five national censuses, defining different social groups in each case. While in 1900 the result was that 48.5% were Indians, the 1950 Census established 63% of the population as indigenous, defining that population as a differentiated race. A statistical leap was notable in Cochabamba: from 22.5% in 1900, the Indians increased to 72.5% in 1950 (Grieshaber, 1985: 45). Grieshaber concludes that the 1900 Census took Indian as equivalent to the tax category of tributary, which excluded hacienda unpaid workers (pegujaleros) in Cochabamba and Chuquisaca. Instead, in 1950 the definition was built upon the language spoken, skin colour and clothing as observed by the enumerators. The percentage of those who

Bolivia: Indigenous Identities and Collective Subjects in the Andes

71

habitually spoke native languages in 1950 was 64% (Albó, 1983). In the 2001 Census, on the basis of self-identification by people aged over 15 years2, the percentage of Indians remained at 62% (Molina and Albó, 2006). In that last study of the census results, applying a matrix of “ethnolinguistic condition”, Molina and Albó claim that the percentage reached 64.9% of the Bolivian population as a whole. The following table shows the distribution of those who self-identify as belonging to an indigenous group, by department. This adds up to a total of 3,142,637 people. Table 3-1. Population self-identified as indigenous in the 2001 Census Departament

Quechua

Aymara

Guaraní

Mojeño

7,972

Chiquitano 395

Chuquisaca

188,427

3,873

La Paz

117,584

Cochabamba

595,629

1,027,549

4,043

62,780

3,071

Oruro

89,702

94,080

Potosí

319,903

Tarija

29,910 206,417

Santa Cruz

289

Other indigenous 1,248

Not indigenous 106,182

1,303

1,668

10,927

338,546

1,533

1,994

4,254

230,759

317

108

65

1,546

65,515

26,283

339

136

49

1,137

66,991

6,377

6,640

551

174

3,523

192,375

48,040

57,587

107,105

13,441

23,512

760,556 135,952

Beni

6,831

7,280

1,086

1,007

28,261

21,752

Pando

1,238

1,619

142

80

395

1,465

25,479

1,555,641

1,277,881

81,197

112,218

46,336

69,364

1,922,355

Total

Source: Molina and Albó, 2006. Data on population over 15 years old

In both the 1950 and 2001 censuses, the category of “indigene” is defined following an individualistic methodology even if using widely different parameters (observation of language spoken and racial traits in the first case, self-identification by the individual in the latter). Although methodological individualism is apparent in both cases, there is an important difference not only between the two censuses’ political context but also in the political intentionality of the two. The political visibility of indigenous collective subjects since the 1980s is associated with the global political re-emergence of indigenous peoples and communities (Hall and Fenelon, 2005). Several authors relate the emergence to a period of crisis and hegemonic decline of Western civilisation (Chase Dunn and Friedman, 2005). This civilisational decline and deep economic and ecological crisis of the West is accompanied by the loss of legitimacy of the nation-state as the political instrument for cultural homogenisation, social disaggregation and individualisation. That loss of legitimacy may

72

Chapter Three

explain the adoption of multicultural or intercultural political frameworks for the re-legitimating of the state. Ruling by cultural homogenisation as a key feature of nation-states was apparently modified and cultural differences were recognised in various governmental areas. Nevertheless, the crisis of the developmentalist model and the reversion to a model of accumulation by dispossession jeopardises the very basis of the production of cultural difference as Harvey (2003) notes. Multiculturalism is sometimes interpreted to be a sleight of hand or even as an appropriate mechanism to expand the new accumulation model (Žižek, 1997). I agree with Friedman that the dialectics of social and cultural forms do not only respond to local histories and political ecologies, but that their dynamics are defined by their positioning in global processes: The overriding argument is that cultural realities are always produced in specific sociohistorical contexts and that it is necessary to account for the processes that generate those contexts in order to account for the nature of both the practice of identity and the production of historical schemes. […] I argue, further, that the processes that generate the contexts in which identity is practised constitute a global arena of potential identity formation. This arena is informed by the interaction between locally specific practices of selfhood and the dynamics of global positioning. (Friedman, 1992: 837)

The indigenous territorial collective subjects reproduce their existence in a sustainable way given the specific ecological, productive and reproductive conditions they need to preserve via collective political action. By the way of collective political action they currently affirm their right to exist as such, as a collective person, actively questioning the notion of individual citizenship as the sole and only political participation model admitted by the liberal political system. The practice of identity by Andean collective subjects “is informed by the interaction between locally specific practices of selfhood and the dynamics of global positioning” (Friedman, 1992–see quotation above) as the reproduction of the community is jeopardised by the current model of capital accumulation (Pacto de Unidad, 2006). On the other hand, new class dynamics and social mobility driven by the ongoing Proceso de Cambio (Process of Change) is currently promoted by the Evo Morales government in favour of specific peasant sectors. This results in social and ethnic fractures and displacements within the indigenous collective subjects and social organisations that partially neutralises the production of identity and reinforces the practice of individualism.

Bolivia: Indigenous Identities and Collective Subjects in the Andes

73

The image of indigenous collective subjects is not a novelty, given that the colonial system defined “Indians” as members of a tax category based precisely on a collective belonging to a jurisdiction under indigenous authority. Effectively, those domestic units that remained on free communities’ land, and as such were under the authority of an indigenous chief, cacique or principal, belonged to this colonial category (Thompson, 2006). In this case we are dealing with collective, not individual identities, human groups within which individuals are socialised and consequently groups that those individuals may identify with. This identification process, however, does not prevent a certain number of individuals from entering and leaving these communities, whether by free will or forcibly. In a voluntary or involuntary way they thus modify their personal identity as they enter other social structures into which they must integrate via processes of secondary socialisation, whether or not the subject remains (mentally) identified, or not, with their “original” or birth identity. This is not only an active “assumption” of identity, developed or negotiated by individuals who develop life strategies for achieving social mobility or are agents of social reproductive strategies prevailing in those indigenous communities, but it is also the process of “endowment” of identity by the socio-political structure where the individual received primary socialisation. In that way, being subject to socialisation processes does not prevent the development of active strategies by individuals and groups. Friedman contrasts the “class of ethnicity” which prevails in Western modernity with that which dominated the society of the Byzantine Empire: This kind of “ethnicity” is, as I have argued elsewhere (Friedman 1991b), of a different order than that typical of Western modernity. The latter is situated in the body as the vehicle or container of identity. One belongs to a group because one is a bearer of a substance of racial identity, one that, in fact, had little to do with biology before the 20th century. The ethnicity of empire is associated with externally defined properties of social life, territory, corporateness, religion–common practices cemented by a political organisation that defines the region as a segment of a larger totality. (Friedman, 1992: 839)

I am interested in identifying the population which is included in the Andean indigenous collective political structures, in order to understand, in the first place, the “dynamics” of identities on a socio-political basis and secondly to compare the data obtained from communal membership with the results of the “individual body” approach used in the censuses. Spedding explains that a dichotomy is built through discourse where each individual may position him/herself at one or other side of an imaginary

74

Chapter Three

dividing line. Discourse become “reality” in the eyes of some researchers; so that an individual who answers affirmatively to the census question (Do you consider yourself indigenous?) “is” indigenous (Spedding, 2006). Evo Morales’ government point of view (probably one of several in his government) understands that “the relevant and defining elements of collective identities–such as blood relationships [sic], skin colour, place of birth, physical features, language and clothing– […] should be understood and complemented […] in order to overcome the vision of identity as an act of confrontation” (Gobierno de Bolivia, 2006: 15). Both understandings, whether that of the “reality” of an individual answer to a census question, or the reified notion of ethnicity equated to race as defined in those quarters of the Morales government, give little or no attention to the existence of a collective subject with cultural and political authority with which the individual identifies him/herself. Identity as a real social matter is not a “thing”, nor it is even the mask an individual carries with him/her, but a process of identification “with”, a social relation in permanent evolution that can be objectively recognised by an observer even if the subject itself is not perfectly conscious–or consciously denies–that such social connection is structuring him. There is no doubt that the state’s interest in integrating the indigenous population into the Creole state via citizenship supposes a process of individualisation that has been promoted ever since the foundation of the Bolivian Republic in 1825. Citizenship was understood, from the state point of view, along the requisite of individual private property (Platt, 1982), as the Republic attempted to annul collective indigenous personality and the commons as collective property. Beginning in the mid twentieth century, the Bolivian state introduced mass citizenship, by recognising women’s and illiterates’ right to vote, promoting Spanish-language schooling and imposing obligatory military service for males. Whitehead points out the limits of these integrative politics: Nevertheless the vertical structures imposed on the countryside both by the MNR3 and then by the MPP [sic] reflected continuing fear that in the absence of central authority the newly organised and (to some extent) armed Indian and peasant masses could unleash anarchy on the towns [...,] the conviction of the urban lower middle classes that universal suffrage would mean an inversion of the social order. Democracy could hardly be accepted as long as the victims of the conquest constituted a majority and remained united by their grievances. Over the past twenty years such fears have abated, and the rural electorate has shown itself responsive to competitive politics and capable of participating within a democratic system of rules. Over the same period, other social changes have mediated

Bolivia: Indigenous Identities and Collective Subjects in the Andes

75

the underlying ethnic division, and strengthened what one might call a shared mestizo symbolic culture […]. Nevertheless, ethnic stratification is still present in contemporary Bolivia and it could be rash to conclude that the fears of the past have been completely dissipated [...] they may be revived, possibly even in a virulent new form. (Whitehead, 2001: 34-5)

According to this point of view, indigenous participation in the system of democratic rules, that is, the definition of citizenship as the process of individualisation and the delegation of sovereignty in a system of political representation, would create some of the necessary conditions for the disappearance of ethnic stratification and indigenous identities. In the following section I discuss the dimensions of the ethnic stratification in Andean Bolivia in an attempt to give dimension to the structural core of the production of indigenous identity.

2. Dimensioning the Andean indigenous collective subjects There is a more or less permanent population living in and around these political-territorial, rural, indigenous cores, even though a good number of these people develop a pattern of dual rural-urban residence. A study I made for the Confederación Sindical Unica de Trabajadores Campesinos de Bolivia–CSUTCB (Unitary Syndical Confederation of Peasant Workers of Bolivia) in 1991 established that around 2,700,000 people constituted that indigenous core (52% of total population at that time). A different situation is that of urban people who, permanently settled in cities, describe themselves as indigenous. A first approximation can be arrived at using the same population census. In the case of Quechua and Aymara-speaking Andean populations, in the Western departments of Potosi, La Paz, Oruro, Chuquisaca and Cochabamba, we have the following data on urban and rural inhabitants who speak one or more native languages.

76

Chapter Three

Table 3-2. Native speaker population in Bolivia’s Andean departments

Department Chuquisaca

Total Population

Total Quechua speakers

Total Aymara speakers

Rural Quechua speakers

Rural Aymara speakers

Rural Native Speakers %

530,853

298,050

4,308

195,244

784

36.7

La Paz

2,346,550

158,260

1,181,593

64,840

602,057

28.4

Cochabamba

1,453,066

872,010

84,921

499,367

27,339

36.2

Oruro

392,147

134,289

127,086

61,341

87,405

37.9

Potosí

707,365

514,421

57,738

395,445

48,303

62.7

5,429,981

1,977,030

1,455,646

1,216,237

765,888

36.5

Total

Source: Molina and Albó, 2006.

While 49.3% of the population (more than 5 years old) in Bolivia speaks a native language, 1,982,125 people in the five western departments (out of the 9 departments into which Bolivia is divided) are native language speakers who live in rural areas, representing 36.5% of the total population of these departments (Molina and Albó, 2006). In my opinion this would be a most conservative figure for the indigenous population subject to community norms in these five departments. This figure increases once those members of rural communities who migrate to the cities or to colonisation areas and still retain land in their communities of origin are included. Another figure, also conservative, can be derived from information on land ownership which is considered below. It should be pointed out that while 62% of the total population of the country self-identify as indigenous4, 13.9% of that population did so despite never having learned a native language and being monolingual Spanish. This positioning of individuals reflects a favourable stance towards the indigenous collective subjects in their relation of forces vis-àvis the state, considering the political and dynamic character of indigenous “identity”. That section of the population is slightly bigger than the percentage of persons who only speak native languages and who do not know Spanish (11.9%) and illustrates the difficulty in establishing any static parameters of identity definition. It would not be surprising if, ten years after that particular census, the Spanish speaking “indigenes” category changes substantially in number, as a result of the changing relations of force between the state and the indigenous collective subjects which would affect the positioning of the urban people who previously identified themselves as indigenous5.

Bolivia: Indigenous Identities and Collective Subjects in the Andes

77

Figure 3-1. Bolivia: Highlands and Lowlands.

Source: Léons and Sanabria, 1997.

In Potosi, monolingual native language speakers are 32% of the total. 96% of these monolinguals are concentrated in the rural areas (Molina and Albó, 2006). There are two groups at the ends of the continuum of monolingualism clearly represented by two poles: those who speak only Spanish and those who speak only native languages. On the one side are the rural indigenous groups with an autonomous system of socialisation, combined with the superimposed jurisdictional legal system of state order (which includes the school system). On the other side of the continuum,

78

Chapter Three

“whitened” middle classes integrated into the political and ideological apparatus of the Bolivian nation-state within which they have been socialised. I assume that even part of this “whitened” or mestizo urban population integrated into, and subordinated to, the state expressed themselves as being indigenous in the 2001 Census, and this is what we understand as self-identification as a result of the change of the relations of force scenario in favour of indigenous movements against the Creole state. A paradox appears now as a result of the fact that the Evo Morales presidency, a government that identifies itself as indigenous, is pushing back the relations of force between the state and the indigenous. What was a dominant situation for the indigenous and popular movements between 2003 and 2005 now puts them on the defensive, as the government is enforcing a developmentalist and extractivist model with a discourse that reduces the indigenous peoples again to “political minorities” which endanger the well-being of the Bolivian population (el pueblo boliviano) as a whole. The previous statistics (Tables 3-1 and 3-2) should be confronted with other data for a less subjective approach to the indigenous population included in Andean politico-territorial structures. In my view, the census provides a quantification of the relative political weight of the indigenous collective subjects at the particular time of the census, but this information has to be understood in light of the data obtained through a different approach, for a less subjective understanding of the issues of the political identity of a differentiated population. As explained below the number of communities, the communities’ membership and the number of households with access to communal land are solid sources for this data.

3. The collective subject On the part of the indigenous and peasant organisations, selfrecognition of their own exercise of territorial jurisdictional power was neither immediate nor straightforward. If we take the example of the 1973 Tiwanaku Manifiesto of the katarista movement, one of the first public political expressions by the Aymaras confronting the Bolivian state’s model of development, it makes no mention of the theme of territory. The CSUTCB was set up in 1979 by a new generation of Aymara katarista leaders, fracturing the vertical union structures under the patronage of MNR and MPP. Their relatively ambiguous position in regard to the relationship between indigenous and peasant movements and the state was probably the result of pressures from the Central Obrera Boliviana–COB (Bolivian Workers’ Central), the national trade union organisation, in

Bolivia: Indigenous Identities and Collective Subjects in the Andes

79

order to adopt the COB’s “labourist” programme. The CSUTCB is a member of the COB. In its political manifesto of 1983, the CSUTCB only stressed the question of the spaces governed by Quechua, Aymara and Guarani communities (Rivera Cusicanqui, 2003). In its second Congress, held that year, with the aim of approving the project of the Ley Fundamental Agraria–LFA (Fundamental Agrarian Law) with which it intended to reorient the process of Agrarian Reform, the CSUTCB declared that: We reject the eminently individualist and smallholder orientation [of the Agrarian Reform] […] we peasants organised as self administrators [autogestionariamente] will promote the move beyond of smallholder farming via associative forms of production and commercialisation, channelled by our mother economic entity, the Peasant Agrarian Corporation CORACA [Corporación Agropecuaria Campesina]. (Cárdenas, 1985: 21)

Criticising the socially and culturally homogeneous Creole state, Cárdenas there maintains that […] the LFA does no more than project on a national level the state vocation which the union organisations possess. These [peasant] unions are not mere class organisations but rather are fundamentally miniature differentiated states. The agrarian union, above all in the Altiplano and the valleys, is simply the community or ayllu under another name. (ibid: 26)

The Confederación Indígena del Oriente Boliviano–CIDOB (Indigenous Confederation of Eastern Bolivia), which covers the indigenous peoples of Amazonia and the Bolivian Chaco, founded in 1982, defines indigenous people in its bill of the Indigenous People’s Law (presented in 1990 but never considered by the national parliament)6: (Article 2) For the purposes of the present Law, “indigenous peoples” are the human groups which possess the following characteristics: 1) their own history, culture and language; 2) they have a specific social organisation, with which they identify, whose members recognise one another as belonging to the same cultural context; 3) they possess their own forms of political organisation and exercise the power of autonomous decision; 4) they possess a historically defined territorial interrelation, which functions in the control of their habitat, their cultural repertory and the character of community life. (CIDOB, 1993)

This definition was adopted by the indigenous leaders under the influence of international organisations who wanted the project to fit into

80

Chapter Three

the established framework of international definitions. The changes introduced are better appreciated when the final document is compared with the first version of that same project; elements of individualisation have been introduced into the second clause of this article. However, the draft of the article still makes it clear that it refers to “human groups […] [which] possess their own forms of political organisation and exercise the power of autonomous decision”. The 1989 ILO Convention n° 169 (International Labour Organization, a specialised branch of the UN) was used as the basis for CIDOB’s definition of indigenous peoples, defining territory as habitat rather than political jurisdiction. It nevertheless recognises the collective relationship with land and territory: Article 13, clause 1. In applying the provisions of this Part of the Convention governments shall respect the special importance for the cultures and spiritual values of the peoples concerned of their relationship with the lands or territories, or both as applicable, which they occupy or otherwise use, and in particular the collective aspects of this relationship. Article 15, clause 1. The rights of the peoples concerned to the natural resources pertaining to their lands shall be specially safeguarded. These rights include the right of these peoples to participate in the use, management and conservation of these resources. Article 17, clause 1. 1. Procedures established by the peoples concerned for the transmission of land rights among members of these peoples shall be respected. [my emphasis]

In its articles 9, 15 and 17, the Convention timidly recognises the existence of an autonomous juridical system and “also” accepts the existence of an educational system proper to each indigenous people. However, it opens the door to the existence of a relation with the land and a transmission of rights to land which is not based on private property. Article 9, clause 1. To the extent compatible with the national legal system and internationally recognised human rights, the methods customarily practised by the peoples concerned for dealing with offences committed by their members shall be respected.7

Thus far I have discussed the need to recognise the categorical core of what is the indigenous Andean in Bolivia as a collective subject that exercises territorial jurisdiction, as is the case of rural highland communities. I use the term “community” in the same way as its members

Bolivia: Indigenous Identities and Collective Subjects in the Andes

81

do, that is, as the political unit, the polity that the same local population recognises as a communal unit with a proper name. A political unit acting as a de facto Andean indigenous jurisdiction with territorial authority may be an agrarian union (sindicato) or an ayllu, both made up of several communities, usually divided into four segments or suyu, each of which has its own proper place name. In the traditional Andean organisation, ayllu is the name for a territorial unit that probably in its origins was a kinship segmentary organisation. Nowadays the ayllu is the basic polity forming part of a Jatun Ayllu, which in turn is part of a Marka8. While the ayllu organisation is still predominant in the Altiplano, in the valleys where the 1953 Agrarian Reform originated the communities have since then organised as rural unions9. The union organisation changes the traditional nomenclature, which becomes adapted to the state’s politico-administrative spatial organisation. Nevertheless, the functions of the sindicato and the ayllu tend to be similar. The aggregate of a few sindicatos constitutes a peasant subcentre, and the peasant subcentres are usually part of one or more Peasant Centres at provincial level. The processes of mobilisation and demobilisation, the political cycles of flux and reflux that the Andean rural population have experienced in the last century at least, and the relations of force which they have established with other social sectors and with the state, have determined processes of social aggregation or, conversely, disaggregation. These processes have emerged in specific forms of territorial reconfiguration.

4. The resilience of the Andean territorial organisations The CSUTCB, with the support of the ILO, co-ordinated a study in 1991 that counted 11,791 Andean communities organised as unions or similar forms, which were part of the pyramidal structure of the Peasant Confederation (Regalsky, 1991: 9). That count does not include many of the communities of Quechua and Aymara speaking migrants from the highlands (colonizadores) established in the lowlands north of Santa Cruz and the Chapare region, nor some 32 different lowland indigenous groups. An interpretation of the CSUTCB’s study may bring a higher figure than that reckoned on the basis of the 2001 census sociolinguistic data. It counted 495,424 affiliated families, although this calculation was done on the basis of statistical samples. Based on an average rural household of 5.1 members, it would result in 2,724,830 people included in Andean peasant communal structures10. The list of affiliates in each agrarian union also includes those heads of households who migrate permanently or temporarily to the city, the

82

Chapter Three

lowlands, or even to a foreign country but still preserve their rights to access land in their community of origin under their name, or maintain land cared for by relatives or by neighbours. That connection between the migrants who maintain access to communal land and their community of origin partly explains the significant number of people in the cities who identify themselves as indigenous people, even if they are not directly under the authority of an indigenous communal structure. Those migrants used to play on an ambiguous social identity as part of the spatial and occupational diversification strategies which led them to continuously switch between jurisdictional spaces such as those of the city or region they live in and their community of origin. The counting of the number of Andean communities through time is relevant and meaningful. It shows the resilience of that political structure even under very unfavourable conditions. The 1950 Census indicated the existence of 3,779 free communities or ayllus and 6,952 haciendas. The hacienda has a particular labour regime, because it maintained the Andean community within the property as a captive community. The captive community will maintain access to their land in exchange for their free labour and all kind of services and tributes which are given to the hacienda owner (Rodriguez, 1978). The aggregate of both free communities and captive communities under the hacienda regime makes a total number of 10,731 demographic and productive units. This number from 1950 is very similar to the number of communities acknowledged by the CSUTCB study in 1991. That is because most of the haciendas were transformed while the captive communities now under the form of agrarian unions took over the territory of the ex-haciendas during the 1953 Agrarian Reform period. Moreover, according to Lavaud (1991, quoted in Assies, 2007) in 1847 there were 11,000 communities, which points to an extraordinary persistence in the number of political and demographic rural communal units. Accordingly, the number of communities which the state had legally recognised under the regime of the Popular Participation Law n°1551/1994 by the end of 1996 is 11,597.

5. Collective land ownership and identity Andean peasant land ownership is related to belonging to and affiliating with a communal polity and is a basis for the resilience of indigenous identity. I refer this communal belonging to common patterns of indigenous space management by Andean community households as a basic element for the reproduction of Andean community jurisdictional

Bolivia: Indigenous Identities and Collective Subjects in the Andes

83

rights of access to land. In this way, a correlation is established between indigenous collective subjects as polities, the specifics of access to communal land, and Andean space management strategies for the reproduction of indigenous households. Through this analysis, it is possible to better understand the meanings of subjective accounting through individual self-ascription, which can only be understood in the framework of the objective parameters I propose here. The current common patterns of space management by Andean Communities’ households are described by an extended literature (see CENDA, 2010; Regalsky, 2005) . What is stressed here, on the basis of a brief description, is the fact that these strategies are the basis for the reproduction of the community’s jurisdiction. The communal jurisdiction has a specific commitment to control access to land, the commons which only members of the community may have access to. Reciprocally, members of the community cannot exercise individual rights to transfer land to people foreign to that community. The jurisdiction exercised upon land by the communal authority and the reproduction of that authority express a differentiated order in the social realm. This order is characterised by the predominance of the collective subject over the individual subject. The collective domain is under permanent stress, both from inside and from outside the community. The current situation under the Evo Morales government presents us with the paradox that the predominance of the collective subject is increasingly neutralised while this government has strongly contributed to re-legitimating the established nation-state. Pacheco and Valda present an interesting synthesis of the characteristics of territoriality in the Andean valleys, in terms of rights of access to land, which at the same time are marked by an ethnic consciousness. The systems of land tenure (formal and consuetudinary) in the communities and settlements of the valleys are characterised by a multiplicity of regimes of proprietary rights. Thus, various proprietary regimes (individual and collective) can be encountered almost simultaneously in those spaces which the community considers to be its property. Likewise, in the cases where it appears, it is insistently proposed that a central element of indigenous identity is related with territoriality in physical, symbolical and political terms, in so far as belonging to an indigenous people signifies having a determinate consciousness of a territory and maintaining special links with the land. (2003: 239)

In Cochabamba and other regions, the agrarian unions have organised on the lands of former haciendas. There is evidence that in the eighteenth

84

Chapter Three

century there were tribute paying Indians with rights to land within the hacienda (Rojas, 2005). Despite almost 400 years of communal submission under landlords (patrones) as captive communities, the current control of access to land still follows “traditional” patterns very similar to the free aymara communities studied by Spedding (2006), which were never taken over by the hacienda regime. In both cases the structures of belonging and authority are based on the collective exercise of control over the peasant families’ access to land. The individuals have some degree of freedom to “leave” their communities without losing rights, while the families are the holders of rights and obligations related to their identification as members of the community. While the status of community originario and that of adjunto or arrimado were colonial categories defining who was a tributary and who was free of paying tribute to the Crown, self-designation as originario points to an ethnic identity and a political definition, establishing a self-ruled political sphere with a degree of juridical autonomy. Platt describes a document he found in a communal archive in Macha, northern Potosí, that illustrates this point: It says that the rights of Indians to all cultivable and grazing land comes from the creation of the world, under the protection of the Inka, the Ordinances of King Carlos, the land compositions of the Quito Bishop in 1590 and the land compositions of Josep de la Vega Alvarado in 1646. These lands will belong to them “until the judgment”, when they will have to give account of their administration to God. These ancient titles were given under oath and inalienable possession under five hundred pesos penalty, twelve years of prison and a hundred [quintal] lashes for any attempt at selling or renting out any land. (Platt, 1992: 67)

Originating from pre-Inka myth, that “great oath” asserts their identity as one and the same as their “originary” right to land. Now sacralised in Christian terms, it legitimises the exercise of communal coercion on any individual in order to protect those collective rights from alienation and private accumulation. Ethnic demarcation is understood as the politicisation of cultural difference directed towards the establishment and legitimising of a relatively autonomous polity. At the same time, political differentiation or politicisation of cultural difference is grounded in the process of territorialisation based on control over resources the community keeps and awards access to: Territoriality can be expected simply because it can be an efficient device for establishing differential access when the resources to be controlled

Bolivia: Indigenous Identities and Collective Subjects in the Andes

85

occur relatively predictably and densely in space and time. Clearing, planting, weeding, and harvesting are all relatively predictable, dense, and stable activities in space and time, and hence agricultural communities, if for no other reason than keeping animals out and preventing people from getting in each other’s way, can be expected to use territoriality [...,] increase if there is also competition for land from outside. (Sack, 1986: 59)

Territoriality, the management of resources, relates to a socio-political organisation. Labour exchange is a constitutive part for the management of resources as the specifics of the Andean organisation of labour are fundamental for the reproduction of the communal collective self. Communal authority and customary rules preserve the orderly exchange of labour and goods between indigenous households. Tensions between collective and family interests are not precluded. These spaces of tension almost always involve processes of differentiation and levels of intermediation between social groups within and without the communities. Gose proposes that the processes of ethnicisation of the peasantry and those of class formation should be considered as part of a single set of practices and should not be separated, even on an analytic level. It seems that commoners’ class and ethnic identity derive from a common, or at least overlapping, set of practices and need not be treated in a dichotomous manner as a result. But the vast majority of the existing literature on the Andes assumes that class and ethnicity can and should be analytically separated. Broadly speaking, there are two main positions: one eliminates class entirely, presenting ethnicity as the only legitimate way of looking at social difference in the rural Andes, and the other assumes that both class and ethnicity are present, but irreducible to each other. (Gose, 2001: 15)

The data shown below on communal land and individual agrarian reform land title distribution is a guide to illustrate that relationship between collective communal land and individual land tenure, and the affiliation of households which usufruct that land to indigenous communities. The information on the number of households affiliated to the indigenous communities who have access to land as collective property is analysed along with the quantification of Andean communities already stated above. These objective parameters allow us to compare the number of indigenous households with the counting of self-identified indigenous people obtained from the 2001 Census and from other sources. The following tables (3-3 to 3-5) set out the distribution of land in Bolivia at three different moments during the twentieth century: before and after the Agrarian Reform, and just before the INRA Law (1996)11,

86

Chapter Three

showing the number of individual and collective landowners. Table 3-5 gives us a clearer idea of the common property titles (proindiviso) awarded to communities, which cover 21.45% of the total land held with legal land titles, in the hands of 333,403 beneficiaries, that is, communal households. There are gaps in this data, because an important number of communities rejected the Agrarian Reform, did not apply for and did not receive land titles at all. The territory of the communities of the Southern Altiplano covers about 4 million hectares, including the Great Salt Flats of Uyuni. They rejected the Agrarian Reform brigades since they demand the recognition of the colonial royal Spanish recomposición12 land title they obtained in the seventeenth century (Quisbert and Huanca, 2001). Considering that the average number of families per community is 30, from the number of beneficiaries quoted above we can assume that at least 11,113 communities hold common property legalised by a land title. The average communal land size for each community would be 1,105 hectares. Once again, we have about the same number of rural communities already mentioned above and so we can infer that these communities all have a certain degree of communal control over access to land. Table 3-3. Distribution of agricultural land in 1950 (before the Agrarian Reform) Size of holdings in hectares

Holdings

no.

Total area (A)

%

Has.

Cultivated area (B) %

Has.

%

B/A

%

< 10

59,988

69.4

132,964

0.41

65,981

10.2

49.6

10 – 500

19,437

22.5

1,467,488

4.48

344,385

52.6

23.5

> 500

6,952

8.1

31,149,398

95.11

343,892

37.2

0.8

Total

86,377

100

32,749,850

100

654,258

100

2.0

Source: Census 1950 elaborated by Rivera Cusicanqui, 2003.

The 1950 census presents a picture of dramatic rural land inequality: 8% of the largest landholders (7,000 property owners with 500 or more hectares of land) owned 95% of available land, while 69% of the smallest landholders (60,000 peasants with 10 hectares or less) owned only 0.4% of available land. The data in Table 3-3 shows cultivated agricultural land that was cultivated by indigenous communities or by an unpaid indigenous labour force including that within large land estates. The next table shows

Bolivia: Indigenous Identities and Collective Subjects in the Andes

87

the results of the Agrarian Reform distribution. As can be seen, huge properties over 5000 hectares, i.e. big landholdings, were established as a result of illegal land distribution by military governments in the 1970s. Table 3-4. Land property distribution according to the 1984 Agricultural Census Class size (ha)

Number of holdings

Cumulative %

Area (ha)

Cumulative %

< 0.5

58,319

18.5

21,368

0.0

0.5 – 5

156,118

68.2

260,769

1.4

5 – 20

57,878

86.6

516,285

3.7

20 – 100

30,125

96.1

1,213,018

9.1

100 – 1,000

8,176

98.7

1,268,673

19.9

1,000 – 2,500

1,997

99.4

3,100,823

33.6

2,500 – 5,000

1,301

99.8

3,999,020

51.3

> 5,000

686

100.0

11,047,809

100.0

Total

314,000

22,670,152

Source: World Bank, 1995.

The next table displays another set of data, adding to the information in Table 3-4, which shows how the new big landowners took hold of more than 39 million hectares during the Agrarian Reform, while communal property was the only significant land the Andean peasant communities could retain. Table 3-5. Land distribution according to type of land title (Agrarian Reform 1953-1993) Type of property Small property Medium property

Beneficiaries (B) 269,179 123,567

% (B) 35.44 16.27

4,850,838.8281 16,231,728.8796

% (A) 8.46 28.32

17,005

2.24

23,011,055.2333

40.16

Agricultural enterprise Peasant holding Communal property No information Total

Source: Balderrama, 2002.

Area ha (A)

Average 18.0209 131.3597 1,353.19 35

3,999

0.53

23,866.1021

0.04

5.9680

333,403

43.90

12,289,511.1437

21.45

36.8608

12,283 759,436

1.62 100

898,322.5654 57,305,322.7522

1.57 100

73.1354

88

Chapter Three

If we take the average number of members for each domestic unit (as in 1993) to be 5.5, multiplied by the 333,403 households that are beneficiaries of communal property titles as shown in Table 3-5, the total population that usufructs communal property would be 1,833,716 persons, 16 years before the 2001 Census. This figure does not take account of the people who have access to collective land without any legal title, as is the case in the South Altiplano region. This figure is very similar to that of the native language speaking population in the rural areas of Bolivia’s five western Andean departments according to the 2001 Census (see Table 32). Considering that this data refers to the years between 1956 and 1980 when those titles were awarded, it can be inferred that the current figure is no doubt higher and is close to that claimed by the CSUTCB quoted above (Regalsky, 1991). I see consistency in the figures related to the number of people with access to the commons through affiliation to Andean communities, the number of people belonging to the Andean national organisations, and the distribution of native language speaking rural people in the five Andean departments. There is evidence that there is a social structure of access to natural resources and for the control of the social and political space that was resilient enough to survive even as captive communities within the hacienda regime. That structure re-emerged after the 1953 Agrarian Reform as Andean peasant communities, after the rupture of the Military Peasant Pact in the 1980s, increasingly identify themselves as indigenous polities that claim rights for self-rule for autonomous territories. The free communities in the Altiplano region that kept the form of ayllus reinforced that tendency to attain political autonomy from the nation-state. The nation-state made every effort against those self-government structures in order to liberalise the land and labour markets (Platt, 1982). That is the thread running through all efforts for individualisation and extension of citizenship to indigenous communities, in order to neutralise the control exercised by Andean communities over land and (partially) over free labour. Meanwhile, the state keeps up the fiction of a monist political and juridical space and uses the census, the school and military conscription as tools for the invisibilisation of those spaces of political autonomy and cultural difference and for the achievement of a homogeneous citizenship. An important transformation took place through the works of a Constituent Assembly in 2006-2007 for recognition of the existence of indigenous nations as pre-existent to the Bolivian state. The new Constitution approved by referendum in January 2009 is explicit in the state’s recognition of indigenous collective rights, including the right to self-government, the validity of a radical juridical pluralism, and the existence of indigenous

Bolivia: Indigenous Identities and Collective Subjects in the Andes

89

institutions for the ruling of their territories (Gobierno de Bolivia, 2009). Against all these basic tenets of the new Constitution the Evo Morales government has taken a nationalistic discursive path and currently denies the indigenous people’s rights established in the new Constitution as attested by the conflict in TIPNIS13 and in other indigenous territories. The discussions which are now taking place for the definition of the questions to be included in the 2012/2013 Census tend to be consistent with the positioning of this government towards invisibilising and neutralising the indigenous collective subjects.

6. Conclusions We can now return to the question of the re-emergence of the indigenous peoples and their classification by the state as individual culture bearers. This chapter argues that methodological individualism invisibilises ethnic emergence as the construction of national social organisations and movements, which not only re-establish ethnic identities as subjective markers but establish territorial control over wide regions of the country through their own polities, as collective subjects. The trend for the politicisation of cultural difference is linked to a global trend that I examined above: the dynamics of global decline of western civilisation and the hegemonic decline of its main political instrument: the nationstate. The legitimacy crisis of the nation-state associated with the hegemonic decline of that global political organisation makes it viable for the indigenous collective subjects to more openly display their own authority over those territories. The Andean communities establish indigenous jurisdictions, with indigenous authority, indigenous norms–all of them with a specific goal: to guarantee the rights of access of indigenous households to usufruct common land, common sources of water, and other vital natural resources for social reproduction. Of course that goal is in deep contradiction with the current needs of capitalism to gain access to all natural resources and, to that end, to dispossess all populations from their rights of access to common resources (Harvey, 2003). That is why it is so important to establish the objective basis for the formation of the contemporary indigenous identity, as I try to do in this chapter. It does not deny the possible existence of abstract and subjective cultural notions alive in individual minds, but it is an alternative and concrete notion of civilisation that it is based on the indigenous working relationships with nature, not currently compatible with predatory globalised capitalism. As a globalised capitalism enters into a spiral of

90

Chapter Three

self-destruction, the indigenous ways emerge as a possible alternative. The struggle for a pluri-national state in Bolivia is not based solely on the right to have access to one’s own culture, but on the need to elevate the jurisdictional indigenous rights of access to natural resources to the level of a new political organisation of the state. A pluri-national state that recognises not only the differences between different individual bodies and different languages, but basically recognises the rights of different peoples as collective subjects to have their own ways of life by exercising their right to self-determination in their own territories.

Bibliographical references Albó, Xavier. 1983. ¿Bodas de plata? o requiem por una reforma agraria. La Paz: CIPCA. Assies, Willem. 2007. Lands, Territories and Autonomy in Latin America: Recognition of Indigenous Territories in Bolivia. In Proceedings of the XXVII International Congress of the Latin American Studies Association. Montreal: LASA. Balderrama, Carlos. 2002. Estadísticas agrarias. La-Paz: DANIDAINRA. Calvo, Luz Maria, Pablo Regalsky, and Carlos Espinoza. 1994. Raqaypampa, los complejos caminos de una comunidad andina. Cochabamba: CENDA. Cárdenas, Victor Hugo. 1985. Ley Agraria Fundamental, algunas notas sobre su origen, protagonistas y fundamentos. Historia Boliviana, 5: 19-27. CENDA. 2010. Estrategias campesinas andinas de reducción de riesgos climáticos. Cochabamba: CENDA. CIDOB – Confederación Indígena del Oriente Boliviano. 1993. Ley de pueblos indígenas del Oriente, Chaco y Amazonía boliviana. Santa Cruz: CIDOB Chase Dunn, Christopher and Jonathan Friedman (eds). 2005. Hegemonic Declines: Present and Past. Political Economy of the World System Annuals. Boulder: Paradigm Press. Friedman, Jonathan. 1992. The Past in the Future: History and the Politics of Identity. American Anthropologist, 94: 837-859. Fundación Tierra. 2012. Marcha indígena por el TIPNIS. La Paz: Fundación Tierra. Gobierno de Bolivia. 2006. Plan nacional de desarrollo. La Paz: Ministerio de Planificación del Desarrollo. —. 2009. Constitución política del estado. La Paz

Bolivia: Indigenous Identities and Collective Subjects in the Andes

91

Gose, Peter. 2001. Aguas mortíferas y cerros hambrientos. Rito agrario y formación de clases en un pueblo andino. La Paz: Mamahuaco. Grieshaber, Erwin. 1985. Fluctuaciones en la definición del indio: comparación de los censos de 1900 y 1950. Historia Boliviana, 5: 4585. Hall, Thomas and James Fenelon. 2005. Indigenous peoples and hegemonic change. Threats to sovereignity or opportunities for resistance? In Jonathan Friedman and Christopher Chase Dunn (eds) Hegemonic Declines: Present and Past, pp. 205-224. Harvey, David. 2003. The New Imperialism. Oxford: Oxford University Press. Keating, Michael. 2001. Plurinational Democracy: Stateless Nations in a Post-Sovereignty Era. Oxford-New York: Oxford University Press. Léons, Madeline and Harry Sanabria. 1997. Coca and Cocaine in Bolivia: Reality and Policy Illusion. In Madeline Léons and Harry Sanabria, (eds) Coca, Cocaine and the Bolivian reality. Albany: State University of New York Press, pp. 1-46. Migdal, Joel. 1988. Strong Societies and Weak State: State-Society Relations and State Capabilities in the Third World. Princeton: Princeton University Press. Mirabal, Fernando. 2009. Estudio sobre la vivienda en Bolivia. Nueva Economia, 760. http://nuevaeconomia.com.bo/productos/revistaarticulos/opinion/estudio-sobre-la-vivienda-en-bolivia/ Molina, Ramiro and Xavier Albó. 2006. Gama étnica y lingüística de la población boliviana. La Paz: Sistema de Naciones Unidas en Bolivia. Pacto de Unidad. 2006. Propuesta de las organizaciones indígenas, originarias, campesinas y de colonizadores hacia la Asamblea Constituyente. Sucre: Asamblea Nacional de Organizaciones Indigenas, Originarias, Campesinas y Colonizadores de Bolivia. Pacheco, Diego and Walter Valda. 2003. La tierra en los valles de Bolivia: apuntes para la toma de decisiones. La Paz: Fundación TIERRA, ACLO, CEDLA, CIPCA, QHANA. Platt, Tristan. 1982. Estado boliviano y ayllu andino. Tierra y tributo en el norte de Potosí. Lima: IEP. —. 1992. Voces de Abya-Yala. Escritura, chamanismo e identidad. UNITAS: 61-73. Quisbert, Francisco and Expectación Huanca. 2001. Los Lipez. Sangre y oro. Cochabamba: CENDA. Regalsky, Pablo. 1991. Proyecto de factibilidad servicio educativo sindical OIT/RLA/87/MO5/ITA. La Paz: ILO-CSTUCB.

92

Chapter Three

—. 2005. Recursos naturales, Estado y sociedad. In Fundación PIEB (ed.) Estados de la investigación en Cochabamba. Investigaciones coeditadas. La Paz: PIEB; CESU-UMSS; ASDI/SAREC, pp. 151-186. —. 2010. Political Processes and the Reconfiguration of the State in Bolivia. Latin American Perspectives, 37: 35-50. Rivera Cusicanqui, Silvia. 2003. Oprimidos pero no vencidos. Luchas del campesinado aymara y qhechwa 1900-1980. La Paz: YachaywasiAruwiyiri. Rodriguez, Gustavo. 1978. Acumulación originaria, capitalismo y agricultura precapitalista en Bolivia (1870-1885). Avances, 1: 119-143. Rojas, Luis. 2005. Historia de la ocupación del territorio en la provincia Ayopaya. In Rene Orellana (ed.) Organización territorial comunitaria en Morochata. Procesos productivos y cultura originaria. Cochabamba: CENDA, pp. 123-136. Sack, Robert. 1986. Human Territoriality: Its Theory and History. Cambridge: Cambridge University Press. Scott, James. 1998. Seeing like a State. London: Yale University Press. Spedding, Alison. 2006. Las fronteras del pensamiento fronterizo. In Mario Yapu (ed.) Modernidad y pensamiento descolonizador. Memoria del seminario internacional. La Paz: PIEB-IFEA, pp. 187204. Thompson, Sinclair. 2006. Cuando sólo reinasen los indios. La Paz: La muela del Diablo-Aruwiyiri. Whitehead, Laurence. 2001. The Emergence of Democracy in Bolivia. In: John Crabtree and Laurence Whitehead (eds) Towards Democracy Viability: The Bolivian Experience. New York: Palgrave, pp. 21-40. World Bank. 1995. Bolivia, National Land Administration Project (Staff appraisal report No. 1360-BO). Washington: World Bank. Žižek, Slavoj. 1997. Multiculturalism, or the Cultural Logic of Multinational Capitalism. New Left Review, 225: 28-51.

Notes 1

http://www.ine.gob.bo/html/visualizadorHtml.aspx?ah=Cronologias.htm The question posed, with seven options to answer, was: “Do you consider yourself part of any of the following originary or indigenous peoples? (Se considera perteneciente a algunos de los siguientes pueblos originarios o indìgenas?)” with one option being “None”. 3 MNR: Movimiento Nacionalista Revolucionario (Revolutionary Nationalist Movement) which led the April 9th, 1952 nationalist revolution. MPP: Pacto Militar Campesino (Military-Peasant Pact). Subsequently to the fall of the MNR 2

Bolivia: Indigenous Identities and Collective Subjects in the Andes

93

regime, the military government established a Pact in 1965 with the Andean peasants “in order to fight communism”. 4 The 2001 census addressed this question only to people over 15 years old. 5 Paradoxically, this relationship took an unfavorable turn against the indigenous peoples less than two years after Evo Morales took office as the first indigenous president of the country in 2006 (Regalsky, 2010). 6 The current term accepted by the Quechua and Aymara highland people living in the five Andean departments to identify themselves is Pueblos Originarios. Many of the Quechua and Aymara peoples consider themselves to be, at the same time, part of the working class so they use the term Trabajadores Campesinos Originarios (First Nation Peasant Workers) or Trabajadores Campesinos Indigenas (Indigenous Peasant Workers). As the term originario is not easily translated into “originary” or “first nation”, I prefer to use in this text the term indigene or “indigenous”, which is accepted by the indigenous people in Bolivia’s lowland for self-designation. The term Indio has a historically racist, pejorative connotation so the highland people do not accept this way of self-identification 7 http://www.ilo.org/ilolex/english/convdisp1.htm accessed 12.12.07 8 Ayllu, Jatun Ayllu, Marka are the vertical levels of integration of Aymara communities, while Quechua communities adopted the unionist nomenclature: sindicato, subcentral, central, federación, confederación. 9 The agrarian sindicato organisation was promoted by sectors of MNR government in the 1950s. 10 The 5.1 personas per household average by years 1990-1995 was obtained through case studies in the Quechua rural area (Calvo et al., 1994). Recent population estimates shows a decline in the number of family members per house to 4.35 (Mirabal, 2009). 11 INRA: Instituto Nacional de Reforma Agraria (Agrarian Reform National Institute). This law, enforced in 1996, introduced land market incentives but at the same time recognised indigenous territorial rights as Tierras Comunitarias de Origen–TCO. 12 The Spanish Crown considered all “vacant” land as its own and restructured the ethnic property. Recomposición implicated ethnic groups should demonstrate their property rights and pay the Crown for the recognition of their entitlements. 13 TIPNIS: Territorio Indígena y Parque Nacional Isiboro Secure (Isiboro Secure Indigenous Territory and National Park) see Fundación Tierra (2012).

CHAPTER FOUR WHO COUNTS INDIGENOUS PEOPLE, HOW ARE THEY COUNTED AND WHAT FOR? CENSUS POLICIES AND THE CONSTRUCTION OF INDIGENEITY IN COLOMBIA GLORIA PATRICIA LOPERA MESA

Population censuses are a prominent device of statecraft. Not only because demographic information provides a necessary input for making diagnostics and designing public policies, but also because subjectivities are constructed through censuses. The power of counting entails the decision on what or who should be counted and for what purposes, and the ability to define relevant criteria to classify the population and to determine which category(ies) everyone belongs to. This, in turn, affects the way in which subjects re-build and mobilise their identities, based on their reasons for being counted within any of these categories of population. The interest of states in adopting methodologies to quantify and highlight ethnic diversity in population censuses has gone hand in hand with the implementation of policies that fall within the multicultural paradigm that has characterised Latin American constitutionalism since the late twentieth century (Yrigoyen, 2010). Such interest has also been fed by demands of indigenous peoples and other ethnic groups not only to be counted with culturally relevant statistical tools, but also to count themselves as a way to self-recognise, be recognised and exercise the power of building indigenous subjectivity. An approach to the ways in which indigeneity is constructed and portrayed in the population censuses should not forget that demography, particularly when it tries to adopt a distinct ethnic approach, struggles between two conflicting claims: on the one hand, the need to leave singularity aside, filtering it through the use of general classificatory concepts, to then quantify and express it in the formal language of

Who Counts Indigenous People, How are they Counted and What For?

95

numbers; on the other hand, the need to capture the specific cultural characteristics and the different ways in which each group, and even each individual, lives their indigeneity. One example of this tension is the growing demand to unify the terminology used to characterise the indigenous otherness in different Latin American countries, in order to build a regional statistical information system to facilitate comparative analysis. At the same time, there is an insistence that the conceptualisation of ethnicity and how it is projected in census exercises cannot be applied uniformly to each national reality. Instead it must track the various ways in which each Latin American nation has built its “others” (Segato, 2007: 28; Lloréns, 2002: 658). Following the proposal of Rita Laura Segato to trace those “national formations of otherness” (Segato, 2007), this article examines the ways in which indigeneity has been constructed and made visible in population censuses in Colombia1. For this purpose, three major periods are set: Colonial, Republican and the “multicultural turn” that occurred after the promulgation of the 1991 Constitution. The analysis of each of these periods attempts to answer three questions: who counts indigenous people in Colombia, how are they counted and what for? This historical account will serve as a basis for reviewing, at the end of the article, some of the challenges the construction and the counting of indigeneity have to face in multicultural Colombia. These are, firstly, the paradoxes arising from the prevalence of the self-recognition criterion and from the relationship between indigeneity and territory; and secondly, the particular dynamics in the counting of ethnicity and in the reconfiguration of identities taking place in the context of the armed conflict in Colombia. Before starting on this path it is necessary to roughly draw the demographic map that has taken shape since the colonial period and largely determined the current patterns of land occupancy and differentiated construction of indigeneity in Colombia.

1. A look at the geography of indigeneity in Colombia The Spanish colonisation proceeded mainly from the coast to the Andean region, which eventually would become the economic, political and administrative centre, concentrating from this point most of the country’s population. However the colonial institutions did not achieve full control over the territory, since large areas such as the eastern plains, the forest regions of the south east (Amazonas, Putumayo, Caquetá, Orinoco), the plains of the Pacific Coast and the jungles of Chocó were out of reach. These areas became internal frontier territories, beyond the

96

Chapter Four

control of colonial authorities which, at most, were present indirectly, through the government of the missions or through individuals (González, 2006: 98-113; Tovar Pinzón et al., 1994: 18-21). Since then, this imagined community that is Colombia has represented itself as divided into a centre, which coincides with the “highlands” of the Andean region, and a periphery, formed by the “lowlands” of coasts, deserts, plains and forests. The periphery is formed by vast stretches of territory where the presence of state institutions diminishes at the same rate as the population density decreases. This division between centre and periphery of the nation, together with the various cultural patterns of land occupation and adaptation to the environment that occur in mountain, coastal, desert, jungle and plain areas, are key factors for understanding the many faces of indigeneity expressed behind the uniformity of census figures. According to data from the General Population Census conducted in 2005 (hereinafter GPC 2005), a total of 87 indigenous peoples are fully identified, a figure that rises to 102 according to data from the Organización Nacional Indígena de Colombia–ONIC (National Indigenous Organisation of Colombia) (DANE, 2007: 20; Andrade, 2010). Figure 4-1. Indigenous Macro-regions according to ONIC’s organisational structure

Source: HREW, http://geographiando.files.wordpress.com/2011/05/macroregiones-onic_en.png

Who Counts Indigenous People, How are they Counted and What For?

97

Figure 4-2. Indigenous territories in Colombia and population density

Source: HREW, http://geographiando.files.wordpress.com/2011/05/ti_densipopen.png

In order to account for such diversity, figure 4-1 shows the division into five indigenous macro regions made by ONIC. Figure 4-2 illustrates the regions where the greatest concentrations of indigenous people are found and contrasts these regions with the map of indigenous territories known as resguardos, which are a form of territorial reservation2. Figure 4-2 shows that the majority of indigenous people living in resguardos are concentrated in the north of the country, mainly in the Guajira Peninsula and the Sierra Nevada de Santa Marta. One can also see that the high density of indigenous peoples in some areas of the Andes and the Caribbean coast is inversely proportional to the size of their resguardos, in contrast to the lower population density and greater landmass of the resguardos in the Orinoco and Amazon regions. To calculate the share of indigenous peoples within the total Colombian population it should be noted that, according to the data from the 2005 GPC, of the total of 40,607,408 people who answered the question on ethnic affiliation, 1,393,623 identified themselves as indigenous, which corresponds to 3.43% of the population. Of this total, 956,679 lived in resguardos (DANE, 2007: 37). This brief look at the figures and distribution of the

98

Chapter Four

indigenous population in present-day Colombia will serve as a starting point to track how the indigenous otherness has been built and made visible in population censuses.

2. The colonial period: from the separation between Spaniards and Indians to the mosaic of peoples “of all colours” The invention of the concept of “Indian”, as used generically to name the inhabitants of these (in the eyes of Europeans) newly discovered lands, is constitutive of the colonial order. This order was structured around the basic separation between Europeans, who arrogated to themselves the right of conquest, and native Americans, who had to play the role of colonized peoples (Todorov, 1987: 15; Quijano, 2000: 374; Ariza, 2009: 73). This social order, founded on the notion of race as a criterion of stratification and attribution of social and legal status, gradually became more complex. The need to distinguish between Spanish-born individuals and their descendants born in America (criollos), Indians, blacks brought from Africa as slaves and the many combinations derived from the racial mix of all the above, would result, in the late colonial period, in the prevalence of the “free neighbours of all colours” (Ots Capdequí, 1941: 24-6; Patiño Millán, 1993; Tovar Pinzón et al., 1994: 26). At first, the insertion of Indians into the social and economic order of the colony took place through institutions such as the mita and the encomienda3. Subsequently, the dramatic decline of the native population during the first decades of the conquest led to the reform of both institutions and, after the second half of the sixteenth century, to the concentration of the native population in Indian villages and resguardos. In this way, a rigid separation was attempted between the “Republic of Spaniards” and the “Republic of Indians”, as a strategy to ensure the physical preservation of the latter, their inclusion as tax-payers in the economic order of the colony and their spiritual discipline. In fact, in their villages, the Indians would be controlled by civil and ecclesiastical authorities (González, 1970; Bonnet Vélez, 2002: 27-8; Herrera Ángel, 2007: 85-90, 173-224; Ceballos-Bedoya, 2011: 231-3). In addition, the building of a legal subjectivity for Indians was necessary, one that would allow their insertion into the rigid status system which characterised the legal rationality that prevailed in Spain and subsequently in the colonies. Within this legal rationality, the rightsbearing individual, typical of legal modernity driven by liberalism, did not yet exist; on the contrary, the rights of each individual were determined in

Who Counts Indigenous People, How are they Counted and What For?

99

a differentiated way, depending on the status he/she occupied in the social order. Thus, the definition of Indians’ legal status was made through their assimilation into the categories of “rustic”, “miserable” and “minor”, derived from Castilian law (Clavero, 1994: 70-72; Ceballos-Bedoya, 2011: 227-31)4. The above-mentioned context provides guidelines to answer the questions about who counts indigenous people in Colombia, how they are counted and what for. It should also be noted that during this period the territory that today corresponds to Colombia was politically integrated with the Audiencia of Santa Fe, which belonged to the Viceroyalty of Peru, and was later included in the Viceroyalty of the Nuevo Reino de Granada in 1739 (Jaramillo Uribe, 1978: 349). Partially following the classification proposed by Woodroh Borah, we can distinguish three periods with respect to the collection of demographic information during the colonial period (Borah, 1972: 16-41). The first one covers the arrival of the Spanish in America until the first half of the sixteenth century. In this period information emerged from the first contacts between the colonists and the native population, primarily recorded in chronicles and in the reports of explorers, settlers and missionaries, made at the request of the Spanish crown. Based on the study of these sources, on statistical extrapolations and other prior research, Colombian historian Hermes Tovar Pinzón estimates that by the year 1500 the Indian population in the territory which corresponds to Colombia today reached 9 million (Tovar Pinzón et al., 1994: 22-3). The second period extends from the late sixteenth century until 1770. It started with the rationalisation of the settling enterprise carried out during the reign of Philip II (1556-1598) and included the implementation of the Bourbon reforms during much of the seventeenth century. This period was characterised by the consolidation of record-keeping systems and reports by civil and ecclesiastical authorities, through which population figures were obtained. However, these counting practices never constituted organised censuses as we understand them today (Borah, 1972: 33). The main sources of information regarding indigenous population during this period came from the Relaciones Geográficas5 (Geographical Relations), from the reports of the Visitors and from the civil and church records obliged to provide information to those officials (Borah, 1972: 35-8; Vidales, 1978: 27-31). Efforts to increase knowledge and control of the population led to the creation in 1758 of the first statistical office of the Nuevo Reino de Granada and in 1770 to the development of a general count of the population of Nueva Granada. It yielded a total of 507,209 inhabitants belonging to the Audiencia of Santa Fe, of which 28.568 were

100

Chapter Four

classified as tax-paying Indians (tributarios) and 98,017 as neighbours (vecinos) (Tovar Pinzón et al., 1994: 57, 63-4). The third period covered the last decades of colonial rule (1770-1810). It was characterised by the implementation of important economic and administrative reforms which required that more attention be given to registration systems and population control, which would result in the undertaking of the first general population censuses (Borah, 1972: 33). This period began with the general census conducted between the years 1777 and 1779 (known as the 1778 census). According to a summary of the information provided by Hermes Tovar Pinzón, the 1778 census showed the following results: Table 4-1. Population census of the Viceroyalty of New Granada, 1778

Ecclesiastics Whites Indigenous Freemen Slaves

Audiencia of Santa Fe 2,999 0,35%

Audiencia of Quito 1,536 0,36%

Total population per stratum 4,562 0,36%

209,265

24,67%

111,058

25,25%

320,323

26,07%

167,885

19,79%

293,643

66,75%

461,528

37,56%

403,377

47,54%

28,937

6,58%

432,314

35,19%

64,890

7,65%

4,700

1,06%

69,590

5,66%

1,228,317

100%

Total 100% 439,901 848,416 population per Audiencia Source: Tovar Pinzón et al., 1994: 68-726.

100%

Although it was not the only one, the 1778 census is considered the most important one of the colonial period (Vidales, 1978: 36-9). Its figures eloquently express some major transformations over the last centuries of the colonial era. One such change was the increase of the total registered population, resulting from the adoption of health and prevention policies implemented to halt the recurrent epidemics, but also from the greater control exercised by the colonial authorities over territories and population groups that were not previously recorded (Tovar Pinzón et al., 1994: 34). Second, the sharp decline of the indigenous population and the corresponding increase in the so-called “neighbours” or “Freemen” is remarkable7. This trend was particularly visible in the case of the Audiencia

Who Counts Indigenous People, How are they Counted and What For? 101

of Santa Fe and it contrasted with the demographic composition of the Audiencia of Quito, where the indigenous population maintained a remarkable dominance with respect to Whites, Freemen and Slaves. In summary, as inferred from the above account of the sources of demographic information, during the colonial period the civil and ecclesiastical authorities were primarily responsible for the counting of indigenous population. In response to the question of how they were counted, the racial and territorial criteria dominated. These correspond to classification patterns which characterised a social order built upon the rigid separation between white neighbourhoods (parroquias), resguardos and Indian villages (Mörner, 1970; Herrera Ángel, 2007: 83-102). This order has since defined indigenous subjectivity based upon an indissoluble bond with the land of the resguardo (Ariza, 2009: 121-31). However, around the mid-eighteenth century, just as statistical efforts were intensified by the colonial authorities, the difficulty of using such criteria for classification became evident. One factor underlying this difficulty was the intense process of miscegenation growing among “pure” Whites, Indians and Blacks, which would give rise to a mosaic of neighbours “of all colours”. The other factor that invalidated the territorial criterion for the identification of the indigenous population, and which was also linked to miscegenation, was the failure of the policy of segregation in resguardos and Indian villages, due to the growing presence of neighbours in indigenous lands and the departure of Indians from their resguardos to avoid paying the tribute-tax and to be hired as labourers on plantations (Bonnet Vélez, 2002: 54-5). Thus, in 1789, the Visitor Francisco Silvestre wrote: “and truth is [...] that the number of Indians has not decreased where they were, but they have progressively become Hispanicized and moved into other castes” (Silvestre, 1968: 68). This dispersion of Indians beyond the boundaries of the resguardos and Indian villages generated the need to create new categories to classify and count them. In addition to the category of the tax-paying Indian (tributario), used to designate indigenous men aged between 18 and 50 who were subject to the obligation to pay a tax (“tributo”)8 in cash or kind to the Crown, the category of forasteros or forajidos was gradually consolidated to designate those who sought refuge outside the resguardos to avoid paying the tributo. The expressions that follow are also frequent: “ladino Indians”, used to refer to the Hispanicized Indian; the “chontal or gentiles Indians”, labelling those who were not subjugated by colonial authorities; the “concertados Indians” for those working on plantations; and the “agregados Indians”, a term that is sometimes confused with that of concertados, but that was used to designate the Indian who had been

102

Chapter Four

taken from one village to another (Ots Capdequí, 1941: 31-2; Fals Borda, 1957: 84, 92; Tovar Pinzón et al., 1994: 38). Finally, what were the indigenous people counted for in the colony? In addition to the reasons directly related to the economic and spiritual purposes that inspired the settling enterprise, that is, the control over labour, the collection of taxes and the maintenance of the new “souls” won for the Church, the counts of indigenous people also had important territorial implications. Indeed, resguardo land allocation was made depending on the number of tributarios. That is why the decline of the latter was the main argument invoked to justify policies to reduce resguardos and Indian villages, initiated from the seventeenth century and widespread after the second half of the eighteenth century (González, 1970; Bonnet Vélez, 2002). The foregoing partly explains the ambivalent attitude of Indians to censuses. Documentary sources account for their reluctance to be counted, as shown in the report of the Bishop of Popayán, who attributes the flaws in the census record of 1777, among other reasons, to the timidity and apprehension of the Indians, who “[...] give sinister interpretations to them, and this leads them into terrible misunderstandings”, and also “hide the number of souls who are in their homes as much as they can” (cited by Tovar Pinzón et al., 1994: 33). However, the archives also house numerous accounts of Indians who demanded to be counted and thus recognised as such for purposes of opposing the diminishment of resguardos or their assignment to other Indians who disputed them. An example of this, in the context of a territorial dispute, is the claim filed in 1721 by the Indians of Lomaprieta against those of La Montaña village. The latter claimed that the territory of the former should be granted to them, arguing that it was inhabited only by “Ten outsider Indians from different towns”9. To counter this claim, the Lomaprieta Indians made the following request to the Mayor of Anserma: Please Your Worship kindly recognise all the Indians who are in the place of Lomaprieta, children and adults, and with the book of baptisms of the Holy Church of Quiebralomo, certify how many of us are currently in that place, also by counting those tributarios […], women, boys and others; likewise will Your Grace kindly certify the care with which we go to mass on holidays to be indoctrinated and all other ministries of our duty; and how useful we are at the place we are, for the conservation of the Royal Mines of Quiebralomo in which our King and Lord is greatly interested (quoted by Escobar, 1999: 15-16).

Who Counts Indigenous People, How are they Counted and What For? 103

3. After independence: the place of the Indian in the project of a unitary nation The decades after 1819, the year in which military victory over Spain was achieved and the construction of Republican institutions began, were characterised by great instability. The confrontation between the various nation-building projects transcended mere political struggle to be fought in the battlefields, thus giving rise to a constant succession of constitutional reforms, “battle charters” that reflected the armistice conditions imposed by the victor of the day (Valencia Villa, 1997). As regards the place of Indians in the Republican social order, it was agreed upon that their fate would be cultural assimilation, but the proposals regarding how and when to carry it out differed. Some proposed their full and immediate integration as citizens and private owners of the land they tilled. This implied the abolition of the tributo and, in general, of the colonial regime of differentiated rights, as well as the dissolution and distribution of resguardo lands. All this reflects two aspects: on the one hand, a political liberalism that postulates formal equality of all individuals before the law and purports to be “blind to their differences”; on the other hand, an economic liberalism that viewed ownership as a condition for citizenship status and sought the insertion of communal lands into the market. This project was also linked to a positive assessment of miscegenation as a mechanism to accomplish the construction of a human type that merged the best of the European, indigenous and African heritage. Others, however, identified miscegenation and integration of the Indian as a threat to their ideal of nation, built upon the rescue and predominance of Hispanic heritage and the shaming and concealment of indigenous heritage. They wanted to halt the accelerated racial integration process, in order to maintain a social border capable of distinguishing between a “white” elite (not just because of the colour of their skin), bound to lead the construction of the nation, and “non-whites”, who would be more subordinate in the social scale, and closer to the “pure” Indian or Black. The maintenance of resguardos resulted in a functional border that fed the alleged superiority of a white “us” and the estrangement of those Indian “others”, with whom relationships ranged between paternalism and contempt (Ariza, 2009: 133-200). The representation of indigeneity, as expressed in law and in the population censuses conducted from the beginning of the Republican period until the multicultural turn of the late twentieth century, was the result of the confrontation and also of the hybridisation between the two

104

Chapter Four

aforementioned alterity construction projects, which could be called liberal and conservative respectively. For example, elements of the liberal project prevailed in the first measures concerning Indians promulgated after independence, as we can see in the Decree of 20 May 1820 issued by Simon Bolívar, who mandated that land be returned to the Indians in order to proceed to further divide and distribute it10. Also, the Law of 11 October 1821 stated that “Indigenous peoples of Colombia, called Indians in the Spanish Code, would hereforth not pay the tax degradingly known as tributo; they will not be put to any service without paying them the corresponding salary previously agreed upon. They are all equal to other citizens and shall be governed by the same laws” (Roldán, 1990: 11-13). The same spirit of assimilation was reflected in the First General Population Census of Gran Colombia11, conducted in 1825, where much of the colonial strata (Whites, Indians and Freemen of all colours) were integrated into a general mass of population. The following year, a study titled “Conditions shown by the independent and uncivilized Indian tribes in Colombia in 1826” was completed and incorporated as an annex to the general population census. In this study only the “independent natives who live wandering in the forests and mountains” were considered and counted as Indians (DANE, 1975: 32-7; Tovar Pinzón et al., 1994: 90-9). Table 4-2. Population censuses in the nineteenth century Period

Year

Source

Gran Colombia

18251827

Population and Housing Census I

Nueva Granada

1835

Population Census II 1843 National Census III 1851 Population Census IV United 1864 Population States of Census V Colombia 1869Population 71 Census VI Republic of 1892 Indigenous Colombia Census Source: Uribe, 1998: 29; DANE, 2007: 31.

Total 2,379,888 (1.228.259 for the current territory of Colombia)

Slaves

Indians

103,892

203,835

1,686,038 1,955,264

111,130 26,777

184,230

2,243,730 2,694,487 2,890,637 160,000

Who Counts Indigenous People, How are they Counted and What For? 105

The trend in general population censuses conducted in the nineteenth century was to exclude the count of the Indians (a category which, as already mentioned, only included the gentiles or independents, i.e., the “not reduced to society”), whose count would be made separately and integrated as an annex to the general population census. According to a law passed in 1834, thereafter that task was in the hands of provincial governors. Later, in a 1842 law, the classification criteria used in indigenous counts were regulated more precisely, requiring that the record show the number of independent Indians differentiated by sex, name of their tribe, place of residence, usual occupation and their “peaceful or fierce” character (Uribe, 1998: 27). As shown in figure 4-2, such indigenous population counts were only carried out in 1827 (as an annex to the General Census of 1825), 1835 and 1843. In all cases the fragmented nature of such information is notorious, since it only included data from certain regions and specifically warned that much of the “wild and wandering” population could not be registered (DANE, 1975: 36-7, 49-50; DANE, 2007: 32; Bastidas, 2007: 137). It would then seem that in the Republic, the former tributarios Indians had already been integrated for all purposes within the overall mass of the population. However, a Decree by Simón Bolívar in 1828 once again established the tributo to be paid by the Indians, in exchange for exemption from military service and other obligations that they had acquired since assuming the status of full citizens. Thus, the Decree ordered a “general census” of all Indians eligible to pay tributo (Uribe, 1998: 25; DANE, 1975: 141). This shows the ups and downs that the process of construction of indigeneity experienced throughout the nineteenth century, in a society struggling between two projects of nationmaking, each of which held a different place for the Indian inside the new social order. By the end of the century, the conservative program known as the Regeneration, based on political centralism and the rescue of the Catholic Hispanic heritage, finally prevailed. It would be reflected in the 1886 Constitution. The Regeneration expressed the conservative project of building indigeneity which we mentioned before, based on a pessimistic view of miscegenation, on maintaining a border between “White” and “non-White” and on a paternalistic relationship with the indigenous people, the assimilation of which would be attempted through the “civilizing” Project embodied in Act 89 of 1890. This law, which largely remains in force today, constitutes one of the legal instruments that has most significantly influenced the construction of indigeneity in Colombia. It classified the Indians into two groups: firstly, the “savages” living in

106

Chapter Four

peripheral areas or “lowland”, governance of whom was given to the missions, entrusted with the task of “progressively reducing them to civilized life”; and secondly, those “already reduced to civilized life”, whose place would be in the resguardos, subjected to an indirect rule through the indigenous councils (cabildos). The latter was premised on the attribution of certain self-governing powers to the cabildos; among those powers was one worth mentioning for the purposes of this article, namely the power to make an annual census of the population of each resguardo. The completion of the annual census, along with other practices such as the appointment of the cabildo authorities before the mayor of the respective municipality or the regulation of land rights by the indigenous cabildos, became structuring elements of indigeneity in the communities of the Colombian Andes to date. The nineteenth century closed with a specific census for indigenous communities conducted in 1892 in order to begin the implementation of the aforementioned Act 89 of 1890. This count, called “Calculation of tribal aboriginals” yielded a total of 160,000 Indians, but did not cover the whole country (DANE, 1975: 113-14). Table 4-3 provides an overview of the 11 population censuses conducted between 1905 and 2005, indicating in which of them indigenous people were included in the count, what identification criteria were used and the total indigenous population resulting from the application of these criteria. These data allow us to track the changes affecting the construction of indigeneity throughout the twentieth century and to this day. Table 4-3. Population censuses in Colombia and identification criteria for ethnic groups (twentieth and twenty-first centuries) Year

Total

Indians

%

Black or AfroColombians

%

1905 1912 1918

4,355,477 5,072,604 5,855,077

344,198 158,428

6.79 2.71

322,499 351,305

6.36 6.00

1928 1938

7,851,110 8,701,816

100,422

1.15

1951

11,548,172

157,791

1.37

Identification Criteria

Not specified. Race. Physical traits as perceived by the census taker. Not specified. Interviewer’s questions and associations, type of language and rural location. Not specified.

Who Counts Indigenous People, How are they Counted and What For? 107 1964 1973

17,484,508 20,666,920

119,180 383,629

0.68 1.86

1985

30,062,200

237,759

0.79

1993

33,109,840

532,233

1.61

502,343

1.52

2005

41,468,384

1,392,623

3.40

4,311,757

10.60

Not specified. Indian is a person belonging to a group characterised by pre-Hispanic cultural traits and a subsistence economy, who lives in previously established areas. Criteria of selfrecognition and belonging to the indigenous people, consideration of living in a specific territory of the community or group. Self-recognition as a feature of identity. Belonging to an indigenous people, ethnic group or black community. Cultural selfrecognition (customs and traditions) or physical features. Belonging to indigenous peoples, Roma*, Raizales of the Archipelago of San Andrés, Palenqueros of San Basilio, and AfroColombian, Black, Mulatto or African descent.

*In GPC 2005 the Roma (Gipsy) people were also counted, with 4,858 people (0.01% of the total population). Source: DANE, 2010: 12.

108

Chapter Four

The first thing to notice about the presence of ethnic groups in the censuses carried out during the first half of the twentieth century is that, with the exception of the 1912 and 1918 censuses, Blacks lose visibility, while Indians were counted in almost all censuses (except those of 1905 and 1928). The Indians of the “highlands” were counted in the resguardos and those of the “lowlands” were counted by the missionaries to whom the government of those populations was delegated (Uribe, 1998: 29; Bastidas, 2007: 137). When discussing the figures for indigenous people in the census of 1951, anthropologist Gerardo Reichel-Dolmatoff drew attention to their poor reliability, arguing that they did not take into account the thousands of Indians living in various central departments of the country. In addition, he argued that the criterion of identification was not clearly established, “defining Indians primarily as those who spoke an aboriginal language and lived in a tribal system”. Thus, “the many individuals who really are Indians from an anthropological viewpoint based on their cultural characteristics and their phenotype, but who speak Spanish, and are sometimes confused with the rural mestizo population, were not included in the official indigenous census” (Reichel-Dolmatoff, 1959: 245-6). Similarly, in a study written in the early 1970s, Dario Fajardo drew attention to the unreliability of the results obtained from the indigenous censuses conducted during this period, due to the inaccuracies in the list and particularly to “the variation of the concepts about indigenous population” (Fajardo, 1971: 39). However, the diagnosis differed depending on whether it concerned the Indians who live in peripheral areas or those living in the central areas of the country. In the case of the former, “from the point of view of identification as a culturally distinct human group, it becomes easier to locate and identify marginal groups, despite the difficulties of access to their settlements”; whereas for the indigenous communities located in highland resguardos, “the difficulties of access are less important than those produced by the same process of acculturation, which tends to destroy the distinctive features of the indigenous population” (Fajardo, 1971: 38). These works are part of a trend which, since the late 1950s and under the influence of indigenismo, proposed a revision of the criteria employed hitherto for identifying indigenous people. These criteria, left to the census taker’s perception, were based on a combination of (i) phenotypic traits associated with the indigenous “race”, (ii) pre-Hispanic cultural features, mainly native language, and (iii) residence in indigenous territories. These criteria were challenged by the claims of recognition raised over several decades by Andean indigenous people who lived in resguardos and by

Who Counts Indigenous People, How are they Counted and What For? 109

many others who, like Manuel Quintin Lame–a pioneer in the indigenous struggle in defence of territory–had been stripped of their land and worked as tenant farmers on large plantations. The 1970s marked the beginning of the indigenous renaissance, in the heat of the struggle for land claims, led by the indigenous peoples of Cauca, and the emergence of the first indigenous organisations12. They called for the application of the old Law 89 of 1890, which had been appropriated and resignified by the indigenous movement, no longer as an accomplishment of the assimilationist project of the Regeneration, but rather as legal support for their defence of territory, culture and autonomy (Friedemann, 1981). Since then, state institutions in charge of dealing with the territorial claims of the indigenous movement have played a major role as agents in charge of counting, and thus building, indigeneity. In particular, the Colombian Institute of Agrarian Reform (INCORA), responsible for conducting socioeconomic studies that endorse the establishment of resguardos, has played a leading role. Such studies have been used to certify the indigeneity of the communities claiming for land titling under the institution of the resguardo, and have even come to be considered by some state institutions, and by the Indians themselves, as proof of membership in a community13. It was precisely the INCORA which funded the Cauca indigenous census, conducted by the DANE between 1971 and 1972, in the context of claims to legalize land. This census model served as a pilot test for the National Population Census that was conducted in 1973. For the first time, a specific census form was used for indigenous people. It would also be the basis of the indigenous census conducted between 1974 and 1976 in the areas known as National Territories14 and which was added to the results regarding indigenous population obtained in the general census of 1973. It is worth noting that both the latter and the general censuses carried out afterwards–1985, 1993 and 2005–made visible and contributed to the forging of the reindigenisation phenomena that became accentuated with the “multicultural turn” of the 1991 Constitution. Table 4-3 shows the progressive increase of the indigenous population with respect to the total of the national population registered in the censuses after the constitutional reform in 199115.

4. After the 1991 Constitution: the multicultural turn The 1991 Constitution, together with the ratification of ILO Convention 169 in the same year, placed Colombia on the path of what some authors have called “multicultural constitutionalism” (Bonilla, 2006) and others, in

110

Chapter Four

a critical tone, “multicultural neoliberalism” (Hale, 2005). Both legal instruments incorporate a number of territorial, social and political rights for ethnic groups, especially for indigenous and Afro-Colombians. Their recognition has been further enhanced by the jurisprudence of the Constitutional Court, which has included these ethnic groups within the category “beneficiaries of special constitutional protection” and in general has kept, up to the present, a jurisprudential development that guarantees and expands the content of their rights. This constitutional design has been justified as a way to repair the historic discrimination indigenous people and Afro-Colombians have suffered and still suffer, both in terms of recognition and territorial distribution (Fraser, 1997; for the indigenous case see CECOIN, 2007). As a first possible explanation, some see the emergence of ethnic citizenship as a project of modernity that no longer passes through assimilation but rather through the proposal of “being different to be modern” (Gros, 2000: 97). This project is useful to the rules of the neoliberal order, rather than being incompatible, since it promotes forms of indirect rule and selfmanagement that can be functional for the typically neoliberal logic of decentralisation, privatisation and deregulation (Gros, 1997: 32-8). Further, multiculturalism, by defining the conditions for these “others” to achieve recognition as subjects of rights, allows the co-optation and domestication of otherness and even its commoditisation by transforming it into a “cultural heritage”, worthy to be preserved and made profitable (Hale, 2005; Comaroff and Comaroff, 2009). In addition, this multicultural project does not conflict with the neoliberal rejection of ensuring universal access to basic goods, instead it focuses on the distribution of these goods among those individuals who are successful in making use of their ethnicity. In terms of censuses, this multicultural turn is reflected in the importance granted to the counting of ethnicity in the censuses of 1993 and 2005 (in Colombia the census for 2010 is planned for 2015). Unlike previous censuses, in 1993 and 2005 the Afro-Colombian population was counted; the 2005 census also included the Roma or Gypsy people for the first time. The attempt to transcend the territorial criterion is also noticeable, because the ethnicity question was applied to everyone counted in the census. Although in the 1993 census the application of the specific ethnic groups form (F2) was limited to the areas with predominance of such groups, in the 2005 census the questions designed to capture ethnicity were applied to all people, through a single data collection instrument (Bodnar, 2005: 246; DANE, 2010: 13-19; Urrea, 2011: 73-75). This represents a significant difference from the 1985 census, in which only the

Who Counts Indigenous People, How are they Counted and What For? 111

indigenous population living in resguardos were included, leaving out of the record not only urban Indians but also those living in other rural indigenous territories not recognised as resguardos (Uribe, 1998: 42; DANE, 2010: 13)16. Finally, the interest in making self-identification prevail over the remaining criteria of ethnic identification stands out, in particular over those based on “politically incorrect” and illegal phenotypic traits, after the discrediting of “race” as a category for classifying populations17. This emergence of ethnic identities is also reflected in the census results for indigenous population: from 532,233 Indians counted in 1993, to 1,378,884 in 2005, with an intercensal growth rate of 159%. Meanwhile, the total Colombian population only increased by 25.24% over the same period (Rey, 2007: 152). Now, who counts indigenous people in multicultural Colombia? The DANE is the entity in charge of carrying out the general census of population and of making the annual projections of indigenous population. The allocation of budgetary resources coming from the General System of Participations (“transferencias”) to the resguardos depends on these projections. However, besides the DANE, many other institutions are in charge of counting the Indians for different purposes. Therefore the answer to the question of who counts indigenous people cannot be separated from the answer to the question of what they are counted for. In terms of counts, not of individuals but of indigenous groups, the role attributed to the following two entities is highlighted: the Colombian Rural Development Institute INCODER (former INCORA)18, the entity responsible for the titling of indigenous land, and the Agustín Codazzi Geographic Institute (IGAC), in charge of mapping the country, including the land of resguardos. As discussed below, the dominant representations maintain a close link between indigeneity and territory. If we take into account that the resguardo represents the main form of indigenous territoriality recognised by the state and that it is the only one that has full constitutional guarantees (inalienable, indefeasible, imprescriptible and receiving budgetary resources from the General System of Participations), the titling of indigenous land as a resguardo has implications for the recognition of indigenous identity, both for the human group that inhabits the land, and for each one of the individuals in that group. At the same time, the Bureau for Indigenous Affairs, Minorities and Roma, under the Ministry of the Interior, is the agency responsible for certifying the existence of indigenous communities and resguardos. It is also responsible for making socioeconomic studies for the establishment, consolidation, expansion and restructuring of resguardos. It also keeps

112

Chapter Four

track of the census information on indigenous communities, from the censuses conducted every year by the indigenous authorities exercising the powers conferred by the old Act 89 of 189019. These census lists, as well as the certificates of community membership that the indigenous authorities issue based upon the lists, are the means through which individuals prove their indigeneity for purposes of benefiting from special rights, such as the exemption from military service, access to subsidized health, and special quotas for higher education, among others. Thus, in multicultural Colombia, the power to count and recognise who is indigenous is shared and often disputed between the state and the indigenous peoples themselves. These disputes express one of the paradoxes that accompany multicultural constitutionalism.

5. The paradox of self-recognition One of the consequences of having self-recognition as a prevailing criterion to build up indigeneity is to privilege self-censuses as an exercise of both individual (for individuals who go before the indigenous authorities to be counted, confirming the sense of belonging to a community) and collective self-identification (a group of people who, through the annual practice of the census, confirms its existence as a collective self). Despite the importance that is currently attributed to this traditional practice of self-recognition, it is viewed with suspicion by some officials who suspect it as an easy way for certain individuals to exploit indigenous identity in order to gain access to the special rights granted to this sector of the population. Such distrust is expressed, for example, in a Circular issued in 2009 by the Bureau of Indigenous Affairs, Minorities and Roma, in which the nature and extent of indigenous self-censuses is defined20. It states that such censuses have become a “contentious issue” because “some indigenous authorities discretionally incorporate non-Indians or Indians from neighbouring communities”, and because “non-indigenous sectors, or at least, with a dubious level of ethnic consciousness (even for their own people), claim themselves as such with an instrumental intent”. According to the Circular, it has been found that some indigenous leaders have even charged for certificates of community membership. There have also been cases of dramatic changes in the composition and size of an indigenous community, either by excessive increase or unjustified decrease of its members. This leads to the activation, through this Bureau, of the state’s function of control and surveillance over self-censuses of communities because:

Who Counts Indigenous People, How are they Counted and What For? 113 Certification of non-indigenous people as members of an indigenous community does not just imply a trivialisation of the notion of community, but rather the denial for many truly indigenous people of goods and services reserved for them, as well as the diversion of public resources whose specific destination is to protect indigenous peoples.

This argument reveals the genuine concern of the head of the state agency that is responsible, in its own words, “for control and surveillance so as to prevent the rights of indigenous people from being denied and, instead, granted to other social sectors”. But this concern, in turn, expresses the paradox that multicultural constitutionalism has to face in a context of marked social inequality, such as the Colombian one. In the Colombian context, satisfaction of claims for redistribution through a policy that ensures access to basic goods for all people is replaced by a social policy that limits access to those goods to vulnerable population groups. Indigenous people are generously included in these groups since they are not required to prove their vulnerability using the identifying instruments that apply for other sectors of the population21. The paradox is that multiculturalism requires identifying (and setting legal standards for) an indigenous subjectivity (individual and collective) in order to establish who are the legitimate beneficiaries of the rights that the Constitution reserves to that subjectivity. But since the prevalence of self-recognition implies attributing the power of definition of the indigenous being to those who claim this identity for themselves, such a criterion threatens to become a sort of “Trojan horse” that destroys from within the regime of differentiated rights on which multicultural constitutionalism is based. Faced with this paradox, the state needs additional criteria, other than self-identification, which allow it to draw the line between the “truly indigenous” and those who are so only in appearance, in order to maintain control over the construction of subjectivity. And the answer to the question about what these “objective”, or at least, “intersubjective” criteria are, in order to identify indigenous people, emerges from the dominant representations of indigeneity. These are more willing to recognise indigenous identity to the individual or the group who are the more different (in physical and cultural terms) and distant from the modern “us” (Ariza, 2009: 300-16). However, in line with the recognition of the reindigenisation processes that have occurred in communities classified as farmers until a few decades ago, the Circular of the Ministry of the Interior attempts to move away from essentialist constructions of indigeneity, to articulate it in rather political terms, when it states that the inclusion of a person into the census conducted by the cabildos must be:

114

Chapter Four The expression of a collective process of ethnic affiliation in which verifiable rights and duties are set by means of, for example, participation in community events, the person’s integral self-recognition as an indigenous individual, the making of contributions, subjection to the internal rules, and so on.

The remaining question is: what does “integral self-recognition as an indigenous person” mean? Furthermore, since the Ministry of the Interior assumes a controlling and monitoring function on the inclusion of a person into the self-census made by the indigenous cabildos, one wonders if this “integral self-recognition” does not translate into a faculty that the state attributes to itself to submit the “indigenous soul” to examination in order to verify that an “apparent” indigenous person is not in disguise (Ariza, 2009: 317). All this supports Christian Gros’ statement when he says that “under these new forms and with the discourse of respect for the cultures, for the traditional forms of collective organisation, and so on, the state has never been as present in the internal affairs of the communities as it is now” (Gros, 2000: 105). In short, the criterion of self-recognition does not completely transfer to the indigenous people the power to define themselves as such, but becomes a new rhetorical space where there is an arm-wrestle between the subject (individual or collective) that struggles to be recognised as indigenous and other social actors, who play the role of interlocutors in that process of mutual recognition which is implied in any construction of identity. In the specific case of indigenous self-censuses, this struggle is between the indigenous authority and the government agency that is responsible for the “control and surveillance” of the indigenous censuses; an agency that, in turn, is consulted by other public and private entities when they have doubts about the validity of the indigeneity certification issued by an indigenous governor. Another government entity, in this case the constitutional jurisdiction (headed by the Constitutional Court), is in charge of settling such disputes and thus has the ultimate power to define the criteria for identifying who is an Indian in Colombia. Hence the study of constitutional jurisprudence is of particular importance when examining the performing power of law to build the social representations of indigeneity, but also the way the law is influenced by such representations. On several occasions the Constitutional Court has dealt with disputes relating to the recognition of indigeneity. An example is Decision T703/2008, where it settled an acción de tutela–a sort of writ for the protection of fundamental rights–filed by an applicant to enter the University of Valle through the special quota for indigenous people. The university denied the indigenous status of the applicant because, despite

Who Counts Indigenous People, How are they Counted and What For? 115

presenting a certificate issued by the governor of the indigenous community of Tacueyó, his name was not recorded in the census that had been reported to the Bureau of Indigenous Affairs of the Ministry of the Interior. In this ruling the Court held that the certification issued by the indigenous governors constitutes full proof of the belonging of an individual to a community, and there is no need to subject it to ratification by the Ministry of the Interior. On the contrary, “the census coming from indigenous communities should have greater weight and serve as a basis for correcting the census performed by the Ministry of the Interior”. But even though in this and other recent decisions the Constitutional Court has given prevalence to the criterion of self-identification, in some earlier decisions (and especially in those of abstract control of constitutionality of laws) it has used a concept of indigeneity in which the former criterion is displaced by others that require that the individual who is recognised as indigenous remains in an indigenous territory and reproduces indigenous cultural practices, which are different from the ones of mainstream society. This happened in Decision C-058/1994, in which the constitutionality of the law providing for exemptions from compulsory military service was endorsed only for the Indians “living within their territories and who preserve their cultural, social and economic integrity”. Although in this case the Court did not expressly deny the indigenous status to those who did not meet the conditions imposed by the rule, it implicitly establishes a border between “authentic” Indians–”those who live with and like Indians”–and the “apparent” Indians, that is, those living near “us” and resembling “us”; for the latter, it would be acceptable to be excluded from the benefits that multicultural Colombia reserves for “true” Indians. Also, in Decision C-394/1995 the obligation to imprison indigenous people in special penitentiaries is declared constitutional, clarifying that it applies “only to those individuals currently belonging to autochthonous indigenous populations”, since a broader understanding would lead to dissolve the borders between these “authentic” indigenous persons and the vast majority of Colombians who could claim to have aboriginal ancestry (Ariza, 2009: 305-10). One of the predominant forms of drawing that border is confining Indians into a territory, preferably the one of the resguardo. However, it expresses another of the paradoxes that the construction of indigeneity in multicultural Colombia faces.

116

Chapter Four

6. The Paradox of the territorial link “Drua ne ea, Embera ne ea” (an Indian without land is not an Indian). That is how Kimy Pernía Domicó, leader of the Embera Katío group of the High Sinú, who was disappeared and killed by paramilitaries in 2001, expressed this vital link between indigenous people and territory (Jaramillo, 2011: 13). Also Otilia Aricapa, an Embera Chamí Indian of the Cañamomo-Lomaprieta resguardo, when asked what indigeneity meant to her, answered bluntly: “Being an Indian is struggling for land”. This earthy sense of indigeneity is a central element of the way in which Indians have built their identity throughout history. It is an important marker of cultural difference with the white, western capitalist society (that “us” in opposition to which indigeneity represents the “other”), where the earth is conceived as an object-resource to exploit. In contrast, for the indigenous peoples such opposition would not exist, but rather an existential continuity with the earth, personalized as the “mother” from which life comes. The link to the land also has an important political role for Indians, since indigeneity, from the conquest until today, has been built upon the struggle for the territory stripped from them with the arrival of the Spaniards. We must also not forget that this link is also the result of the colonial policy–maintained until today–of segregation of indigeneity in the space of resguardos. In the general population censuses of 1993 and 2005 the question regarding ethnicity was not only applied in the indigenous territories but also in the whole country, which allowed identifying indigenous people living outside of the former. However, the fact is that staying on an indigenous territory is a decisive criterion for inclusion in the self-censuses that communities carry out, on which the recognition of the rights reserved for Indians depends. As Aquileo Yagarí, governor of the Embera Chamí resguardo of Karmata Rua, says “one gives recognition to the people inside”. In applying this criterion, twenty indigenous families from this resguardo that had lived in the city for several years and no longer spoke the Embera language were eliminated from the internal census. Only those who leave the territory to study are counted, or those who, being outside, maintain links with the organisational process of the community. Like the indigenous authorities, state authorities are also reluctant to acknowledge an indigenous character to those people living outside the resguardos, especially those who live in cities22. This difficulty to visualise and recognise urban Indians contrasts with the figures yielded by the 2005 census, which reveals that the proportion of the indigenous population residing in municipal urban areas increased

Who Counts Indigenous People, How are they Counted and What For? 117

from 7.42% in the 1993 census to 21.42% in 2005 (DANE, 2010: 29). As Schkolnik and Del Popolo point out, Latin America is one of the most urbanised regions of the world; a trend which also applies for the indigenous peoples; especially when the inequalities in access to basic goods, which pushes migration to urban areas, is added, in Colombia, to the scarcity of land in many of the resguardos in the Andean region and to the high rates of forced displacement that the armed conflict produces (Schkolnik and Del Popolo, 2005: 121; Rey, 2007: 161; Urrea, 2011: 779). The paradox to which I refer is that, while on the one hand the link between indigenous people and territory has been and continues to be a decisive factor to socially strengthen and legitimise the indigenous struggles for land titling, on the other hand, it represents an obstacle to recognising indigeneity in urban contexts. The existence of indigenous individuals in cities is barely acknowledged, and is excluded for indigenous communities, who as a collective self could claim collective rights before the state, such as ownership of territory and prior consultation23. A recent case decided by the Constitutional Court highlights the difficulty in recognizing the presence of indigenous communities in cities. Decision T-282/2011 solves a set of acciones de tutela filed by 120 indigenous families who were displaced by violence and settled in a public lot located in a slum in the city of Cali, and who were about to be evicted by the city administration. The families, belonging to the Nasa and Yanacona ethnic groups, constituted an indigenous cabildo in order to reproduce their culture, their modes of social and political organisation, thus building territoriality in an urban space. The Constitutional Court granted the requested protection to these indigenous families, but not because it admits the possibility of establishing indigenous land in cities, rather as a temporary measure (a temporary shelter situation) to be maintained only until state authorities guarantee the return of these families to their communities of origin or, failing that, until they can collectively be relocated in a (rural) territory, where they can reproduce their indigeneity. The key reason that led the Court to grant protection is not the indigenous identity claimed by the applicants, but their double condition of indigenous and displaced persons.

118

Chapter Four

7. Multiculturalism at war: towards the construction of the “indigenous–victim of the armed conflict” The aforementioned decision illustrates the drift that is evident in the construction and counting of ethnicity in recent years, i.e., that construction and census recording is no longer of the Indian (or AfroColombian) as such, but also of the “indigenous (or Afro-Colombian)– victim of the armed conflict”. Decision T-025/2004 opened this trend, since the Constitutional Court declared the existence of an “unconstitutional state of affairs” (estado de cosas inconstitucional) in relation to the protection of fundamental rights of displaced people and issued a series of structural rulings, which are the basis for the construction of public policies of support to this sector of the population. Mindful of compliance with the orders issued in this decision, the Court has issued a series of follow-up Court orders requiring adoption of a differential ethnic approach to address displaced people that would give special attention to the indigenous groups (Court Order 04/2009) and Afro-Colombians (Court Order 05/2009) which are in this situation. To justify this differential approach, the Court appeals to the fact that, according to census results, ethnic groups are among the sectors most affected by forced displacement. The Court also appeals to the need to design specific measures to protect their cultural identity and their collective relationship with the territory (DANE, 2007: 44; Rodríguez and Lam: 2011:9). In the recently issued Law 1448 of 2011 (Victims Act) this differential approach is also incorporated, ordering a specific regulation to provide reparation to Indians, Roma and Afro-Colombians through Decrees 4633, 4634 and 4635 of 2011 respectively. One of the main challenges in incorporating this differential ethnic approach is to define the criteria to construct the subjectivity of the “indigenous (or Afro-Colombian)–victim of the armed conflict”. The difficulties of identifying who is a Victim are now added to the those of building an indigenous (and Afro-Colombian) subjectivity that captures the multiple ways in which both ethnicities express themselves in Colombia, but that also allows the building of the “ethnically and culturally differentiated” subject claimed by multicultural constitutionalism. Although in the case of indigenous people the conceptual difficulties in defining indigeneity are actually solved through the traditional practice of self-censuses, the situation becomes more complex in the case of AfroColombians, whose organisational process is more recent and who do not have a consolidated practice of self-censuses24. Today, a great deal of the institutional efforts of the state and of the indigenous and Afro-

Who Counts Indigenous People, How are they Counted and What For? 119

Colombians organisations themselves concentrates on defining the identification criteria of this new subject of differentiated rights. Its construction combines claims of collective ethnic justice, which inspire recognition of differentiated rights for indigenous and Afro-Colombian people, with those of transitional justice, aimed at compensating the victims of the armed conflict (Rodríguez and Lam, 2011). In conclusion, a parallel can be drawn between the colonial legal order and the order that is typical of multicultural constitutionalism. In fact, both steer away from the construction of a single subject of law, typical of liberalism, to instead claim the construction of differentiated identities as a criterion to grant differentiated rights. Indigenous subjectivity is thus not surprisingly crucial to both the colonial and the multicultural order. The construction of this subjectivity in both orders is based upon a different rhetoric: the consideration of the indigenous people as similar to the “rustic”, “miserable” and “minor” of Castilian law differs from their definition as a culturally diverse subject that enriches the cultural heritage of the nation. However, some of the elements that law uses, yesterday and today, to identify who is an Indian, highlight both inferiority (under the guise of “vulnerability”) and the physical and cultural distance as typical of the indigenous being. Nonetheless, the criteria for counting and classifying Indians have changed. While the distinction between tributarios, forasteros and gentiles was important in the Colony, it was displaced in the Republican period by the distinction between Indians of resguardos and wild Indians of the lowlands. In multicultural Colombia, affected by an internal armed conflict of long duration, this distinction has become less important for legal purposes, and has been replaced by the distinction between urban Indians, Indians from resguardos and Indians living in rural areas without a legally recognised territory. To this distinction the one of “Indian victim of the armed conflict” is now added. The latter, if the current trend continues, will displace “mere indigeneity” as a condition for access to those basic goods that a decent society should guarantee all of its members.

Bibliographical references Andrade, Luis Evelis. 2010. ¿Cuáles son, cuántos son y dónde se ubican los pueblos indígenas de Colombia? http://onic.org.co/Pueblos.shtml, consulted on September 5, 2011. Arango, Raúl and Enrique Sánchez. 1997. Los pueblos indígenas de Colombia. Bogotá: Tercer Mundo Editores – Departamento Nacional de Planeación.

120

Chapter Four

Ariza, Libardo. 2009. Derecho, saber e identidad indígena. Bogotá: Siglo del Hombre. Bastidas, Edith. 2007. Conteo y caracterización de los indígenas de Colombia: una mirada a los procesos de censo. In Elena Rey (ed.) Indígenas sin derechos. Situación de los derechos humanos de los pueblos indígenas. Informe 2007. Bogotá: CECOIN. Bodnar, Yolanda. 2005. Pueblos indígenas de Colombia: apuntes sobre la diversidad cultural y la información sociodemográfica disponible. Notas de población, 79: 231-262. http://www.eclac.org/publicaciones/xml/5/23525/notas79-cap8.pdf Bonilla, Daniel. 2006. La constitución multicultural. Bogotá: Siglo del Hombre. Bonet Vélez, Diana. 2002. Tierra y comunidad. Un problema irresuelto. El caso del altiplano cundiboyacense (Virreinato de la Nueva Granada) 1750-1800. Bogotá: Instituto Colombiano de Antropología e Historia – Universidad de los Andes. Borah, Woodroh. 1972. La demografía histórica de la América Latina: fuentes, técnicas, controversias, resultados. Bogotá: Universidad Nacional de Colombia – Cuadernos de Historia Social y Económica. Ceballos-Bedoya, Nicolás. 2011. Usos indígenas del derecho en el Nuevo Reino de Granada. Resistencia y pluralismo jurídico en el derecho colonial. 1750-1810. Revista Estudios Socio-Jurídicos, 13(2): 223-247. CECOIN – Centro de Cooperación al Indígena. 2007. Indígenas sin derechos. Situación de los derechos humanos de los pueblos indígenas. Informe 2007. Bogotá: CECOIN. Clavero, Bartolomé. 1994. Espacio colonial y vacío constitucional de los derechos indígenas. Anuario Mexicano de Historia del Derecho, VI: 61-86. Comaroff, John and Jean Comaroff. 2009. Ethnicity, Inc. Chicago London: The University of Chicago Press. DANE – Departamento Administrativo Nacional de Estadística. 1975. Estadísticas históricas. Bogotá: DANE. —. 2007. Colombia: una nación multicultural. Su diversidad étnica. Bogotá: DANE. http://www.dane.gov.co/files/censo2005/etnia/sys/colombia_nacion.pdf —. 2010. La visibilización estadística de los grupos étnicos colombianos. Bogotá: DANE. http://www.dane.gov.co/files/censo2005/etnia/sys/visibilidad_estadisti ca_etnicos.pdf Escobar, María Elvira. 1999. Sírvase reconocer todos los indios que estamos en el partido de Lomaprieta. Virajes, 3(1): 6-16.

Who Counts Indigenous People, How are they Counted and What For? 121

Fajardo, Darío. 1971. Datos censales sobre la población indígena colombiana. In Darío Fajardo, Alfredo Cadavid, Francisco Arango Montoya y Julio Tobón (eds) Ayer y hoy de los indígenas colombianos. Bogotá: DANE. Fals Borda, Orlando. 1957. El hombre y la tierra en Boyacá. Bogotá: Ediciones Documentos Colombianos. Fraser, Nancy. 1997. La justicia social en la época de la política de la identidad: redistribución, reconocimiento y participación. Bogotá: Universidad de Los Andes [Estudios ocasionales del CIJUS. Catedra Ciro Angarita]. Friedemann, Nina S. de. 1981. Niveles contemporáneos de indigenismo en Colombia. In Nina S. de Friedemann, Juan Friede and Darío Fajardo (eds) Indigenismo y aniquilamiento de indígenas en Colombia. Bogotá: CIEC. González, Fernán E. 2006. Ciudadanía, ley y presencia diferenciada del Estado. In Fernán E. González and Gloria Isabel Ocampo (eds) Globalización, cultura y poder en Colombia: una mirada interdisciplinaria. Medellín: Universidad de Antioquia – Colciencias. González, Margarita. 1970. El resguardo en el Nuevo Reino de Granada. Bogotá: Universidad Nacional. Gros, Christian. 1997. Indigenismo y etnicidad: el desafío neoliberal. In María Victoria Uribe y Eduardo Restrepo (eds) Antropología en la modernidad. Bogotá: Instituto Colombiano de Antropología. —. 2000. Políticas de la etnicidad: identidad, estado y modernidad. Bogotá: ICANH. Hale, Charles. 2005. Neoliberal multiculturalism: the remaking of cultural rights and racial dominance in Central America. PoLAR, 28 (1): 10-28. Herrera Ángel, Marta. 2007. Ordenar para controlar. Ordenamiento espacial y control político en las llanuras del Caribe y en los Andes Centrales Neogranadinos. Siglo XVIII. Medellín: La Carreta. Jaramillo, Efraín. 2011. Kimy, palabra y espíritu de un río (Kimy bed’ea jauri ome dod’ebena). Bogotá: IWGIA – Colectivo Jenzerá. Jaramillo Uribe, Jaime. 1978. Manual de historia de Colombia. Tomo I. Bogotá: Instituto Colombiano de Cultura. Jiménez de la Espada, Marcos. 1897. Relaciones geográficas de Indias. Perú. Madrid: Tipografía de los hijos de M. G. Hernández. Lloréns, José A. 2002. Etnicidad y censos: los conceptos básicos y sus aplicaciones. Bulletin de l’Institut Français d’Etudes Andines, 31(3): 655-680.

122

Chapter Four

Morales Folguera, José Miguel. 2001. La construcción de la utopía. El proyecto de Felipe II (1556-1598) para Hispanoamérica. Madrid: Editorial Biblioteca Nueva. Mörner, Magnus. 1970. La corona española y los foráneos en los pueblos de indios de América. Estocolmo: Almqvist & Wiksell. Ots Capdequi, José María. 1941. El estado español en Las Indias. México: El Colegio de México. Patiño, Víctor Manuel. 1983. Relaciones geográficas de la Nueva Granada (siglos XVI a XIX). Cespedesia, 45-46, suplemento No 4. Patiño Millán, Beatriz. 1993. Indios, negros y mestizos. La sociedad colonial y los conceptos sobre las castas. In Amado Guerrero Rincón (ed.) Ciencia, cultura y mentalidades en la historia de Colombia. Memorias 8º congreso nacional de historia. Bucaramanga: Universidad Industrial de Santander: 41-76. Quijano, Anibal. 2000. Colonialidad del poder y clasificación social. Journal of World-System Research, VI(2), Special Issue: Festschrift für Immanuel Wallerstein – Part I: 342-386. Reichel-Dolmatoff, Gerardo. 1959. Indígenas de Colombia. América Indígena. Vol. XIX(4): 245-253. Rey, Elena. 2007. ¿Qué hay detrás de un número? Análisis de los “síntomas” sobre la situación de los pueblos indígenas desde una lectura del censo de 2005. In Elena Rey (ed.) Indígenas sin derechos. Situación de los derechos humanos de los pueblos indígenas. Informe 2007. Bogotá: CECOIN. Rodríguez, César and Yukyan Lam. 2011. Etnoreparaciones: la justicia colectiva étnica y la reparación a pueblos indígenas y comunidades afrodescendientes en Colombia. Bogotá: DEJUSTICIA. http://www.dejusticia.org Roldán, Roque (comp.). 1990. Fuero indígena colombiano. Bogotá: Presidencia de la República. Scholknik, Susana and Fabiana del Popolo. 2005. Los censos y los pueblos indígenas en América Latina: una metodología regional. Notas de población, 79: 101-132. http://www.eclac.org/publicaciones/xml/5/23525/notas79-cap4.pdf Segato, Rita Laura. 2007. La nación y sus otros. Buenos Aires: Prometeo. Silvestre, Francisco. 1968. Descripción del Reyno de Santa Fé de Bogotá (1789). Bogotá: Universidad Nacional. Todorov, Tzvetan. 1987. La conquista de América. El problema del otro. México: Fondo de Cultura Económica.

Who Counts Indigenous People, How are they Counted and What For? 123

Tovar Pinzón, Hermes, Camilo Tovar M. and Jorge Tovar M. 1994. Convocatoria al poder del número. Censos y estadísticas de la Nueva Granada 1750-1830. Bogotá: Archivo General de la Nación. Uribe, Margarita. 1998. Los grupos étnicos de Colombia: intentos de cuantificación y criterios para el censo de 1993. Bogotá: Departamento Nacional de Estadística. ftp://190.25.231.247/books/LD_10320_1993_EJ_5.PDF Urrea, Fernando. 2011. Contar y ser contados. El censo 2005 y las minorías étnicas. In Margarita Chaves (ed.) La multiculturalidad estatalizada. Bogotá: Instituto Colombiano de Antropología e Historia ICANH. Vidales, Luis. 1978. Historia de la estadística en Colombia. Bogotá: Banco de la República – Departamento Nacional de Estadística. Valencia Villa, Hernando. 1997. Cartas de batalla: una crítica del constitucionalismo colombiano. Bogotá: CEREC. Yrigoyen, Raquel. 2010. Pueblos indígenas. Constituciones y reformas políticas en América Latina. Lima: IIDS.

Notes 1

This work is the final result of the research project “Interactions between State Law and Indigenous Law in Colombia: A view from Law and History”, carried out in 2010 under the auspices of EAFIT University. In addition to this entity, the author is grateful to Fidel Mignorance from HREW, Nicolás Ceballos Bedoya, Laura Posada Correa, Yulieth Hillón, Libardo Ariza, Carlos Ariel Ruiz and Nestor Fabio Garcia, who provided valuable cartographic and documentary information; to Lina Lorenzoni, Tatiana Lopera, Robert Dover and Fernando Crespo for their assistance with the translation. 2 The resguardo, an institution that traces its origins to colonial times, is currently the main form of indigenous territoriality recognized in Colombia. The law defines it as “[...] a legal and socio-political institution of a special nature, made up of one or more indigenous communities, who own a territory under a collective title on land that benefits from guarantees as private property, and who administer it as well as their internal life by means of an autonomous organisation with its own jurisdiction and its own regulatory system” (Article 21 of Decree 2164 of 1995). 3 The mita was the duty imposed on the Indians to work regularly for a certain time at the service of the Spanish, in mining, agricultural, pastoral or domestic activities. The duration of the mita depended on the type of work done and a salary was set as compensation for services rendered. The encomienda entailed the delivery of a group of Indian families to the authority of a Spanish encomendero, who had the obligation to protect them and provide for their religious instruction; in return, the Indians were required to provide personal services and to pay their encomendero for various economic benefits (Ots Capdequí, 1941: 28-34).

124

Chapter Four

4 As explained by Bartolomé Clavero, “rusticity” meant the lack of participation in literary culture, which is why the subjects so qualified were outside the scope of much of the legislation and could be governed in these matters according to their customs. The status of “miserable” was assigned to those who “not being socially able to govern themselves required special protection”; in Castilian law this was the case for the destitute, orphans and poor widows. Meanwhile, the concept of “minority” supposed legal incapacity of the individual, due to a limitation of their rationality: “they were not really people of full reason” (Clavero, 1994: 70-2). 5 The Geographical Relations are historical documents containing the answers to questionnaires sent by the Spanish Crown to be completed in each city or subordinated political unit in America. The answers were obtained from interrogating Indians and Spaniards of long time residence (Borah, 1972: 18). The first questionnaires were developed in the sixteenth century by Philip II as part of a project called the “Book of Descriptions of the Indies”, commissioned to Juan López de Velasco. The main questionnaire used to prepare the answers in the form of Geographical Relations dates back to 1577 and it demanded extensive information on “topographical and biophysical characteristics of land, number of Spanish and Indians, distances between the different ecclesiastical and civil jurisdictions, animals, plants and mineral resources, as well as ethnographic issues” (Morales Folguera, 2001: 39-40). Although this is one of the main sources of demographic information and for documenting the integration of the indigenous population into the Spanish colonial regime in the sixteenth century, its fragmented nature is notable. The largest number of these Relations corresponds to the nuclei of the Viceroys of Nueva España (present day Mexico) and Peru, with very little information on the peripheral areas. The information on the territory which corresponds to Colombia today, which in the sixteenth century was part of the Viceroyalty of Peru, is found in the compilations made by Marcos Jiménez de la Espada (1897) and, particularly for the territory of Nueva Granada, by Victor Manuel Patiño (1983). 6 Table 4-1 summarises the total population by strata and Audiencia, which Hermes Tovar Pinzón also presents disaggregated by provinces and population strata. In contrast with Tovar Pinzón’s figures, other sources indicate that the 1778 census showed a total population of only 828,775 (without specifying whether it is the total population of the Nuevo Reino de Granada or just of the Audiencia of Santa Fe. This seems to be the case, given the proximity to Tovar Pinzón’s figures) (DANE, 2007: 30). 7 The terms “neighbours” and “free” were used interchangeably from the mideighteenth century to designate people of mixed descent, Black, Freeman origin or poor White, born from the many racial mixtures resulting from very different shades of miscegenation and who, by then, were trying to find accommodation in a colonial order that had not yet provided a place for them (Bonnet Vélez, 2002: 13, 39-42; Herrera Ángel, 2007: 193-8). 8 This author points out that, among others, the caciques (chiefs) and their older children, the Indian mayors of resguardos, women, and the Indians who submitted peacefully to the colonial order were exempt from paying the tributo; for the latter the exemption was extended for a period of 10 years (Ots Capdequí, 1941: 31-2).

Who Counts Indigenous People, How are they Counted and What For? 125 9 The communities involved in this dispute have maintained until the present their character of resguardos of colonial origin, located in the Andean region in the jurisdiction of the municipalities of Supía and Riosucio (Caldas). 10 Despite the adoption of several laws ordering the liquidation of the resguardos and the distribution of their land among the Indians, for various reasons its application did not became effective in many areas of the country. However, this did not mean indigenous peoples retained their land rights, due to the increasing occupation of the lands of resguardos by mestizo settlers and landowners, as well as to the intense process of assimilation of the Indians to the peasantry which took place in many of the resguardos in the Andean region. 11 After independence, an initial political unit was formed called La Gran Colombia (Great Colombia), with jurisdiction over the present territories of Colombia, Venezuela, Ecuador and Panama. Gran Colombia disintegrated after the separation of Venezuela and Ecuador in 1830. In 1832 the rump political unit was renamed the Republic of Nueva Granada. In 1858 it became the Granadine Confederation and in 1863, the United States of Colombia. In 1886 it acquired the definitive name of Republic of Colombia, from which the territory now corresponding to Panama was split in 1903 as an independent state. 12 The creation of CRIC (Regional Indigenous Council of Cauca) in 1971 marked the first milestone of this organisational process, followed in the early 1980s by the founding of ONIC (National Indigenous Organisation of Colombia) and other regional organisations such as OIA (Indigenous Organisation of Antioquia), CRIDEC (Regional Indigenous Council of Caldas) and CRIT (Regional Indigenous Council of Tolima), among others. 13 An example is found in Constitutional Court Decision T-1253/2008, which ruled upon a dispute arising within a Pijao community of Tolima regarding which of its members were entitled to vote to elect the indigenous cabildo. While a sector claimed that only the founding families were entitled to vote (based on the census conducted by the INCORA in order to constitute the resguardo) another sector wanted to grant the right to vote based upon the census conducted by the community. 14 This denomination was given during much of the twentieth century to the internal frontier areas located in the Amazon and Orinoco regions, considered to be vacant land belonging to the nation. Until the constitutional reform of 1991, those territorial entities did not have the rank of departments, but of administrative divisions and provinces. 15 Some studies suggest that the high volume of indigenous population recorded in the census of 1973, compared with the ones in 1964 and 1985, was due to the fact that the indigenous people of the resguardos in Cauca were counted twice (Uribe, 1998: 38). Others attribute the decrease in the number of Indians in the 1985 census to the fact that the indigenous population that lived outside resguardos was not counted as such (DANE, 2010: 13). A factor that could have influenced the high numbers of Indians in the 1973 census is the social context of land claims and revival of indigenous identity that took place at the beginning of the 1970s. 16 Colombian law recognizes other forms of indigenous territoriality, which do not have the same guarantees granted to the resguardos. They are: (i) the reservas

126

Chapter Four

indígenas, consisting of land portions demarcated by INCORA/INCODER and delivered to one or more indigenous communities in usufruct and excluding third parties, which have not yet been titled as resguardos but have the character of “communal lands of ethnic groups” for purposes of benefitting from the guarantees provided in article 63 of the Constitution and in ILO Convention 169; (ii) the comunidades or parcialidades, defined by Decree 2164 of 1995 as “groups of Amerindian descent who are aware of identity and share values, traits, cultural traditions, customs, forms of government and endogenous regulatory systems that distinguish them from other communities, whether or not they have titles to own the land” (Arango and Sánchez, 1997: 213-215). Within this latter category, some distinguish between comunidades civiles (made up of indigenous communities settled in the lands of ancient resguardos that were dissolved) and asentamientos, which are lands without legal established limits, occupied by one or more indigenous communities without titles on it, or government authorisation to usufruct (Uribe, 1998: 57). 17 However, it is noteworthy that in the 2005 census the physical traits associated with skin colour were incorporated as a criterion for identifying Black population or people of African descent. This was done following recommendations made by Afro-Colombian groups, after finding that people who identify themselves as “Black” from physical features outnumber those recognizing themselves as “AfroColombian” on the basis of cultural elements (Bodnar, 2005: 245; DANE, 2010: 19; Urrea, 2011: 75-7). 18 In 2003 INCORA was replaced by the Colombian Institute for Rural Development (INCODER). 19 Article 3 of Decree 1720 of 2008 regulates the functions assigned to the Bureau of Indigenous Affairs, Minorities and Roma. They can be seen in the website: http://www.mij.gov.co/econtent/CategoryDetail.asp?idcategory=52&IDCompany= 2&Name=Asuntos+Ind%EDgenas+Minor%EDas+y+Rom&idmenucategory=52 y http://www.gobiernoenlinea.gov.co/tramite.aspx?traID=3468 20 External Circular No. CIR09-301-DAI-0220 of December 29, 2009, issued by the Bureau of Indigenous Affairs, Minorities and Roma of the Ministry of the Interior. 21 The main instrument used to identify the population benefiting from social programs is the SISBEN Survey (System of identification of potential beneficiaries of social programs), through which the state determines who is entitled to have access to subsidized health, nutrition aid, housing subsidies, education credits, and so on. Indians do not need to be classified in the SISBEN since most of these social programs set special conditions on access for them. In turn, the Constitutional Court has developed the category of “beneficiaries of special constitutional protection” to establish preferential access conditions to basic goods for Indians, Afro-Colombians, children, female heads of households, the elderly, displaced people, victims of armed conflict and other vulnerable population sectors. Being recognized within any of these population groups becomes for many individuals a form of struggle for survival in a country like Colombia, where the precarious social investment budget contrasts with the alarming figures of

Who Counts Indigenous People, How are they Counted and What For? 127 inequality. According to the GINI index, Colombia ranks first in social inequality in Latin America and 4th place in the world. See: http://www.semana.com/nacion/desigualdad-extrema/153207-3.aspx 22 An example of the internal disputes that arise from implementing this census criterion is seen in Decision T-514/2009. Here the Constitutional Court solved the acción de tutela filed against the governor of Chenche-Buenos Aires-Tradicional resguardo, by a member of the community who was excluded from the census and, as a consequence, from the benefits of the General System of Participations. The indigenous authority justified his exclusion because the petitioner and his family migrated to Bogotá on business. In this case, the Court declined to rule on this conflict, arguing that it should be resolved by their own indigenous jurisdiction, under the principle that mandates maximising the autonomy of indigenous communities to settle their internal affairs. 23 However, the growing presence of urban cabildos in some of the major cities of Colombia should be noted. Such cabildos serve as a space for collective organisation for indigenous people. In addition, they carry out some administrative functions, such as preparing the annual census and the certification of membership to the cabildo, for purposes of access to certain social rights for indigenous people. Some of these urban cabildos are recognized by local governments (mayoralties), but have not yet official recognition from the Division of Indigenous Affairs of the Ministry of the Interior. 24 An eloquent example of the difficulties to build this collective subject in the case of Afro-Colombians, are the decisions of the Constitutional Court relating to the protection of the communities of Curvaradó and Jiguamiandó. While in the first decisions (Court Order 05 of 2009) the Court ordered the design of a specific plan of protection and care for these communities, in the most recent ones (Court Order of 18 May 18 2010 and 384 of 10 December 2010) the orders of the Court concentrate on calling on the authorities to conduct a census enabling definition regarding who, among the current inhabitants of the area, is entitled to be included as a member of the communities of Curvaradó and Jiguamiandó.

CHAPTER FIVE THE CONVERGENCE OF THE “INDIGENOUS” AND “LADINO” CATEGORIES IN THE GUATEMALAN CENSUS OF 2002 GEMMA CELIGUETA COMERMA In the most recent population census carried out in Guatemala in 2002, new classifications were introduced in line with changes in what “indigenous” was perceived to mean at international level. In an attempt to break away from the binary classification of the past that divided the Guatemalan population into indigenous and non-indigenous (or ladino), separate categories were established to record ethnic group (grupo étnico, i.e. indigenous/non-indigenous), ethnic belonging or people (pertenencia étnica o pueblo, i.e. Maya, Xinka, Garifuna, ladino and other) and the language in which the respondent had learned to speak (Maya, Xinka, Garifuna, Spanish and other), thus revealing an ethnic complexity not found in previous censuses. In contrast to other Latin American countries such as Bolivia or Brazil (Canessa, 2007: 208), in Guatemala these changes did not reflect an increase in the proportion of the country’s inhabitants who reported in the census as indigenous (41%), which, incidentally, though on a downward trend, has remained more or less stable since the decline found in the 1964 census (42.2%) (see Table 5-1). Nevertheless, there are two striking features of different kinds in the new census data. On the one hand, the number of those reporting as indigenous (4,610,440) is greater than the overall number (4,433,218) for those groups who in Guatemala are officially deemed to be indigenous: the Mayas (4,411,964), Xinkas (16,214) and Garifunas (5,040)1. On the other hand, more people reported as ladino (6,750,170) than non-indigenous (6,626,756), the two terms having traditionally been viewed as synonymous (see Table 5-2). The 2002 census therefore reveals the presence of over 100,000 individuals who, though self-designating as indigenous in the ethnic group box, chose to self-designate as ladino in the box referring to ethnic belonging or

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

129

people. This chapter contains some initial reflections on the appearance of dual indigenous/ladino self-categorisation, or indígenas ladinos, in the 2002 census data. Figure 5-1. Percentage of indigenous population by department

Source: INE (2002: 5 [Thematic map Nº 5]).

Table 5-1. Changes in the relative size of the indigenous, nonindigenous, ladino, Maya, Xinka and Garifuna populations of Guatemala NonLadino indigenous 1950* 53.6% 46.4% 1964* 42.2% 57.8% 1973* 43.8% 56.2% 1981** 41.9% 58.1% 1994** 42.8% 57.2% 2002** 41.0% 59.0% 60% Sources: *Early (2000: 115); **INE (2002: 30). Year

Indigenous

Maya 39.3%

Xinka and Garifuna 0.2%

130

Chapter Five

Table 5-2. Total population, ethnic group, ethnic belonging and language in which the respondent learned to speak, by department, according to the 2002 census Department Country total Guatemala El Progreso Sacatepéquez Chimaltenango Escuintla Santa Rosa Sololá Totonicapán Quetzaltenango Suchitepéquez Retalhuleu San Marcos Huehuetenango Quiché Baja Verapaz Alta Verapaz Petén Izabal Zacapa Chiquimula Jalapa Jutiapa

Total population 11,237,196 2,541,581 139,490 248,019 446,133 583,746 301,370 307,661 339,254 624,716 403,945 241,411 794,951 846, 544 666,510 215,915 776,246 366,735 314,306 200,167 302,485 242,926 389,085

Ethnic group Indigenous 4,610,440 343,154 1,250 104,802 352,903 40,297 8,373 296,710 333,481 338,055 208,200 54,811 248,639 551,295 581,996 127,061 720,741 113,462 73,151 1,574 50,427 46,766 13,292

Non- indigenous 6,626,756 2,198,427 138,240 143,217 93,230 498,449 292,997 10,951 5,773 286,661 195,745 186,600 546,312 295,249 73,514 88,854 55,505 253,273 241,155 198,593 252,058 196,160 375,793

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

131



Department Country total Guatemala El Progreso Sacatepéquez Chimaltenango Escuintla Santa Rosa Sololá Totonicapán Quetzaltenango Suchitepéquez Retalhuleu San Marcos Huehuetenango Quiché Baja Verapaz Alta Verapaz Petén Izabal Zacapa Chiquimula Jalapa Jutiapa

Ethnic belonging Maya 4,411,964 294,757 766 100,992 350,757 33,746 3,427 295,899 333,438 323,848 189,558 49,607 228,444 531,970 579,067 125,694 718,223 109,068 68,504 948 45,558 26,279 1,414

Xinka 16,214 1,322 36 18 52 148 3,592 12 9 95 391 539 207 69 48 38 22 92 84 155 76 33 9,177

Garifuna 5,040 704 8 16 23 99 45 8 4 604 110 24 147 40 8 20 26 67 2,958 35 20 57 17

Ladino 6,750,170 2,229,846 138,640 146,018 94,779 503,750 294,168 11,507 5,640 297,995 209,949 190,749 564,193 300,011 76,044 89,646 57,692 257,238 242,292 198,915 255,921 206,850 378,327

Other 53,808 14,952 41 975 522 1,003 138 235 163 2,174 3,937 492 1,960 14,454 343 517 283 270 468 114 910 9,707 150

132

Chapter Five

Language in which the respondent learned to speak (people aged 3 and over) Total Maya Xinka Garifuna Spanish Other Country total 10,283,387 3,174,383 1,283 3,564 7,080,909 23,248 Guatemala 2,366,743 160,924 375 515 2,198,166 6,763 El Progreso 128,822 553 15 12 128,231 11 Sacatepéquez 228,418 39,883 8 21 187,713 793 Chimaltenango 405,177 185,826 15 24 219,074 238 Escuintla 496,268 23,471 89 80 472,368 260 Santa Rosa 276,840 2,505 101 48 274,126 60 Sololá 281,987 252,291 17 1 29,465 213 Totonicapán 311,081 255,287 4 7 55,710 73 Quetzaltenango 575,309 182,110 61 77 392,646 415 Suchitepéquez 369,283 60,994 37 40 308,031 181 Retalhuleu 222,157 20,770 23 35 201,257 72 San Marcos 725,353 163,568 95 62 561,480 148 Huehuetenango 767,594 464,292 52 56 291,151 12,043 Quiché 588,822 499,293 18 16 88,581 914 Baja Verapaz 195,895 99,793 12 25 96,006 59 Alta Verapaz 691,339 603,042 12 9 88,114 162 Petén 330,452 86,563 34 48 243,649 158 Izabal 285,340 57,351 34 2,340 225,343 272 Zacapa 183,396 675 14 30 182,590 87 Chiquimula 275,222 11,548 31 39 263,486 118 Jalapa 220,300 2,743 28 45 217,335 149 Jutiapa 357,589 901 208 34 356,387 59 Source: INE (2002: 75 [Table A2]). Department

1. Censuses, concepts and history Despite being rather fervent supporters of qualitative methods, when starting our ethnographies we anthropologists usually turn to censuses to determine the places and subject matter to be researched. We generally consider the data they provide to be one of many authoritative voices from which the ethnographic story is formed and use them without having conducted any kind of prior critical analysis. Nevertheless, some anthropologists have put considerable effort into challenging census figures and categories relating to the places in which they were doing their field work, as, for example, in the case of Guatemala in order to try to resolve the chronic undercounting of the indigenous population in official

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

133

statistics (Early, 2000). A few others have taken demography itself as their subject of study. Scheper-Hughes, for instance, criticised the “neutral and objective” status of the categories used by that discipline, accusing it of having a “lack of awareness of the ways in which culture […] structures the questions asked and overdetermines the findings” (2000: 267-8). However, as also pointed out in the same article, such mistaken “neutrality and objectivity” serve to reveal the particular classification system a society uses, the ideas inherent in its thinking and the policies that lie behind them (ibid: 271). This chapter falls into the second camp in that it is more interested in offering some reflections than in reifying a new classification term such as indígena ladino. In fact, my intention is rather the opposite. I am interested in emphasising the idea of having a range not only of categories but also of meanings for any specific category such as “indigenous”, thereby trying to get away from the idea of separating, purifying and clarifying identities by suppressing heteroglossia, as zealously advocated by certain politicians, demographers and even anthropologists (De la Cadena, 2006: 53). What is more, by bringing together the apparently mutually exclusive categories of indígena and ladino, we are not only adding nuances and increasing complexity, we are developing meanings and interpreting history. In short, indígena ladino is a term that is “good to think”, as Lévi-Strauss would say.

1.1 The difficult task of counting Indians and ladinos in Guatemala Counting the indigenous population of Guatemala has been and is a complicated task for three main reasons that are worth outlining here. The first, which is similar to what happens in other Latin American countries, has to do with the meanings of the terms used to classify the population. However much it may these days be the individuals themselves who selfidentify as indigenous, non-indigenous, Maya, Xinka, Garifuna, ladino or other, they have not chosen the terms on offer and such self-designation fails to reveal what De la Cadena (2006: 53-5) calls the conceptual policy. In her view, it is necessary to explore the many meanings that feature in the genealogy of classification terms, the political ideologies they contain and the historical circumstances that have given potency to some meanings while leaving others marginalised. What is the interviewee thinking when he or she answers yes or no in response to the question of whether or not he or she is indigenous? Analysis of conceptual policy may reveal suppressed meanings and show what from a different perspective is self-

134

Chapter Five

evident. In Guatemala, the main terms used to differentiate and classify the population have been Indian (or indigenous) and ladino (or nonindigenous). Although the terms “Indian” and ladino originated in the colonial period, it was in the nineteenth century that they acquired the binary opposition around which Guatemalan identity has been structured ever since, the result being that ladino has become equated with “nonindigenous”. That is why it is very difficult to talk about one term without the other. As Grandin (1999: 24) states in his thesis on the transformations that the K’iche’ of Quetzaltenango underwent between the mid-eighteenth and mid-twentieth centuries, indigenous identity was maintained and transformed in the course of the same historical process of colonialism, capitalism, state formation, social crisis and political terror under which ladino identity was forged. This process will be examined in greater depth in the course of this chapter. The second reason is directly related to the first and has to do with the classification criteria used. How is it decided who is and is not indigenous, and who makes that decision? In the course of Guatemalan census history, these decisions have been made mainly on the basis of three criteria: according to the perception of the interviewer, according to the perception the interviewer has of the social esteem in which the interviewee is held, and according to how the interviewee self-designates. Nevertheless, it is somewhat more complicated than that because both the definitions of the classification terms and the classification criteria employed have a direct relationship with the ways in which relations between the state and the indigenous population are being approached at the time the census is carried out. In this respect anthropology, national and international legislation and lastly the indigenous movement itself have played an extremely important role in determining both the meaning of “indigenous” and the role that indigenous people play in the nation. We shall look at these criteria more closely. The third reason has to do with the political significance of census results. Indeed, deciding whether a country is or is not mainly indigenous has had, and continues to have, important consequences. Some people, especially those within the Maya movement, argue that the number of indigenous people in the country has been and continues to be underestimated. According to them, indigenous people account for over half the population of the whole country, whereas the figure given by the Instituto Nacional de Estadística–INE (National Institute of Statistics) is lower. Apart from the fact that knowing whether a majority of the Guatemalan population is or is not indigenous is a political issue of paramount importance, determining the size of the indigenous population

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

135

has been and is directly contingent on the reasons already given: the meanings of the classification terms used and the type of classification criteria employed. As will have become apparent, the boundaries between “indigenous” and ladino are not clear and raise difficult issues. The problem of the undercounting of the indigenous population, the reasons for which are many, ranging from intentional falsification, such as the alterations made to the 1940 census on the orders of the Presidency (DGE, 1950)2, to the difficulties faced by enumerators in accessing the mainly indigenous rural communities, has been a constant in the country’s census history. Critical voices have included that of Early (2000), who expressed the need to question and contrast the official figures contained in Guatemalan censuses with the figures available from registry offices, which were deemed to be more reliable because they had been gathered by officials who were closer to the population and who therefore had greater knowledge of the identity of the people being recorded. Applying this method, Early (ibid: 115) corrected the undercounting of the indigenous population of Guatemala in the censuses of 1950, 1964 and 1973 (with its share increasing from 53.6% to 56.2%, 42.2% to 50.3% and 43.8% to 48% respectively). If these figures are correct, there was not such a marked drop in 1964 and the indigenous population of Guatemala was still slightly in the majority in the mid-1960s, when the acculturation model was in full swing. Also, as Early himself points out (2000: 25), some anthropologists, such as Adams and Smith, reported that in the communities where they were doing their field work the numbers of indigenous people were considerably greater than those reported in the 1950 or 1964 censuses. More recently, this undercounting has been denounced by indigenous intellectuals who have put forward considerably more generous figures for that sector of the population (Tzian, 1994). Nevertheless, and despite the difficulties inherent in counting the indigenous population, from colonial times the benefits that accrued to the government by doing so made it expedient. It is well known how in the colonial period the government and the Church registered the indigenous population in order to generally facilitate their management and the payment of taxes in particular. Perhaps less well known is the role played by the indigenous authorities who were responsible for collecting taxes and recruiting workers. This role was vital for their authority since it was closely bound up with determining what “indigenous” signified and consequently the size of the indigenous population in the villages they administered. In fact, where there was a significant non-indigenous presence early on these authorities tended to lose legitimacy3.

136

Chapter Five

From the start the tasks of defining, counting and governing the indigenous population were inter-related and, as a result, became an issue in the official censuses conducted by successive Republican governments, despite their assimilationist ideology. The first Statistics Department in Guatemala was set up in 1879 (Early, 2000: 24). Since then there have been ten censuses (1880, 1893, 1921, 1940, 1950, 1964, 1973, 1981, 1994 and 2002), which we could group according to the prevailing paradigm for interpreting who was indigenous and the resulting role it was envisaged that the indigenous population would play in the nation. There have essentially been three paradigms: the racial paradigm, the acculturation/social integration paradigm and, lastly, the multiculturalism paradigm.

1.2 The racial paradigm in Guatemala or the ambiguity of a concept The term ladino, used in Guatemala to designate the non-indigenous population, suggests a hispanicised identity rather than biological miscegenation, although its meaning has changed over time and according to the context in which it is being used. Employed during the colonial era to denote Indians who spoke Spanish and later to refer to the hispanicised rural population, by the end of the nineteenth century it had come to encompass the whole non-indigenous population. In fact, following the liberal victory of 1871, the many different social divisions that had existed in colonial times (Spaniards, Indians, mestizos, blacks, ladinos) were reduced to just two: Indians and ladinos, as can be seen in the liberal censuses of 1880 and 1893 (Taracena, 2002: 104-5). These two censuses established the hegemonic division of the Guatemalan population into the two mutually exclusive categories of ladino and indígena. This meaning is connected not only with the important role mestizos played in the collapse of the colonial structure known as Pueblos de Indios (Rodas, 1999: 14) but also with the fundamental role played by an anti-indigenous ideology in the building of the Guatemalan nation. Indeed, the nineteenth century in western Guatemala was characterised by a particular political dynamic of resistance, struggle, alliances and negotiations between indigenous and non-indigenous authorities over the control of resources (land, taxes and indigenous labour) in the local area. In this dynamic, the indigenous authorities sought to preserve their own institutions in an attempt to retain as much autonomy as possible vis-à-vis the non-indigenous authorities while the latter, by contrast, wanted to abolish such institutions and set up new ones under their control. Such

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

137

attacks on local native institutions in fact gradually eroded the distinct, protected and corporate status of the Indian populations. Sheltering behind a liberal ideology that identified their interests with those of the region (and later the nation) and opposed, in the name of equality and civilisation, to specific political representation for indigenous people, the liberal elites went on to advocate the privatisation of communal property and even the abandonment of those characteristics (language, clothing) that allowed Indians to be seen as a distinct social group. In reality, the new nineteenthcentury institutions embodied the contradiction between egalitarian theory and segregationist practice inherent in liberalism. In other words, although in theory the whole population should have been represented, in practice the indigenous majority was excluded. Such a deviation from the political equality extolled by liberalism needed a theoretical legitimation that was to be found in concepts such as civilisation and racial degeneration. Although the liberals believed, at least in theory, in the equality of all human beings, in their view not everyone had reached the same stage of civilisation. And civilisation was necessary in order to enjoy citizenship and the benefits of progress (Alda, 1999: 847). In seeing indigenous people as partly or completely lacking in civilisation, they believed them to be incapable of knowing their own interests let alone exercising political responsibilities, which needed to be assumed by others on their behalf as they set off on the path to civilisation. Despite often also being poor and illiterate, individuals akin to the old colonial castes who had more of a hispanicised identity were, for their part, able to become–thanks to the possibility of ladino “fusion”–new historical subjects of the nation. Indigenous people, on the other hand, came to represent at most that pre-Hispanic past that differentiated them from the recently abandoned metropolis and, in particular, that present that prevented them from achieving the progress for which they yearned (Taracena, 2002). For all these reasons, and unlike the social Darwinism of the nineteenth century, Latin American racialism did not (to start with) ascribe an innate character to this supposed indigenous inferiority (Favre, 1996: 31). Interpreting its own past as being free and glorious, the drunkenness, criminality, idiocy, dirtiness and laziness it attributed to indigenous people were interpreted mainly as being the result of the colonial yoke to which they had been subjected for centuries and from which they could only be freed with the help of the state. While the evolution of racism in the twentieth century unquestionably ended up adding eugenic measures (such as “whitening” policies) to the well-known nineteenth century proposals

138

Chapter Five

calling for separate schooling and the extinction of languages and customs, such measures were never the only ones or the most important. We could carry on examining the local meanings of race and how they evolved (including also some indigenous versions) but I believe that what I have described so far is sufficient to underscore the notion of “epistemological hybrid” that De la Cadena (2006: 62-8) ascribes to the concept of race in Latin America–a concept that stresses moral defects rather than phenotypes, that mixes faith with science and the notions of evolution and civilisation with heredity and descendancy. Here we should also highlight the “difference” found in Latin American racism that some authors have been trying to explain since the mid-twentieth century (Nogueira, 1954; Hale, 2007) and the racial ambivalence these same authors have identified among ladinos (Hale, 2007) and mestizos (De la Cadena, 2006). Illustrative of the issue that concerns us here was the 1921 census which demonstrates how difficult it was to incorporate the racial paradigm that prevailed in the early twentieth century into Guatemalan reality. As the “scientific” notion of race gained strength, it was considered that census takers should not just tick the race of the respondent according to what they deemed it to be but that they could and should learn how to distinguish one race from another. The instruction manuals compiled for census takers provide interesting information in this regard. Just as De la Cadena (2006: 53) explains in the case of the 1912 census conducted in the city of Cuzco, where a US economist taught the interviewers to look at skin colour, in Guatemala specific instructions were given about how to correctly classify the country’s races, suggesting not only that the local interviewers did not know how to do so but that those being interviewed would have been upset if they were asked a direct question on this subject. In the 1921 census, the instructions to census takers specified the following: Race, box 4. The characteristic features of each race are well defined. Make a discreet note. For example, it is not easy to confuse an individual of the yellow race or a black with an Indian. It should be written down without asking the person being registered what their race is; because in previous censuses such individuals have been wrongly recorded as “ladinos” and “Indians” and this must be avoided now for the sake of ensuring greater accuracy in the entries. (DGE, 1921: LXXXII).

It should be noted that at that time the races in question were “ladino”, “Indian”, “pure black”, “white”, “yellow” (the offspring of a Guatemalan woman and a Chinese man) and “pure Chinese”. Such racial classifications

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

139

clearly took no account of the local meanings of the term ladino which at that time–as stated earlier–already allowed for non-indigenous “fusion”. The political dynamic of the nineteenth century was therefore conducive to the possible joint identification of a heterogeneous population (made up of criollos, peninsulares [people from mainland Spain], hispanicised indígenas and mestizos) who together would be recognised as being in opposition to the indigenous population. It was a political identification which in the second half of the nineteenth century would elevate those in the ladino category to the position of hegemonic representation of the new Guatemalan nation and leave those in the “indigenous” category with a collective stigma to be overcome (Taracena, 1997: 401-8). From then on “Indian” and ladino were the main terms used to classify the Guatemalan population. Although the two categories were seen as opposing races that could be identified simply by looking, the ambiguity of the concept of race in the region would give it a flexibility that not even the “scientific” pretensions of the 1921 census would succeed in changing.

1.3 Anthropology, culture and social integration Change eventually came with the 1950 census–the first professional one according to Early (2000: 7)–which was conducted under the guidance of institutions including the Asociación Estadística Interamericana (InterAmerican Statistics Association) and the Instituto Indigenista Nacional (National Indigenist Institute). Following guidelines drawn up by anthropologists, the 1950 census dissociated itself from the concept of race and adopted that of “ethnic group” for the ladino and indigenous groups, with “the yellow and black races” being included in the ladino group, therefore identifying ladinos as non-indigenous4. Nevertheless–and in line with the anthropological theory of the time–the census specified that, given that there was no single definition of ladino or “indigenous”, census takers should take account of the social esteem in which the person was held in the places where the census was being conducted5. If the person was a servant, the head of the family or woman of the house should be asked and only if none of that were possible was it suggested that the interviewee should be asked directly. The 1950 census also included a whole section on “cultural characteristics”, including use of language, clothing, footwear, food and type of housing, which was supposed to supplement the information concerning the ethnic group to which the person belonged (DGE, 1950: XI-XII). The concern shown in the 1950 census to reflect the local meanings of the terms ladino and “indigenous”, despite the fact that doing so required

140

Chapter Five

resources that would have been hard to mobilise, is striking. It would have been necessary virtually to carry out an ethnography of each place. However, the question remains: given the difficulties in covering the whole population, how was it possible to know the social esteem in which a person was held? Although the censuses indicate that efforts were made to recruit census takers who were natives of each place, Early (2000: 38-9) states that the census system was supervised by teachers from the national education system (mainly ladinos who did not come from the villages in which they taught), making it very likely that in 1950, as well as in 1964 and 1973, there were many doubtful cases which interviewers decided based on their own opinions or according to the clothing, footwear or language used by the interviewee, as suggested in the 1950 and 1964 instructions. The problem is that at that time many indigenous men were abandoning traditional dress, and in some cases also their languages, thereby automatically swelling the numbers recorded in the ladino box. Thus, the acculturation paradigm, according to which indígenas would eventually turn into ladinos, significantly determined the census results. The effort entailed in adapting the country’s social reality to statistical method would turn the 1950 census into an example for practically all the other twentieth century censuses. As acknowledged in the 1981 census6, the censuses of 1964, 1973 and 1981 would continue to adopt social esteem as a criterion for determining ethnic group. Nevertheless, there was one significant variant which is worth highlighting here. In 1964, the ladino group came to be directly designated “non-indigenous”. This change suggests a nuance in the representation of ethnicity which is worth pointing out and one which became an attribute solely of indigenous people7. Furthermore, the state of being “indigenous” was essentially assumed to be an anomaly that needed to be supplanted, preferably as soon as possible. In the official commentary on the results of the 1964 census, the relative 10.3% reduction in the indigenous population that had occurred since the 1950 census was interpreted as being due to the tendency for indigenous people to become socially integrated, possibly as a result–it was noted with satisfaction–of factors such as literacy campaigns, the expansion of the education system and modernisation of the modes of transport (DGE, 1966: 22). Some anthropologists, such as Melville and Melville (cited in Early, 2000: 25), condemned the 1964 ethnic totals as being a “deliberate falsification by a military dictatorship” in order to make the country appear “modern”. It is interesting to note the importance anthropology had both in terms of replacing the concept of race with that of culture and ethnic group and in ensuring the predominance of the acculturation and social integration

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

141

paradigm. North American anthropologists working in Guatemala during the 1930s and 1940s had established the community and municipio (municipality or township) as the basic unit of meaning, and therefore of study, of Guatemalan indigenous societies. The problem of classifying, and therefore studying, indigenous people in Guatemala, as well as one of the possible solutions to it, had already been raised in some of the earliest ethnographic works, when the German scholar Otto Stoll published his Ethnography of Guatemala in 1884 (newly published in 1958)–“the first study of the current Guatemalan indígena”, according to Goubaud Carrera (1964: 110)–and acquired a following with his concept of “cultural mosaic”, which identified and established municipios as autonomous social and cultural units. This idea was reinforced by Redfield and Tax8 and taken up again later by many other ethnologists, including Gobaud Carrera himself9. However, for all that the Guatemalan municipio had been established as the cultural unit, there was no escaping the fact that there were people living there who did not consider themselves to be indigenous. Even though anthropologists from the area were to find defining indigenous to be their main headache10, their analyses returned to the distinction between ladinos and indígenas, determined it to be cultural–and not racial–and granted it an importance that far exceeded other forms of social differentiation (Tax, 1959: 103-34). For these anthropologists, ladinos and indígenas shared a single society albeit with different cultures. They affirmed in their ethnographies11 that indigenous societies were homogeneous and unchanging as far as beliefs and customs were concerned but shifting, and including impersonal relationships (similar to those found in a large city, according to Tax), with regard to other aspects12. Given the supposed permeability of ethnic boundaries and the evidence of ladino hegemony, they wondered what was preventing indígenas from turning into ladinos (Tax, 1959: 103-34). And since the boundaries were seen as cultural, the answers usually came from education. Not only did mid-twentieth century anthropologists defend the social integration of indigenous people in their respective countries, albeit respecting some of their cultural characteristics, they encouraged it by working in government institutions13 that were trying to reconcile anthropological theory with the national debate on the Indian. Nevertheless, in the same way as the Guatemalan “cultural mosaic” had ended up being reduced to a photograph in black (indígena) and white (ladino), the problematic of social and cultural change in exhaustively monographed communities ended up being turned into a general theory about acculturation and the shift from one classification term to another

142

Chapter Five

(indígena-ladino) which in Guatemala went under the name of “ladinisation”. Change was identified with modernisation and the latter with “ladinisation”, which was supposed to be the final destination for indigenous people. Indigenous people were even beginning to be classified according to what stage in that process they had reached (Adams, 1956: 24-5). Curiously, although it had been determined that the boundary between “indigenous” and ladino was cultural in order to be able thereby to explain a change that, according to those authors, did not allow the concept of race, the dominant cultural essentialism ended up anchoring the meaning of “indigenous” around certain characteristics (language, clothing, place of residence) which, if changed, made it impossible to belong to that group. Fortunately the social sciences eventually gave up focusing on the problematic of social classifications as such (namely, describing their nature and the shifting from one classification term to another) in order to concentrate on the contexts within which such people were classified in their own societies. Nevertheless, and despite this change, this essentialist representation of indigenous people has left a certain legacy behind it as well as a degree of confusion between the classification terms used (ladino, indigenous, campesino (peasant), natural (native), K’iche’, Maya) and the concepts of ethnic group, race, social class, and the people or culture with which they are associated, making it difficult to distinguish between the contexts in which such terms arise, the contexts in which they are used and the lives of the people who identify or are identified with them. As has already been extensively pointed out, in the end it was events, especially the emergence of the indigenous movement, later to become known as the Maya movement, which scotched the “ladinisation” theory. As a result of indigenous mobilisation in the 1970s, the debate around who was indigenous and the possible transformation from indigenous to ladino became bound up with the struggle for new indigenous representation at both local and national and international level. As stated earlier, the nineteenth century political dynamic, as well as structuring Guatemalan identity around the Indian/ladino opposition, had fostered the consolidation of what we might call the stigma attached to being Indian. During almost the whole of the twentieth century, this stigma rendered national development incompatible with the development of an indigenous identity, and a range of different assimilationist policies were put forward, of which the censuses were clearly reflections (or instruments). However, such policies never achieved the success hoped for, not only because of indigenous opposition or because the Guatemalan government was incapable of mobilising the resources required to implement them but also

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

143

because the prevailing racism and exclusion made such assimilation extremely difficult. As a result, in the second half of the twentieth century, when the old indigenous institutions were already in serious decline and all indigenous people seemed to be in danger of disappearing as a result of “ladinisation”, the indigenous movement was able to raise the indigenous question again but in a different way. Their proposals, put forward mainly through indigenous organisations, sought simultaneously to reconcile indigenous involvement in state institutions with recognition of that sector of the population. This difficult balancing act between two political traditions, one assimilationist and the other segregationist, which were seen–in the words of the team from the Centro de Investigaciones Regionales de Mesoamérica–CIRMA (Centre for Regional Research of Mesoamerica)–as being the only possible ways for the indigenous people of Guatemala to relate to the state, expected to find a possible ally in the multiculturalism and interculturality paradigms (Taracena, 2002). However, such ideologies were later criticised as being politically correct forms of the traditional assimilationism (García-Ruiz, 2006: 87) or a “neoliberal” multiculturalism which, though promising equality, only conferred selective cultural recognition that failed to threaten the privileges of the ladino racial hierarchy (Hale, 2007: 41).

1.4 Indigenous recognition of multiculturalism The 1990s in Guatemala, which saw the establishment of the Peace Accords between the government and the guerrillas as well as a degree of indigenous visibility at international level, were conducive to a rethinking of the indigenous question at national level that moved away from social integration theories towards those of cultural rights. The period surrounding the negotiation and signing of the Peace Accords was historic in that it entailed widespread discussion of the situation of the country’s indigenous peoples as well as the type of policies needed to improve it. So much so that the said Accords, together with international law, have so far been the main reference point for matters related to indigenous legislation and institutionalisation in the country (Ley Marco de los Acuerdos de Paz, Decreto 52-2005), although this has not meant any substantial change in the often marginal situation of that population. In fact, the victory for the “No” camp in 1999 referendum on constitutional reform (among the amendments proposed was that Guatemala should be recognised as being a pluricultural, multi-ethnic and multilingual nation) and the disintegration shortly afterwards of the Coordinadora del Pueblo Maya de Guatemala–

144

Chapter Five

COPMAGUA (Coordinating Committee of the Maya People of Guatemala)14, immediately showed the limits of this recognition. Nevertheless, despite those limits, that period also opened up new possibilities. Within ten years of the signing of the Peace Accords in 1996, several ministerial, government and legislative agreements concerning indigenous peoples had been signed, the main consequence of which was the establishment of a significant though limited framework of indigenous institutions at national level. Acuerdo Gubernativo (Government Agreement) 390-2002, for example, set up the Defensoría de la mujer indígena (Office for the Defence of Indigenous Women) and the Comisión presidencial contra la discriminación y el racismo (Presidential Commission to Combat Discrimination and Racism), and Acuerdo Gubernativo 526-2003 established the technical Vice-ministerio de educación bilingüe intercultural (Vice-Ministry of Intercultural Bilingual Education). The country’s secondary legislation, established as a result of the Peace Accords, also reflected this trend. The 2002 Código Municipal (Municipal Code), for example, contained some changes compared to the 1988 version, including recognition (with right to legal personality) of the “communities of indigenous peoples” (art. 20) and of “alcaldías indígenas” (“indigenous mayoralties”) (art. 55). Thus, although not defined at constitutional level, the meaning of “indigenous”–influenced by the Maya movement and certain international bodies such as the United Nations and the ILO (Convention 169)–is to a certain extent discernible at the level of secondary Guatemalan legislation and also, of course, in the 2002 census. The third and last significant transformation in official statistics has to do with multiculturalism and the sensitivity to indigenous issues that developed at the end of the twentieth century. These changes began to be seen in the 1994 census, when self-adscription was adopted as a classification criterion, culminating (to date) in the 2002 census, in which a distinction was made between ethnic group, on the one hand, and ethnic belonging or people, on the other. This last change not only means recognition of the Mayas and Xinkas as peoples, as well as the (re)appearance of the old “black race”, now known as the Garifuna people, but also the ethnicisation of the term ladino and the inclusion of ladinos as yet another Guatemalan people. Over and above the political implications inherent in the concept of people, we can see that this signifies an important change in that, first of all, it breaks away from the binary classification (indigenous/ladino-non-indigenous) that had been a feature of most previous Guatemalan censuses (with the exception of that of 1921) and, secondly, for the first time the ladino category can be thought of as

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

145

indigenous. Unlike the 1921 census, which had endeavoured to distinguish ladino heterogeneity (whites, blacks, ladinos and yellows) by trying to make the matter more complex through the adoption of racial categories that were alien to local classification, the 2002 census does so by focusing on indigenous heterogeneity. This also means (for the time being) that ethnic categories take official precedence over racial ones. These data, and in particular the appearance of indígena ladino as a category (as a result of the convergence of “ethnic group” and “ethnic belonging”), might make us think that there has been a certain degree of “indigenisation” in the country similar to that which has occurred in other countries of the region, where the language of indigeneity has become key either as a new language of national identification or for challenging global capitalism (Canessa, 2007). However, and unlike other countries such as Bolivia, the figures for the overall indigenous population continue to be very similar to those found in the second decade of the twentieth century. The next section contains some initial reflections in this regard.

2. Censuses, mestizaje and indigeneity We said at the beginning of this chapter that one of the most significant pieces of data revealed by the 2002 census was the appearance of over 100,000 indígenas ladinos. Unlike the “ladinised” indígenas theorised by Adams (1956: 24-5), who emerged under the social integration paradigm, the indígenas ladinos made their appearance under the multiculturalism paradigm. These are therefore no longer indígenas in the final stages of becoming acculturated but individuals who, from among the options presented to them, self-designate as being both indígena and ladino. What does this dual categorisation mean? Although the 2002 census did not provide for a mestizo category, the appearance of the indígena ladino could be demonstrating that what was hitherto an unprecedented category in Guatemalan censuses does in fact exist. Or should we instead be looking more closely at the arguments postulated by indigeneity theories? Hale (2007: 236) asserts in his book Más que un indio (More than an Indian) that one of the ladino responses to what he calls the “Maya efflorescence” is that some of them are recognising that they are mestizos. He distinguishes three categories within this group: the “mestizo universalists”, influenced by the 1944-1954 period and the ideal of a universal citizenship that promotes an equal (racially blind) society in which there are just Guatemalans; the “new mestizos”, who usually inhabit the poorest urban neighbourhoods and assume mestizo identity as a way of silently rejecting the indigenous/ladino binary; and the “militant mestizos”,

146

Chapter Five

who identify as such in solidarity with indigenous people, acknowledging their indigenous ancestry and renouncing the racism associated with ladino ancestry. Nevertheless, Hale also states that many still identify as ladinos, be it (the least common case) in the traditional racist way, or in ways that have been reshaped to a greater or lesser extent by the influence of a neoliberal multiculturalism that recognises cultural rights but does nothing to change racial privilege, or, lastly, because they are what he calls “ladino dissidents”. The latter identify as ladino but retain a permanently critical attitude towards that category, seeking to transform it from within. Quite a few of the indígenas ladinos revealed in the 2002 census would probably identify with one of Hale’s categories but I believe there are still more options. Although in Guatemala, unlike other Latin American countries, ladinisation has replaced mestizaje as a political ideology15, Hale (2007: 261) reveals that, when he carried out a survey of identity in a Chimaltenango school, 45% of the pupils reported as indigenous, 38% as mestizo and only 17% as ladino. The author uses this example to urge analysts of Guatemalan identity policy to finally accept “mestizo” as a category. Including it in the census would have undoubtedly changed the results, adding greater complexity to the diversity we are sketching here, but if we simply see mestizos as “non-indigenous” (Hale, 2007: 235), we are excluding other possibilities, for example, that the mestizos might see themselves as “indigenous”. De la Cadena (2006: 51) criticises the association between “mestizo” and “non-indigenous”, arguing that this is further facilitated by the conceptual hybridity of this category. Thus, the indígenas ladinos found in the 2002 census could be individuals who, though they see themselves as indigenous, do not see themselves as Mayas, Xinkas or Garifunas. Unlike the indigenism that emphasised mestizo heritage, the indigeneity theory places “indigenous” as a category at the heart of national ideologies, though it also has a significant profile at the global level (Canessa, 2006). This interpretation of the meaning of “indigenous” allows such identity to be displayed with pride without the need to keep up indigenous practices. In other words, on the one hand it allows individuals who (with all their differences) might also fit into Hale’s non-indigenous categories to be identified while, on the other hand, embracing those who continue to identify as indigenous despite having adopted many practices that are not. Indigeneity, unlike indigenism, allows them to change without ceasing to be indigenous. While the appearance of indígenas ladinos in the 2002 census could also be interpreted from that point of view, it does not explain why the size of the indigenous population in the 2002 census remained the same or

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

147

even slightly decreased. In our interpretations, we cannot rule out the historic tradition of undercounting Indians or the probable on-going tendency to choose to abandon an indigenous identification that still has pejorative associations; however, Hale (2007: 100-2) believes that, for the department of Chimaltenango, the fact that the ethnic totals have remained the same conceals a trend towards the bifurcation of racial demography. In other words, it is only in municipios where the indigenous population is by far in the majority (over 85%) that the trend towards the Mayanisation of all the inhabitants can be observed whereas that trend is virtually nonexistent in municipios where ladinos have been in the majority for some time. If this bifurcation were to take place in other departments of the country, we could say that some regions of the country are becoming indigenised while others are taking different paths as far as identification is concerned. A quick glance at Table 5-2 seems to suggest that the indígena ladino phenomenon is prevalent in places where each of those two identities is present in significant numbers. By contrast, in Totonicapán (one of the departments with the largest proportion of indigenous inhabitants in the country) the number of those reporting as indigenous (333,481) is similar to the number reporting as Maya (333,438) while, in the department of El Progreso (which has a very small indigenous population), the number of those reporting as non-indigenous (138,240) more or less tallies with the number reporting as ladino (138,640). Another problem that arises in applying indigeneity theses is that, being so influenced by indigenous movements at international level, they do not shed light on how people who do not identify with those movements perceive their own identity. In this regard, Canessa (2006: 253-4) warns Bolivia of the paradox that could arise if indigenous people from rural communities end up being deemed “inauthentic” by comparison with the “hyperreal” Indian theorised by Ramos (1994). The contradiction would be that those who speak a Mayan language, live in a rural community and wear indigenous clothing, namely those who for the anthropologists of the 1950s were clearly indígenas, may not identify as Mayas under the multicultural paradigm. The fact that in Guatemala the number of those reporting as “indigenous” is greater than the total number of Mayas, Xinkas and Garifunas not only indicates that some ladinos may be identifying as indigenous but also that some indigenous people may be denying that they belong to those peoples. In Guatemala, for example, evangelical Indians might not identify with the ways in which the Maya movement defines “Maya”.

148

Chapter Five

In 1998, when starting my field work in the city of Quetzaltenango (in the Guatemalan Altiplano), I had the following conversation with a middle-aged woman who wore indigenous dress every day: Doña X: I’ve been married for 25 years and tried to lead an honest life. And we’ve done that as a couple. We’ve been able to understand our differences even though I’m an evangelical and he is Maya. Myself: But aren’t you Maya? Doña X: No, I’m not Maya, I’m an evangelical. Myself: But you wear the typical dress? Doña X: Yes, because I’m indígena.

Given the importance for the Maya movement of their so-called world view (cosmovision) in general and their ceremonies in particular (which many evangelicals still associate with witchcraft), some of them probably reject an identification that they associate mainly with a religious choice. Even more interesting was the remark made to me a few days later by this woman’s daughter when I sought to pursue the issue. Look, I can’t say that I am not. We are indígenas because we wear our costume, we have our customs but our spirituality is not really Maya. That is what they don’t understand […] I think we are no longer pure Mayas, we aren’t pure Mayas anymore, because, like it or not, there has been a bit of blood crossing. Whether we like it or not, we have ladino blood […] If the Maya religion was pure, we would probably have been convinced by now but it’s a mixture with the Catholic religion […] Not long ago my dad brought a video and we were watching a Maya ceremony in which they were praying a lot. And so I think that that is a mixture because praying is Catholic.

Curiously the different reasons this woman and her daughter gave me for not identifying as “Maya” were couched in a language centred on purity and mixture which, in my view, resembles De la Cadena’s view of the mestizo (2006: 69). For her, the mestizo is above all a person who rejects the purification of identities. Like the indígenas mestizos of De la Cadena (2006: 77), the indígenas ladinos found in the 2002 census could above all be those individuals who have chosen not to be converted to either of the two projects, and whose ways of being indigenous repudiate purity and authenticity. Lastly, only ethnography can help us to delve more deeply into the variety of meanings behind each form of identification. However, the data revealed in the 2002 census show how, despite the fact that the total number of indigenous people in the country has remained the same or even

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

149

slightly decreased, Guatemala has not been entirely immune to the changes that have taken place at international level with regard to the representation of what it means to be indigenous. In my view, one of the most important is that it is now possible for ladino to also mean “indigenous”. This amounts to a break with the past the consequences of which remain to be seen.

Bibliographical references Adams, Richard N. 1956. Encuesta sobre la cultura de los ladinos en Guatemala. Guatemala: Seminario de Integración Social Guatemalteca. Alda Mejías, Sonia. 1999. Indígenas y política en Guatemala en el siglo XIX: conflicto y participación en la administración local. Madrid: Universidad Autónoma de Madrid [PhD Thesis in Historia]. Canessa, Andrew. 2006. Todos somos indígenas: Towards a New Language of National Political Identity. Bulletin of Latin American Research, 25(2): 241-263. —. 2007. Who is Indigenous? Self-Identification, Indigeneity, and Claims to Justice in Contemporary Bolivia. Urban Anthropology, 36(3): 195237. De la Cadena, Marisol. 2006. ¿Son los mestizos híbridos? Las políticas conceptuales de las identidades andinas. Universitas humanística, 61: 51-84. Early, John D. 2000 [1982]. La estructura y evolución demográfica de un sistema campesino: la población de Guatemala. Miami: Cirma/Plumsock Mesoamerican Studies. Favre, Henri. 1996. L'indigénisme. Paris: Presses Universitaires de France. García Ruiz, Jesús. 2006. Lealtades en disidencia. La construcción de la identidad política en los movimientos mayas de Guatemala. In García Ruiz, Jesús (ed.) Identidades fluidas. Identificaciones móviles. Guatemala: ICAPI. Goubaud Carrera, Antonio. 1964. Indigenismo en Guatemala. Guatemala: Ministerio de Educación. Grandin, Greg. 1999. The Blood of Guatemalans: The Making of Race and Nation, 1750-1954. New Haven: Yale University [PhD Thesis]. Hale, Charles R. 2007. “Más que un indio”. Ambivalencia racial y multiculturalismo neoliberal en Guatemala. Guatemala: Avancso. [English version: 2006. Más que un Indio (More than an Indian): Racial Ambivalence and the Paradox of Neoliberal Multiculturalism in Guatemala. Santa Fe: School of American Research Press].

150

Chapter Five

Nash, Manning. 1958. Machine Age Maya: The Industrialization of a Guatemalan Community. Glencoe: The Free Press. Nogueira, Oracy. 1954. Preconceito racial de marca e preconceito racial de origem: sugestão de um quadro de referência para interpretação do material sôbre relações raciais no Brasil. Anales del XXXI Congreso Internacional de Americanistas, São Paulo: 409-427. Ramos, Alcida. 1994. The Hyperreal Indian. Critique of Anthropology 14(2): 153-171. Rodas, Isabel. 1999. Identidad, asentamiento y relaciones de parentesco de los españoles de Patzicía (siglos XVI-XVIII). In Jean Piel and Todd Little-Siebold (eds) Entre comunidad y nación: la historia de Guatemala revisitada desde lo local y lo regional. Guatemala: CIRMA. Scheper-Hughes, Nancy. 2000. Demografía sin números. El contexto económico y cultural de la mortalidad infantil en Brasil. In Viola, Andreu (ed.) Antropología del desarrollo. Teorías y estudios etnográficos en América Latina. Barcelona: Paidós. Stoll, Otto. 1958 [1884]. Etnografía de Guatemala. Guatemala: Seminario de Integración Social. Taracena, Arturo. 1997. Invención criolla, sueño ladino, pesadilla indígena: los Altos de Guatemala: de región a estado, 1740-1850. San José-Costa Rica: Porvenir. —. 2002. Etnicidad, estado y nación en Guatemala 1808-1944. Guatemala: CIRMA. Tax, Sol. 1937. The Municipios of the Midwestern Highlands of Guatemala. American Anthropologist, 39(3): 423-444. —. 1941. World View and Social Relations in Guatemala. American Anthropologist, 43(1): 27-42. —. 1953. Penny Capitalism: A Guatemalan Indian Community. Washington: Smithsonian Institution - Institute of Social Anthropology. —. 1959. La visión del mundo y las relaciones sociales en Guatemala. In Richard Adams et al. (eds) Cultura indígena de Guatemala: ensayos de antropología social. Guatemala: Seminario de Integración Social Guatemalteca. Tax, Sol (ed.). 1952. The Heritage of Conquest. The Ethnology of Middle America. Glencoe: Free Press. Tzian, Leopoldo. 1994. Mayas y ladinos en cifras: el caso de Guatemala. Guatemala: Cholsamaj.

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

151

Censuses consulted DGE–Dirección general de estadística. 1921. IV Censo de población. Guatemala: DGE. Available at: http://ccp.ucr.ac.cr/bvp/censos/guatemala/1921/index.htm [consulted 25 January 2012]. DGE–Dirección general de estadística. 1950. VI Censo de población. Guatemala: DGE. Available at: http://ccp.ucr.ac.cr/bvp/censos/guatemala/1950/ [consulted 25 January 2012]. DGE–Dirección general de estadística. 1964. VII Censo general de población. Guatemala: DGE. Available at: http://ccp.ucr.ac.cr/bvp/censos/guatemala/1964/ [consulted 25 January 2012]. DGE–Dirección general de estadística. 1966. Censo de población 1964. Resultados de tabulación por muestreo. Guatemala: DGE. Available at: http://ccp.ucr.ac.cr/bvp/censos/guatemala/1964/64c00gt.pdf [consulted 25 January 2012]. DGE–Dirección general de estadística. 1973. Manual de instrucciones para el empadronador. VIII Censo general de población. Guatemala: DGE. Available at: http://users.pop.umn.edu/~wlt/col/gt1973ei_guatemala_ interviewer_manual.es.pdf [consulted 25 January 2012]. INE–Instituto nacional de estadística. 1981. IX Censo nacional de población. Guatemala: INE. Available at: http://ccp.ucr.ac.cr/bvp/censos/guatemala/1981-t1/ and http://ccp.ucr.ac.cr/bvp/censos/guatemala/1981-t2/ [consulted 25 January 2012]. INE–Instituto nacional de estadística. 1994. X Censo nacional de población y V de habitación. Guatemala: INE. INE–Instituto nacional de estadística. 2002. XI Censo nacional de población y VI de habitación. Guatemala: INE.

Legislation consulted AIDPI–Acuerdo de identidad y derechos de los Pueblos Indígenas. Peace accord of 31 March 1995. Available at: http://www.congreso.gob.gt/Docs/PAZ/Acuerdo%20sobre%20identida

152

Chapter Five

d%20y%20derechos%20de%20los%20pueblos%20indígenas.pdf. [consulted 9 April 2006]. Acuerdo sobre aspectos socioeconómicos y situación agraria. Peace accord of 6 May 1996. Available at: http://www.congreso.gob.gt/Docs/PAZ/ACUERDO%20SOBRE%20A SPECTOS%20SOCIOECONÓMICOS%20Y%20SITUACIÓN%20A GRARIA.pdf [consulted 9 April 2006]. Acuerdo Gubernativo 390-2002 del 8 de octubre del 2002. Crea la Comisión presidencial contra la discriminación y el racismo contra los Pueblos Indígenas en Guatemala. Diario de Centro América, Tomo CCLXX(18): 4. Available at: http://www.congreso.gob.gt/archivos/acuerdos/2002/gtagx3902002.pdf [consulted 12 April 2006]. Acuerdo Gubernativo 526-2003 del 12 de septiembre de 2003. Crea un tercer viceministerio en el Ministerio de Educación como Viceministerio de educación bilingüe e intercultural, encargado de los temas de la lengua, la cultura y multietnicidad del país. Diario de Centro América, Tomo CCLXXII (53): 1-2. Available at: http://www.congreso.gob.gt/archivos/acuerdos/2003/gtagx5262003.pdf [consulted 12 April 2006]. Código Municipal de 1988. Decreto Nº58-88. Guatemala: Jiménez & Ayala. Código Municipal de 2002. Decreto Nº12-2002. Guatemala: Secretaría de Coordinación Ejecutiva de la Presidencia. Constitución Política de la República de Guatemala de 1985. Guatemala: Secretaría de Coordinación Ejecutiva de la Presidencia. Ley Marco de los Acuerdos de Paz. Decreto 52-2005 del 3 de agosto del 2005. Diario de Centro América, Tomo CCLXXVII (56): 1-3. Available at: http://www.congreso.gob.gt/archivos/decretos/2005/gtdcx52-2005.pdf [consulted 12 April 2006].

Guatemalan Census of 2002: “Indigenous” and “Ladino” Categories

153

Notes 1 Even if we add to it half (26,904) of all those who reported as “other” under ethnic belonging (53,808), the result (4,460,122) is significantly lower than the “indigenous” total (4,610,440). Indeed, we cannot know whether the individuals who reported as “other” under ethnic belonging (53,808) also considered themselves, in the ethnic group section, to be “indigenous” or instead put themselves down as “non-indigenous”. Some may have considered themselves to be “non-indigenous” but did not self-identify as ladino. Others, on the other hand, may have considered themselves to be “indigenous” but did not self-identify as “Maya”, “Xinka”, “Garifuna” or ladino. 2 The sixth population census (1950), which was conducted during Guatemala’s ten-year period of democracy, accused the government of General Ubico of altering the figures obtained from the 1940 census. In the 1950 census, it is claimed that orders were found from the Presidency and the Finance Ministry of the time stipulating in advance what the population of each department should be. These alterations increased the total population of Guatemala by at least 900,000 (DGE, 1950). 3 As happened, for example, in Quetzaltenango (see Grandin, 1999: 72-98) 4 “The General Directorate of Statistics has tried to retain solely the concept of ethnic group, regardless of any racial considerations. Ever since the first population census the designations of indigenous and ladino (the latter equating to non-indigenous) have appeared within the concept of race. The appearance of such designations has not yet been properly studied, although it has been pointed out that in some dictionaries from the last century the word ladino, in addition to the meaning it currently has, meant someone who spoke two languages (Spanish and an indigenous language). However, the term has been traditionally used in our field by anthropologists to designate the non-indigenous group, though not considering them to be a racial unit but rather a set of cultural characteristics” (DGE, 1950: XI). 5 “For the reason previously established, efforts were made, wherever possible, to use the services of census takers from the place itself who were fully conversant with the local way of categorizing people” (DGE, 1950: XII). 6 “The difficulty of formulating a precise definition of the indigenous population having been recognised by specialist bodies, the criterion used in the 1950 census was adopted and the social esteem in which the person was held in the place where the census was being taken was taken as the basis” (INE, 1981: 26). 7 On the 1973 census form, for example, the interviewer already had to mark yes or no in response to the written question “Are you indigenous?” even though that interviewee was not yet being directly asked that question. 8 According to Sol Tax, the municipio was the primary social and cultural unit of the western Altiplano (Tax, 1937). 9 “Within the sphere of western Guatemalan culture there are, not on the basis of a homogeneous indigenous culture but on one of many cultures belonging to the many different Guatemalan indigenous groups, what Stoll would call a “cultural mosaic”. This mosaic is made up of 200 indigenous ethnic entities that are to a

154

Chapter Five

greater or lesser extent distinguishable from one another, but who because of that differentiation must be considered as distinct ethnic groups. Each of these indigenous ethnic groups constitutes a social community, a set of customs, attitudes, feelings and ways of thinking that are shared by the group of people occupying a geographical area, a territory, which in Guatemala is the municipio” (Gobaud Carrera, 1964: 23-4) 10 “We have a little trouble in defining the Indian and the non-Indian... The situation seems pretty complicated in that area, but I suggest that for convenience at the moment, we assume there is a major distinction–however it is going to be defined–between Indians and non-Indians” (Tax, 1952: 94-6). 11 For example, Tax’s Penny Capitalism (1953), an evocative title for describing the indigenous economy of Panajachel in Lake Atitlán, or Machine Age Maya: The industrialisation of a Guatemalan community, in which Nash (1958) devotes many pages to showing what a good thing it was that the spinning mill opened in Cantel in 1876 had been incorporated into indigenous culture. 12 “We may say that in respect to their world view the Guatemalan Indians are of the primitive type, while in respect to their kind of social relations, they are of the civilised type” (Tax, 1941: 37). 13 For example, the Instituto Nacional Indigenista (National Indigenist Institute) founded in 1945. 14 Saqb’ichil-COPMAGUA, which was responsible for representing the Maya sector in the peace process negotiations, emerged in 1994. Although COPMAGUA was wholly funded through international aid, its establishment encouraged the bringing together of all strands within the Maya movement and it was seen as very representative, especially because of the signing in 1995 of the Acuerdo de identidad y derechos de los Pueblos Indígenas (Agreement on Identity and Rights of Indigenous Peoples). Nevertheless, the limits of the process itself, such as the failure of the 1999 referendum on Constitutional reform as well as a number of internal disputes, led to its dissolution in 2000. 15 There are clearly people who are mestizos, the offspring of a ladino and an indígena. In Quetzaltenango, these types of individuals were recognised but they identified mainly as either ladino or indigenous for a variety of reasons that it would be interesting to analyse on another occasion.

CHAPTER SIX THE SOCIAL AND POLITICAL CONSTRUCTION OF RACIAL AND ETHNIC CATEGORIES IN NATIONAL CENSUSES OF PANAMA, 1911-2010 MÒNICA MARTÍNEZ MAURI In Panama, one century separates 19111, the year in which the young Republic carried out its first population census, from 2010, the date of the most recent census2. During this period ten censuses have been carried out. All of them have categorised the national population according to racial and/or ethnic characteristics, making possible a historical analysis of how such differences are conceptualised and integrated into national statistics. The account offered here starts from the premise that socio-cultural classification categories of this sort are not the product of moral and cultural differences per se, but are rooted instead in political and ideological principles (Stolcke, 2008: 20). Censuses help to constitute racial discourse (Nobles, 2002: 43) and, as Panama’s experience also shows, their study reveal the relationship between racial ideas, censustaking and public policy. Panama’s official classifications thus cannot be separated from the history of the state’s relations with racialised and ethnicised groups, specifically with the indigenous peoples and the descendants of African slaves who have lived within its present borders for centuries. In the indigenous case, these relations are characterised by contradictory government policies and by failure on the part of indigenous peoples to organise at a national level. In reflecting on the political and social nature of these ethnic classifications, and in the process considering how the Panamanian state has managed internal cultural diversity, I analyse three issues related to the construction of socio-political categories. Taking as my starting point the census results and other publications produced by the Comptroller General of the Republic over the last century (1911-2010)3, I analyse changes in: a) the distribution of cultural difference and ethnic boundaries, b) criteria for

156

Chapter Six

identifying indigenous and Afro-Panamanian populations, and c) census methodologies. The first issue, the distribution of ethnic groups, leads to consideration of how Panamanian territory has been conceptualised. The relationship between metropolitan hubs and outlying rural areas, as well as national integration projects, can be studied by looking at the construction of social categories that have been employed in the censuses to classify the inhabitants of these areas. This focus will enable us to ascertain if the indigenous populations have always been the target of identification in urban areas, how many indigenous groups are recognised in the censuses, and what establishes the boundaries between them. The second topic–how indigenous and Afro-Panamanian populations are defined and identified–highlights the permeability and fluidity of these categories. Focus on changes in these terms reveals how the country’s cultural diversity has been conceptualised, as well as how the regulatory frameworks imposed by international organisations have influenced the criteria by which groups are identified. The third issue examined here is the methods, mechanisms and resources inherent in Panamanian census-taking, notably the role played in the production of census data by census-takers and by the Indians and Afro-Panamanians being censused. Changes in the stances assumed towards the census, both by the state and by indigenous organisations, and the consequences of such stances, are evident in Panamanian census methodology, once again pointing towards the contingent, political nature of ethnic identification. All these three topics will help us to show that censuses do not simply reflect social reality; they play a key role in the construction of that reality (Kertzer and Arel, 2002: 2). Ethnic or racial categories in Panamanian censuses do not capture demographic realities, but rather help to create political realities and ways of thinking the difference.

1. The Panamanian state and the administration of cultural diversity Panama’s national census must be understood in the context of the policies and ideologies that have dominated the management of cultural diversity over the last one hundred years. Panama’s geostrategic position and its interoceanic railroad and canal have attracted waves of immigration, creating an ethnically complex national population. In addition to the indigenous survivors of conquest and colonisation and the Hispanic-indigenous mestizos often called interioranos4, the contemporary

Social and Political Construction of Racial and Ethnic Categories (Panama) 157

population includes the descendants of several immigrant populations: Spanish-speaking African slaves; Chinese labourers on the transisthmian railway; French- and English-speaking Afro-Antilleans brought to work on canal construction; and, most recently, Gujaratis and other East Indians and Sephardic Jews and Europeans attracted by commercial expansion on the isthmus. Ever since independence from Colombia in 1903 Panama has stood out in the Latin American panorama for its recognition of indigenous territories and governments through the establishment of comarcas, or semi-autonomous indigenous territorial reserves. This positive indigenous policy has been compromised, however, by two contradictory practices: efforts to “civilise” indigenous peoples through schools, missions, and participation in electoral politics; and concessions of partial sovereignty to some indigenous peoples without providing the financial means or state support needed to consolidate their position. One of the first such entities, the Comarca de San Blas (today Kuna Yala) which was created in 1938 and shaped by negotiations between 1925 and 1953, established a (not always fruitful) model for later indigenous policies. The Kuna case was in key respects non-replicable. Its unique features include a history of negotiations and alliances in the colonial period; an accord from 1871 with the Colombian government; intervention by North Americans during and after the so-called Dule Revolution of 1925; location on a national border; formation of a talented cohort of native scribes and lawyers; and the evolution of the Congreso General Kuna, a governing body capable of reconciling internal factions and forging united policy and action. Nonetheless, beginning in the 1970s, attempts were made to replicate Kuna successes by government agencies and other indigenous peoples. These initiatives, reinforced by indigenous political mobilisation, led successively to the formation of the General Congress of the Emberá-Wounaan Comarca in 19835, the General Congress of the Kuna de Madungandi Comarca in 19966, the General Congress of the Ngäbe-Buglé Comarca in 1997, and the General Congress of the Kuna de Wargandi Comarca in 2000. These new territories and governing bodies formed part of a larger reorganisation of traditional regional and local political authority structures. The very recent nature of this process underlines Kuna uniqueness as the only indigenous group that actively negotiated with the state during the first half of the twentieth century. This political engagement was made possible by formation of a group of young literate men, who acted as mediators between traditional authorities and the government (Martínez Mauri, 2007; Howe, 2009). Professionals of this sort have not appeared

158

Chapter Six

among other indigenous groups until very recently, and the largest group, the Ngäbe, actively avoided political contact and engagement for fear of subjugation. Such avoidance has been evident even in census-collection: in 1950, while Kuna leaders cooperated with census-takers, in Chiriquí Province, Ngäbe inhabitants refused to be registered (DEC 1954, vol. IV: xiii). Despite the differences among indigenous systems of governance in Panama, the comarcas are generally considered an effective guarantor of cultural diversity. Their creation has furthered a reconceptualisation of indigenous-state relations, promoting the election of native authorities and local control over natural resources. Some of them, notably the Comarca de Kuna Yala, have even secured further rights, including partial delimitation of their territory and de facto recognition of internal legislation. The Kuna General Congress, responding to growing dissatisfaction with the 1945 Constitution or Carta Organica of the comarca and with the governing legislation (Law 16 of 1953), drafted and passed a series of legislative initiatives: General Congress Regulations (1993); the “Fundamental Law” (1995); Comarca Bylaws (2000), and Tourism Regulations (2007). Although these bodies of legislation have never been formally ratified by the national government, they are internally enforced within the comarca and tacitly accepted by government functionaries, thus, as Turner (2008) notes, significantly advancing indigenous self-determination in Panama. The real but partial successes of the comarca system offer few protections to the 52% of the indigenous population who, according to the latest national census, live outside comarca boundaries on lands lacking effective legal protection. Their situation has grown worse in recent years, as recent national governments have undercut rather than strengthened indigenous rights. Unlike a number of neighbouring countries, which have passed legislation respecting and promoting cultural diversity, Panama has distanced itself from international measures in this area and failed to protect its indigenous citizens (Martínez Mauri, 2011). The only country in the region without formal constitutional recognition of multiculturalism, Panama is also one of the few countries in Latin America that has ratified Convention 107 of the International Labour Organization (ILO) but not ILO 1697. Panama’s indigenous policies thus embody a contradiction: creation of effective indigenous territories vitiated by non-recognition of multiculturalism and rejection of international protective measures, exacerbated by repeated violations of indigenous rights, both inside and outside the comarcas.

Social and Political Construction of Racial and Ethnic Categories (Panama) 159

Another salient characteristic of indigenous-state relations in Panama is bilateral particularism. Unlike their counterparts in other countries of the region, such as Ecuador and Colombia, Panamanian indigenous peoples have never had a pan-tribal spokesman or a successful, enduring national organisation to represent them in negotiations with the government. Each group has negotiated its rights directly with the central government, with moral support from other peoples but without national coherence or national-level goals. The state’s relations with Afro-Panamanians are strikingly different from those described above for native Indian groups. While the AfroPanamanian or “Afro-descendent”8 population of Panama has long been recognised as one of the primary elements in the “racial melting pot”, it has never enjoyed territorial rights or positive official policies. Panamanians have traditionally made a sharp distinction between Colonial blacks and Antilleans, that is, between the Spanish-speaking descendants of Colonialera African slaves on the one hand and the mostly French and Englishspeaking Caribbean labourers brought to Panama for railway and canal construction and their descendants (Maloney, 1989). Both groups have been subjugated and exploited, one by centuries of slavery, the other by discrimination in the banana plantations and the Canal Zone, where blacks and whites were separated into so-called Gold and Silver Rolls, as well as in urban Panama. Although both groups have experienced prejudice, special animus focused on the Antilleans, who were stigmatised as intrusive non-Hispanic interlopers, and they were legally disenfranchised for several years by a national Constitution passed in 1941. Beginning in mid-century, Afro-Panamanians began to organise against racism and to demand public policies on their behalf. Some of the most significant developments occurred during the 1980s, including three successive Panamanian Black Congresses (in 1981, 1983 and 1988) and the creation of two organisations, Acción Reivindicadora del Negro Panameño–ARENEP (Advocacy Action for Panamanian Blacks) and the National Conference of Organized Panamanians (NCOP). Although both organisations collapsed after the invasion of Panama in 1989, the Panamanian Committee against Racism was later founded in the 1990s, and the Coordinadora Nacional de Organizaciones Negras Panameñas– CONEGPA (National Coordinator of Panamanian Black Organisations), an organisation that brings together 20 groups of Afro-descendants, was established in 2004 (Priestley and Borrow, 2010). The Panamanian state has engaged with indigenous peoples and AfroPanamanians to markedly different extents, which helps explain why efforts have been made to include and distinguish them in population

160

Chapter Six

censuses over the last century. The differences between the indigenous and black populations are clear, particularly with respect to the negotiation of territorial rights, although both have been thought of as culturally and socially-homogenous entities requiring state control. To illuminate the importance of the censuses for the conceptualisation of racialised and ethnicised groups in the state’s political praxis, I examine three pertinent issues: the distribution of ethnic groups and the boundaries between them; the criteria for assigning people to particular groups; and the methodology employed in the censuses.

2. Ethnic boundaries and the distribution of the indigenous population according to the censuses From the outset, it must be stressed that these censuses determine the boundaries between the country’s ethnic groups. At the end of the twentieth century, the Panamanian state had recognised within its territorial borders eight indigenous groups (Ngäbe, Kuna, Emberá, Buglé, Wounaan, Naso Tjerdi, Bri Bri and Bokota) and four categories of Afrodescendants (“Colonial black”9, “Antillean black”10, “black”11 and “other black”12). The first indigenous groups to be recognised by name, in the population census of 1940, were the Kuna, Guaymí and Chocó. The latter two names, however, actually encompass multiple populations– in the case of “the Guaymí”, the Ngäbe and the Buglé, and in that of “the Chocó”, the Wounan and Emberá. In the 1950 Census, the indigenous survey form added a question for the first time about the language–referred to as “dialect”–normally spoken in the family, without providing a list of suggested responses. In the census results, this open-ended procedure yielded a larger number of alternative “dialects” than the three groups specified in the previous census: Bogotá or Bocatá, Cricamola or Valiente, Chocó, Guaymí, Plains Guaymí, Mountain Guaymí, Savannah Guaymí, Cuna13 and Teribe (DEC 1954, vol. IV: 15). This amplification, however, did not lead to expansion of the number of officially recognised indigenous groups. A decade later, the 1960 Census, which invoked social and geographic rather than linguistic definitions of indigeneity, continued to treat the Republic’s native population in terms of three basic cultures: the Kuna, found primarily in the Comarca of San Blas, with smaller populations in Alto Bayano region, the district of Chepo, and the province of Panama; the Guaymí, concentrated in the highland regions of Chiriquí, Bocas del Toro and Veraguas Provinces; and the Chocó, who

Social and Political Construction of Racial and Ethnic Categories (Panama) 161 for the most part live along the rivers of the Province of Darién. (DEC 1961: 136).

Two decades later, the 1980 Census form once again included linguistic criteria to differentiate the country’s indigenous groups. To answer the census question about natal language, respondents could choose between Kuna, Ngobere, Murire, Naunana, Emberá, Teribe, Buglere, Spanish and English. This recognition of linguistic multiplicity in the censuses of 1960 and 1980 did not, however, expand the number of peoples recognised until 1999, when the Teribe–today called Naso–were first included, the “Bokota” were separated from the Guaymí, and the Chocó category was broken down into Emberá and Wounan. Finally, in 2000 the Census recognised the presence in Panama of Bri Bri (who had previously been identified exclusively with Costa Rica) and discarded the Guaymí category altogether, substituting the terms Ngäbe and Buglé. Today, as a result, the Panama Directorate of Statistics and Censuses recognises eight indigenous populations in Panama, corresponding to eight different socio-linguistic categories. The reliability of the terms–the degree to which they are confirmed by anthropological research–is not uniform, however: no one questions Kuna, Ngäbe, Buglé, Naso (Teribe), Emberá, Wounan and Bri Bri, but the evidence confirming the separate identity of the Bokota is scant at best. The Bokota case, in fact, illuminates the fallibility and contingency of ethnic categorisation. Generally known as sabaneros, or savannah dwellers, the Bukuete or Bukueta and Murire were first identified as “Guaymí sub-tribes” in Pinart’s pioneering studies from the late nineteenth century (1882, 1887, 1892 and 1897), a terminology later used by Johnson (1948a and 1948b). In 1927, the Swedish anthropologist Erland Nordenskiöld tried to visit a community of about two hundred near the Calovébora River, at the border between Bocas del Toro and Veraguas Provinces. Although stymied by unhelpful indigenous guides, Nordenskiöld claimed a separate identity for the group he called Bogotá on the basis of linguistic differences (Nordenskiöld, 1928). The term Bogotá, first introduced by Nordenskiöld, was accepted shortly afterwards by the Panamanian geographer Manuel María Alba to designate some indigenous populations dwelling along the Calóvebora River (1928 and 1929). In the 1960s, however, Nordenskiöld’s successor, Henry Wassén, cast doubt on the uniqueness of Bogotá. Wassén, with the help of the linguist Nils Holmer, argued persuasively that the differences between Bogotá and Murire pronouns and numerals were minimal. These slight linguistic variations thus defined the Bogotá as a Guaymí subgroup. Wassén concluded that the Guaymí could be seen as divided into two

162

Chapter Six

groups: eastern Guaymíes (Murire, Bukueta and Savannah) and western Guaymíes (Valiente, Mové, Northern and Penonomeño) (Wassén, 1963: 91). In 1964, Francisco Herrera and Raúl González, two students of the Panamanian anthropologist Reina Torres de Araúz, visited the region between the Chutara and Calovébora rivers to investigate Nordenskiöld’s claims for a distinct Bogotá people. This effort paralleled studies from the previous year by Torres de Araúz of the Teribe (later termed Naso) on the Teribe River in Bocas del Toro. Confirmation of the presence of Bogota in the region led to a subsequent trip a year later to the Las Palmas district in Veraguas province by Francisco Herrera, who visited the so-called Savannah Guaymíes, or Buglé. Based on these short reports, Torres de Araúz revised the nomenclature of Panama’s indigenous groups14. In a later text on the Guaymí, Richard Cooke (1982) distinguished between the Bocotá and Sabaneros and mentioned the cultural divisions among the Guaymíes. At this time, the term Ngawberé was used for the languages spoken by western Guaymíes, while Buglére designated the language used by the Bocotás and Sabaneros. Although Buglére-speakers also used the term as an ethnic self-designation, academics like Cooke argued that since the suffix “ére” means language, the name Buglére should not designate a human group: thus Cooke and others continued to reserve the names Bocotá and Sabanero for Buglére-speaking Guaymí. The confusion and disagreement over these terms has continued. In further developments, during the indigenous political ferment of the 1990s, some Bokota, Bokueta and Sabanero insisted that they were not Ngäbe but Buglé, a claim that has since gained ground. Unity has not prevailed, however, since some groups rejected the latter term, organising a committee dedicated to preserving the name Bokota. Resolution of this tangle of nomenclatural issues awaits further indepth study, as Cooke (1982), Young (1970), Wassén (1963) and Torres de Araúz (1980) have all concluded. The few anthropological studies available suggest that the Bokotá, Muri, Murire and Buglé speak a single language with dialectal and phonetic variants, but to date the ethnographic evidence is so scarce that the internal boundaries between these groups remain unclear. To make matters even more confusing, many Buglére (or Murire) speakers live intermixed–and even married–with Ngäbe. Regardless of how these questions are ultimately resolved, official statistics identify eight primary indigenous groups in Panama:

Social and Political Construction of Racial and Ethnic Categories (Panama) 163 Fig. 6-1. Indigenous peoples of Panama (%) 2010

Ngäbe Kuna Buglé Emberá Wounaan Naso Tjerdi Bri Bri Bokota

Source: By author based on data from the 2010 DEC Population Census.

Table 6-1. Number of indigenous and Afro-descendants in the Republic of Panama Population or ethnic group Total indigenous population Ngäbe Kuna Emberá Buglé Wounaan Naso Tjerdi Bokota Bri Bri Other Indian group Not declared Afro-descendent population Total population of Panama

Total number of people 417,559 260,058 80,526 31,284 24,912 7,279 4,046 1,959 1,068 460 5,967 313,289 3,405,813

Percentage of total population 12.3 % 7.6 % 2.4 % 0.9 % 0.7 % 0.2 % 0.1 % 0.05 % 0.03 % 0.013 % 0.1 % 9.2 % 100 %

Source: By author based on data from the 2010 DEC Population Census.

164

Chapter Six

According to the latest national census from 2010, indigenous peoples represent 12.26% of the total national population, 417,559 people (DEC, 2010). Of these groups, the Ngäbe are most numerous (representing 62.38% of all indigenous people), followed by the Kuna (19.28%), Emberá (7.49%), Buglé (5.97%), Wounaan (1.74%), Naso Tjerdi (0.97%), Bri Bri (0.26%) and, lastly, the Bokota (0.23%). In addition to establishing the boundaries between the country’s ethnic groups, the censuses also placed them geographically. The first census to define “indigenous population” (in 1940) was, in particular, highly essentialist: it limited indigeneity to the inhabitants of peripheral regions “locked into” their “traditions” (OC, 1945: 54). This definition, obviously based on prejudicial assumptions about Indians, excluded indigenous workers on the banana plantations, as well as indigenous families in cities and the Canal Zone. Given all these omissions, it comes as no surprise that the census identified only five percent of the national population as indigenous. This incomplete and biased result subsequently shaped official planning for infrastructure and services, being invoked to justify the lack of public investment in indigenous zones (Heckadon, 1982). The idea that the indigenous peoples lived exclusively in inaccessible regions far from urban hubs endured until the 1990s. In recent years (as I will argue in the following section), identity has not been tied so closely to geography, and the presence of native Indians has been recognised throughout the country, in both rural and urban areas. The comarcas, however, are still seen by all parties as Indian homelands par excellence. At present, comarcas represent 23.2% of the country’s territory (17,651 km2) and encompass some 48% of the Panamanian indigenous population. The 52% living outside the comarcas is scattered throughout the rest of the country–on lands adjacent to the comarcas excluded during the delimitation process, as well as in the Panamanian “interior”, in the cities of Colon and David and the surrounding zone15, in Changuinola and other banana-growing areas, and in tourist hubs such as Contadora Island–where they support themselves as farmers, fishermen and salaried employees. In the multiethnic spaces of Panama, inter-ethnic relations are not always smooth. Friction and separation follow in part from differences in intra-group dynamics–Kuna migrants in Colon and Panama City tend to live together and reproduce their socio-political organisation, founding suburban enclaves, appointing authorities and building congress halls–but they are also encouraged by the discriminatory policies of outside change agents and transnational companies, which have organised production by playing off one group of workers against another (Bourgois, 1988).

Social and Political Construction of Racial and Ethnic Categories (Panama) 165

Table 6-2. Panama’s indigenous comarcas: Legislation, size and population 2000-2010

Comarca

Creation: Law no. & date

km2

2010 Population (preliminary data)

2000 Population

Kuna Yala 2 (16 February 1953)

3206

32,446

31,577

EmberáWounaan

4383

8246

9544

2319

*

-

6968

110,080

154,355

775

**

**

17,651

150,772

195,476

22 (8 November 1983)

Madungandi 24 (12 January 1996) (Kuna) Ngäbe-Buglé 10 (7 March 1997) Wargandi 34 (25 July 2000) (Kuna) Totals

* Demographic data included in the Chepo district. ** No data available, as the comarca was created prior to 2000. Source: By author, based on data from the national population censuses (DEC, 2000 and DEC, 2010).

The latest censuses find indigenous population cores both in rural areas far from the metropole and in urban areas. The rural sector is found in territories of great biodiversity or as-yet unexploited natural resources (fast-flowing rivers, mineral deposits) or in regions highly attractive to tourists (with coral islands, beaches, etc.). The very large urban indigenous population–which, at least among the Kuna, now outnumbers comarca inhabitants–underscores their mobility when services or jobs are scarce at home.

3. The definition of indigenous and Afro-descendent populations in censuses from 1911 to 2010 In the light of my concern with the mechanisms and assumptions governing ethnic categorisation, the discrepancy is striking between the efforts made to identify indigenous peoples and the lack of similar interest in the black population, a difference that probably follows from regional indigenist policies and from persistent efforts by indigenous leaders to secure territories and resources. Thus I focus particular attention on two

166

Chapter Six

questions: What terms did the state use to denominate indigenous peoples? And what mechanisms and practices generated those terms? In Panama, though the concept of “indigenous people” is recognised in jurisprudence, there is no accepted official definition of the term (CEPAL/CELADE/BID, 2005: 24). The 1972 Constitution, which was heavily influenced by Reina Torres de Arauz and other anthropologists, explicitly recognised the existence of native Panama, mentioning “indigenous communities” in several articles concerning land ownership, political-administrative divisions of the government, promotion of language conservation, and the development of traditional material, social and spiritual values. The document failed, however, to offer a clear workable definition of indigenous. Indigeneity is thus whatever the national census says it is. But the census neither defines the term nor applies it consistently to communities or other collectivities. Since census questions ask responders to identify themselves according to ethnic categories, indigenous names are thus primarily individual identifiers. Similarly, Afro-Panamanians have been identified by race or ethnicity only in the census, and between 1950 and 1990 they were ignored altogether. The census has also been the only government mechanism that has defined indigeneity. It should be noted that Panama differs from other countries in which efforts to name and count indigenous peoples have been motivated by the demands of international organisations (Peyser and Chackiel, 1994). After 1925, Panama’s enduring (if inconsistent) commitment to racial and ethnic taxonomy16 reflects, I would argue, a non-homogenising model of the Nation State, one that seeks not to make the population uniform but rather to organise and manage its multiplicity for the sake of social peace and the country’s primary revenue source, the Canal and the transit zone. My analysis of the censuses makes this peculiar national model clear, at the same time that it reveals the dynamics of twentieth century ethnic policies. I analyse in turn each of the censuses, demonstrating the criteria employed in identifying the indigenous population and the racial and ethnic categories that have accumulated over the last one hundred years to record and administer the social diversity of the Isthmus. Although the large majority of the indigenous population was not counted in the first censuses of 1911, 1920 and 1930 due to logistical problems17 (Heckadon, 1982: 90), they did use categories of race, gender and age to classify the country’s ethnic and social diversity. The racial categories first used in the 1911 census were white, mestizo, black, yellow [sic, i.e. Chinese], indigenous and [East] Indian. Information was also

Social and Political Construction of Racial and Ethnic Categories (Panama) 167

collected on national origin and religion: Catholic, Protestant, Buddhist, idolater (almost certainly referring to indigenous religions, since the totals for the category derive from districts populated by indigenous people), Muslim, Christian, Jewish, Orthodox, Mormon, Deist and miscellaneous (CG, 1917). The 1920 Census kept the notion of race but defined it in a selfcontradictory way. On the one hand, it posited that “human races can be divided into three main families: white or Caucasian, yellow or Mongol and black or Melanian” and that “the principal traits that separate human races can be basically reduced to physiological considerations”; on the other hand, it recognised the category of “mixed race”, arguing that it was “the effect of crossing one race with another race, such that mulattos are the offspring of black and white, zambos are the offspring of Indian and black and mestizos are the offspring of white and Indian” (DGC, 1922 bulletin n°1: 17). Perhaps the most notable contradiction in these racial categories is that indigenous people were not listed anywhere in their own right, although oddly enough, new racial types (mestizos and zambos) were recognised consisting of indigenous-white and indigenous-black crosses. This self-contradictory classification is best understood in terms of Verena Stolcke’s thesis that the category of mestizo, like those of mulatto and zambaigo, did not arise spontaneously during colonisation to recognise new racial entities, but rather it was constructed to reinforce already existing socio-political distinctions. As Stolcke notes, the social categories labelling people of mixed race in Spanish America “were not the result of pre-existing differences between their ancestors. Only when the children of Indian mothers and Spanish fathers were designated by the special label of mestizo did the socio-political distinctions and inequalities between their parents come to light and be institutionalised” (Stolcke, 2008: 50). From this perspective, it should come as no surprise that although individuals of mixed race predominated in Panama in the early twentieth century18, the Census Office continued to use the categories “white, black, yellow, mestizo, Indian and Hindu”, so that census-takers could classify the population. Mixed-race people, rather than diluting or calling into doubt the essence of these categories, did just the opposite, heightening the differences between them. The 1930 Census contained no significant differences from the previous ones. The racial categories of “white, black, mestizo, yellow and Indian Panamanian” continued in use, with the addition of the category “mulatto” (DGC, 1931).

168

Chapter Six

The 1940 Census employed for the first time the concept of “indigenous tribes” (Heckadon, 1982: 90). Unlike previous censuses, which only referred in general terms to the “Indian race” or the “indigenous”, the new census broke the category down into three groups: Guaymí, Kuna and Chocó, as well as alluding to the Cholos or Coclesanos, as Coclé Indians who had “lost their racial and cultural identity and did not live in tribes”. As noted above, this census also claimed explicitly that the indigenous population had “largely withdrawn to the valleys of inaccessible mountains and coastal islands” (OC, 1943: 54). Kuna, Guaymí and Chocó recorded in urban areas were not defined as indigenous, since they did not fit the assumption that all indigenous people lived outside state control. Even in the rural hinterland, in “the remotest and most inaccessible mountain and coastal regions” in the provinces of Darien, Panama, Colon, Chiriquí, Veraguas and Bocas del Toro (OC, 1945: 59), the census drew social distinctions. In the words of the census report, “Indians who live within the Republic’s social and political structure, speak Spanish and practise Catholicism are defined as part of the civil population” (OC, 1943: 41) and classified racially as mestizos. Thus 556,589 persons were recorded as part of that civil population, and only 55,957 as unassimilated Indians. Despite this broad distinction between the civil and indigenous populations, it is noteworthy that racial categories were still employed. Census organisers, well aware of the difficulties in operationalising these categories, noted that the concept of race was “open to subjective interpretations by the census-takers, as they did not have any reason for applying a rigid and strict scientific criterion, since they were not anthropologists” (OC, 1943: 59). They decided, nonetheless, to retain the distinction between white, yellow, mestizo and Indian races. The only change made was to fold mulatto into mestizo, which radically broadened the latter category beyond its traditional usage as a label for Indian-white admixture (OC, 1943: 60). One paradoxical result of these changes was that the civil population could be classified under four categories, “white, black, mestizo and other races”, but not as indigenous. These contradictions testify to the permeability and flexibility of categories based on racial criteria. Despite the shift towards social categorisation, the Census Office did not discard racial distinctions, because it did not want to represent the country’s cultural diversity in political and social terms. Doing so might have meant acknowledging the equality of all citizens, an equality that, according to the census publication, was impossible to achieve in the Panama of the 1940s, given the country’s position in world trade and the multiplicity of its ethnic

Social and Political Construction of Racial and Ethnic Categories (Panama) 169

groups, it could not achieve social homogeneity, and thus it needed to develop racial classification mechanisms, which would in turn gave rise to “social legislation” placing “restrictions on several ethnic groups” (OC, 1945: 94)19. After World War II, racial criteria were abandoned on the 1950 Census, in favour of the category “indigenous population” to designate “Indians who live primordially separated from the country’s socioeconomic structure and at times even under tribal organisation” (DEC, 1954, vol. IV: xiii). Indigenous populations were increasingly seen, in Panama and elsewhere, as a challenge to national integration, reflecting the influence of Pan-American indigenism, and specifically the InterAmerican Indigenist Institute and its national counterparts20. For the 1960 Census the definition of “indigenous population” was slightly modified. It was defined as “made up of the isthmus’ aboriginal inhabitants, who live in populated sites situated in regions primarily inhabited by them. In general, they speak a dialect and have maintained their traditions and customs” (DEC, 1962: vii). This definition closely follows Convention 107 of the International Labour Organization (ILO), which was adopted by that body in 1957, though Panama did not in fact ratify Convention 107 until 1971. A complex, multi-part definition in the first article of the Convention applied the term “indigenous” to (a) members of tribal or semi-tribal populations in independent countries whose social and economic conditions are at a less advanced stage than the stage reached by the other sections of the national community, and whose status is regulated wholly or partially by their own customs or traditions or by special laws or regulations; (b) members of tribal or semi-tribal populations in independent countries which are regarded as indigenous on account of their descent from the populations which inhabited the country, or a geographical region to which the country belongs, at the time of conquest or colonisation and which, irrespective of their legal status, live more in conformity with the social, economic and cultural institutions of that time than with the institutions of the nation to which they belong.

and to semi-tribal groups, namely, groups and persons who, although they are in the process of losing their tribal characteristics, are not yet integrated into the national community.

In Panama, census employees in this era were sensitive to the integration problem. The official returns for 1960 record a decrease in the indigenous population in the provinces Bocas del Toro, Colón and Panama

170

Chapter Six

(above all in the Alto Bayano), a change attributed to a “gradual integration movement of the Panamanian Indians into the country’s social and economic structure, especially among youth from the Guaymíe and Kuna tribes” (DEC, 1962: ix). Ten years later, the 1970 Census reflected a change in data collection. Although the old definition of indigenous population was carried forward (DEC, 1971a: vii), a new concept was introduced, that of “indigenous populated site”, meaning a location completely occupied by an indigenous population. A more geographic outlook began to take hold in that year. The concept of “populated site” replaced the more overtly social categories, and the indigenous versus non-indigenous dichotomy played a smaller role in data collection. Although the information on indigenous communities appeared separately in reports, census workers used a single form for all localities. Only the separate agricultural survey collected information separately for indigenous zones. It should not be forgotten that discussion of indigenous issues became increasingly prominent after General Omar Torrijos came to power in 1968. They were debated, among other places, in a newly created populist body, the assembly of representatives of corregimientos, which itself reflected a new territorial division of the country. During discussions of a preliminary draft of the National Constitution adopted in 1972, polemic heated up concerning Article 116 of Chapter VII, “on the agrarian system”, which stipulated that the state would guarantee lands to indigenous peoples under collective ownership21. This provoked discussion of who precisely these indigenous peoples were and where they lived, thus returning the discussion to geographic concerns. The next census, in 1980, preserved this geographical perspective, emphasising the population’s spatial distribution and the size of populated sites, but it also listed the number of indigenous and non-indigenous inhabitants of all populated sites in the country. For the first time, the term “indigenous area” was used to refer to those inhabited by the aboriginals of the isthmus who live in populated sites situated in regions principally reserved for them. In general, they speak a dialect and have maintained their tribal traditions and customs (DEC, 1982: 2).

The 1990 Census marked a watershed in the history of the census and its relationship to indigenous peoples, thanks to pressure from several anthropologists and demands from indigenous organisations: the Indians themselves now understood that, to obligate the state to invest in their communities, they had to prove they represented more than 5% of the

Social and Political Construction of Racial and Ethnic Categories (Panama) 171

national population. In that year the concept of “indigenous populated site” was abandoned once and for all, and the definition of “indigenous population” was modified. Thereafter, indigenous was defined for census purposes as “all those who state that they belong to an indigenous group”, without regard to where they were censused. By the new criteria, an indigenous population could be found in “any geographic area in the country” (DEC 1991b: 6). With the geographic and social criteria withdrawn for indigeneity, self-identification prevailed (GuionneauSinclair, 1992). Although it might seem obvious that many indigenous people might live outside official indigenous communities22, cutting the previously assumed connection between location and social identity has important implications. Business and political elites have often exerted pressure to restrict indigenous rights to the comarcas. One frequently hears comments along the lines of “the indigenous already have their lands in the comarcas, and they shouldn’t move out of them” or “if the indigenous want their rights to be respected, then they should have to go to their comarca”. These arguments are particularly dangerous concerning lands surrounding the comarcas with numerous indigenous inhabitants, which are coveted for their natural resources. Regarding these spaces, business interests argue that indigenous people already enjoy abundant land in the comarcas; that more should not be granted them; and that the indigenous population should be concentrated within existing lands23. These pressures to de-indigenise Indians and to restrict their land rights are especially acute where Indians come between the interests of transnational companies and the resources in their territories. Despite the persistence of these highly biased and pernicious preconceptions, legally, being indigenous in Panama is not synonymous with living in a comarca, according to either the census or the Constitution. These documents, moreover, are in line with ideas expressed in the United Nations Declaration of the Rights of Indigenous Peoples and in ILO Convention 16924. In the most recent population census from 2010, indigenous categories remained as before, but ethnic categories were changed. For the first time, respondents were allowed to self-identify as “black or Afro-descendant”. This category was defined as “the social group originating in Africa initially brought to America by the Europeans and that is divided into subethnic groups, in accordance with the different periods in which they arrived on the isthmus of Panama”. Here identity is a matter of descent: respondents are asked about their ancestors, with the option of identifying themselves as “Colonial black” or “Antillean black”, as well as by the

172

Chapter Six

miscellaneous categories of “black” or “Other black”. The published results show that out of a national population of 3,405,813 people, 12.3% (417,559) claimed membership in an indigenous group, and that 313,289, constituting 9.2%, identified themselves as blacks or Afro-descendants. The inclusion of Afro-descendants in the 2010 Census, which followed United Nations recommendations, was also promoted within the government by the National Council of Ethnic Blacks, which reports to the office of the President of the Republic. Outside the government, the initiative was also supported by the CONEGPA, which hoped that the new data would encourage diagnosis of the situation of Afro-Panamanians and thus promote policies favouring social inclusion. Afterwards, however, CONEGPA spokespersons claimed that many census-takers did not follow instructions, failing to ask all respondents relevant questions and thus significantly undercounting the Afro-descendant population25. Table 6-3. Change in Panama’s indigenous population according to censuses, 1911-2010 Census 1911 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010

Total population 300,564 446,098 467,459 566,589 805,285 1,075,541 1,428,082 1,831,399 2,329,329 2,839,177 3,405,813

Indigenous population 11,028 15,369* 42,897 55,987 48,654 62,187 75,738 93,091 194,269 285,231 417,559

Percentage 3.7 % 3.4 % 9.2 % 9.9 % 6% 5.8 % 5.3 % 5.1 % 8.3 % 10 % 12.3 %

* Does not include data on the Chiriquí and Veraguas provinces. Source: By author based on 1911-2010 census data.

4. Census methodology and recognition of ethnic differences in Panama, 1911-2010 In understanding the relationship of Panamanian state with racialised and ethnicised groups, the methods employed in censusing population have been just as important as the criteria for defining and setting the boundaries of ethnic categories. In this respect, it is important to highlight

Social and Political Construction of Racial and Ethnic Categories (Panama) 173

what might be termed ventriloquism–eliciting results from third parties and putting words in their mouths–a procedure that characterised field methods in the first three Panamanian censuses. In the first census of 1911, the instructions given to census employees were explicit on this point: In those districts in which there are savage Indians, set this on record. When the census registry of these Indians presents difficulties that are insurmountable, try to obtain the approximate number of Indians in the tribes from truthful and prudent parties who are experts on the tribes in question.

As the excerpt makes clear, this census was de jure, as were the ones that followed in 1920 and 1930, that is, no direct information was obtained from indigenous respondents, but instead from non-indigenous third parties. Census-taking, moreover, extended throughout the dry season (January to April). The year 1940 marked a significant change in census procedures, inaugurating the country’s first true modern population count. Unlike the three previous censuses, it was carried out on a single day, the 8th of September. The count, moreover, was de facto rather than de jure: censustakers gathered information directly from the heads of families rather than from third parties. Lastly, extensive information was collected about the social and economic structure of the country’s population. Beyond counting the country’s inhabitants, the Census Office took as a primary objective to contribute to studying “the indigenous tribes that still live outside the Panamanian social and cultural structure”. Reflecting this wider scope of interest, several historical and anthropological articles based on census results were published in the magazine Lotería (OC, 1948a and 1948b), which was a primary outlet for disseminating the results of research in the humanities and social science. These advances were hampered, however, by difficulties faced in the field by census-takers. Although they were provided with a special identification card, employees had many problems travelling in indigenous-occupied areas and finding suitable people to provide valid, uniform census information. In the end, they often did not use their cards, and they could only estimate the numbers of indigenous peoples very roughly, classifying them by geographic region, gender and age (minor or adult). In 1950, the now-traditional population census was complemented by censuses of housing and agriculture. If the 1940 effort represented the first modern census in Panama’s history, the 1950 Census was the first detailed

174

Chapter Six

investigation of the Panamanian indigenous population (DEC, 1950: xii). This was the first to record the de facto indigenous population, along with several features of that population. An entire volume of the census was devoted to the indigenous population, and a total of 48,654 native Indians were counted, distributed over 874 towns, villages, and hamlets, 185 of which were also inhabited by non-indigenous people. Census organisers went beyond geographical criteria, counting the indigenous population outside of Indian reserves (DEC 1954, vol. IV: xi). By the criteria they used, they found indigenous populations in seventeen districts. Despite this expansion of territorial scope, however, the 1950 volume showed an improbable decrease of 13% in the indigenous population since 1940. The procedures established in previous censuses–direct census-taking and the use of the indigenous survey–were continued in 1960. Special forms included questions about socio-demographic subjects (age, birthplace, birth-rate, infant mortality), educational and economic characteristics, customs and needs (dialect, foods eaten, dress, access to western medicine, and other social characteristics). In the 1970 Census, a change can be seen in data collection. Beginning in that year, a more geographic outlook prevailed. Although information on indigenous communities appeared in different sections of the report, data was collected on a single form, except for information on agriculture, which was collected on a separate form. Ten years later, the 1980 Census kept this geographic perspective, adding the term “indigenous area” to refer to lands inhabited by the isthmus’s aboriginals who live in populated sites located in regions primarily reserved for them. In general, they speak a dialect and have maintained their tribal traditions and customs. (DEC, 1982: 2)

An indigenous survey was carried out once again in these areas, to collect data on population, housing and agriculture, using linguistic criteria to distinguish between different groups (DEC, 1981). As noted above, the indigenous survey disappeared in 1990, replaced by the question “Which Indian group do you belong to?”. The last census also included a question on Afro-descendant identity: “Does anybody in this household consider themselves Black or Afro-descendant?”. If the person answered affirmatively, he or she was then asked to choose between Colonial black, Antillean black or other. If one takes into account these changes in methodology, it is apparent that the census has been transformed from a mere numerical tally to a complex characterisation of indigenous and Afro-descendant groups, reflecting the state’s need to know the salient traits of these groups, traits

Social and Political Construction of Racial and Ethnic Categories (Panama) 175

like economic, educational and healthcare conditions, which in the second half of the twentieth century have shaped official policies on development. It is no coincidence that changes in data collection occurred in the 1960s, during the consolidation of an economic model based on the expansion of commercial farming into indigenous territory and the exploitation there of natural resources. Like the other socio-economic indicators published by the government and by international organisations–above all in the reports offering poverty indexes for indigenous areas26–census information is functional for a specific development model. Survey design is no exception, being guided by the dominant ideologies of national development. At the same time, however, it is influenced by political demand, in this case the demands of indigenous and Afro-descendant organisations. The inclusion of questions asking respondents to choose from a list of ethnic identities led to demands in the 1980s that such questions be retained in future censuses. To make their national presence visible and to make demands for more public investment, most ethnic organisations–first indigenous, then Afro-Panamanian–demanded population counts based on responses to census questions asking for self-identification.

5. Conclusions In this account, population censuses have stimulated reflection on the ways in which identities and cultural difference have been conceptualised in Panama. Beginning with the distribution of difference and the boundaries between ethnic groups, I have shown that geographic criteria were used in the first censuses to classify the indigenous population, while racial difference was invoked to characterise urban groups. In this way, the censuses reinforced the idea that Panama encompassed two extremes, an urban centre with a civilised population of diverse origins, and a hinterland populated by savages. Over time, this dichotomy was reconceptualised in new terms, civil population versus indigenous population, until it was abandoned in the late twentieth century. The dynamic seen in the conceptual border between indigenous and non-indigenous is also apparent in the boundaries between indigenous groups. The three indigenous entities recognised in the first censuses expanded to eight, after questions calling for self-identification were added to census forms. Another feature of the censuses facilitating analysis was the set of criteria used to define Panama’s indigenous and Afro-descendant populations. As noted above, the national population was classified by race until the 1950s. Out of this classification “mixed-race” categories developed, such as mestizo, mulatto and zambo. Far from blurring or

176

Chapter Six

negating differences, such terms assumed the prior existence of pure cultural identities (Wade, 2004; Stolcke, 2008). Later, the ethnic classification criteria adopted until the 1990s combined geographic and social characteristics that derived from ideologies, notably indigenism, or from regulations of international scope such as ILO Convention 107. Such external influence continues into the present. The criterion of self-identification adopted by the Census Office in 1990 was demanded by indigenous organisations, and at the same time conformed to the policies of organisations such as the Inter-American Development Bank (IDB)27 and the World Bank28, which exert great influence over Panama’s internal policies. Census methodologies also reveal the way that Indians and Afrodescendants have over the years gained recognition as individuals and citizens. Information about indigenous populations was not gathered directly from them until the 1940 Census. Much more recently, at the turn of the twenty-first century, indigenous and Afro-descendant organisations have been mobilising to ensure that they are accurately counted and thus gain visibility. All this underlines the fact that as cultural identities and differences become issues for regulation, they also emerge as political arguments for demanding rights (Kuper, 2003; Kenrick and Lewis, 2004). Paradoxically, the move to differentiate is imposed by the neoliberal state and the international development organisations that influence it, and at the same time by members of the populations being censused. How has this odd conjunction of forces developed? This question leads back to the construction of indigenous categories, and the way in which they are grounded less in culturally objective criteria, and more in policies favouring a particular model of development. During the last 20 years, these policies have been marked by the neoliberal cultural plans of the post-Washington Consensus. At the same time that such policies are aimed at preventing indigenous peoples from blocking neoliberal resource extraction (Bretón, 2005), they have been embraced by indigenous organisations because international cooperation promotes an assistancebased model at a local level. In this context, the integrationist and indigenist models implemented throughout Latin America since the 1940s have become obsolete; the ethnic issue has overshadowed exclusion and financial inequality; and legal debates have replaced the political struggles of the past (Comaroff and Comaroff, 2011). All of these changes can be traced through changes in the design of Panama’s population censuses and, it should added, in those of all the countries of the region.

Social and Political Construction of Racial and Ethnic Categories (Panama) 177

Bibliographical references Alba, Manuel María. 1928. Etnología y población histórica de Panamá, Panamá: Imprenta Nacional. Alba, Manuel María. 1929. Geografía descriptiva de la República de Panamá, Panamá: Benedetti Hermanos. Bourgois, Philippe. 1988. Conjugated Oppression: Class and Ethnicity among Guaymí Banana Workers. American Ethnologist, 15(2): 328248. Bretón, Víctor. 2005. ¿Más allá del neoliberalismo étnico? Enseñanzas desde los Andes del Ecuador. Boletín del Instituto Científico de Culturas Indígenas, 7(78/79): 19-36. CEPAL/CELADE/BID. 2005. Los pueblos indígenas de Panamá: diagnóstico sociodemográfico a partir del censo del 2000. Santiago de Chile: CEPAL. CG–Contraloría General. 1917. Censo General de la República en 1911. Boletín del censo de la República de Panamá 1911. Anexo al boletín número 30. Comaroff John L. and Jean Comaroff. 2011. Etnicidad S.A. Madrid: Katz editores. Cooke, Richard G. 1982. Los Guaymíes sí tienen historia. In Foro Guaymí y Ceaspa (eds), El pueblo Guaymí y su futuro: 27-64, Panamá. DEC–Dirección de Estadística y Censo, Contraloría General de la República. 1950. Boletas y principales formularios utilizados en los censos nacionales de 1950. Panamá: Imprenta Nacional. DEC–Dirección de Estadística y Censo, Contraloría General de la República. 1954. Censos Nacionales de 1950. Panamá: Imprenta Nacional. DEC–Dirección de Estadística y Censo, Contraloría General de la República. 1961. Informe General sobre el levantamiento de los censos nacionales 1960-1962, Panamá: Imprenta Nacional. DEC–Dirección de Estadística y Censo, Contraloría General de la República. 1962. Censos Nacionales de 1960, volumen I, Lugares poblados de la República, Panamá: Imprenta Nacional. DEC–Dirección de Estadística y Censo, Contraloría General de la República. 1971a. Censos Nacionales de 1970, volumen I, Lugares poblados de la República, Panamá: Imprenta Nacional. DEC–Dirección de Estadística y Censo, Contraloría General de la República. 1971b. Informe Metodológico sobre el levantamiento de los censos nacionales de 1970-1972, Panamá, Imprenta Nacional.

178

Chapter Six

DEC–Dirección de Estadística y Censo Contraloría General de la República. 1981. Informe metodológico sobre el levantamiento de los censos nacionales de 1980-82, Panamá: Imprenta Nacional. DEC–Dirección de Estadística y Censo, Contraloría General de la República. 1982. Censos Nacionales de 1980, Lugares poblados de la República, volumen I, Panamá: Imprenta Nacional. DEC–Dirección de Estadística y Censo, Contraloría General de la República. 1991a. Informe metodológico sobre el levantamiento de los censos nacionales de 1990-93. Parte I, Panamá: Imprenta Nacional. DEC–Dirección de Estadística y Censo, Contraloría General de la República. 1991b. Censos Nacionales de Población y Vivienda 1990, Resultados finales ampliados, lugares poblados de la República, volumen I, Panamá: Imprenta Nacional. DEC–Dirección de Estadística y Censo, Contraloría General de la República. 2000a. Resultados Finales Censos Nacionales de Población y Vivienda 2000 [on line]: http://www.contraloria.gob.pa/dec/Aplicaciones/POBLACION_VIVIE NDA/index.htm DEC–Dirección de Estadística y Censo, Contraloría General de la República Contraloría General de la República. 2000b. Censos Nacionales de 2000, X Censo de Población y VI de vivienda, Informe metodológico, Panamá, Imprenta Nacional. DEC–Dirección de Estadística y Censo, Contraloría General de la República. 2010. Censo de población 2010. http://www.censos2010.gob.pa/ DGC–Dirección General del Censo. 1922. Censo Demográfico de la Provincia de Panamá 1920. Boletín núm 1. Panamá: Imprenta Nacional. DGC–Dirección General del Censo. 1931. 1930 Censo Demográfico, tomo I y II. Panamá: Imprenta Nacional. Guionneau-Sinclair, Françoise. 1992. Los amerindios de Panamá en el censo de población 1990. Scientia [Universidad de Panamá], 7(1): 3143. Heckadon, Stanley. 1982. Quiénes son los Indígenas? Foro Guaymí y Ceaspa (eds), El pueblo Guaymí y su futuro: 32-44, Panamá. Howe, James. 1998. A People Who Would Not Kneel: Panama, the United States and the San Blas Kuna. Washington, D.C.: Smithsonian Institution Press. Howe, James. 2009. Chiefs, Scribes and Ethnographers: Kuna Culture from Inside and Out. Austin: University of Texas Press.

Social and Political Construction of Racial and Ethnic Categories (Panama) 179

Johnson, Frederick. 1948a. Central American Cultures: an Introduction. Bureau of American Ethnology, Bulletin, 143(4): 43-68. Johnson, Frederick. 1948b. The Caribbean Lowland Tribes: the Talamanca Division. Bureau of American Ethnology, Bulletin, 143(4): 231-251. Jordan Ramos, Osvaldo. 2010. Indigenous Mobilization, Institutionalization and Resistance: the Ngöbe Movement for Political Autonomy in Western Panama. Ph.D. Dissertation, University of Florida. Kenrick, Justin and Jerome Lewis. 2004. Indigenous Peoples’ Rights and the Politics of the Term “Indigenous”. Anthropology Today, 20: 4-9. Kertzer, David and Dominique Arel. 2002. Census, Identity Formation, and the Struggle for Political Power. In David Kertzer and Dominique Arel (eds) Census and Identity: the Politics of Race, Ethnicity and Language in National Censuses. Cambridge: Cambridge University Press, pp. 1-42. Kuper, Alan. 2003. The Return of the Native. Current Anthropology, 44: 389-402. Maloney, Gerardo. 1989. El canal de Panamá y los trabajadores antillanos Panamá 1920: cronología de una lucha. Panamá: Ediciones Formato. Martínez Mauri, Mònica. 2007. De Tule Nega a Kuna Yala. Mediación, territorio y ecología en Panamá, 1903-2004. Ph.D. Dissertation. Barcelona – Paris: Universitat Autònoma de Barcelona and Ecole des Hautes Etudes en Sciences Sociales. Martínez Mauri, Mònica. 2011. Panamá: pocos avances y muchos retrocesos en el reconocimiento de los derechos indígenas sobre tierras, territorios y recursos naturales. In Marco Aparicio Wilhelmi (ed.) Los derechos de los pueblos indígenas a los recursos naturales y al territorio conflictos y desafíos en América Latina. Barcelona: Editorial Icaria, pp. 223-269. Nobles, Melissa. 2002. “Racial Categorization and Censuses”. In David Kertzer and Dominique Arel (eds), Census and Identity: the Politics of Race, Ethnicity, and Language in National Censuses. Cambridge: Cambridge University Press, pp. 43-71. Nordenskiöld, Erland. 1928. Indianerna pa Panamañaset, Stockholm: Ahlén & Akerlunds Forlag. OC–Oficina del Censo, Contraloría General de la República. 1943. Censo de Población 1940, informe preliminar. Panamá: Imprenta Nacional. OC–Oficina del Censo, Contraloría General de la República. 1945. Censo de Población 1940 Compendio General, volumen X. Panamá: Imprenta Nacional.

180

Chapter Six

OC–Oficina del Censo. 1948a. La provincia de Chiriquí. Revista Lotería, 82: 14-25. OC–Oficina del Censo. 1948b. Tribus indígenas de Panamá (los Guaymíes). Revista Lotería, 82: 25-29. Peyser, Alexia and Juan Chackiel. 1994. La población indígena en los censos de América Latina. In Estudios sociodemográficos de pueblos indígenas. Santiago de Chile: CELADE, CIDOB, FNUP, ICI. Pinart, Alphonse L. 1882. Colección lingüística y etnografía americanas, part IV “Noticias de los indios del departamento de Veragua y vocabularios”, San Francisco: A. L. Bancroft. Pinart, Alphonse L. 1887. Les Indiens de l’Etat de Panamá. Revue d’Ethnographie, 33-56: 117-32. Pinart, Alphonse. L. 1892. Vocabulario Castellano-Guaymíe, dialectos Move-Valiente, Norteño y Guaymí-Penonomeño. Paris: Petite Bibliothèque Américaine. Pinart, Alphonse. L. 1897. Vocabulario Castellano-Guaymíe, dialectos Murire-Bukueta y Sabanero. Paris: Petite Bibliothèque Américaine. PNUD–Programa de las Naciones Unidas para el Desarrollo. 2011. Atlas de desarrollo humano y objetivos de desarrollo del milenio (ODM) de Panamá. Panamá: PNUD. Priestley, G. and Alberto Barrow. 2010. El movimiento negro en Panamá: una interpretación histórica y política, 1994-2004. In Odile Hoffmann, Política e identidad afrodescendientes en México y América Central. Mexico: INAH-UNAM-CEMCA-IRD. Stolcke, Verena. 2008. Los mestizos no nacen, se hacen. In Verena Stolcke and Alexandre Coello, Identidades ambivalentes en América Latina (siglos XVI-XXI). Barcelona : Edicions Bellaterra. Torres de Araúz, Reina. 1980. Apuntes etnohistóricos sobre guaimíes en el siglo XVIII. Panamá: Actas del I Congreso Nacional de Antropología, Arqueología y Etnohistoria, INAC. Turner Y., Anayansi. 2008. El derecho de autodeterminación de los pueblos indígenas de Panamá. Universidad de Panamá: Instituto de Estudios Nacionales. Wade, Philip. 2004. Images of Latin American Mestizaje and the Politics of Comparison. Bulletin of Latin American Research, 23(3): 355-366. World Bank. OP 4.10–Indigenous Peoples, Washington D.C.: World Bank. Wassén, Henry. 1963. Algunas observaciones sobre la división de los indios guaimíes, Lotería, 8(86): 86-94. Young, Philip. 1970. Notes on the Ethnohistorical Evidence for Structural Continuity in Guaymí Society. Ethnohistory, 17(1-2).

Social and Political Construction of Racial and Ethnic Categories (Panama) 181

Legislation and normative frameworks ILO–International Labour Organization. 1957. Convention No. 107 on Indigenous and Tribal Populations. Geneva. ILO–International Labour Organization. 1989. Convention Nº 169 on Indigenous and Tribal Peoples in Independent Countries. Geneva. United Nations Declaration on the Rights of Indigenous Peoples (2007) Ley 2 (16 febrero 1953) Por la cual se organiza la Comarca de San Blas, Gaceta Oficial Nº 12042 de 7 de abril de 1953. Ley 22 (8 de noviembre de 1983) Por la cual se crea la Comarca EmberáWounaan, Gaceta Oficial N° 19,976, del 17 de enero de 1984. Ley 24 (12 enero de 1996) Por la cual se crea la comarca kuna de Madungandi, Gaceta Oficial Nº 22951 de 15 de enero de 1996. Ley 10 (7 de marzo de 1997) Por la cual se crea la comarca ngobe bugle, Gaceta Oficial Nº 23242, 11 de marzo de 1997. Ley 34 (25 de julio de 2000) que crea la Comarca Kuna de Wargandí, Gaceta, Gaceta Oficial N° 24106, 28 de julio de 2000.

Notes 1

Between the sixteenth and nineteenth centuries, efforts were made to quantify the indigenous population on the isthmus. When Panama was a department of Colombia in 1870, the last census of the 19th century was taken, although the first attempt to systematically measure the population in Panama was not carried out until 1911, eight years after independence. 2 For the production and translation of the present chapter, the author has received support from the following: “Ambivalent Identities: Comparative Study on Social Classification Systems”, financed by the Autonomous Government of Catalonia (SGR00658) and “Ambivalent Identities: Comparative Study on Social Classification Systems” from the National R&D&I Plan of the Ministry of Science and Innovation (MICINN) of the Spanish government (HAR2008-04582/HIST). The author would like to thank the members of the Research Group on the Anthropology and History of the Construction of Social and Political Identities (AHCISP), James Howe (MIT) and the publishers of the present volume for their perspicacious and stimulating comments during the writing process. 3 Documentary research was carried out during August 2011 in the library of the Comptroller General of the Republic of Panama, thanks to the support of the two research projects acknowledged in Note 2. 4 “Interior” means the south-western region, also known as “the Central Provinces”. The mestizos from that region are called interioranos in Panama. 5 Congresses of the Emberá, Wounaan, Ngäbe-Buglé, Kuna de Madungandi and Kuna de Wargandi had previously been organised, but they were not legally recognised by the state until these laws were approved.

182

Chapter Six

6 For a comprehensive recent analysis of Ngäbe socio-political structure and Ngabe relations with the Panamanian state cf. Jordán (2010). 7 To date, ILO Convention 169 has been ratified by the following Latin American countries: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominica, Ecuador, Guatemala, Honduras, Mexico, Paraguay, Peru, Venezuela and Nicaragua. 8 During the 1911-1940 period, the census category used to denominate the populations of descendants of African slaves and Antillean immigrants was “black race”. The 2010 census substituted the term “Afro-descendent”. 9 This category was used in the 2010 Census for the descendants “of the African slaves brought to the isthmus during Spanish colonisation. The descendants of the latter can be identified in the central provinces, in areas like Natá, Parita and Monagrillo, and in Chiriquí, in areas such as Puerto Armuelles and Alanje, in Colon province in areas like Costa Arriba and Costa Abajo, and in Pacora, San Miguel and Chepo in Panama province”. 10 According to the 2010 Census: “Descendant of French- and English-speaking Antilleans or other languages that came to Panama, primarily during the construction of the Transisthmian Railway and the French canal project at the end of the 19th century, and during the canal construction by the Americans. The majority are located in areas of the cities of Panama and Colon and in the province of Bocas del Toro”. 11 According to the 2010 Census, “black” is defined as: “people with ancestors who are descendants of the enslaved or Colonial blacks, and/or descendants of Antillean blacks or French- or English-speaking Afro-Antilleans or speakers of other languages, migrants during the different periods of national development who selected this option for their self-identification”. 12 According to the 2010 Census, “Other” is defined as “some other black or Afrodescendant group not included in the previous categories”. 13 Refers to the Guna group. Until the 1970s, the spelling “Cuna” predominated; afterwards, up until March 2011 “Kuna” was standard. Since then, the Guna General Congress has mostly used “Guna”. Law 88 of 22 November 2010, however, mandates “Kuna”. In this article, I have retained the spelling “Kuna” because it was used in the most recent census publications. 14 Francisco Herrera, personal communication of June 2011. 15 Specifically, in the outskirts of David, the concentrations of the Ngäbe population in Las Lomas must be taken into account (personal communication by Phil Young). 16 As Howe (2009) has noted this was not the policy followed by liberal governments before 1925. In the era of Porras, the policy was homogenising, and thus the government avoided recognising the existence of multiple indigenous peoples. They were all lumped together as Indians or indigenas, and they were all expected to give up that identity forthwith. 17 As Howe has shown, starting in 1920 the Kunas of San Blas were counted in the national censuses and, in the province of Colon, the variable “Indian race” was an option. In San Blas censuses were elements of government practice. The police and administration noted births and deaths on official forms to control the Indians, but

Social and Political Construction of Racial and Ethnic Categories (Panama) 183 Kuna people interpreted such documentation as a kind of symbolic claim-making. To count and record was to incorporate and appropriate (Howe, 1998: 180). In his latest book (2009: 79-80) Howe argues that not publishing a count for the Kuna was part of a policy of down-playing or even erasing a separate indigenous identity during an era in which the government was committed to complete assimilation of the Kuna. A complete census count was indeed taken in San Blas in 1920, but the total was released unofficially. Only a very small part of the indigenous count was published officially in the volume for Colon province. In doing so the national census influenced Indian identity by ignoring it. 18 The census bulletin stated that: “Of these mixed-races, there are a considerable number on the isthmus and it is almost a surety that when considering the races in their legitimate pure states, the percentage of specimens that could resist the thoroughness of the exam would be very low” (DGC, 1922: 18). 19 The Constitution that came the following year (1941) disenfranchised Englishspeaking Afro-Antilleans, Chinese, and people from India, North Africa and Asia Minor. They were all called “prohibited immigration” (inmigración prohibida). 20 In 1956, Panama held its first and only Indigenist Congress (Howe, 2009: 191). 21 This article ended up being transformed into the current Article 127, Chapter VIII: “The state shall guarantee the reservation of the lands that are necessary and their collective ownership to the indigenous communities in order to achieve their financial and social wellbeing”. 22 Those non-comarca Indians actually enjoy legal possession as individuals and families under Panamanian land law (Law no. 72, 2008 Collective Lands Act) but outsiders often succeed nonetheless in stealing their lands. 23 These arguments were popularised in the 2000s by Haydeé Milanés de Lay. This Darién legislator repeatedly criticised the creation of comarcas and the approval of the Collective Lands Act. 24 The definition of “indigenous population” in Panama is in line with Article 33 of the United Nations Declaration on the Rights of Indigenous Peoples, which specifies that “indigenous peoples have the right to determine their own identity or membership in accordance with their customs and traditions” and Article 1.2 of ILO Convention 169 states that “self-identification as indigenous or tribal shall be regarded as a fundamental criterion for determining the groups to which the provisions of this Convention apply”. 25 “9.2% of those participating in the census stated they were Afro-descendants”, La Prensa, 30-12-2010. 26 See, for example, the 2008 Standard of Living Survey by the Panamanian Ministry of Economy and Finance; the National Human Development Report drafted by the UNDP in 2002, which states that “the indigenous peoples have disturbing levels of poverty, which not only situates them as a human group in a condition of vulnerability, but also as a high-priority human group” (2002: 112); or the Human Development ATLAS and Panama’s MDO (millennium objectives) (PNUD, 2011), which also make it clear that the indigenous zones have a very low Human Development Index. 27 The IDB’s operating policy, following the indications of Convention 169, define indigenous people based on three criteria: (i) they are descendants of the peoples

184

Chapter Six

who inhabited the region of Latin America and the Caribbean in the period of the Conquest or colonisation; (ii) whatever their legal status or current location may be, they partially or totally conserve their own institutions and social, economic, political, linguistic and cultural practices; and (iii) they self-describe as belonging to indigenous or pre-colonial peoples or cultures (IDB, 2006: 5). 28 Although the World Bank does not define the term “indigenous peoples”, it uses the term “in a generic sense to refer to a distinct, vulnerable, social and cultural group possessing the following characteristics in varying degrees: (a) self-identification as members of a distinct indigenous cultural group and recognition of this identity by others; (b) collective attachment to geographically distinct habitats or ancestral territories in the project area and to the natural resources in these habitats and territories (c) customary cultural, economic, social, or political institutions that are separate from those of the dominant society and culture; and (d) an indigenous language, often different from the official language of the country or region” – http://web.worldbank.org/WBSITE/EXTERNAL/PROJECTS/EXTPOLICIES/EX TOPMANUAL/0,,contentMDK:20553653~menuPK:4564185~pagePK:64709096 ~piPK:64709108~theSitePK:502184,00.html

CHAPTER SEVEN FROM PRE-MODERN “INDIANS” TO CONTEMPORARY “INDIGENOUS PEOPLE”: RACE AND ETHNICITY IN PERUVIAN CENSUSES 1827-2007 DAVID SULMONT AND NÉSTOR VALDIVIA

1. Introduction The purpose of this chapter is to present a critical review of the various strategies employed by the state to classify the Peruvian population in terms of ethnic or racial categories. We will focus our attention on the challenges and debates of the present, but we will describe how the official statistics, particularly in the population censuses, have reflected both the political agenda and the cultural and social processes that have framed the debate about the ethnic and cultural diversity of Peruvian society in different moments of its Republican history1. Contemporary discussion in Latin America about the measurement of the ethnic characteristics of the population has focused on the four dimensions that have oriented the design of national censuses in some countries of the region since 2000: self-identification, common origin, culture and territory (Schkolnik, 2009: 67-8). In comparison with other Andean countries (Bolivia, Ecuador, Colombia), the inclusion of a more complex array of indicators in the official statistics seems to suffer some “delay” in Peru. Between 2000 and 2007, 16 Latin American countries had already included an ethnic or racial self-identification question in their censuses (Schkolnik, 2009: 77-9). Despite this regional trend, the 2007 Peruvian national census did not ask this kind of question. However, the last ten years have witnessed several research projects and studies presenting theoretical and methodological innovations to test and use

186

Chapter Seven

different variables and questions to tap the ethnic and racial dimension of Peruvian society in official, academic and public opinion surveys2. The results of those studies and projects represent important inputs for the current debate on how to present a more accurate image of a society profoundly marked by ethnic and racial distinctions. In the Peruvian case, public officials, academics and several social organisations representing ethnic groups are beginning to engage in a debate on how to include those dimensions in the next census due in 2017. The last census of 2007 reproduced the same kind of indicators used in Peruvian censuses since 1940 to identify indigenous and non-indigenous populations, based almost solely on a linguistic marker (maternal language). This methodological choice is clearly unsatisfactory in the light of the current debates: it ignores important dimensions such as selfidentification; it makes insufficient distinctions among indigenous populations; and it invisibilises groups like the mestizos, cholos3 or Afrodescendants, who have played an important role in the social imagination and the cultural and political historical narratives of the nation-building process of Peruvian society. As Néstor Vadivia (2011) points out, the use of ethnic or racial categories in official and academic surveys in the last decade in Peru reveals four types of objectives: to acknowledge the situation of poverty and social exclusion of different ethnic groups; to characterise and describe their culture and values; to recognise their individual and collective rights; and to analyse the issues of ethnic and/or racial discrimination. The methodological framework to produce and analyse statistical data on ethnic and racial groups varies according to those objectives. In some studies, supposedly “objective markers” of ethnicity (such as language) have been more important to visibilise situations of poverty and exclusion. In other cases, self-identification indicators are employed to analyse and discuss the issues of individual and collective rights, or the processes of cultural change related to social mobility and social modernisation. External categorisations (how someone is seen or sees “others”) have been used in studies on discrimination (Valdivia, 2011: 13-14). However, beyond the academic discussion or practice in using ethnic and racial variables in surveys and research projects, the production of official statistics, especially in censuses, implies important political decisions about how and why the state or the different social and political groups in the society want to present an ethnic portrayal of the nation: decisions regarding the recognition of citizenship rights (civil, political and social; individual or collective); public programme design (education,

From Pre-Modern “Indians” to Contemporary “Indigenous People”

187

health, poverty alleviation, “social inclusion”, anti-discrimination, affirmative action programmes); or territorial and population management (economic development projects; immigration policies; reproductive policies). Most recently, in 2011 the Peruvian government approved a Ley de consulta previa– Law on prior consultation–(Normas Legales, 2011) in order to adapt the national legislation to the provisions of ILO Convention 169 concerning indigenous people’s rights. This law establishes a process of consultation of indigenous people on economic development projects (mainly private investments in natural resources or energy projects) that might affect indigenous territories. The implementation of this law has reactivated a debate about which groups or collectivities should be considered “indigenous people” in Peru. According to Rogers Brubaker, the state is a powerful identifier that can enforce an ethnic or racial classificatory system through official statistics, public records, territory management, administrative procedures or even the public education curriculum. These processes do not necessarily create “ethnic identities” or “ethnic groups”, but they make “certain categories readily and legitimately available for the representation of social reality, the framing of political claims, and the organisation of political action” (Brubaker, 2004: 54) An official ethnic and racial classificatory system might create an illusion of a society divided into clearly identifiable groups, reinforcing an essentialist view of ethnic or racial differences. By contrast, social and cultural processes show how ethnic or racial identities or categorisations can become quite fluid and porous, changing depending on history, the context of social interactions, or on their relevance for the individuals and groups that might be classified in ethnic or racial terms. In the history of official censuses in Peru we can identify four different phases of ethnic and racial statistics, according to the criteria the state has used to classify the population using ethnic or racial indicators. During the first phase, from Independence in 1821 to the end of the 1940s, the censuses used racial categories to classify individuals. In a second phase, specifically in the census of 1961, an “anthropological” approach was employed to identify cultural differences in the population, recording indicators about what were considered “cultural traits” of the indigenous populations (language, customs, clothing). In the next phase, from 1972 until the 1990s, national censuses “silenced” the issues of ethnic or racial distinctions, and focused on linguistic indicators (maternal language) that were interpreted in relation to the advances and limitations of the education system to alphabetise the population and improve their educational levels. In the current phase, from 2001 to the present, the

188

Chapter Seven

persistence of social inequalities and practices of racial discrimination among different ethnic groups (along with other factors) has motivated academics and public official to analyse more closely the issues of ethnic distinctions, introducing for the first time in some official surveys questions about ethnic and racial self-identification. The main surveys in which those questions were introduced were the Encuesta Nacional de Hogares–ENAHO (National Household Survey) and the Encuesta Demográfica y de Salud Familiar–ENDES (Demographic and Health Survey)4. Despite those developments, the last census of 2007 did not include this kind of questions, leaving the debate on those issues for the next round of national censuses in 2017. In the next sections we will describe those phases in more detail, discussing the relationship between the methodological strategies employed in the ethnic or racial official statistics and their historical context (cultural, social and political). Finally, we will focus our attention on the current practices and debates that will define the next phase of the “official” statistical representation of ethnicity in Peru.

2. Race in censuses from 1827 to 1940 During the colonial period, racial categories, identified as “castes”, corresponded to different legal statuses in the society. Depending on the caste they belonged to, individuals and communities were entitled to some rights or subjected to specific duties. The racial classifications in official statistics reproduced the hierarchical organisation of the colonial society, and censuses were oriented to identify indigenous populations for fiscal and evangelisation purposes. By 1791, the categories used in the censuses were: Indians, Mestizos, Spaniards, Pardos (brown), slaves and “others”. With Independence, the liberal ideologies that inspired the libertadores led to the abolition (in theory) of colonial population categories in the official language of the new Republic. Shortly after the proclamation of Independence in 1821, General San Martin issued a decree by which the use of expressions like “Indians” or “aboriginals” was banned from official language and replaced by the more “modern” terms “citizens” and “Peruvians” (Gaceta del Gobierno de Lima Independiente, 1950: 67). However, in practice, many institutions in charge of making population statistics in the first decades of the Republic (the new government agencies but also the Catholic Church), continued to employ colonial categories (such as “castes”) in their accounts, as can be traced in the civil records of marriages and births of the parishes in Lima in 1841 (Centro de Estudios de Población y Desarrollo, 1972: 107). As under colonial rule,

From Pre-Modern “Indians” to Contemporary “Indigenous People”

189

the official record of the population continued to have fiscal purposes. In the first years of Independence, the tributo indígena (“indigenous contribution”) was abolished, but the libertador Simón Bolívar quickly restored it in 1826, constituting an important source of tax revenues for the state until its final abolition in 1854. The first Republican population census was made in 1827, basically oriented to fiscal and electoral objectives. Most of the records of this census are lost but it is known that it classified the population according to categories inherited from the colonial period: Indians, “castes” and slaves. Historical analysis of the few partial records of the 1827 census that were preserved in the archives led to an estimate of 1.5 million inhabitants, from which 61.6% would have been classified as “Indians” (Gootemberg, 1995: 25). The following official censuses of the nineteenth century were performed in 1850 and 1862, recording a population of roughly 2 million and 2.5 million inhabitants respectively. There is no information in those censuses that made it possible to distinguish the population in terms of racial or ethnic categories (Gootemberg, 1995: 12-13). The census of 1876 is considered the first “modern” census in Peruvian history (Gootemberg, 1995: 13); it was conducted by the newly created Dirección de Estadística del Ministerio de Gobierno (Statistical Office of the Ministry of Interior), with the assistance of a French specialist. This census reported a total population of 2,699,106 inhabitants and included a question about “race”. The information was recorded directly by the census enumerators, who had to classify the individuals by visual observation in the following categories: White, Indigenous, Black, Mestizo, and Asian. The “Mestizo” race was defined as “any mix between the other groups” (Dirección de Estadística - Ministerio de Gobierno, 1878: XXXIII). Using those categories, 57.6% of the population were Indians, 38.6% Whites and Mestizos, 1.9% Blacks and 1.9% Asians5. According to the historical research, the implementation of the question of race in the 1876 census had to be done with “precaution, avoiding hurting the sensibility of the interviewee” (Centro de Estudios de Población y Desarrollo, 1972: 334). The enumerators were instructed not to address the question directly, trying to make their notes in a discrete manner; only in extreme cases could this question be formulated openly, but those situations should be avoided, by deducing the race from the characteristics of the parents or close relatives (Dirección de Estadística Ministerio de Gobierno, 1878: XXXIII). After 1876 Peru did not perform another census until 1908, mainly because the national defeat in the “War of the Pacific” with Chile (18791883) left the state in a situation of complete bankruptcy and disarray. It

190

Chapter Seven

was only at the end of the nineteenth century that the economy and the official institutions showed some signs of recovery. Between 1908 and 1935 the state made eight partial censuses, circumscribed to some regions or departments: Lima (1908 and 1920), Cusco (1912), Arequipa (1917), Huancayo (1917), Callao (1905 and 1920) and Tacna (1935). A national electoral census was conducted in 1931 in preparation for the elections of the same year, but it did not include demographic variables (moreover, at that time, only adult males had the right to vote). Some of those partial censuses included the same question on race used in 1876, which, in some cases, allowed analyses on the evolution of the racial composition of the population between 1876 and 1920. For example, comparison of the census data available for the capital of the country–Lima–shows that the proportion of the white population remained almost constant at around 38% between those years. In the case of the mestizo and indigenous populations, the trend was quite different: the percentage of mestizos rose in Lima from 21.2% in 1876 to 39.9% in 1920; inversely, the percentage of “Indians” decreased from 21.8% to 14.1% in the same period. At that time immigration from rural areas to the capital was not a widespread phenomenon, so it is possible that the increase in interracial marriages and the cultural assimilation of “urban indigenous” might be among the factors behind the increase in the mestizo category in the Lima censuses at the beginning of the twentieth century. The first national census of the twentieth century took place in 1940, showing that the Peruvian population had almost tripled in 64 years, to a total of 7 million inhabitants. Practically the same racial categories of 1876 were used on that occasion6. In 1940 the census enumerators had to classify individuals using the following categories: White, Indian, Black, Yellow and Mestizo. It is interesting to note that the “Mestizos” were defined in the methodological report of the census as “the descendants of the races White and Indian, White and Black, White and Yellow; as well as the descendants of two Mestizos or Mestizo and White, Mestizo and Indian, Mestizo and Yellow, Mestizo and Black (…) the racial types generally known as ‘Zambos’, ‘Injertos’ [grafted], ‘Cholos’, ‘Mulattos’, etc., should be noted simply as ‘Mestizos’” (Ministerio de Hacienda y Comercio, 1944: 587). According to the census report, in 13% of the cases the racial question was directly put to the head of the household, who supplied the information for all members of his family. The remaining 87% of the cases were classified by the census enumerator by direct observation; the census manual stipulated that “this question should not be asked when the individuals can be seen by the interviewer, who has to fill

From Pre-Modern “Indians” to Contemporary “Indigenous People”

191

in this information according to his personal appreciation” (Ministerio de Hacienda y Comercio, 1944: 598). The most important result concerning the racial question in the 1940 Census was the inversion of the racial composition of the Peruvian population: the percentage of indigenous people decreased from 57.6% in 1876 to 45.8% in 1940, while the proportion of “Whites and Mestizos” rose from 38.6% to 52.9% (Ministerio de Hacienda y Comercio, 1944). At this point, it is important to mention that the official report of the 1940 Census merged the “White” and “Mestizo” categories into a single one. To justify this decision the report indicated that it was difficult for the census enumerator to clearly distinguish “White” from “Mestizo” phenotypes, because in Peru, as in many Latin American countries, it was impossible to apply “the European criterion to differentiate the white race” (Ministerio de Hacienda y Comercio, 1944: CLXXIX). According to John Rowe, the Peruvian historian Jorge Basadre indicated that the high percentage of “Whites and Mestizos” (53%) was used as an argument to refute the idea promoted by the indigenista movement of the 1920s and 1930s that Peru was mainly an indigenous nation (Rowe, 1947: 207-8). More recently, Nelson Manrique has argued that merging both categories in the census report was a strategy to hide the fact that by 1940 the white population was clearly a minority in Peruvian society (Manrique, 1999: 6). As will be seen in the following paragraphs, we consider that this methodological choice is more related to the intellectual debate that took place in the 1940s concerning the “Mestizo” nature of Peruvian national identity. Even if they used the same kind of racial categories, the historical and cultural context of the 1876 and 1940 censuses were quite different. During the nineteenth century, despite their liberal language, the “founding fathers” of the Peruvian Republic were all members of the Spanish-descended elite and well aware of their superior social status but also of their minority status relative to the indigenous population. However, in theory, the colonial division between the “Republic of Spaniards” and the “Republic of Indians” could no longer continue in the legal framework of the new Republic. The social (re)construction of the category of “Indian” during the nineteenth century would be different from the former legal and fiscal categories of the Spanish Empire. The Peruvian liberal elites considered that the “backwardness” and “ignorance” of the “Indians” in postcolonial Peru was the consequence of the prolonged Spanish colonial domination, which destroyed the “great Inca civilisation”, transforming their descendants into inferior human beings. This “racialisation” of the indigenous population allowed the Peruvian elites to exclude them from

192

Chapter Seven

the status of full citizens (Larson, 2002; Méndez, 1993; Rowe, 1954; Thurner, 1997; Walker, 1988). In this sense, “Indians” constituted a “burden” for the new nation, a remnant of the colonial domination incapable of becoming modern Peruvians in their current condition. The task of the Republic was to (re)civilise the “Indians” under the enlightened direction of the white elites. By the second half of the nineteenth century, this racial thinking had become more present in the debates concerning the construction and fate of the nation and the racialisation of the social differences extended to other groups like African descendants (sons and daughters of former slaves)7 and Asian immigrants. The 1876 Census took place in a context when, in contrast to the colonial “caste categories”, a different form of racial thinking was being established, but also at a moment when the state still had the impulse of a liberal project to modernise and civilise the country. The wealth of the “guano era” (1840-1879) allowed the government to undertake several modernisation projects (for example the construction of railroads and the consolidation of some government institutions). Manuel Pardo, the first civilian elected president of Peru, in whose government the census was launched, was the head of a liberal political party. Besides the racial question, the census asked about the education level and the occupations of the population, so it could be used as an instrument to measure the level of “civilisation” attained by the country and their different social and racial groups. That could explain the “sensitivity” of the methodology used to record the racial question in 1876. However the profound economic crisis caused by the end of the “guano era” and the war of 1879 truncated this embryonic modernisation project. Defeat in the war led to a “national identity” crisis, questioning the capability of the Peruvian population, particularly of the indigenous population, to stand up to the challenges of a modern world and even preserve the integrity of the territory of the nation8. Along with this “identity crisis”, the influence of “scientific racism” at the end of the nineteenth century radicalised a racialised conception of the roles and characteristics of the different ethnic groups of Peruvian society (see for example León García, 1909; Palma, 1897; Prado, 1941). It is possible that in this context the first partial censuses of the twentieth century were interpreted in the light of a concern on how much less “Indian” (less backward) and how much whiter and mestizo (more modern) the population had become. In 1940 the situation was quite different. After the reconstruction of the post-war period, during the 1920s the country experienced a new modernisation period, with a modest increase in the working and the

From Pre-Modern “Indians” to Contemporary “Indigenous People”

193

middle classes in the main cities. The celebration of the first centenary of the Republic in 1921 was accompanied by new debates on the issue of national identity and the role of indigenous people in the society. By those years, the “hispanist” movement promoted by some conservative intellectuals like José de la Riva Agüero and José Santos Chocano (Martínez, 1994; Riva Agüero, 1907) vindicated the Spanish inheritance (particularly the language and the Catholic religion) as a central feature of national identity. Almost at the same time, the indigenista movement brought into public debate the importance of the prehispanic9 and Andean origins of the Peruvian nation. The confrontation between “hispanists” and “indigenists” was followed by a reconciliatory discourse in which mestizaje was exalted as the main characteristic of Peruvian society. Writing in the late 1930s and early 1940s, the conservative intellectual and politician Victor Andrés Belaunde elaborated a vision of “Peruvianhood” as a “living synthesis” of Spanish and Andean cultures and races in a superior and harmonic historical reality (Belaúnde, 1987). In the 1940 Census, the use of racial categories reproduced the common sense of the racial distinctions in Peruvian society at that time; however, the census officials acknowledged that they did not want to make a “scientific categorisation of the population based on their racial characteristics” (Ministerio de Hacienda y Comercio, 1944: CLXXVIII). One of the purposes of classifying the population by racial categories and making cross-tabulations with other socioeconomic variables was to contribute to the elimination of certain social prejudices that attribute a “racial” content to issues that are fundamentally socioeconomic, as the socalled “indigenous problem in Peru” (Ministerio de Hacienda y Comercio, 1944: LXXI). As we have said, one of the main findings of the 1940 Census was the absolute and relative increase in the “Mestizo” and “White” categories in the population. This fact was interpreted in the census report as an indicator of the rhythm of [racial] fusion […] revealing a tendency towards the formation of the specific type of national race: the Mestizo, in which racial crossbreeding is synthesised, with a predominance of the ethnic characteristics of Whites and Indians (Ministerio de Hacienda y Comercio, 1944: CLXXX).

The census of 1940 introduced another important ethnic indicator: language, which was used as the main variable to differentiate indigenous from non-indigenous populations in Peruvian society during the second

194

Chapter Seven

half of the twentieth century. The census report stated that the increase in Spanish speakers and literacy levels among the population were “promising signs of a vigorous social transformation” leading towards a better “demographical and cultural integration or mesticisation [sic] of the Peruvian population” (Ministerio de Hacienda y Comercio, 1944: CLXXXIII). The government’s intention to highlight the advances in the “castellanisation” of the indigenous population can be seen in the priority given to the results regarding the “official” language. As John Rowe wrote a few years after the publication of the first census results: It is difficult to look over these tables without feeling that they deliberately present the figures in such a way as to exaggerate the apparent importance of Spanish in Peruvian life (Rowe, 1947: 210).

3. The census of 1961: a cultural anthropological approach to the issue of ethnicity The 1940 Census was the last one in which racial categories were used to classify the Peruvian population. After the Second World War, the country experienced an important period of economic growth and social modernisation (industrialisation, urbanisation, expansion of the education system, etc.). Social modernisation and development theories and discourses emphasised the importance of literacy and education not only as the key elements to modernise society, but also as important tools to achieve national integration. Those were the years during which the social sciences began to develop in Peru, and the 1940s and 1960s witnessed the first modern anthropological studies on indigenous populations, mainly in the Quechua and Aymara10 regions of the country (Degregori, 2009). Those developments in Peruvian social sciences, particularly in cultural anthropology, might have influenced the design of the questions in the 1961 Census. The census report explicitly indicated that the criterion of racial classification was discarded because it was no longer in use in modern social science on account of its “inadequacy”. However, several other elements and variables were designed in order to measure the ethnic dimension of Peruvian society. For instance, an explicit effort was made to estimate the size of native populations in the Amazon region, who were defined as human groups with “sporadic or no contact with civilisation, on the margin of the socioeconomic structure of the country” (Instituto Nacional de Planificación, 1966: II). Secondly, the census questionnaire included questions concerning language for each member of the household: maternal language and

From Pre-Modern “Indians” to Contemporary “Indigenous People”

195

spoken language. Those linguistic markers were considered useful to know the ethnic composition of the population, although the interest in recording those variables served an assimilationist purpose. It was assumed that language was a “more precise indicator […] to measure the degree of linguistic unification” in order to achieve “cultural assimilation, which is fully accomplished when linguistic differences cease to exist” (Instituto Nacional de Planificación, 1966: VI). The main indicators of cultural assimilation were the proportion of the population who spoke Spanish, and the literacy rate. In fact, the 1961 Census showed that the percentage of the population older than five who spoke an indigenous language dropped from 51.6% in 1940 to 38.7% in 1961; in the same period, illiteracy went down from 56.6% of the population in 1940 to 37.2% in 1961. The “indigenous condition” was defined as a “mere social and economic condition, where a person shows cultural backwardness and economic weakness”, and “castellanisation” and better levels of education were indicators that “determine the level of cultural progress” (Instituto Nacional de Planificación, 1966: VI). In other words “castellanisation” and literacy were both indicators of “disindiginisation”. Along with language and literacy, the census questionnaire included a question about the “place of birth” of the members of the household. The census report indicated that this information was a proxy to distinguish some ethnic groups and useful to understand the process of immigration from rural areas to urban centres, although no analyses were made on this last issue. Finally, some data was recorded concerning “regional customs” that might help to identify indigenous people, particularly four items: if the individual “walks barefoot”; “uses ojotas”; “uses poncho or lliclla”; “chews coca leafs”11. In the common sense of the epoch, those items were considered ethnic markers as they constituted “customs of the autochthon indigenous population”. According to the census report, 22.7% of the respondents showed at least one of those “ethnic markers” in 1961. As in previous censuses, the framework to analyse the ethnic dimension in Peruvian society in 1961 continued to consider “indianness” as a remnant of the past, a “backward” condition that could be overcome by the assimilation of indigenous people into modernity through the extension of education and economic development. This modernisation process was no longer framed in terms of racial categories, like “mesticisation” in 1940; instead, more “politically correct” and neutral concepts began to be used, such as language, literacy, educational levels, occupational status, etc.–categories more “appropriate” for a developing country on the road towards a modern capitalist society. In fact, much of

196

Chapter Seven

the social science research in anthropology and sociology during the following years framed discussion of the social issues of a developing society in terms of tensions and conflict between “tradition and modernity”. One consequence of this analytic and cultural turn was the statistical invisibilisation of other ethnic groups, like Afro-peruvians and Asian descendants, who were considered more integrated or assimilated into modern urban Peruvian society than indigenous populations.

4. From the 1970s to the 1990s: the “statistical silencing” of the ethnic dimension of Peruvian society The next twentieth century censuses (1972, 1981 and 1993) silenced the issue of ethnicity and race in official statistics. After 1961 no more questions were included regarding race or the “customs” of ethnic groups. The only variable that could be used to estimate ethnic diversity in Peruvian society was language. Several political and social processes could explain this methodological choice. During the 1960s, the Peruvian population made the transition from a rural society to a predominantly urban one: in 1961 the percentage of people living in urban areas was 47.4%; eleven years later, in 1972, it was 59.4%. At the end of the 1960s the Revolutionary Government of the Armed Forces (1968-1980) undertook an ambitious programme of social and economic reforms. One of the most important was the expropriation of the haciendas by the Agrarian Reform of 1969, which radically undermined the oligarchic regime and enfranchised the indigenous population within the hacienda regime in the Peruvian Andes, modifying the relationship between the state and social classes in the country. A new political discourse, promoted by the government, replaced the expression “indigenous” and “Indians” by the classist categories of “peasants” and “peasantry” in the official language. Those changes were relayed by left-wing political parties (Marxist and Maoist), which became more present in Peruvian politics. From their point of view, the “indigenous problem” in Peru was not a problem or conflict among different ethnic groups, but the product of the oligarchic regime and the semi-feudal nature of the hacienda economic system. This could be solved by economic and social development (capitalist or socialist) or by the political mobilisation of the peasantry conceived as a social class with revolutionary potential. The 1972 Census included a question on “maternal language”, defined as the “language or dialect that a person learned during his/her childhood, regardless of whether he/she currently speaks that language” (Oficina

From Pre-Modern “Indians” to Contemporary “Indigenous People”

197

Nacional de Estadística y Censos, 1974: XVI). According to the census results, 30.5% of the population had an indigenous language as their maternal language, a decrease of over 8% in comparison with the 1961 Census. That variable was not analysed as an ethnic indicator but as another dimension related to the educational level of the population. The census report showed the results in the section “Other cultural and educational characteristics of the population”. A similar pattern occurred with the questions about “place of birth” and “place of residence five years ago”, which were only used to analyse the issue of internal migrations. The 1981 Census followed a similar design, with more precise indicators to analyse the migration phenomenon. The closest question to an ethnic marker was the one concerning the “spoken languages” of the population. However, this question could be considered a step backwards in comparison with previous censuses, since it did not allow recording the “maternal language”. It was not conceived as an ethnic variable but as an indicator of the “plurilingual nature” of Peruvian society (Instituto Nacional de Estadística, 1984: XXXI). The census results showed that 24.8% of the population (aged 5 or over) spoke a native language in 1981. The last census of the twentieth century took place in 1993, just after the most violent years of the internal armed conflict initiated by the Partido Comunista Peruano–Sendero Luminoso (PCP-SL) in 198012. In fact, the census should have been launched in 1991, following the plan to perform a population census every 10 years, but the severe economic crisis13 at the end of the 1980s and the armed conflict prevented the government from undertaking the census in the scheduled year. As in 1972, the specific question asked in the 1993 Census was the “maternal language”, but in this case the results showed that 19.5% of the population over five years of age had an indigenous language as their maternal one. In contrast with 1972, a more precise list of languages was provided in the questionnaire14. All through the twentieth century, census reports show a massive relative decline of native speakers, from 51.6% in 1940 to 19.5% in 1993. Those trends show profound socioeconomic, cultural and political transformations in the structure of Peruvian society. One can read in those figures the impact of the extension of formal education and the urbanisation process in Peruvian society in general, and among indigenous people in particular. Another interpretation of this statistical trend is the progressive “disindigenisation” of the country as it went through its “modernisation process”. From constituting a majority of the population at the beginning of the twentieth century, indigenous people (using language as an indicator of “indianness”) have become a “minority group” in the

198

Chapter Seven

twenty-first century, although a minority of more than 4 million people. Those statistical patterns cannot be explained by demographic factors like fertility or mortality; in fact if we use the linguistic variable, the indigenous population has remained almost constant around 4 million people since 1961. Figure 7-1. Peru 1876–2007: total population (in millions) by census year and estimates of indigenous population

Source: Population censuses. Indicators to estimate indigenous population: in 1876, race categories; from 1940 to 2007, spoken or maternal language.

An important innovation regarding the issue of ethnicity was introduced in 1993 with the First Census of Native Communities in the Peruvian Amazon Region (CNCPAR). The vast Amazon region in Peru accounts for 62% of the national territory but less than 15% of its population. It is home to a large variety of native ethnic groups. The census was undertaken by the National Statistical Office, in collaboration with representatives of indigenous organisations of the Amazon region. The main tool for this census was an ethnolinguistic map that allowed identification of the different linguistic families and ethnic groups within them. According to the census results, 239,674 people lived in native communities belonging to 53 different Amazon ethnic groups (Instituto Nacional de Estadística e Informática, 2010: 17). This represented 1.06%

From Pre-Modern “Indians” to Contemporary “Indigenous People”

199

of the total population of the country in 1993 (22.6 million), and 5.5% of the “indigenous” population, that is, of the population who had a native language as their maternal language (4.4 million). The CNCPAR was one of the first signs of the changes to come in the next decade concerning the statistical study of the ethnic dimension in Peruvian society. In December 1993, the government ratified the International Labour Organization Convention 169, binding subsequent governments to respect and recognise certain rights to indigenous peoples. Those international obligations motivated the statistical offices and academics to undertake an effort to develop a better set of tools to represent and analyse the situation of indigenous people in official statistics. However, this legal and “technical” endeavour has not been accompanied by a large political mobilisation of indigenous people in Peru. As many scholars have pointed out, Peru constitutes a sort of “exception” in indigenous political mobilisation in the context of other Andean countries, especially considering its large indigenous population. Among the explanations of why Peruvian indigenous people have not been able to form an important ethnic political movement in the country, the literature on this issue has mentioned the following factors15: the absence of an indigenous intelligentsia, in part due to the ruthless repression of the indigenous elite by the colonial rule after the rebellion of Tupac Amaru in 1780; the process of social mobility and the subsequent disindigenisation related to the massive immigration from indigenous rural areas to urban centres, and the expansion of formal education and castellanisation during the second half of the twentieth century; the social and political reforms of the Revolutionary Government of the Armed Forces (1968-1980), with a discourse that proscribed the terms “Indians” and “indigenous” and promoted the organisation of peasant unions linked with the government; the influence of Marxist and Maoist left-wing parties in the 1960s and 1970s which contributed to the consolidation of the classist category of “peasantry” over the ethnic category of “indigenous” in the organisation of rural social movements; and the political violence generated by the internal armed conflict initiated by the communist party Sendero Luminoso (Shining Path) during the 1980s which particularly targeted indigenous rural areas and their social leaders. During the last decades of the twentieth century, most of the political claims of indigenous peoples in Peru have not been framed in “ethnical” terms, and very few important ethnic social movements have emerged. “Indigenous identity”, particularly in the Andean regions of the country, has not constituted an important resource for political organisation or

200

Chapter Seven

mobilisation; people in those areas have organised rather in “classist movements” such as peasant organisations or unions, or have been demobilised because of the internal arming of the country. According to the Truth and Reconciliation Commission, most of the violence between 1980 and 1994 took place in rural and indigenous areas, causing an estimate of 70 thousand deaths16, of which roughly 74% were indigenous language speakers. An important percentage of the victims were political authorities and social leaders in those regions (12% of the dead) who had begun to enter electoral politics after the democratic transition of 1980, the year of the beginning of the conflict (Comisión de la Verdad y Reconciliación, 2003). The authoritarian regime of Alberto Fujimori (1992-2000) further contributed to disincentivise party politics and social movements in general. The end of the twentieth century did not present the most favourable political context for the emergence of a broad indigenous movement in Peruvian society.

5. The twenty-first century and the new developments in official ethnic and racial statistics in Peru By the end of the 1990s significant developments were made in the conceptualisation and operationalisation of ethnic variables in official statistics. The main impulse to undertake those innovations came from the new agenda of international cooperation agencies regarding those issues. In that context, the National Statistical Office, the Instituto Nacional de Estadística e Informática (INEI), introduced for the first time in 2000 a self-identification ethnic variable in the ENAHO using the following question: “Considering your ancestors and your customs, do you consider yourself as: a) an indigenous of the Amazon region; b) from Quechua origin; c) from Aymara origin; d) from Black / Mulatto / Zambo origin; e) from Mestizo origin; f) Other?”17 According to the information given by key officials inside the INEI in some recent interviews conducted by Néstor Valdivia, at the beginning of 2000 an internal debate took place inside that office questioning the traditional conceptions and visions of the institution’s technical staff on the issues of ethnicity and race. A key actor in this process was the World Bank, which promoted several meetings in which the necessity to include an ethnic approach into national statistics was discussed. The main event was an international seminar which took place in the Colombian city of Cartagena in 2000, under the name of Todos contamos: Los grupos étnicos en los censos (We all count: Ethnic groups and censuses). The second meeting of this seminar was organised by the INEI in Lima in 2002. That

From Pre-Modern “Indians” to Contemporary “Indigenous People”

201

same year, the anthropologist José Antonio Llorens was hired by the INEI as an important technical advisor on the issues of ethnicity in censuses and surveys. Llorens played a significant role discussing and challenging some widespread “essentialist” conceptions on ethnicity among INEI’s staff. Other factors facilitating the changes inside the INEI were some research projects and surveys promoted by international cooperation agencies. For example, the World Bank financed a study by the INEI on Afrodescendants in Peru (Benavides et al., 2006). The preparation of that survey involved important debates and discussions on how to measure the ethnic identity of this specific group, not only among INEI’s officials but also with representatives of Afro-peruvian organisations who were invited to participate in the research design. In 2006, a research project financed by the International Development Research Centre of Canada (IDRC) and designed by Peruvian academic institutions (particularly the Grupo de Análisis para el Desarrollo [GRADE] and the Universidad Nacional Cayetano Heredia [UPCH]) allowed INEI to introduce a multiple choice question in an ethnic module of the National Demographic and Health Survey (ENDES). The categories of that module were similar to those used by ENAHO in its ethnic selfidentification question since 2000, but it included two new labels: “cholo” and “criollo”18. Also, instead of asking which category the respondent belonged to, it asked “how much” the respondent identified her/himself with each of those ethnic groups (a lot, somewhat, not at all). At the same time, additional questions were asked in order to measure some “external” ethnic markers, such as clothing and the relationship between the respondents and their original communities. Those questions were included in a special module of the 2007 and 2009 ENDES, as part of a joint project between the INEI, the GRADE, and the UPCH. During 2006 and 2007, GRADE and the UPCH had been implementing a research project on racial discrimination and they found that some ethnic markers activated a series of prejudices among some inhabitants of the cities, which lead the latter to display acts of discrimination against indigenous people from the Andean regions of the country, mainly poor peasants. Those markers were associated with indigenous language; indigenous “accent” when those peasants spoke Spanish; and with some traditional clothing of the Andean and rural areas (skirts known as polleras, hats and braids). For that reason, GRADE and UPCH urged the INEI to include some of those markers in the ENDES questionnaire, in order to analyse possible associations between those variables and other social and health variables, and personal experiences of discrimination reported by the survey’s respondents.

202

Chapter Seven

The main methodological innovations concerning the measurement of the ethnic dimension during the first decade of the twenty-first century were limited to the official national social surveys (ENAHO, ENDES). The 2007 Census did not include any of those new developments, continuing to use the maternal language question as the main ethnic indicator. In fact, in the census and household surveys where the linguistic indicator is employed, it is considered an “educational indicator”. The 2004-2009 instruction booklets for the ENAHO stipulate that the question on maternal language has the purpose of “identifying possible unfulfilled needs for bilingual education in primary schools in rural areas”. However, academic research concerning the issues of indigenous groups and social exclusion has used this question as an ethnic indicator to identify indigenous population. The results of the 2007 Census showed that 15.9% of the population aged three or over had an indigenous maternal language, confirming the declining trend of that variable since 1940 (see Table 7-1). Table 7-1. Peru 1876 - 2007: Ethnic and racial indicators used in national censuses and estimations of indigenous population Census

Indicator

Estimated indigenous population (%)

Population base

1876

Racial classification

57.6

Total

1940

1972

Racial classification Spoken language Maternal language Customs and clothing Maternal language

45.9 51.6 38.7 22.7 30.5

Total Total 5 or more years Total Total

1981 1993

Spoken languages Maternal language

24.8 19.5

5 or more years 5 or more years

2006(*)

Maternal language Self-identification

18.3 27.0

12 or more years 12 or more years

2007

Maternal language

15.9

3 or more years

1961

(*) ENCO–Encuesta Nacional Continua Sources: National censuses (Instituto Nacional de Estadística e Informática).

Concerning the data in Table 7-1, as we can see, different bases associated with the lower age limit have been used to estimate the size of the indigenous population, particularly when the language questions were

From Pre-Modern “Indians” to Contemporary “Indigenous People”

203

used to make such estimations. In 1961, 1981 and 1993, the lower age limit was set to five years because the language question was included in the education module of the questionnaire, and five was the lower limit to enter formal education (kindergarten). However in 1972 no age limit was set in the questionnaire (the reasons are still unclear). In the late 1990s, the government made an effort to expand preschool education to include younger children from the age of three, which motivated the corresponding changes in the maternal language question in official surveys. The 2006 Encuesta Nacional Continua–ENCO (National Continuous Survey, see infra) respected this age limit (maternal language was recorded for individuals from age three onwards), but the selfidentification question was put only to respondents aged 12 and above (Instituto Nacional de Estadística e Informática, 2006: 112). In order to compare the figures of those different questions of the ENCO in Table 7-1, we have set the lower age limit in both cases to 12. In official surveys using the self-identification question, the estimations of indigenous population are much higher. For example, the ENCO of 2006 recorded that 27% of the population aged 12 or more identified with one of the “indigenous categories” (Quechua, Aymara or from the Amazon). Furthermore, unlike maternal language, selfidentification does not show a declining trend through time. Using the data of the 2009 ENAHO survey concerning the heads of households (aged 18 or more), we can appreciate that while the percentage of people with an indigenous maternal language decreases in the younger cohorts, the percentage of people who identified themselves with an indigenous category in the self-identification question remains quite constant around 35–37%, indicating that those two variables refer to different social phenomena. Even if younger people no longer speak indigenous languages, they still self-recognise some sort of “indigenous” background when they are asked about their “ancestors and customs”. The remaining issue is how “strong” or “weak” this kind of “indigenous identity” is, or if it can become a source of a collective identity as the basis of an “indigenous people” in Peruvian society.

204

Chapter Seven

Figure 7-2. Peru ENAHO 2009: % of head of households identified as indigenous, by type of indicator (self -identification or maternal language), and age groups

Source: Sulmont, 2010.

Despite the innovations introduced in the measurement of ethnic identities in official statistics (particularly the inclusion of the selfidentification question in official surveys), this process had several weaknesses. First of all, it did not correspond to an institutional effort of systematic methodological research to include ethnic indicators in official statistics in an orderly manner. That would have allowed a better integration of the different units inside the INEI, and a more fluid and permanent relationship with the academic institutions (universities, social research centres) and the international cooperation agencies. This would also have prevented what finally happened: the coexistence of different conceptions inside the same institution, managed by different units without uniform criteria to operationalise the ethnic variables in the several official surveys and statistical reports. Another weakness was the lack of proper government funding and expert technical advice to validate the new methodologies aiming at the inclusion of self-identification as an important dimension to measure and analyse ethnicity in official statistics. It has to be said that this kind of weakness (lack of resources) is shared by many other government agencies in charge of social policy and social programmes in Peru.

From Pre-Modern “Indians” to Contemporary “Indigenous People”

205

As Néstor Valdivia points out (2011), the conceptual and methodological limitations of the ethnic self-identification question in official social surveys led to several problems that need to be addressed in a more systematic way. The first is the absence of an explicit conceptual framework that justifies the methodological choice to use a selfidentification question and the specific categories employed in that question. A second problem is the lack of a process validation for the actual phrasing of the question and its categories, which is a sensitive issue with important repercussions in the kind of result that they produce (Burton et al., 2010; Spencer, 2006). For instance, the inclusion of the word “origin” in the question can change the answer of the respondent who does not identify himself as a “Quechua”, but who recognises that he has “Quechua” ancestors; this issue has been explored in a qualitative study on ethno-racial identities among people of Andean origins in Lima and Cusco (Planas and Valdivia, 2007). The use of expressions such as “Quechua”, “Aymara” of “from the Amazon”, is in fact a methodological strategy to circumvent the problems around the use of the “indigenous” label in quantitative social research in Peru (Sulmont, 2010). As many scholars have pointed out, the terms “indigenous” and “Indians” are still marked with a social stigma, since they continue to refer in many cases to “backwardness”, “lack of education” and “poverty” that are associated with the “indigenous condition” (Cadena, 2004; Callirgos, 1993; Degregori, 1998; Paredes, 2008; Planas and Valdivia, 2007; Sulmont, 2005 and 2011). When an ethnic self-identification question includes the label “indigenous” among its categories, very few people in Peru identify themselves with such an “identity”. For example, the Latin American Public Opinion Project (LAPOP) 19 survey of 2010 used the following question: “Do you consider yourself as White, Mestizo, Indigenous, Black, Mulatto, Moreno, Other?” Only 3.3% of respondents interviewed in Peru chose the “Indigenous” category, in comparison of 15.8% of the Bolivian sample. As we have seen, in the 2006 ENCO 27% of the respondents chose the “Quechua”, “Aymara” or “from the Amazon” categories. However the categorisation of those respondents in an aggregated “indigenous” category is a researcher’s decision, and as Lavaud and Lestage point out, it entails the risk that ethnic measurements using those kind of survey questions will help to reproduce social preconceptions of what is “an Indian” or “an indigenous person”, transforming a supposedly “scientific” statistical estimation into an ideological and political one (Lavaud and Lestage, 2009: 66).

206

Chapter Seven

Another problem with the self-identification criterion is the lack of conceptual or methodological justification for the changes experienced in the phrasing of questions. For example, in 2001 the question introduced the category “White or Caucasian”, which was later replaced by “White”. Another example is the change of the category “indigenous of the Amazon region” to “from the Amazon region”, which led “mestizo” people living in large cities in the Amazon to choose that category, as well as indigenous people living in native communities. Those modifications from one survey to another were not properly justified at any time. In some cases, more systematic efforts were made to address those problems. In 2007, the INEI conducted a set of “meaning tests” as part of the activities in preparation for the national census of that year. The results of those tests showed the lack of resonance of the concept of “indigenous people” in Peru, in contrast with other Latin American countries (such as Bolivia, for instance). Also, the tests brought to light that the words “ethnic” or “ethnicity” were quite unfamiliar to the respondents, in comparison with other more common-sense terms such as “race”. In fact, other research projects had already pointed out this issue (see for example: Planas and Valdivia, 2007; Sulmont, 2005 and 2011). Those limitations and problems might explain why, despite the innovations made in the official household surveys (ENAHO, ENDES, ENCO), the INEI decided not to include a self-identification question in the 2007 Census, even after the government had issued a bill (Ley Nº 27778) in 2002 that explicitly stipulated the obligation to include the ethnic dimension in the design of the national censuses. It is possible that, inside the statistical office, the technical staff were not convinced of the necessity or adequacy of the self-identification question used in previous studies. However, another important factor for that decision might have been the lack of social and political pressure from ethnic organisations to make this kind of decision. An important issue on this matter is the political context affecting the implementations and continuity of some policies inside the INEI. As the National Statistical Office, the INEI is in charge of producing important indicators for social and economic policies. Social surveys are used to measure the evolution of the poverty rate and the living standards of the population. Also, economic surveys performed by this institution are the main source for measuring some economic development indicators (economic growth, unemployment, income distribution, inflation rates, etc.). Those indicators are politically sensitive since they can present either a “good” or “bad” image of the government’s performance in those areas (especially in a context of rapid economic growth), and become subject to

From Pre-Modern “Indians” to Contemporary “Indigenous People”

207

questioning by opposition political parties or groups. The deficits of political institutionalisation and the lack of confidence in the main political institutions (political parties, parliament, judiciary, government officials) in a country like Peru tend to polarise political debate and project a shadow of suspicion of political manipulation of official statistics. During the administration of President Alejandro Toledo (2001-2006), important methodological innovations were introduced in the design and management of the population census due in 2005. A new kind of methodology was used, inspired by the model of “continuous censuses”. Until 1993, the census interviews were conducted in a single day for most of the country (people were ordered to remain in their houses until the census enumerator visited them). That required a huge number of census enumerators (recruited among university students, senior high school students and “volunteers”) with the corresponding challenges and problems concerning their training. In 2005, the census interviews and visits to homes were made over a longer period of time (two months), with fewer and supposedly bettertrained staff. Furthermore, the census questionnaire was simpler and included fewer variables than in 1993, its main purpose being to estimate the size and distribution of the population and update the sampling frame for the national household surveys. The variables not included in the 2005 Census (like ethnic self-identification, for example) were supposed to be measured in ENCO following the model of the ENAHO but with much larger samples, allowing better statistical representation for smaller geographical areas in the country. The first and only ENCO was effectively carried out in 2006, but the new administration of President Alan García (2006-2011) discarded those changes and the model of “continuous censuses” was abandoned. The 2005 Census results and methodology were widely criticised by the new administration officials and many scholars of the academic community. Apparently, the innovations introduced in 2005 were considered too “radical” and the measured variables insufficient to portray an accurate image of the country’s population characteristics. The 2005 Census was supposed to be the first census of the twenty-first century and comparison with the 1993 results could have shown the social transformation of the country after 15 years of neoliberal economic reforms (something that could also be done using the national household surveys). Besides, inside the INEI some “underground” opposition to the new methodology opened many flanks to questioning of the results. Facing those criticisms, in 2007 the government decided to launch a new national census (just two years after the previous one) using a similar

208

Chapter Seven

questionnaire and methodology to those of 1993, which did not include the ethnic self-identification question. Despite the fact that the last census questionnaire did not include other ethnic indicators beyond maternal language, in other official national surveys (ENAHO, ENDES and ENCO) the INEI has indeed made important improvements to measure the ethnic dimension in Peruvian society, like the inclusion of the self-identification question, and some other questions related to the respondent’s ethnic or racial discrimination experiences. This valuable information has been used in several studies to analyse the issues of exclusion and social discrimination of indigenous people and Afrodescendants in Peru. A recent review on this field has identified more than 30 recent academic research projects (in universities and social research centres) that have used the data on ethnicity reported by those official surveys. The availability of those survey’s micro data on the website of the INEI (www.inei.gob.pe) has allowed many scholars and students to use this information for their own purposes.

6. Current challenges and issues for an ethnic portrayal of the Peruvian population in official statistics As we have seen in this chapter, the statistical representation of the ethnic diversity in Peruvian society has undergone several phases and approaches since the beginning of the Republic. Those phases are reflected in the different questions used in the censuses (see infra, Table 7-2). From race to language, customs and self-identification (the latter dimension used mostly in surveys); the methodological strategies used to measure the ethnic dimension in the country have reflected the debates and conceptions of their times, and also the different challenges faced by the nationbuilding process in Peru: Independence; state building; war and reconstruction; social and economic modernisation and development; “national integration”; the persistence of inequalities and the issues of a multicultural society. A constant factor in those processes has been the “indigenous issue”: how to deal with the status and integration of indigenous peoples into a “modern” society and into full citizenship. The different statistical representations of the indigenous people have also been subjected to the preconceptions of political elites, to lack of definition and to debates about what constitutes an “indigenous people”, especially when modernisation allows people to acquire new “resources” to build their own identities, or else forces them to relinquish some other identity markers (like language). A constant factor has been the persistence of social, economic and political inequalities between indigenous and non-

From Pre-Modern “Indians” to Contemporary “Indigenous People”

209

indigenous populations (and between these and other minority groups like Afrodescendants). Before the current wave of national censuses in Latin America, the production of official statistics that might reflect the ethnic diversity of the country seems to have suffered some “delay” in including the various dimensions of the contemporary concepts of ethnicity and indigenous people (most notably self-identification), especially in the last census of 2007. To a certain degree, this delay is the consequence of the way ethnic identities have been defined in Peruvian society, and of the absence of social groups assuming and actively demanding “ownership” of some ethnic, racial or cultural identities. This issue does not concern only indigenous people, but also other ethnic groups like Afrodescendants. It has to be noted that, as in other Latin American countries, the state has historically paid more attention to indigenous people than to other ethnic groups (Hooker, 2005; Valdivia et al., 2007; Wade, 2000) and, since the elimination of the “race question” in national censuses after 1940, the statistical representation of the Afro-peruvian population–estimated around 5% of the total population (Benavides et al., 2006)–has been circumscribed to official surveys like the ENAHO, which after 2000 began to use the self-identification question. Until the 1940s, the Peruvian state conceived the “Indian” or “indigenous” person as a social category identifiable by his “race”. This form of categorisation went into crisis during the 1960s, but it was not replaced by an alternative. The socioeconomic and political changes associated with the end of the oligarchic regime, along with the populist and reformist military government of the 1970s, facilitated the emergence of new urban and popular social actors and identities: the cholos or urban “indigenous-mestizos” (Cadena, 2004), reluctant to be categorised as “indigenous” and seeking better chances of social, cultural and geographical mobility outside the framework of “ethnic” mobilisation or “ethnic” political demands. Table 7-2. Peru 1876-2007: Dimensions and questions used in censuses and surveys to measure ethnicity and race Censuses and Surveys 1876 National Census

Ethnicity as Language Not used

Ethnicity as Culture Not used

Race Racial categorisation by direct observation of the census enumerator. Categories: White, Indian, Mestizo, Black, Asian.

210

Chapter Seven

1920 Partial Census of Lima and Callao provinces

Not used

Not used

1940 National Census

Question 10. Language: Do you speak Spanish? (Yes / No)

Not used

1961 National Census

1972 National Census

1981 National Census

Question 11. Other languages: ¿Which indigenous or foreign languages do you speak? (write down all the answers) Question 10. (Only for people from 5 years old) Maternal language (write down the answer) Question 11. (Only for people from 5 years old) Do you speak Spanish? (Yes / No) Question 12. ¿Which is your maternal language learned during your childhood? (Write down all mentioned) Question 13. (Only for people from 4 years old) Do you speak Spanish? Question 11. (Only for people from 5 years old). Do you speak Spanish, Quechua, Aymara, another aboriginal language or foreign language? (Mark all that correspond)

Racial categorisation by direct observation of the census enumerator. Categories: White, Indian, Mestizo, Black, Asian. 13% by selfidentification; 87% by direct observation of the census enumerator. Categories: White, Indian, Black, Yellow, Mestizo.

Identification of regional customs by observation: Walks barefoot; uses ojotas; uses poncho or lliclla; chews coca leafs. (Mark all that correspond)

Not used

Not used

Not used

Not used

Not used

From Pre-Modern “Indians” to Contemporary “Indigenous People” 1993 National Census

2007 National Census

2011 ENAHO Survey

Question 8. (Only for people from 5 years old). Your maternal language learned during your childhood is: Quechua, Aymara, another native language, Spanish, a foreign language? (Mark only one) Question 9. (Only for people from 3 years old). Your maternal language, the one you learned to speak with is: Quechua, Aymara, Ashaninka, another native language (specify), Spanish, a foreign language? (Mark only one) Question 300A. (Only for people from 3 years old). Your maternal language, learned during your childhood is: Quechua, Aymara, Ashaninka, another native language (specify), Spanish, English, Portuguese, another foreign language (specify), deaf-mute? (Mark only one)

Not used

Not used

Not used

Not used

Questions 46 – 47 (Only for the household head and his/her spouse). From your ancestors and according to your customs, do you consider yourself from: Quechua origin, Aymara origin, the Amazon origin, Black/Mulatto/Zambo origin, White origin, Mestizo origin, another origin (specify)? (Mark only one)

Not used

211

Sources: For 1876 and 1920 censuses, see: Dirección de Estadística–Ministerio de Gobierno (1878); Gootemberg (1995). The questionnaires from 1940 thru 2007 censuses are reproduced in Valdivia (2011: 214-40). The ENAHO 2011 questionnaire can be downloaded from INEI’s webpage, Instituto Nacional de Estadística e Informática (2011).

The denial of “indianness” could have influenced the formation of contemporary ethnic identities during the last decades of the twentieth century. In the political and cultural discourses of the 1970s and 1980s, national integration was conceived as a process of social recognition of “Peruvian citizenship”, leaving aside or minimising ethnic or cultural

212

Chapter Seven

differences. Those kinds of differences were seen as “particular cultural traits” inside a wider (mestizo or “cholo”) “popular sector”. In this context, neither society nor the state seemed to have the intention (or the need) to categorise or identify indigenous people and Afrodescents as particular “ethnic groups”. Even inside those groups, there was not an identifiable pressure or a political demand to be recognised in such way. The prevalence of a discourse of national integration privileging the centrality of a “mestizo national identity” might have prevented the emergence of such kind of identity politics. Others factors, like the internal armed conflict of the 1980s and 1990s, which particularly affected densely indigenous populated regions, fed into this process by discouraging political organisation and mobilisation of potential indigenous leaders. This situation underwent important changes at the end of the 1990s. International cooperation agencies, which began to play an increasingly important role in the design of social policies in the context of neoliberal economic reforms, introduced into the political agenda the issues of poverty and social exclusion in relation to the ethnic condition of the poorest and most excluded groups in the country. However, this new emphasis on the “ethnic dimension” of social exclusion was not yet correlated with a strong civil society mobilisation or an organisation of ethnic groups in the country. With the exception of a few functionaries at the INEI and some scholars interested in those issues, until a few years ago there were not many political and social actors interested in being “ethnically represented” and visibilised in official statistics. Changes accelerated during the government of Alejandro Toledo (2001-2006) and indigenous and other ethnic organisations became more active in the debates, although weak representativeness and lack of institutionalisation continued to be important obstacles. There are some indications that the contemporary dynamics of Peruvian indigenous organisation responds more to incentives from abroad than to a national indigenous social movement. A recent paper by the sociologist Omar Cavero argues that the organisations that are trying to represent indigenous people in the Andean regions of the country lack a solid base at a local level and that, instead, they are sustained by transnational networks promoted by foreign NGOs or international cooperation agencies (Cavero, 2011). As we have said, during the administration of President Alan García (2006-2011) the national census 2007 reproduced the methodology used in previous censuses to measure the ethnic dimension of Peruvian society, based on the maternal language variable. However, other official surveys continued to implement the self-identification question and several

From Pre-Modern “Indians” to Contemporary “Indigenous People”

213

academic research projects have looked into this dimension in more detail. At a political level, during this period social conflicts involving claims concerning the impact (particularly environmental impacts)20 of private economic development projects in rural areas and indigenous lands in the Andes and in the Amazon region have brought into the public debate the issue of indigenous rights and their relationship with the state, the economic development model and the situation of social exclusion of the majority of the indigenous population in the country. Some of those conflicts have had highly dramatic consequences, as in the case of the regional struggle organised in 2009 by Amazon indigenous organisations against some government decrees aiming at facilitating private investment in Amazonian indigenous territories. The government’s repression led to a violent clash in June 2009 between police and protesters near the Amazon city of Bagua, resulting in the death of 24 policemen and 10 civilians. This violent event caused the resignation of the Prime Minister and his cabinet and the withdrawal of the government’s decrees. The issue of “social inclusion” was one of the most recurrent and important ones in the 2011 national elections. After a decade of high economic growth rates, and despite the reduction in the poverty rate in the country, the increase in social conflicts and the persistence of profound social and regional inequalities have marked the electoral debate. The new administration of President Ollanta Humala has stressed in its political discourse its intention to vigorously address those issues in its public policies. This new political context might offer an opportunity to include in this debate the situation and the agenda of the different ethnic groups in Peruvian society. Such an agenda could include the following dimensions: • The issue of social exclusion and inequality between different ethnic groups (access to and quality of education, health or other public services; inequality in the labour market, etc.). • Political rights (individual and collective), including political representation (electoral quotas) and the right of consultation when collective rights are at stake (especially concerning indigenous lands). • Racial and ethnic discrimination in public and private services, the labour market and the mass media. • The government’s cultural policies, the status of native languages in society and in the educational system (bilingual education) and the recognition of traditional practices in some areas (for example in health care, child education, etc.). In August 2011, one of the first measures of the Humala administration was to approve the new Ley de consulta previa a los pueblos indígenas u

214

Chapter Seven

originarios (Law on prior consultation with indigenous peoples) when economic development projects might affect their territories (Normas Legales, 2011). The implementation of this law has launched a new debate about who should be considered an “indigenous people” and how the state should identify it in public records or official statistics. Since this law entails some political rights, it is possible that its enforcement might create institutional incentives for indigenous people to organise themselves or to “become an indigenous people” in order to participate in consultation processes with the government. On this matter, the government is now facing important pressures and critiques from indigenous organisations which demand more participation in the debate and a wider definition of “indigenous people” for the implementation of this law. This new social and political context might also open a new phase in the discussion of the statistical representation of the ethnic dimension for the next wave of official surveys and censuses in the country. However, in order to make significant improvements, some challenges should be addressed. For instance, the INEI has to define a clearer strategy to include theoretical and methodological innovations in the design of ethnic indicators, enhancing its relationship with academic and social research institutions, reinforcing common definitions and analytical approaches inside the institution. Better funding and proper technical advice are necessary conditions for that purpose. Since the inclusion of a specific approach to ethnic statistics is a politically sensitive issue, it is necessary to undertake an open national debate, allowing the different stakeholders, particularly ethnic organisations, to participate in the discussion. This also involves the adoption of measures to disseminate and promote the use and discussion of the statistical data concerning ethnic indicators, not only within governmental, academic or scientific institutions, but also (and perhaps more importantly) among civil society organisations and the mass media. A specific agenda in the preparation of the next national census (due in 2017) should include those issues. The participation of grassroots organisations of indigenous people and other ethnic groups (like Afroperuvians) might play an important role in the promotion of debates on the ways of identifying ethnicity in Peruvian society. However it is important to bear in mind that, as history has shown, ethnic identification and categorisation is and will continue to be a contingent phenomenon, depending on political, cultural and social changes.

From Pre-Modern “Indians” to Contemporary “Indigenous People”

215

Bibliographical references Belaúnde, Víctor Andrés. 1987 [1942]. Peruanidad. Lima: Comisión del Centenario de V. A. Belaúnde. Benavides, Martín, Máximo Torero and Néstor Valdivia. 2006. Más allá de los promedios: Afrodescendientes en América Latina. Pobreza, discriminación social e identidad: el caso de la población afrodescendiente en el Perú. Washington DC: GRADE, Banco Mundial. Brubaker, Rogers. 2004. Ethnicity without Groups. Harvard: Harvard University Press. Burton, Jonathan, Alita Nandi and Lucinda Platt. 2010. Measuring Ethnicity: Challenges and Opportunities for Survey Research. Ethnic and Racial Studies, 33(8): 1332-49. Cadena, Marisol de la. 2004. Indigenas mestizos: raza y cultura en el Cusco. Lima: Instituto de Estudios Peruanos. Callirgos, Juan Carlos. 1993. El racismo: la cuestión del otro (y de uno). Lima: DESCO. Cavero, Omar. 2011. Movimiento indígena en el Perú: ¿transnacional antes que local? In Diana Flores (ed.) Nuevas miradas al Perú contemporáneo: movimientos sociales, identidades y memoria. Lima: Programa y Transformación Global. Centro de Estudios de Población y Desarrollo. 1972. Informe demográfico. Perú 1970. Lima: CEPD. Comisión de la Verdad y Reconciliación. 2003. Informe final. Lima: CVR. Defensoría del Pueblo. 2011. Reporte de conflictos sociales No. 93. Noviembre 2011. Adjuntía para la Prevención de Conflictos Sociales y la Gobernabilidad. Lima: Defensoría del Pueblo. http://www.defensoria.gob.pe/conflictos-sociales/home.php. (Accessed: 21/12/2011). Degregori, Carlos Iván. 1993. Identidad étnica, movimientos sociales y participación política en el Perú. In Democracia, etnicidad y violencia política en los países andinos. Lima: IEP - IFEA. Degregori, Carlos Iván. 1998. Movimientos étnicos, democracia y nación en Perú y Bolivia. In Claudia Dary (ed.) La construcción de la nación y la representación ciudadana en México, Guatemala, Perú, Ecuador y Bolivia. Ciudad de Guatemala: FLACSO. Degregori, Carlos Iván. 1999. Estado Nacional e identidades étnicas en Perú y Bolivia. In Konings Kees and Patricio Silva (eds) Construcciones étnicas y dinámica sociocultural en América Latina. Quito: Ediciones Abya Yala.

216

Chapter Seven

Degregori, Carlos Iván (ed.). 2009. No hay país más diverso. Compendio de antropología peruana. Lima: Red para el Desarrollo de las Ciencias Sociales en el Perú. Dirección de Estadística - Ministerio de Gobierno. 1878. Censo general de la República del Perú formado en 1876. Lima: Imprenta del teatro. Fukumoto Sato, Nancy Mary. 1997. Hacia un nuevo sol: japoneses y sus descendientes en el Perú. Historia, cultura e identidad. Lima: Asociación Peruano Japonesa del Perú. Gaceta del Gobierno de Lima Independiente. 1950. Decreto del 27 de agosto de 1821. La Plata: Universidad Nacional de la Plata. Gootemberg, Paul. 1995. Población y etnicidad en el Perú republicano (siglo XIX): algunas revisiones. Lima: Instituto de Estudios Peruanos. Hooker, Juliet. 2005. Indigenous Inclusion / Black Exclusion: Race, Ethnicity and Multicultural Citizenship in Latin America. Journal of Latin American Studies, 37(2): 285-310. Instituto Nacional de Estadística. 1984. Censos Nacionales VIII de Población y III de Vivienda 1981. Lima: Instituto Nacional de Estadística. Instituto Nacional de Estadística e Informática. 2006. Manual del encuestador: Encuesta Nacional Continua 2006. Lima: INEI. Instituto Nacional de Estadística e Informática. 2010. Perú: análisis etnosociodemográfico de las comunidades nativas de la Amazonía, 1993 y 2007. Lima: INEI. Instituto Nacional de Estadística e Informática. 2011. Encuesta Nacional de Hogares 2011. Encuesta de opinión. Módulo: gobernabilidad, democracia y transparencia. Lima: INEI. http://www.inei.gob.pe/srienaho/descarga/DBF/275-Modulo85.zip (Accessed: 2/2/2012). Instituto Nacional de Planificación. 1966. VI Censo nacional de población - 1961. Lima: Instituto Nacional de Planificación - Dirección Nacional de Estadística y Censos. Larson, Brooke. 2002. Indígenas, élites y estado en la formación de las Repúblicas Andinas. Lima: IEP, PUCP. Lavaud, Jean-Pierre and Françoise Lestage. 2009. Contar a los indígenas (Bolivia, México, Estados Unidos). In Valerie Robin Azevedo and Cármen Salazar Soler (eds) El regreso de lo indígena: retos, problemas y perspectivas. Lima: IFEA - CBC. León García, Enrique. 1909. Las razas en Lima: estudio demográfico. Lima: Universidad Mayor de San Marcos, Facultad de Medicina.

From Pre-Modern “Indians” to Contemporary “Indigenous People”

217

Manrique, Nelson. 1999. Algunas reflexiones sobre el colonialismo, el racismo y la cuestión nacional. In Nelson Manrique (ed.) La piel y la pluma. Lima: Casa Sur. Martínez, Ascención. 1994. El Perú y España durante el Oncenio: el hispanismo en el discurso oficial y en las manifestaciones simbólicas (1919-1930). Histórica, XVIII (2): 355-63. Méndez, Cecilia. 1993. Incas sí, Indios no: apuntes para el estudio del nacionalismo criollo en el Perú. Lima: Instituto de Estudios Peruanos. Ministerio de Hacienda y Comercio, Dirección Nacional de Estadística. 1944. Censo Nacional de Población y Ocupación 1940. Vol. I. Lima: Ministerio de Hacienda y Comercio. Normas Legales. 2011. Ley N° 29785, Ley del Derecho a la Consulta Previa a los Pueblos Indígenas u originarios reconocidos en el Convenio Nº 169 de la Organización Internacional del Trabajo (OIT). El Peruano, Sept. 7. Oficina Nacional de Estadística y Censos. 1974. Censos Nacionales VII de Población y II de Vivienda 1972. Lima: Oficina Nacional de Estadística y Censos. Pajuelo Teves, Ramón. 2004. Identidades en movimiento. Tiempos de globalización, procesos sociopolíticos y movimiento indígena en los países centro andinos. Caracas: CIPOST, FaCES, Universidad Central de Venezuela. Palma, Clemente. 1897. El porvenir de las razas en el Perú. Lima: Imprenta Torres Aguirre. Paredes, Maritza. 2008. Weak Indigenous Politics in Peru. CRISE Working Paper 33. Oxford: University of Oxford. Planas, María Elena and Néstor Valdivia. 2007. Identidad étnica en el Perú: un estudio cualitativo sobre los discursos de autoidentificación en tres zonas del país. Lima: GRADE/UPCH. Prado, Javier. 1941. [1894]. Estado social del Perú durante la dominación española (estudio histórico - sociológico). Lima: Librería e Imprenta Gil. Quijano, Aníbal. 1980. Dominación y cultura: lo cholo y el conflicto cultural en el Perú. Lima: Mosca Azul. Riva Agüero, José de la. 1907. Prólogo a Problemas ético sociológicos del Perú. El Comercio, Nov. 3. Rodriguez Pastor, Humberto. 1989. Hijos del Celeste Imperio (18501900): migración, agricultura, mentalidad y explotación. Lima: Instituto de Apoyo Agrario. Rowe, John. 1947. The Distribution of Indians and Indian Languages in Peru. Geographical Review 37(2): 202-215.

218

Chapter Seven

Rowe, John. 1954. El movimiento nacional Inca del siglo XVIII. Revista Universitaria, 107: 17 - 47. Sánchez, Rodrigo. 1994. Procesos culturales e identidad étnica en Ecuador y Perú. Allpanchis, XXVI(43/44): 321-378. Sánchez, Rodrigo. 1996. Evolución agraria y protagonismo indígena: los casos de Perú y Ecuador. Revista de Sociología, 9(10): 87-117. Schkolnik, Susana. 2009. La inclusión del enfoque étnico en los censos de población de América Latina. Notas de Población - CEPAL (89): 57100. Spencer, Stephen. 2006. Race and Ethnicity. Culture, Identity and Representation. New York: Routledge. Sulmont, David. 2005. Encuesta nacional sobre exclusión y discriminación social. Informe de investigación para DEMUS Estudio para la defensa de los derechos de la mujer. Lima: Demus. http://www.manuelaenelcongreso.org/files/Encuesta_discriminacion.p df (Retrieved: February 2, 2012). Sulmont, David. 2010. Raza y etnicidad desde las encuestas sociales y de opinión: dime cuántos quieres encontrar y te diré qué preguntar. Paper presented at Taller organizado por el Centro de Investigación de la Universidad del Pacífico: La discriminación social en el Perú, 24 June. Lima. Sulmont, David. 2011. Race, Ethnicity and Politics in Three Peruvian Localities: An Analysis of the 2005 CRISE Perceptions Survey in Peru. Latin American and Caribbean Ethnic Studies 6(1): 47-78. Thorp, Rosemary and Maritza Paredes. 2010. Ethnicity and the Persistence of Inequality: The Case of Peru. Basingstoke, Hampshire: Palgrave Macmillan. Thurner, Mark. 1997. From Two Republics to One Divided: Contradictions of Postcolonial Nationmaking in Andean Peru. Durham: Duke University Press. Valdivia, Néstor, Martín Benavides and Máximo Torero. 2007. Exclusión, identidad étnica y políticas de inclusión social en el Perú: el caso de la población indígena y la población afrodescendiente. In Investigación, políticas y desarrollo en el Perú, pp. 603 - 655. Lima: GRADE Valdivia, Néstor. 2011. El uso de categorías étnico/raciales en censos y encuestas en el Perú: balance y aportes para una discusión. Lima: GRADE. http://www.grade.org.pe/upload/publicaciones/archivo/download/pubs/ ddt604.pdf (Retrieved: February 2, 2012) Wade, Peter. 2000. Raza y etnicidad en Latinoamérica. Quito: Ediciones Abya Yala.

From Pre-Modern “Indians” to Contemporary “Indigenous People”

219

Walker, Charles. 1988. The Patriotic Society: Discussions and Omissions about Indians in the Peruvian War of Independence. Americas, 55: 275 - 298. Yashar, Deborah J. 2005. Contesting Citizenship in Latin America: The Rise of Indigenous Movements and the Postliberal Challenge. Cambridge: Cambridge University Press.

Notes 1 Many of the ideas and arguments presented in this chapter come from recent research projects conducted by the authors, particularly a study by Néstor Valdivia on racial and ethnic categories in Peruvian censuses (Valdivia, 2011), and the work by David Sulmont and Juan Carlos Callirgos in the Peruvian chapter of the Project on Race and Ethnicity in Latin America–PERLA (forthcoming publication). PERLA is directed by Edward Telles at the University of Princeton. For more information see: http://www.grade.org.pe and http://perla.princeton.edu/. 2 For a discussion of some of those studies see: Sulmont, 2010; Sulmont, 2011; Valdivia, 2011. 3 Cholo is a term used to denote an indigenous person with urban experience and some degree of education (Paredes, 2008: 8). Anibal Quijano (1980) considered the cholos a transitional social group, constituted by indigenous immigrants from rural areas who settle in urban centres during the twentieth century. 4 The ENAHO is a national monthly household survey; it includes both urban and rural areas and it is oriented towards gathering information on living conditions and measuring poverty levels among the national population. The ENDES is a specialised national survey on demographic and health variables; it has been carried out since 2003 and the target population is constituted by women from 15 to 49 years old, and children under 6 years old. 5 In the second half of the nineteenth century many Chinese coolies were “imported” from China to work in the haciendas of the coast (sugar and cotton) and in the exploitation of guano in the islands of the littoral. For a history of Chinese immigration in Peru, see Rodriguez Pastor (1989). 6 The only difference in the racial categories used in the 1876 and 1940 censuses concerns the “Asian” label. In 1876 the label for this groups was “Asians” (asiáticos), and in 1940 it was “Yellow” (amarillos), referring to the supposedly “characteristic” skin colour of Chinese and Japanese people. In the nineteenth century, almost all “Asians” in Peru came from China and the first groups of Japanese people arrived in Peru in 1899. For a history of Japanese immigration in Peru, see Fukumoto Sato (1997). 7 The first steps to abolish slavery in Peru began with independence, when the newborn of the slaves were freed. Full enfranchisement from slavery was only attained in 1854.

220 8

Chapter Seven

One of the most important and traumatic consequences of defeat in the war was the loss of the southern provinces of the country (Tacna and Tarapaca), which now constitute northern Chile. 9 New archaeological discoveries, particularly the work of Julio César Tello during the first three decades of the twentieth century, proved the antiquity of the prehispanic cultures, showing that high civilisations had existed in the Peruvian territory since at least 1500 BC. 10 Quechua is the most widespread native language in the Peruvian Andes. Aymara is the language of native populations in the Andean southern regions surrounding Lake Titicaca and the Altiplano (high plateau) between Peru and Bolivia. 11 Ojotas are a sort of rudimentary sandals commonly used by peasants in rural areas in Peruvian Andes. Poncho (for men) and lliclla (for women) are considered “typical” clothing of indigenous people in the Andes. 12 The PCP- Sendero Luminoso, known in English as the “Shining Path”, was a Communist party of Maoist inspiration, the most radical of the many Marxist Peruvian left wing parties that existed between the 1970s and 1980s. 13 The inflation rate in 1990 reached more that 7000% and per capita GDP fell to the level of 1975. 14 In 1972 the question on maternal language was a multiple open question and the enumerator had to write down all the languages mentioned by the interviewer for each of the members of the household. In 1993, the census questionnaire used a semi-closed question and the enumerator had to indicate only one language per person in the household (aged 5 or over) using the following categories: a) Quechua, b) Aymara, c) Other “native” language, e) Spanish, f) A foreign language. 15 See for example: Degregori, 1993, 1998 and 1999; Pajuelo Teves, 2004; Sánchez, 1994 and 1996; Sulmont, 2011; Thorp and Paredes, 2010; Yashar, 2005. 16 The rebel groups, mainly PCP-SL, caused more than half of the victims. Peru is a sui generis case in internal armed conflicts, since the rebel groups, rather than the government security forces, were responsible for a majority of the victims (Comisión de la Verdad y Reconciliación, 2003). 17 The Spanish wording of the question was: ¿Por sus antepasados y de acuerdo con sus costumbres, usted se considera: a) Indígena de la Amazonía; b) de origen Quechua; c) de origen Aymara; d) de origen Negro / Mulato / Zambo; e) de origen Mestizo; f) Otro? 18 Criollo was the term used during colonial times to designate Spanish descendants in the New World. It now designates “plebeian” white inhabitants of cities in the coast. 19 LAPOP is a project of the University of Vanderbilt, which conducts comparative public opinion surveys across all Latin American countries. The respondents are at least 18 years old, drawn from a probabilistic sample representative of each country. The average sample size is 1500 people. The project is directed by Michael Seligson. For more information see: http://www.vanderbilt.edu/lapop/. 20 As an example, we can mention that in the November 2011 Ombudsman Office report on social conflicts in the country, from 220 identified conflicts, 57% were related to socio environmental issues (Defensoría del Pueblo, 2011).

CHAPTER EIGHT NATIONAL CENSUSES AND INDIGENEITY IN VENEZUELA LUIS FERNANDO ANGOSTO FERRÁNDEZ

1. Introduction In 2011 Venezuelans demonstrated a growing public concern about the political implications of the national census. Within state organs, bureaucrats took every opportunity to emphasise their commitment to improving the quality of the census and making it as “realistic” and resourceful as possible; within civil society, a variety of actors engaged in debates on, or directly participated in, the elaboration of the census questionnaires; political parties approached this statistical exercise as yet another occasion to display confrontational discourse and prepare the electoral battlefield. The census was never a “trending topic”, but even in the Internet realm it had its share of popularity: the Twitter account created by the Instituto Nacional de Estadística–INE (National Institute of Statistics) had over 7300 followers by mid-September–barely two weeks after the enumerators started their work. Albeit commented on for spurious reasons on occasion, this most recent national census did deserve attention for its significant novelties. It included new ethnic and racial categories and a reformulation of the manner in which citizens could self-ascribe to already existing ones. This chapter will examine these issues1, with an emphasis on the identification of contemporary approaches to indigeneity as reflected in and around the last Venezuelan census2. It is argued that the ongoing process of state reform in the country is proving fertile for a wide range of politics of recognition and that this is reflected in the (increased) degree of sensitivity that the census holds for population diversity. However, evidence suggests that this process does not run smoothly, and in fact it is dependent on the

222

Chapter Eight

difficult conciliation of diverging political perspectives even within state organs. This will be illustrated by contrasting the ideological orientations that guide the INE and the Ministerio del Poder Popular para los Pueblos Indígenas–MINPI (Ministry for Indigenous Affairs)3 and by analysing the practical consequences of those orientations. The analysis of this critical juncture will be preceded by a diachronic exploration of the history of relations between the state and indigenous peoples in Venezuela as both reflected in and conditioned by census politics. The chapter is structured into seven sections. After the introductory one, the second one presents an overview of approaches to indigenous population in Venezuelan censuses. The third undertakes a contextualised analysis of the design and implementation of the census round of 2011. The fourth section identifies contested political perspectives on indigeneity within state organs and discusses their practical implications. The fifth one critically comments on the variances in the numbers of indigenous peoples registered in different national censuses and statistical records. The sixth section examines what is at stake in censuses for indigenous population. The final one is dedicated to conclusions.

2. The indigenous population in Venezuelan censuses The national census of 2011 was the fourteenth official one of the republican period. The pioneering count took place in 1873 during the first government (1870-1877) of Antonio Guzmán Blanco, whose autocratic rule was pervaded by nationalist ideals of centralisation and modernisation of the state (Nava, 1965). The Dirección Nacional de Estadísitica (National Bureau of Statistics) was created in this period and at least one national census has been conducted in every subsequent decade (with the exception of the initial one of the twentieth century). All national censuses, save in the 1970s, have included quantifications of population labelled as “indigenous” (see Table 8-1 below). But, before drawing attention to contextual and methodological features that influenced and characterised different census rounds, some general clarification is required. In spite of the continued presence of the indigenous population as a distinctive quantification in them, Venezuelan national censuses were produced with such differences in terms of logistical resources, operational concepts, geographical scopes and timing that a comparative demographic analysis of their data is impossible (Allais, 2004). Who is considered “indigenous” has been a matter of disagreement over time, and official state responses to that question related in turn to dilemmas as to where indigenous people are to be found.

National Census and Indigeneity in Venezuela

223

Table 8-1. The indigenous population in Venezuelan national censuses Year of National Indigenous Relative percentage4 census population population 1873 3.22% 1,732,411 55,811 1881 2,005,139 70,154 3.49% 1891 2,221,572 94,627 4.25% 1920 2,479,525 48,855 1.97% 1926 2,814,131 136,147 4.83% 1936 3,364,347 103.492 3.07% 1941 3,850,771 100,600 2.61% 1950 5,034,838 98,682 1.92% 1961 7,523,999 75,604 0.99% 1971 10,721,522 ------------------1981/25 14,516,735 140,040 0.96% 1990/26 18,105,265 315,815 1.71% 2001 23,232,553 534,816 2.3% Source: Prepared by author using data from the National Institute of Statistics (INE, 2011) and Allais (2004).

Up until 2001, in accordance with prevalent conceptions of indigeneity, national censuses only registered indigenous population in certain regions of the country. In that view, urban and most rural locations were not compatible with authentic, officially sanctioned indigeneity7. General and indigenous censuses were conducted separately from 1950 until 1990/92. Differences in timing went on occasion beyond two years, which obviously undermined accurate calculations of the indigenous proportion of the total national population. In spite of those distorting conditionings, Table 8-1 evinces two points that are significant for the study of the relation between the state and indigenous peoples in Venezuela: 1) the state never consummated a total “statistical genocide”8 in relation to the indigenous population, which has always been at least testimonially present in national censuses; 2) according to these official records, the indigenous population was never above 5% (and in fact generally below 3%) of the national one, so it has always been considered a “minority group”. This did not determine but of course did influence the degree of marginalisation experienced by indigenous people in the country and their very limited presence in official narratives of nationality up to the present.

224

Chapter Eight

But let us look with more detail at the characteristics of different censuses and to general trends that can be found among them.

2.1. Censuses until 1971: “civilised” and “uncivilised” indigenous people As regards the state approach to indigenous peoples, the decades preceding the first national census were characteristic of the liberal nationbuilding ideology. After independence in 1821, liberal principles provided theoretical legitimacy for a succession of governmental attempts to dismantle resguardos9 and indeed all forms of indigenous collective property, which were portrayed by the dominant class as an obstacle for economic development and for legal equality among republican citizenry. By 1882, in an attempt to materialise what had been unsuccessfully pursued since the 1820s, Guzmán Blanco sanctioned the Ley Sobre Reducción, Civilización y Resguardos Indígenas (Law of Reduction, Civilisation and Indigenous Resguardos). This new attempt did not accomplish the expected partition of the resguardos and the inclusion of indigenous population into “civil life” (Samudio, 2005: 252), and by 1884 legislation consolidated a paradigm that would linger as a long-lasting state perspective on indigeneity in Venezuela: the distinction between “integrated” and “civilised” indigenous peoples, on the one hand, and “non-integrated” or “uncivilised” on the other. These conceptual poles would indeed be pivotal for legislation and public policy for over a century, and in time led to the confining of officially sanctioned indigenous identities within territorial spheres defined in the negative by their lack of integration into “civilisation”. In 1884, for the incipient state bureaucracy “indigenous communities” were only those located in Amazonas Territory, Upper Orinoco and Guajira (areas regarded as selváticas and inhabited by people not included in the “civilised life” of society), along with those that possessed the title of fundación doctrinaria fidedigna (indigenous settlement founded under the auspices of catholic religious orders). In addition to the territorial limitation, the legal provision implied that indigeneity was only granted through descent, concretely “to the legitimate or natural descendants, in straight or collateral line, of the aborigines of this part of America” (ibid). The ideological distinction between “uncivilised” and “civilised” indigenous populations was reflected in the constitutional corpus of the period: the Constitution of 1874 referred to “indigenous people not civilised” (Art. 43.22); those of 1881 and 1891 maintained a specific disposition for “indigenous people not reduced or civilised” (Art. 43.22);

National Census and Indigeneity in Venezuela

225

in 1901, the Constitution addressed the indigenous population living in “savage state”, only to exclude it from calculations of the population quotas established to grant federal state representation in the Cámara de Diputados (National Assembly); in 1909, Art. 80.18 stipulated that the government could commission missionaries to establish themselves in those areas of the Republic where there were “indigenous people to be civilised”. It is to be noted that the concept of “civilisation” did not only set cultural boundaries between dominant and subaltern populations, but also tacitly demarcated the existence of economic ones: the “uncivilised” Venezuelan peoples and territories had not been inserted into the predominant mode of production (a pre-industrial form of capitalism). During this period indigenous population estimates were presented as total numbers, as an indication of an undifferentiated mass not incorporated into national, “civilised” life. The lack of a scientific methodology and the vague conceptualisations of census categories demand caution in the analysis of available data (OCEI, 1993: 13). Only in 1926 did the census begin to present disaggregated data on the indigenous population, showing its distribution by Municipios (Municipal Units) and, for approximately a tenth of said population (15,192 individuals), by sex as well10. From that count onwards, censuses projected a continuous decrease in the indigenous population that was only reversed in 1982. Although in-depth studies would be required to shed light on the phenomenon of indigenous population decrease in a country where the national population numbers were growing exponentially (see Table 8-1), in this case it is only natural to look at non-demographic factors when pursuing causal explanations. That historical period between the censuses of 1936 and 1971 was still pervaded by the notion of indigeneity that equated it to segregation from “civil life”, but it was also the time during which regional indigenismo emerged as a powerful ideological and institutional continental influence11. Indigenismo proposed the integration of the indigenous population into nationality through its incorporation into national production while claiming to respect its cultural distinctiveness–in opposition to previous forms of assimilationism. Along with the indigenista goals, the ideology of mestizaje was consolidating as one of the basic pillars of the imagined national community. There was therefore a convergence of two hegemonic political currents aimed at reducing economic and cultural diversity respectively in the name of constructing a homogeneous social substratum upon which the modernisation of the state and the consolidation of nationality could be achieved.

226

Chapter Eight

The impact of conceptual (and in this case ideological) factors on the fluctuations in indigenous population numbers is well illustrated by contrasting two censuses conducted in the 1940s by institutions holding different approaches to indigeneity. In 1941, when according to the national census there were 100,600 indígenas in Venezuela, the International Labour Organization (ILO) quantified a total of 538,000. The ILO divided this population into “pure indigenous” (136,147) and “predominantly indigenous” (402,000) (Marroquin, 1977: 3). The number of “pure indigenous” people was, according to the ILO, a third higher than the total indigenous population number provided by the Venezuelan census of that year (see Table 8-1); but what stands out more powerfully is the difference between the two measurements if one takes into account the total indigenous population estimated by the ILO: with a methodology that included “predominantly indigenous” (i.e. mestizos) into the “indigenous” category, the numbers sprang up. As mentioned above, between 1950 and 1992 the indigenous censuses were conducted separately from the general ones. In 1950 the census registered two categories: “accessible [and] integrated in civil life” indigenous population and “inaccessible” indigenous population (OCEI, 1993: 13). For this census programme, the country was divided into five geo-ethnographic regions, and only 41,977 out of the total 98,682 indigenous persons registered were identified directly by enumerators; the rest (56,705 individuals), considered as población selvática, were included as the result of estimations made by specialists. That census round was the first conducted after the foundation of the Instituto Indigenista Interamericano (Inter-American Indigenist Institute) and was in tune with its ideology. Acosta Saignes, a key figure in the institutionalisation of anthropology as an academic discipline in Venezuela and promoter of indigenist policy in the country, representatively voiced that ideology when stating that “in a country with such a poor demographic growth as Venezuela, it would be most fruitful to utilise and incorporate the indigenous population. Equally, it is necessary to think of the utility that [this] could yield to national production” (1948)12. Integration into national production and economic development were two of the pillars of the indigenista view, articulated through a top-down state approach to designing and implementing public policy for indigenous peoples. Though Acosta Saigne’s sensitivity towards indigenous cultures and diversity was always remarked upon, like most indigenista intellectuals of his time in the continent he nonetheless promoted that type of national “integration” as one of the state’s goals (cfr. Díaz Polanco, 1982). If to these concerns we add the influence of the ideology of

National Census and Indigeneity in Venezuela

227

mestizaje, which with the appearance of Acción Democrática (AD) in the 1940s as a multiclass and multiracial political organisation claimed again with considerable success the value of racial democracy in Venezuela (Wright, 1990: 97-100), it is unsurprising that national censuses of the period reflected a decline in indigenous population numbers. The 1961 census continued to reflect a decrease in indigenous population numbers, and by 1971 this population was obliterated in the national statistics: it was integrated into the general numbers through the concept of “indigenous peoples incorporated to civil life” in the areas accessed by enumerators (urban and rural areas). The indigenist goal of including indigenous people into “civil life” (tantamount to integration into market economy) still prevailed and, again, it was consistent with national goals as expressed by the Constitution in force (1961): “[t]he law shall establish the regime of exception that the protection of the communities of indigenous people and their incorporation into the life of the Nation may require” (Art. 77 [my emphasis]). In the absence of official census data, in 1974 the state organ in charge of indigenous affairs (by then the Direction of Indigenous Cults and Affairs–DICA, ascribed to the Ministry of Justice) obtained an estimate of 100,000 indígenas belonging to 28 ethnic groups located in seven regional entities (Clarac, 2001: 342). The DICA commissioned the report to a prestigious academic body, the Institute of Economic and Social Research (Ethnographic Section) at the Universidad Central de Venezuela. This report illustrates the influence of central academic views in the shaping of state-sanctioned approaches to indigeneity, and in fact it is still echoed by dominant views in state organs nearly four decades later: the regions pointed out in the report as those with “traditional [indigenous] occupation” are basically the same ones thus regarded today13. It is also worth remarking that, after this report, the distinction between two types of indigenous population would not be articulated any longer through the poles of the “civilised” and the “uncivilised”, though these poles were basically overlapped by the establishment of two new ones: those that distinguish between “traditional” and “non-traditional” indigenous people. The distinction between these latter poles is still conceptually rooted in “culture” (as it was in the previous conceptual polarisation), though now there is a valorisation of the non-dominant culture (previously categorised as “uncivilised”) that springs from the social worth with which “tradition” is invested.

228

Chapter Eight

2.2. Censuses from 1982 onwards: “traditional” and “non-traditional” indigenous population The main coordinators and advisors of the indigenous census of 1982, still conducted separately from the general one, were renowned anthropologists (see credits in OCEI, 1985). The indigenous census program included for the first time trained indigenous enumerators, in an example of a technical decision with a political background. By the 1980s the so-called “new” or “avant-garde indigenism” had gained terrain as an ideological alternative to official indigenist views. The “new indigenism” advocated a displacement of “the indigenous” from the position of “object” to that of “subject” of state politics. The principles of this ideological move had been well outlined in documents such as the Declaration of Barbados of 1971, the drafting of which featured prominent academic allies of the embryonic indigenous movement in Venezuela (such as Nelly Arvelo and Esteban Emilio Mosonyi). The first Venezuelan indigenous organisations appeared in the 1970s, smoothing the path for other platforms of indigenous representation in civil society. According to the operational definition of the census of 1982, an indigenous person was one “who speaks, or who spoke as a child, an indigenous language, or whose mother or grandmother speaks an indigenous language or spoke it as a child” (OCEI, 1985). But the census was only conducted in those areas previously labelled as regions with “traditional occupation”, and it thus established a total of three essentialist, externally-decided criteria for the identification of indigeneity in that round: direct descent, knowledge of the vernacular language and “habitation in indigenous communities” [i.e. población selvática]. As an outcome of this last criterion, individuals and groups who fulfilled the first two criteria but dwelt in urban, or even in certain rural areas, were not taken into account. Only those communities located in “traditional areas of occupation” and at the same time preserved features that were considered “traditional” (from the type of housing to the means of production) would be targeted by the indigenous census. Besides those conceptual boundaries in the approach to indigeneity, a concurrent methodological circumstance contributes to explaining the relatively low numbers of indigenous population registered by the census (see Table 8-1): most of the population of Zulia state was only roughly estimated (indeed, underestimated, as subsequent censuses would show), and thus the Wayuu population, basically located in that region and the largest indigenous one in Venezuela, was largely undercounted.

National Census and Indigeneity in Venezuela

229

This census did not provide information on age or sex either, but in spite of its pitfalls and the lack of resources faced by the organisers, it established a landmark in terms of coverage (within the regions of “traditional occupation”) and participation of indigenous people in its implementation. The indigenous census of 1992 introduced remarkable conceptual novelties. Firstly, the criterion of self-ascription became core for indigenous identification and the conceptual definition of indígena was accordingly transformed: indigenous then embraced “all persons who at the time of the census self-declared as a member of an indigenous ethnic group with current presence in Venezuela and/or in neighbouring countries” (OCEI, 1993: 25). Secondly, the census coverage included urban areas too, albeit only those within those eight federal units associated with the notion of “traditional indigenous occupation”. This inclusion was nevertheless far from obviating the distinction between different types of indigenous population: in “traditional” indigenous communities the applicable questionnaire captured information clearly distinguishable from the general census questionnaire, in which “urban” indigenous population was registered through a question that enabled selfidentification (“to which indigenous group do you belong”?) and then enquired about knowledge of indigenous languages as the only distinctive cultural feature to be made salient. The Wayuu population was included in this round (through a special separate census coordinated with Colombia), but there was however another notable omission: some areas of the Upper Orinoco were not accessed by the enumerators, and therefore a considerable part of the Yanomami people was only estimated (Allais, 2011). There were indigenous enumerators in the census teams, all of them native speakers of the relevant indigenous language. In regard to results, the substantial increase in absolute numbers (see Table 8-1) was again principally conditioned by non-demographical factors. Besides the quantitative impact of including the Wayuu population, other indigenous peoples saw how their official numbers jumped up. The criterion of self-ascription, the widening of the coverage to urban areas, the elimination of the language proficiency requirement and a greater participation of indigenous surveyors in the undertaking of the census were crucial factors. The inclusion of urban population in particular had a massive impact: it amounted to 41.9% of the total indigenous population. A total of 39 different etnias and ethnic sub-groups were registered (OCEI, 1993: 17), but only 29 of them were identified as

230

Chapter Eight

“traditionally inhabiting in national territory” by the census authorities (see Table 8-3)14. In the round of 2001, “indigeneity” pervaded the whole national territory. Unlike in previous rounds, it was then neither exclusively secluded in áreas selváticas nor confined in a limited number of federal units. The influence of various forms of globalisation, trans-national campaigning and supra-national organisations concerned with indigenous rights (Brysk, 2000: 18-20), as well as the lobbying of national activists, had indeed been already noticed in 1992 with the inclusion of selfascription as a core census criterion (Allais, 2011). However, in 2001 there were also other political conditions that need to be taken into account. The Constitución de la República Bolivariana de Venezuela–CRBV (Constitution of the Bolivarian Republic of Venezuela) of 1999 had recognised differentiated rights for indigenous peoples and both government and state organs had started to make intensive use of an imaginary of indigeneity in an accelerated process of national identity reformulation (Angosto, 2008). There was a significant conceptual shift in the denomination of the indigenous population, and “indigenous ethnic groups” [étnias] became “indigenous peoples”, and the national census echoed for the first time a political terminology that can be tracked back to the influential ILO Convention 169 (1989)15. Indigenous claims were largely dissociated from the sphere of “minority group issues” and instead conceptualised as springing from groups with inherent rights to territory and free determination, rights that precede the establishment of the Nationstate. The Venezuelan indigenous movement, in tune with most of its continental counterparts, had taken the issue of denomination as one of the strategies to advance its cause in an international context strongly influenced by the ideology of multiculturalism (Hooker, 2005: 303). At an internal level, the Consejo Nacional Indio de Venezuela (CONIVE) consolidated its position as the legitimate interlocutor with state organs since the constituent period (Van Cott, 2002; Angosto, 2010), and this in turn increased indigenous political influence, even if moderately, in census design. Census results reinforced the thesis that political factors informing the conceptual and technical aspects of the count have been more influential than actual demographical changes in the fluctuations of official indigenous population numbers. In 2001 the total indigenous population rose to 534,816 individuals. In urban areas, where indigenous selfidentification is more sensitive to issues of social discrimination or revalorisation, there was an indicative increase in absolute and relative

National Census and Indigeneity in Venezuela

231

numbers: 66% of the total indigenous population registered in 2001 dwelt in urban areas, as compared with the 42% registered in census 1992. In addition, it is most significant that in 2001 surveyors registered selfascription to a total of 54 indigenous peoples–though only 40 would afterwards be recognised by the National Assembly as currently “existing and identified” Venezuelan peoples (see First Final Disposition, LOPCI 2005). We shall come back to this point in Sections 5, but let us first examine more closely the last census round (2011).

3. The 2011 census round Recent research on censuses has drawn attention to the ways through which global and national politics and social context exert influence over results. Particularly when the option of self-ascription to ethnic or racial categories is introduced in national censuses, researchers suggest the existence of under-declaration tendencies in countries with strong structural discrimination and opposite tendencies in societies undergoing processes of ethnic re-vitalisation, where over-declaration may even be identified (Llorens, 2002; Gundermann et al., 2005: 92-93; Del Popolo, 2008: 26). In the latter cases, both national and international factors have been identified as propitious for ethnic and racial self-identification in the census. Thus Gundermann et al. (2005), analysing the Chilean case, pointed out the process of return to democracy and the strength of indigenous movements in the continent as two political currents that, from the national and the international spheres respectively, contributed to shape a social context in which identification with the indigenous ingredient of national identity was particularly appealing in the census round of 1992. From this perspective, in the following section I will examine salient features of the contemporary Venezuelan political context in relation to the design, implementation and results of the past census programme. Focus will be placed upon two contextualising forces within the national sphere: 1) the “indigenisation” of national identity parameters that is taking place in the ongoing process of republican refoundation; 2) the so-called “new geometry of power” as a governmental proposal for structuring national development.

3.1. Guaicaipurismo and socialism of the twenty-first century Any political system, regardless of its ideological filiations and its institutional arrangements, is a social construct that requires legitimising discursive narratives and at least minimum levels of identification and

232

Chapter Eight

cohesion among its members. In Venezuela, a discursive current that draws intensively on an imaginary of indigeneity stands out for its centrality in the contemporary process of national identification (Amodio, 2007; Angosto, 2008). After the collapse of the party system and the political crisis that peaked in the late 1990s, the republican refoundation inscribed in the project of the CRBV spurred a process of national identity re-arrangement in which notions of indigeneity have played an outstanding role. This distinctive current of symbolic and discursive production which I have named guaicaipurismo16 pervades the action of many government and state agents and, in general, of supporters of the so-called Bolivarian revolution (Angosto, 2008). Actors self-identified with this political process have made of guaicaipurismo one of their banners and this has facilitated a convergence of the discourses of revolution and indigeneity which is epitomised in slogans such as “the revolution began 500 years ago”. This slogan exemplifies the systematic semantic and symbolic association of the concepts of “indigenous resistance to colonisation” and “Bolivarian alternative to neo-liberalism” that is today central to the political discourse of government and indigenous movement actors alike. This discursive narrative runs in parallel to a factual increase of indigenous presence in the public arena, largely dependent on the institutionalised minimum representation that the CRBV grants to indigenous peoples in legislative organs and on the creation of new state organs with competences in the field of indigenous policy (such as the MINPI or the National Institute of the Indigenous Peoples). It is difficult to disregard these features of the contemporary Venezuelan political context when comparing the results of the past two national censuses. The 2001 round, conducted after the approval of the CRBV and with guaicaipurismo in full display, showed an increase of 69.3% in the national indigenous population, while also capturing selfascription to indigenous denominations that had never appeared in previous censuses (see Tables 8-1 and 8-3). According to the provisional results of the last census round (2011), a 2.7% of the total Venezuelan population is “indigenous” (in 2001 it was 2.3%), but it is likely that a detailed analysis of the final data may reveal some significant changes in its composition. Before the census was conducted the MINPI was already circulating information remarking the existence of 44 peoples in Venezuela (four more than in the list sanctioned by the National Assembly in 2005, see Table 8-3) and some of its highranked officials openly estimate that the indigenous population will reach one million in the new census results17. We shall come back to these points in Sections 3.2 and 5, but let us continue with the examination of another

National Census and Indigeneity in Venezuela

233

influential contextual feature in contemporary Venezuela: the so-called “new geometry of power”.

3.2. The new geometry of power and the MINPI The Ley Orgánica de Pueblos y Comunidades Indígenas–LOPCI (Organic Law of Indigenous Peoples and Communities), sanctioned in 2005, marked a climax in the identification between the Venezuelan indigenous movement and the government. By then, the country’s political polarisation had not yet projected a clear reflection within the realm of indigenous politics and, in general terms, opposition parties had totally neglected that political field. Moreover, National Assembly elections in 2005 were marked by the abrupt voluntary withdrawal of those parties in an orchestrated attempt to delegitimise the forthcoming parliament, and as a result the election of indigenous representatives linked to the government front was not contested (Angosto, 2011). This contributed to maintaining the sphere of indigenous politics as an apparently homogeneous field with no discrepant interests among the actors who constitute it. But also in 2005 the Venezuelan government explicitly declared a shift towards socialism, and then in the presidential elections of 2006 Hugo Chávez was re-elected. The socialist banner had been openly aired during the electoral campaign and, right after its victory, the government launched the so-called “five revolutionary engines” as priority fronts on which to focus for socialist transformation of the country. The fourth of those engines was the “new geometry of power” (NGP), a conceptual frame for a project of public and economic power redistribution underlined by a re-arrangement of the country’s territorial administrative units. It contemplated the possibility of creating new types of political-territorial units, while in parallel promoting communal self-management within those new units. From the perspective of the NGP, “the community” becomes the territorial core for state organisation, the Communal Council being the community’s basic political institution. From Communal Councils and through aggregation, larger territorial units may emerge, such as “the Commune”. The NGP soon became the governmental referent in the management of indigenous affairs, and indeed it has been at the core of the current promotion of political-territorial units that are categorised as “indigenous” (i.e. the indigenous Communal Councils) but do not match the model that the CRBV stipulated for “indigenous lands and habitats”18 (Angosto, 2010). The NGP supports an ongoing government-fostered project of incorporating indigenous peoples into nationality with a strong focus on

234

Chapter Eight

the development of their productive forces and with the indigenous Communal Council as the political institution that links the indigenous and the national polities within the state structure. Although a legitimate, plausible and potentially beneficial alternative for a part of the national indigenous population, the consolidation of the NGP perspective as the only government-supported option has left aside the possibility of formalising constitutional principles related to the rights to free determination. Concession of collective property over ancestral territories and the recognition of distinctive forms of indigenous political organisation have in practice become subsidiary issues for the government. The MINPI is in charge of articulating this policy and indeed has set the creation of indigenous Communal Councils as one of its principal tasks (Angosto, 2010: 121-125). This ministry has also become an important organ of economic redistribution for indigenous people, who now receive state resources (credits, productive or assistance equipment) precisely (and nearly exclusively) through Communal Councils. This strong redistributive potential of the MINPI has situated it as an institutional focus that stimulates indigenous ethnic revitalisation in Venezuela. And it is important to note that the MINPI, along with other state and civil society actors, participated to some extent in the design and implementation of the census of 2011: As soon as the [preparatory census] meetings started [the INE] sent a call to the Ministry of Indigenous Peoples so that they could appoint an official […]; and an assistant vice-minister attended the meetings. Another call was sent to the Indigenous Peoples Comission at the National Assembly and they also appointed a person, who by the way had already worked in previous censuses. There was another call for CONIVE, and CONIVE appointed a person too. [Although] these people did not have all the expected regularity in meeting attendance and participation, […] they all knew the instruments and the methodology through which the indigenous census was to be conducted […], and some of them indeed made observations that were incorporated in the census questionnaires, basically in the community questionnaires. (Allais, 28-07-2011; interview).

The MINPI and the other participating actors bring their own views on to the INE census meetings but, as we shall see in Section 3, the existence of inputs (and institutional pressure) coming in particular from the MINPI, go beyond the contributions that this ministry might have made to the design of community questionnaires: those inputs also concern the orientation given to the self-identification criterion in the process of defining indigeneity. Before elaborating on this point, let us analyse these

National Census and Indigeneity in Venezuela

235

contextual political influences related to conceptual and methodological transformations in the 2011 census round.

3.3. Conceptual and methodological novelties in the 2011 census round A technological novelty drew considerable attention in the most recent census round. An electronic mobile device for data capture replaced the former paper questionnaires. This simplified the collection and processing of data and facilitated some public accompaniment of the programme–data was centrally gathered on a daily basis and the INE’s website provided occasional updates. But, however important, technology was far from being the most transcendent element in this census round: politics not technology conditioned the results. Arguments about the favourable climate for politics of recognition in contemporary Venezuela are buttressed by the inclusion in 2011 of a census question that for the first time in this country enabled selfascription to non-indigenous ethnic and racial categories. The question posed in the general questionnaire was: “According to your physical features, family ascendency, culture and traditions, do you consider yourself: Black / Afrodescendent / Dark [Moreno/a] / White / Other (which one?)”19. This question changed a census history that, as Wright put it in his study of race, class and national image in Venezuela, was “symptomatic of the professed racial harmony found here” (1990: 4). Racial discrimination never disappeared after the abolition of slavery in 1854, but it was veiled by the hegemonic official ideology of racial democracy and the elite’s hopes of “whitening” society and absorbing blacks into a pardo, criollo majority. These new ethnic and racial categorisations require farther analysis that, in order to tackle the immediate concerns of this chapter, will not be undertaken here, but their appearance is in itself symptomatic of important changes. Instead two other questions will be discussed next: • •

the distinction between “peoples” and “ethnies” in the denomination of indigenous groups in the 2011 census round; the (national) boundaries set to the possibilities of self-identifying as “indigenous”.

236

Chapter Eight

3.3.1. From ethnies to peoples to ethnies Political and sociological times do not always synchronise. Over the past two decades, but particularly since the constituent period of 1999, the use of the term “people” emerged in Venezuela as the prevalent official denomination for indigenous groups within the state. The constitutional chapter that grants distinctive rights to this population is dedicated to “indigenous peoples”, which confirmed an official upgrading of their legal status in the country: they completed the transit from “ethnic groups” to “peoples”. This conceptual journey was setting Venezuela in tune with the terminological grounds established by the ILO Convention 169 (1989), whose definition of “indigenous peoples” gradually permeated the socalled “new Latin American constitutionalism”. However, “people” was not always the term in vogue, and in fact “ethnies” (etnias) was the term with which indigenous groups were generally referred to in Venezuelan media, colloquial speech, state records and even in the academic realm before their constitutional recognition. The shift towards the new official denomination was not devoid of controversy, and during the constituent period a part of the Assembly was reluctant to accept the proposal of distinctive indigenous rights if it was articulated around the concept of “peoples”, confronted for its potential nationalist resonances. During that period the term etnias still predominated among those who opposed distinctive indigenous recognition, as for instance illustrated in newspaper headings such as “The rights of the ethnies fracture the Republic” (Angosto, 2006: 147). Once the controversy was overcome, the term “people” immediately took its place in public administrative procedures, but only gradually gained acceptance in everyday popular speech. The general questionnaire in the 2001 census round already reflected that terminological shift: the question posed to enable self-ascription to the indigenous category was “Do you belong to an indigenous people?”, while in the previous round (1990/92) the question had been “To which indigenous group do you belong?”. However, unexpected census results in 2001 led the census designers to reconsider again the wording of this question, and indeed in 2011 it presented a terminological mixture finally shaped as “Do you belong to an indigenous people or etnia?”20. Behind this modification there was also a combination of political and semantic reasons. As mentioned above, the term “indigenous people” had only had a couple of years in official usage by 2001 and as we shall see below not everyone was familiar with (or subscribed) such new usage. In addition, the term “people” (pueblo) poses semantic ambiguities to the Spanish speaker: it has both the meaning of “people” and that of “village” or “town”. The outcome of this convergence of political and semantic factors

National Census and Indigeneity in Venezuela

237

in census 2001 is well explained by María Luisa Allais, who closely supervised the results as consultant to the INE: In 1992 the question [for ethnic self-ascription in the census] was “to which indigenous group do you belong?”. Etnia was the concept in use by indigenous organisations and the concept that these organisations transmitted to their supporters. It was etnia not “indigenous people”. When in 2001 the concept was changed to “indigenous people” we had many equivocal declarations because many people thought they were being asked about their place of origin. People would be asked: “what people do you belong to?” [sic], and they would say, for instance: “Sinamaica”21. And we had a good number of declarations of the like. Some of them we could clarify [by finding that the provided answers corresponded to the names of indigenous communities and then checking the indigenous people to which those communities belonged]. But there were cases we could not clarify, because [the answers retrieved the names of] urban centres [or mixed villages] and we had to leave those unidentified. What did we decide in this census round [2011]? We would accompany “people” with “ethnic group” so that the former is not confused with “village”. (Allais, 28-07-2011; interview).

The relevance of semantics is made obvious in this census experience, but also the impact of socio-cultural context on semantics. It is worth comparing this situation with what had occurred in the Chile of 1992. With the country undergoing a process of democratic normalisation after dictatorship, the term “people” (pueblo) created rejection among both opposition and government supporters and had to be discarded from the census question which aimed at quantifying the indigenous population through self-ascription (Gundermann et al, 2005: 111). The question finally posed was “If you are Chilean, do you consider yourself as belonging to any of the following cultures? Mapuche – Aymara – Rapanui – None of those”. “People” was thus substituted by “culture” in the question, and this is today considered to have facilitated an overdeclaration in ethnic self-ascription. Analysts argue that “culture” and “people” address different (but related) entities in the common national imaginary (ibid.: 110), the term “culture” being less semantically restrictive than that of “people”. Interviewees would therefore self-ascribe more unproblematically to an indigenous “culture” than to an indigenous “people”. With the former one might feel affinities or share certain practices, whereas with the latter the issue of belonging is less flexible and much more politically loaded. If empirical evidence certainly supports that argument in the Chilean case22, the Venezuelan one provides material to additionally argue that the restrictiveness of the term “people” is not only

238

Chapter Eight

relative to comparison with other terms such as “culture”, but also to other elements of social context: in a country where the term “people” had not pervaded social usage nor the particular realm of indigenous politics, it became the source of semantic and political confusion and as a result a factor for under-declaration in self-ascriptions to indigenous categories. 3.3.2. Indigenous, but nationals In the past census round self-ascription to the indigenous category was exclusively open to people born in Venezuela. This condition set boundaries to a certain trans-national quality with which, in practice, indigeneity had been invested in previous rounds. In the censuses of 1982 and 1992 indigeneity was indeed shaped by that trans-national character. According to their operational definitions of indígena (see Section 2.2), those censuses registered individuals who belonged to both indigenous peoples ancestrally based in the territory of contemporary Venezuela and indigenous peoples located in the territory of neighbouring countries. Given those normative conditions, it is most significant that census results in 1982 and 1992 did not register as many indigenous peoples from neighbouring countries as did the results of the 2001 round (see Table 8.3). The coverage was certainly larger in this latter census, which expanded the possibility of self-ascription to the indigenous category over the whole national territory; but, more determinantly, that count of 2001 came in a period of positive ethnic re-valuation. The influence of guaicaipurismo was in full force in Venezuela, and in the continent in general the influence of the ideology of multiculturalism had been advancing, as reflected in Latin American constitutionalism of the 1990s. However, in 2001 there were also unexpected results in regard to the indigenous peoples registered by enumerators: not only did peoples considered “extinct” appear in the records, but also a considerable number of non-Venezuelan, non-neighbouring and even “non-existing” peoples too. In the preliminary results there were Chibcha, Kechwa, Pygmy, Sioux and Apache, among others. These results were facilitated by the questionnaire design, which asked about belonging to an indigenous people and, when a positive answer ensued, left the field for specific ascription open to the respondent. That is, there was not a list of officially recognised indigenous peoples to which the respondent needed to stick to. Queried about this point and the political implications of answers such as the ones mentioned above, Maria Luisa Allais summarised the positioning of the INE remarking that “these other things such as the appearance of some ‘crazy’ answers, well, simply you check them [in the processing of

National Census and Indigeneity in Venezuela

239

data] and discard them. They are very few, I tell you, [but indeed] we had Sioux, Apache, Redskins [...]” (Allais, 28-07-2011; interview). This makes obvious that, at the stage of data processing, census technicians and INE functionaries are obliged to make decisions concerning unexpected or “irregular” data. However, setting boundaries to valid or “good sense” answers is not merely an administrative task. Although pejorative and discriminatory denominations such as “Redskins” can be straight-forwardly discarded in data processing, other cases pose more complicated questions to the census technicians. For instance, the enumerators obtained self-ascriptions to categories such as Kechwa or Chibcha. The self-denomination “Chibcha” could hypothetically come from a member of the Bari people in Western Venezuela, for instance, or perhaps from someone belonging to one of the indigenous peoples in Colombia whose language is also part of the Chibcha linguistic branch (e.g. Kowi, Wiwa, U’wa). Registered as they were in Carabobo and Barinas states, far from both Bari territory and the territories of neighbouring Colombian Chibcha speaking peoples, these two Chibcha self-ascriptions in census 2001 might have come from people with a somewhat libertarian sense of humour and no link whatsoever with a Chibcha group, or on the contrary from people with actual Chibcha ancestry who spoke a Chibcha language; we simply cannot know. The enumerator’s job was to register the answers of the interviewee and he/she took note of the Chibcha response, a totally legitimate one in a round in which self-ascription was the operating criterion for indigenous identification and there was not a list of “recognised” indigenous peoples to which respondents should adhere. Even if “Chibcha” is not a denomination adjusted to a given people within Venezuela, it is not an absolutely “crazy” answer considering the existence of the Chibcha linguistic family and of the Venezuelan Bari and other indigenous peoples in Colombia who could qualify under that linguistic-political umbrella. Notwithstanding all these possibilities, that response could have easily been discarded under the argument that it is a broad linguistic denomination not exclusive to a single self-recognised people. Yet the Chibcha case creates scope for some complex political and technical discussion. And, far from being an exception, other cases blur even more the thin line of “common sense” in census ethnic identification: let us think of the case of the “Apache”. One cannot completely discard the possibility that a response such as “Apache” came from an actual member of one of the Apache peoples. It could be someone who had settled in Venezuela due to a migration spurred by labour, politics, romance or some combination of these forces. Should

240

Chapter Eight

the Apache be automatically considered an “abnormal” case and discarded? One would not think so, given the fact that other original foreign indigenous peoples’ denominations were accepted in the preliminary list of results of the 2001 census (e.g. Inga, Kechwa). Accordingly, the Apache could have been unproblematically eliminated only if the criterion of including neighbouring indigenous peoples had been in force in the 2001 census (which it was not) and the concept of “neighbouring” was literally taken (which tacitly it seems to have been). The “Apache” case thus evinces that there are decisions to be made in data processing that go well beyond technical procedure. Indeed, not all (ab)normal indigenous self-ascriptions were equally treated in the processing of data in 2001. Some of them were included in the preliminary results, while some others were not. In the absence of an official publication of results by the INE, the self-ascriptions listed in the preliminary results ended up being treated as official results and reproduced uncritically in all types of publications (from academic to divulgatory)23. And the decisions made around these issues have implications beyond the symbolic value of recognition, as the following sections will show (Sections 4 to 6). In any case, the dilemmas posed by the results of 2001 led the census designers in the 2011 round to introduce conceptual limits to indigeneity along national boundaries: the question posed in 2011 was “Do you belong to an indigenous people or ethnie?”, but it was only asked to those who in a previous question stated that they had been born in Venezuela. In addition, a list of (recognised) indigenous peoples was introduced this time in the mobile device of data capture, appearing when self-ascription to the indigenous category took place. Though that list had a box open to nonlisted ascriptions too, it had of course, along with a practical dimension, a degree of sanctionable indigeneity filtered by political criteria. Data tabulation in the previous census round proved to be problematic, since it had originally been captured in hand-writing by the enumerators and, given the variety of denominations, the general lack of familiarity with many of them and the occasional lack of clarity in the writing, it was often difficult to decipher the written names. Before closing this examination of the 2011 census round, it is worth commenting on another phenomenon, though it goes beyond the confines of regular census politics discussion.

National Census and Indigeneity in Venezuela

241

3.4. Party politics and polarisation in the 2011 census round The distinction between census politics and party politics occasionally faded during the past round. Like many other aspects of social life in contemporary Venezuela, the national census was transformed into a propitious terrain for political party confrontation and for the deployment of pre-electoral manoeuvring. Controversy arose when, barely a week before the commencement of the surveying process, spokespersons of an opposition party launched a media campaign questioning the contents of the census questionnaires. José Curiel of COPEI (Comité de Organización Política Electoral Independiente [Social Christian]) initially voiced the critiques in a press conference widely covered and reproduced by private media. He remarked that “the census advanced by the government is more an investigation of personal data than a real census of population and habitation. These data can be used, like in Cuba and in other totalitarian regimes [sic], to obtain information that permits controlling people”24. Those remarks generated a prompt response from INE authorities, who explained the reasons why the existing questions did not erode citizen rights nor breach international standards, additionally emphasising that none of the questions was compulsory for the respondents. A noisy controversy ensued nevertheless. Some opposition leaders, allies of COPEI in the so-called Mesa de la Unidad25 (MUD), quickly detached themselves from the alarmist declarations, but this had a lasting media impact nonetheless. The INE director, Elías Eljuri, showed up in several public media programmes projecting a calming institutional message that could be summarised in the main points of an article he signed in a government-funded newspaper three weeks after the controversy was sparked26: public response to the census was excellent, which demonstrated the “legitimacy that society has granted to this statistical operation”; all the personnel of the INE were always duly identified; “the information provided is protected by statistical secret [and that] data is transformed into codes in order to preserve anonymity” and is to be used for public policy; “this census […] counts on advice from international organs27, is undertaken with practically the same questions it always had plus some new ones that resulted from suggestions by public organisms, statistical sub-committees and social movements, oriented to obtain information for the amelioration and strengthening of public policies”. The concern with the loss that the whole country could experience if the census was not conducted successfully stimulated demonstrations of civic consciousness in various sectors of society. While some commercial newspapers continued echoing voices of distrust towards the census, others published informative sections on census contents and methodology, along

242

Chapter Eight

with statements of support. These publications recommended things such as giving “a kind treatment to the enumerator, who is a worker hired to carry out a task that benefits the country. If there is a discrepancy with the INE or with the government it does not mean that the official should be mistreated”28. The fact that a national commercial newspaper felt the necessity of publishing a “guide of behaviour” of this type gives an indication of how hostile some people can be towards state employees, especially when there are politically orchestrated campaigns aiming at generating distrust in institutional structures. A generalising classification of the openly militant commentaries of actors involved in the controversy indicates that they replicated the country’s political polarisation around one of the axes that structures discursive tactic for supporters of each of the antagonistic poles: thus the actors of the so-called Polo Patriótico29 accused the rivals of a lack of patriotism and of aligning with foreign, imperial interests, while the members of the MUD actively involved in the controversy denounced partisan use of the state apparatus and authoritative government practice. Those who wanted to cast doubt on the legitimacy of the census also drew on what I have elsewhere called the “Tascón list effect”, a phenomenon that has also been identified as the inspiration of electoral tactics among members of some opposition parties (Angosto, 2008: 32). The so-called “Tascón list” was a record with personal data (names and identity card numbers) from citizens who in 2004 signed a petition to legally activate a referendum through which citizens could revoke or maintain (as eventually happened) the mandate of Hugo Chávez for presidential office30. This record was illegally filtered and ended up being widely circulated and even sold by street vendors. Many people have since criticised that filtration and alleged illegitimate use of that list as a tool for political retaliation against opposition members through exclusion from civil service ranks or public benefits. Government members have publicly denied that this is an approved practice and have asked their supporters not to engage in it, but the controversy still resonates today and is in fact used as part of the tactical arsenal that some opposition members resort to in order to undermine the legitimacy of public institutions in Venezuela. The “Tascón list effect” certainly affected the controversy generated around the past census round31.

National Census and Indigeneity in Venezuela

243

4. Indigeneity from different angles: the INE and the MINPI Nowadays a considerable variety of social actors require detailed statistical information on the indigenous population in order to pursue their institutional and/or political goals. Albeit associated with different procedures and ideological sensitivities, many civil society and state organs have developed a growing concern with the relation between human rights, cultural diversity and ethnic and racial discrimination. In parallel to that concern there has been an increased demand for diversity-sensitive statistical data. Among the Venezuelan state agencies that pay particular attention to this differentiated data are the Ministry of Education, which seeks the improvement of intercultural programmes, or the Ministry of Health, which has a department dedicated to indigenous health. This ministry in particular has introduced a variable for ethnic or racial identification in the datacapture instruments of primary health and medical services, though at a national level the procedures still need to be refined and conceptually homogenised: on occasions, the concept used for these statistics is “race”, on occasions “ethnic group” and on occasions the criterion is the self-ascription to an existing indigenous people32 (Del Popolo, 2008: 19). However, the organ that currently concentrates more executive and coordinating power in the area of indigenous policy, and the organ that therefore requires more specific data, is the MINPI, created in 2007. And it is significant that, shortly after its creation, this ministry embarked on the elaboration of its own census: more than a practical inclination to update the results of the 2001 census, it revealed a new perspective on indigeneity as the foundation of its strategic plans. The MINPI conducted its own census in order to obtain data consistent with its own conceptual approach to indigeneity and adjusted to the orientation of its public policy. The results of that census differed considerably from the national ones obtained in 2001 by the INE, while the degree of coincidence with the results that the census of 2011 will yield is still to be seen. The causes behind the differences in these census results are fundamentally two, neither of them related to demographics. The first one is that the INE and the MINPI diverge in their conceptual approach to the category “indigenous community”.

4.1. What is an indigenous community? The definition of the INE states that this “refers to those settlements, generally located in traditional areas of occupation, where the main part of

244

Chapter Eight

the population has self-identified as indigenous. These settlements are recognised by a particular name which can be different in Spanish and the indigenous language”. Along with this definition, the INE establishes a condition: “Its members belong to one or more indigenous peoples and possess forms of social organisation and cultural expressions of their own which permit them to be distinguished from other communities”33. This definition can be contrasted with the one included in the LOPCI (2005, art. 3.2): “[Indigenous communities] are human groups constituted by indigenous families associated with each other, belonging to one or more indigenous peoples located in a particular geographical space and organised according to cultural patterns specific to each people, with or without modifications coming from other cultures”. The INE thus operates with a more territorially restrictive definition, since in general terms it associates “indigenous community” with “traditional areas of occupation”. This is not a conceptual confinement that the MINPI shares, as its political praxis demonstrates, so this ministry draws on the LOPCI definition in order to back its policy orientation. The MINPI has been pursuing the development and integration of the indigenous population into nationality through the conceptual frame of the NGP, as explained in Section 3.2. Within the NGP perspective, the basic political-territorial unit is the Communal Council, which ought to arise from a self-recognised and self-demarcated local community. The Ministry has dedicated many efforts and resources to the constitution of indigenous Communal Councils and Communes (Angosto, 2010) and, in the pursuit of its goal, it has adopted its own approach to determine where “indigenous communities” can be found. Besides long-time recognised indigenous communities in the eight federal states in which “traditional occupation” is conventionally assumed to exist, the MINPI has focused on the identification of groups of indigenous people living in urban areas and in federal states previously neglected by officially sanctioned indigeneity. The views of the INE and the MINPI were prone to clash during the preparation of the 2011 census, and they did. Their positioning was conciliated to a large extent, but a total agreement was not possible. The words of Allais, consultant to the INE, help us again to comment on the divergences existing in relation to the possibility of dealing with indigenous communities outside the so-called “areas of traditional occupation”: The Ministry of Indigenous Peoples has a list of communities. They call it “census of indigenous communities”, and that census is managed with a concept of “community” that is not the concept that the INE has been working with. The INE works with the concept of “settlement”: an indigenous community is a settlement with a majority of indigenous

National Census and Indigeneity in Venezuela

245

population. Basically, in the census tradition [this settlement was in] the indigenous [territories] demarcated generally by specialists. [Thus for the INE], “indigenous community” is a settlement, a centro poblado with a majority of indigenous population. [But] it is possible that [one finds] a populated centre [(centro poblado) in which] there is a corner where some indigenous families are located; that is possible, but for the INE that is not the same as a community which has its own authorities, which is located in its traditional area of occupation, which has a historical [record in the censuses]. With the process of ethnic revitalisation we might now have […] a centro poblado in which a majority of the population self-declares as indigenous. This might happen, and if it does we [the INE staff] will check it. The INE is open [to methodological and conceptual modifications] in accordance with results […]. We will have a great advantage with the mobile device [for data capture], knowing the progress of the census, and in those cases where there might be doubts about whether [the indigenous population] is a majority or a minority, if it happens to actually be a majority, we can [come back and] apply the questionnaire of [indigenous] communities. We can apply this questionnaire after a negotiation (we have already talked about this with some organisations) on the following grounds: we will apply the community questionnaire, but if the results tell us that that is not an indigenous community as such, then … (Allais, 2807-2011; interview).

For the time being the INE continues to operate a factual distinction between “traditional” and “non-traditional” indigenous population. The Institute is however open to recognising (through its practice) indigenous communities outside the traditional areas of occupation, but not exclusively on the grounds of the self-identification of a majority of the population in a given settlement: if that occurs, the INE will additionally check, through the community questionnaires, if the features of such communities match the profile of what a “traditional” indigenous community is according to the Institute (i.e. it has a certain type of housing, economic activity, form of family organisation or other traits sanctioned as “indigenous”). Nevertheless it is clear that the influence of the MINPI and of other social actors who point towards a broadening of state approaches to indigeneity are contributing to the generation of conditions for a certain degree of change within the INE, as the following remarks suggest: [The INE contemplates] the possibility of changing the concept of “community”. It is possible that the INE will move on to use the concept of centro poblado indígena. A redefinition of the concept might be a way of overcoming this issue […]. But what is certain is that the INE is not going to permit that a street in Caracas, or in Miranda state, be called “indigenous

246

Chapter Eight community”, [which is a type of request] of which we have already had [instances]. (Allais, 28-07-2011; interview).

The second cause of divergence between the INE and the MINPI is the difference in the nature of their institutional work. The MINPI, with a strong political motivation and a structure well articulated all over the country, has both the interest in and the possibility of capturing the existence of organised groups which self-identify as indigenous outside traditional areas of occupation. That interest is fuelled by several forces. On the one hand, there is the aforementioned ideological NGP frame and its emphasis on “the community” (for the MINPI, the “indigenous community”) and its Communal Council as the basic unit of state political organisation. This relates to the government project of socialist transformation which the MINPI articulates within its sphere of responsibility. At this stage it is clear that the minister Nicia Maldonado strongly adheres to that project as the avenue for the advancement of the indigenous interests she represents. Communal Councils are indeed essential institutions for the NGP, but in addition they have played a central role in the government discourse around the concept of “popular power”; albeit in principle non-partisan administrative units, these Councils have been strongly associated with the revolutionary process and the Partido Socialista Unido de Venezuela–PSUV (United Socialist Party of Venezuela). Communal Councils can be created in both urban and rural locations, and there are specific rules for the creation of these Councils in indigenous communities, according to the Organic Law of Communal Councils (2009). In urban centres a Council can be created in any given self-recognised community with a population base between 150 and 400 families, and in rural communities with 20 or more families. Thus, a sector of a city or a large town with a considerable concentration of indigenous population in one of its neighbourhoods could create a Communal Council observing the stipulations of the law, and in these cases the MINPI is recognising a type of “indigenous Community Council”. In these cases, the Councils obviously do not spring from an “indigenous community” in the sense that it is understood, for instance, by the INE. That is a potential source of divergence between those state organs. On the other hand, there is the power struggle that the MINPI maintains with certain civil indigenous organisations. The creation of the MINPI put an end to the factual power that some of these organisations had gained as the only existing connection between the state apparatus and a part of the indigenous population. For certain types of state policy, those organisations were indispensable mediating actors between the national polis and the indigenous polities. Today the MINPI, with its regionalised

National Census and Indigeneity in Venezuela

247

structure, enables the government to articulate its policies in relation to the indigenous population without necessarily resorting to those civil society organisations (Angosto, 2008: 19-21). This fact, united to the strategic perspective articulated by the ministry, which prioritises productive development and insertion into the nation through Communal Councils rather than free determination within demarcated and officially recognised indigenous territories, has alienated a part of those organisations, who do not see themselves supported or represented by the ministry. The MINPI has tried to side-step the potential threats of this rift by seeking political support among indigenous population sectors that openly share its approach to development and which are not under the influence of the old organisations. The urban indigenous population, and in general that which is now (re)creating their indigenous identities in areas of non-traditional occupation, has thus become the predominant political supporters of the ministry line. The existence of this pool of potential political allies removed from the classical “indigenous community” sphere also contributes to explaining the MINPI’s approach to that concept. Beyond the conceptual and practical challenges around “the community”, the ideological differences between the INE and the MINPI are also reflected in the number of indigenous peoples recognised by those two state organs, as we will see next.

5. The number of indigenous peoples in Venezuela Oscillations of indigenous peoples’ numbers in censuses and other counts originating in state organs are a good illustration of the influence of political context and conceptual and methodological arrangements over results. That number of “peoples” is even more dependent on political factors than the total population numbers, upon which demographics have a more regular impact. The abrupt appearance of a people in the census records, for instance, can never be explained solely by demographics. This section compares results from official censuses and other surveys made by state organs since the 1970s. In Table 8-3 there is a list of peoples registered in the following sources: 1) DICA’s estimate (1974); 2) official indigenous census (1982); 3) official indigenous census (1992); 4) official indigenous census (2001–preliminary results); 5) LOPCI (2005); 6) MINPI. These lists shall be approached as a reflection of political context and census politics on census results. One of the points worth noting is the difference in numbers between the lists derived from censuses produced before and after the CRBV was approved in 1999. Between 1974 and 1992 the variance is certainly small

248

Chapter Eight

if we compare those results with the ones that are to come in the 2000s. The total number of indigenous peoples registered remains in between 28 and 30, if we take into account that in the 1992 round, when 32 were listed, there was a clarification that three of them, concretely the Japreira, Kuiva and Maco, were classified as sub-groups of the Yukpa, the Guajibo and the Piaroa respectively. The official publication of results of the census of 1992 pointed out that there were only 29 “traditionally inhabiting in national territory”. The official lists of the censuses of 1982 and 1992 are practically the same, save for the inclusion of the Arawak and Ñengatu in the latter and the separate mention of the Makuchi and the Japreira in the former. This proximity in results between different rounds vanished with the completion of census 2001, and from then onwards the number of indigenous peoples registered has been above 40. What is the difference between one set of results and the other? When comparing methodologies, one finds in the first place a different approach to indigeneity. In 1974 and 1982 there were externally-imposed criteria for the definition of indigeneity and also a clear territorial limitation (see Sections 2.1 and 2.2). In 1992 things changed substantially with the introduction of the self-ascription criterion and the inclusion of some urban areas in the census coverage (see Section 2.2). The number of registered indigenous people increased but, nevertheless, it was far from matching the number resulting from the subsequent count (see Table 8-3). In 2001 the criterion of self-ascription operated too, but this time with national coverage. This is the most substantial methodological difference, but it seems to be secondary if one attends to the existence of powerful contextual political influences over citizenry. The impact of constitutional recognition and the wave of guaicaipurismo is evident in the census of 2001. In 2001, the number of registered indigenous peoples jumped to 54, though they constitute a list that did not go through a final process of revision for an official publication of results (never produced for the indigenous census 2001). Thus the preliminary list of 54 peoples has been treated as an official INE list and reproduced in a wide range of publications. That number of peoples is indicative of the predisposition that existed in Venezuela to self-identify as indigenous in that census round. Although official results were not finally published, one can analyse what happened in relation to the approval of the LOPCI in order to unveil the state’s approach to those results. The LOPCI (2005) included a list of indigenous peoples which was approved by the National Assembly, thus obtaining political legitimacy. In-depth analysis shows that that list was basically the one from the 2001

National Census and Indigeneity in Venezuela

249

census after receiving a triple distillation (never publicly revealed or openly conceptualised): first, there was a filtering of what one may carefully call “abnormal” cases and naming (in)correctness (thus Pygmy, Waika and Caribe can be eliminated); second, there was the filter of nationality, which would later become explicit in the census questionnaire in 2011, as explained in Section 3.3.2 (thus Chibcha, Kechwa and Makuchi can be eliminated); thirdly, there was the filter that eliminated all those denominations which only appeared in the general census (therefore denominations derived from individual acts of self-ascription) and also those denominations that, though appearing in the community questionnaire, did so in mixed settlements in which such self-ascribed denominations were a minority: Caquetio, Gayon, Guanano, Jirahara, Matako, Piritu, Tukano, Tunebo, Waikeri, Wapishana34 could thus be eliminated from the list to be ratified by the National Assembly. A separate look at the census results 2001 for all these peoples who were not included in the list of the LOPCI is clarifying: Table 8-2. 2001 census results for denominations not included in the list of indigenous peoples backed by the LOPCI Indigenous people Caquetio

Total number 10

In census of communities

In general census 0

10

Caribe 165 3 162 Chibcha 2 0 2 Gayon 5 0 5 Guanano 6 4 2 Jirahara 14 0 14 Kechwa 1 0 1 Makuchi 83 81 2 Matako 1 1 0 Pigmeo 1 0 1 Píritu 236 1 235 Tukano 11 2 9 Tunebo 11 0 11 Waika 78 52 26 Waikeri 2,839 1 2,838 Wapishana 17 16 1 Source: prepared by author from INE data (provisional results census 2001)

250

Chapter Eight

The filter of nationality was however ambiguously applied. The Inga, coming from Colombia but with a relatively large group settled in Venezuela for a considerable time (to the extent that some of them have been born in the country), achieved National Assembly recognition as one of the “existing and identified” indigenous peoples in Venezuela. This case is most exceptional, considering that the Inga were not targeted by the census of communities. It constitutes a clear example of the scope for political negotiation and decision existing before apparently technical issues. It is also to be noted that the list approved for the LOPCI included one denomination that had not been registered in 2001 (Amorua), and that in its final provision the law clarifies that “The enunciation of these indigenous peoples does not imply the negation of the rights and guarantees, nor the undermining of the rights that other indigenous peoples not identified in this very law might have”. The list of the MINPI incorporates four peoples who do not appear in the LOPCI’s list. Of these, two correspond to peoples located in the Andean Mérida state (Guazabara and Quinaroe), and the other two are the Camentza (Barinas state) and the Gayón (located in Lara, Portuguesa and Falcón). It is interesting to note that these peoples are all located in federal states that are not included in the “traditional areas of occupation”, and therefore states in which up until now the census enumerators did not apply the so-called questionnaires of communities. Besides the political processes described in Section 3 as stimuli for indigenous selfidentification, it is worth remarking here on the effects that the procedures for the constitution of Communal Councils are generating. These procedures stimulate the recovery of local history and on occasion are contributing to providing foundations for processes of ethnic revitalisation. A good example is provided by the work conducted around the community La Loma de los Indios (Portuguesa state), where the collection of local history within the process of documentation gathering for the registration of its Communal Council has been used to support the reconstruction of a now self-proclaimed indigenous identity35. Let us now look at the connection between inclusion in the national census and the enjoyment of differentiated rights in Venezuela.

6. What is at stake Censuses are widely recognised as essential tools for effective public policy making and as symbolically important for minority groups, but of course they may have other properties. For indigenous population in

National Census and Indigeneity in Venezuela

251

Venezuela, appearance in the census is also key for the enjoyment of certain constitutional rights and practical benefits. Two of these rightsrelated questions will be examined next: 1) access to collective property of territory; 2) access to political representation in deliberative state organs.

6.1. The census and access to collective property over territory At certain points in its history, the census has also become the only source of authority to determine which are and which are not indigenous “lands and habitat”. Given the current stagnation of the process of indigenous territorial demarcation and titling (Caballero, 2007; Angosto, 2010), government agencies that, observing the legislation in force, work on the basis of the existence of those territories have resorted to the censuses to determine which areas of the national territory are to be considered “indigenous habitat” (Angosto, 2006: 166). For instance, the Dirección de Asuntos Indígenas–DAI (Direction of Indigenous Affairs), the state organ coordinating indigenous affairs until the creation of the MINPI in 2007, explicitly established in its “Norms and Criteria” for granting fieldwork permits for indigenous territories that it would resort to the national census in order to determine which were such territories. Insofar as for census officials there were only “indigenous territories” where “indigenous communities” were located, and taking into account that from the institutional perspective of the INE these communities were only to be found in regions that were labelled beforehand as regions with “traditional occupation” due to a census tradition based on academic advice, the DAI and other state agencies were in fact engaging in a sort of bureaucratic tautology. Along the same lines, the Ley de Demarcación y Garantía del Hábitat y Tierras de los Pueblos Indígenas (2001) (Law of Demarcation and Guarantee of Indigenous Peoples’ Habitat and Lands) set boundaries to its territorial scope of enforcement by stating that “[t]his Law shall be applied in the regions identified as indigenous within national territory, according to the last indigenous census and other referential sources which identify them as such” (Art. 13, my emphasis). The census thus became an initial filter for any indigenous people or community that wishes to obtain titles of collective property over the land they occupy or claim. Although this “catch 22” effect of the census was overcome in the LOPCI (2005), whose territorial scope of application is national (without restriction) and even contemplates the possibility of restituting lands to indigenous peoples or communities which had been violently or forcefully displaced (Art. 24), it is evident that the census (and therefore the census designers) determined

252

Chapter Eight

the options of a given community (and therefore a given population) to be legally treated as legally indigenous. However, the census was not only affecting the possibility of being considered a legitimate applicant in the process of obtaining collective titles over (indigenous) territory; it also conditions the right to political representation that the CRBV grants to indigenous peoples.

6.2. Political representation The LOPCI displaced the determinant role of the census from the process of indigenous territory recognition, but maintained it as a determinant for other rights-related questions. In its “First Transitory Disposition” the LOPCI established that, as regards the election of the indigenous representatives that the CRBV guarantees for deliberative state organs, the constitutional transitory arrangements would be in force until a specific law regulated the process. The relevant constitutional arrangement (Seventh Transitory Disposition) stipulated that “for the effects of indigenous representation in Legislative Councils and in the Municipal Councils of the states and Municipios with indigenous population, the official census of 1992 of the Central Office of Statistics and Informatics will be taken as a referent”. Ten years later, the Ley Orgánica de los Procesos Electorales–LOPE (2009) (Organic Law of Electoral Processes) established in this regard that any federal state whose indigenous population is 500 or above will elect an indigenous representative (Art. 151). Although this regulation has not been enforced yet (since so far only the federal states with “traditional” indigenous occupation according to the census parameters chose such representatives), it is obvious that the official census results will be maintained as the authorised source to validate population quotas.

7. Conclusions Venezuelan censuses have always labelled a part of the national population as “indigenous”, although the characteristics of the population categorised under that rubric, along with the political motivations behind that categorisation, have been substantially different over time. From the state’s perspective, a part of the indigenous population was always conceptually situated “outside” mainstream national society. When in 1873 the first modern national census was undertaken, segregation was articulated around the conceptual poles of the “civilised” and the “uncivilised”. The population categorised under the latter rubric was

National Census and Indigeneity in Venezuela

253

obviously not recognised as a part of the republican polis, and the task set for the state, inspired by the values of the liberal project, was to incorporate that population into “civil” society. This state approach to indigenous peoples in fact spanned over a century, though of course with practical modulations and different ideological legitimisations: in spite of the influences of regional indigenismo, with its particular advocacy of indigenous cultures, it was only in the 1970s that that approach would be essentially transformed through a new paradigm that affected, among other things, the design and implementation of national censuses. In the 1970s there continued to be, from the state’s perspective, categorically distinguishable types of indigenous population. But with the new ideological paradigm there was a revalorisation, built upon the discursive source of “tradition”, of that part of the indigenous population that had still resisted integration into mainstream national society (precisely the part previously regarded as “uncivilised”). “Traditional” indigenous population started to gain a new status within national society and, gradually, the state was driven towards shifting its general policy from the focus on integration to the focus on recognition: at least in theory, it would start assuming a sense of responsibility for protecting this population and its (traditional) forms of life. Accordingly, the definition of indigeneity in the censuses of the period was built upon externally-decided “anthropological” criteria. The examination of the censuses conducted in the 21st century indicates that, in practice, the distinction between categorically separate indigenous populations has not disappeared from state administrative practice and from national ideology, in spite of the introduction of subjective forms of identification such as the self-ascription criterion. For indigenous population this has been a door to recognition and also a basis for substantiating, through national statistics, claims for the re-orientation of state policies of redistribution, which only exceptionally addressed social inequality with ethnic substrata. However, it is clear that the possibility of self-identification in censuses does not imply that the state renounces its own forms of “objective” definition to discern what part of the national indigenous population can claim the differentiated rights that the CRBV grants to indigenous peoples. It was shown how, despite the existence of self-ascription as a central census criterion for ethnic (and now also racial) identification in the whole of the national territory, for the time being only the part of the indigenous population that inhabits certain regions and matches cultural traits that the INE associates with “indigenous tradition” receives distinctive administrative attention.

254

Chapter Eight

The diachronic analysis of Venezuelan censuses also buttresses the thesis that the fluctuations in official indigenous population numbers are to be understood, in the first place, as reflections of changing political climates. The influence of political context over technical decision in census programs was emphasised throughout the chapter, but the 2011 census in particular was approached as an immediate scenario in which that thesis could be more closely examined. There was an identification of two contextual political forces that influenced census implementation and facilitated indigenous self-identification in that census round. One of them is guaicaipurismo, a current symbolic and discursive production that, inspired by a lax imaginary of indigeneity, plays a central role in the process of national identity reconstruction that has been started with the so-called Bolivarian process. The other one is the new geometry of power (NGP), a term which conceptually frames a governmental project of political and economic power redistribution that is underlined by a potential transformation of the territorial administrative units of the nationstate. It was additionally argued that, as a political referent for the praxis of the MINPI, the NGP influences a conception of “indigenous community” that challenges the existing working definition for that concept at the INE. This clash of conceptions within different public organs demonstrates that the state apparatus is not as ideologically monolithic as some theoretical perspectives assume it to be, and also that there is a relatively open terrain for political negotiation and the readjustment of paradigms in forthcoming census programs. In the identification of concrete conceptual and methodological shifts in the census of 2011 there was also mention, among other things, of a significant “nationalist” shift in the conceptualisation of indigeneity. For the first time in the country’s history, indigeneity was only associated with nationals, a fact in principle associated with technical issues but which is also symptomatic of the challenges that are posed to states when they recognise differentiated rights to indigenous peoples on the bases of identity. The exploration of causes underlying the variances in the numbers of indigenous peoples registered and/or recognised in state records of different historical contexts contributed to shedding light on how decision-making within state organs is often regulated by unwritten hegemonic views. Though presented as devoid of ideological leanings and springing from “common sense”, those views are in fact heavily charged with political meanings and implications. The final section showed how, in fact, for indigenous peoples there is (even) more at stake in censuses than prospects for more equitable redistributive policy making and symbolic recognition; given the centrality

National Census and Indigeneity in Venezuela

255

of the national counts to determining who can legitimately claim to be indigenous, the enjoyment of certain differentiated rights such as the access to collective titles of property over territory and political representation in deliberative organs of the state depends on census data. Table 8-3. Indigenous peoples registered in Venezuela according to different official census and other sources Indigenous people Akawayo/Akawaio/Akawai Amorua Añú/Paraujano Arawak/Arawako Ayamán Baniva Baré Barí Camentza Caquetío Caribe Chaima Chibcha Cumanagoto/Kumanagoto Curripaco/Kurripaco Eñepá/Panare Gayón Guanono/Guanano Guazabara Hoti/Hodi/Jodi/Chikano Inga Ingariko Japreira Jivi/Jiwi/Guajibo Jirahara Kariña Kechwa Kubeo/Cubeo Kuiba/Kuiva Mako/Maco/Maku

1974 X X

1982 X

1992 X

200136 X

X

X X

X X X

X X X

X X X X X X

X X X

X38 X

X X X X X X X X X

2005 X X X X X X X X

201137 X X X X X X X X X

X

X

X X X

X

X

X X

X X

X X X X X X X X

X

X X

X39 X

X X

X X

X

X

X

X X X X X X X X

X

X

X X X

X X X

X

X

X X

X X

X

X

X40 X41

256 Indigenous people Makushi/Makuchi Mandahuaca Matako Ñengatú/Yeral Patamona Pemón Piapoco/Piapoko Pigmeo Píritu Puinave/Puninave/Punave Pumé/Yaruro Quinaroe Sáliva Sanemá/Shirianá Sapé/Sape Timotes Tukano Tunebo Uruak/Urak/Arutani Waika Waikerí Wanaí/Mapoyo Wapishana Warao Warekena/Guarekena Wayuu/Guajiro Wotuha/Wotjuja/Piaroa Yanomami Yarabana/Yabarama Yaratani42 Ye’kwana Yuana43 Yukpa TOTAL NUMBER

Chapter Eight 1974 X X

1982 X

1992

X X X

X X

X X

X

X X

X X

X

X X X

X X X

X

X

X

X

X

X

X X X X X X X

X X X X X X X

200136 X X X X X X X X X

2005

201137

X

X

X X

X X

X X X X X X

X

X

X

X

X X X X X X

X X X X X X X X X X X X X X X X X

X X X X X X X

X X X X X X

X X X X X X

X

X

X

X

X

X X X X 32 54 40 44 (29)44 Source: prepared by the author from different sources (DICA’s estimate, 1974; official indigenous census, 1982; official indigenous census, 1992; official indigenous census, 2001 (preliminary results); LOPCI, 2005; MINPI). 28

X 29

National Census and Indigeneity in Venezuela

257

Bibliographical references Acosta Saignes, Miguel. 1948. Noticia sobre el problema indígena en Venezuela. Caracas: Imprenta Nacional (Publicaciones de la Comisión Indigenista). Allais, María Luisa. 2011. Visibilización de la población indígena en los censos de Venezuela a partir de 1980. Paper presented at the international seminar Avances en la visibilización de los pueblos indígenas en América Latina y el uso de información para políticas públicas. Cartagena de Indias (Colombia), 2-4 August. Allais, María Luisa. 2004. La población indígena en Venezuela según los censos nacionales. Paper presented at the Segundo encuentro nacional de demógrafos y estudiosos de la población. Caracas, 24-26 November. Amodio, Emanuele. 2007. La República Indígena: pueblos indígenas y perspectivas teóricas en Venezuela. Revista venezolana de economía y ciencias sociales, 13 (3): 175-188. Angosto, Luis Fernando. 2010. Pueblos indígenas, multiculturalismo y la nueva geometría del poder en Venezuela. Cuadernos del Cendes, 73: 97-132. Angosto, Luis Fernando. 2008. Pueblos indígenas, guaicaipurismo y socialismo del siglo XXI en Venezuela. Antropológica, LII (110): 933. Angosto, Luis Fernando. 2006. Negotiating polities, ethnicity and citizenship in the Pemon community of Tuauken: an analysis of the relations between the Bolivarian revolution and indigenous peoples in Venezuela. PhD thesis, Anthropological Studies, Queen’s University of Belfast. Brysk, Allison. 2000. From tribal village to global village: Indian rights and International Relations in Latin America. Stanford: Stanford University Press. Caballero, Hortensia. 2007. La demarcación de tierras indígenas en Venezuela. Revista Venezolana de Economía y Ciencias Sociales, 13 (3): 189-208. Clarac, Jacqueline. 2001. Análisis de las actitudes de políticos criollos e indígenas en Venezuela (de los años 60 hasta el 2001). Boletín Antropológico, 53: 335-372. Cojtí, Demetrio. 2009. Marco conceptual para enfocar estadísticas de los pueblos indígenas. Guatemala: Instituto Nacional de Estadística. www.ine.gob.gt/np/generoypueblos/documentos/Marco_conceptual_p ueblos_indigenas.pdf.

258

Chapter Eight

Del Popolo, Fabiana. 2008. Los pueblos indígenas y afrodescendientes en las fuentes de datos: experiencias en América Latina. Santiago de Chile: CEPAL – Naciones Unidas. Díaz Polanco, Héctor (1982 [translated by S. Gorman]). Indigenism, populism and Marxism. Latin American Perspectives, 9(2): 42-61. Gundermann, Hans, Vergara, Jorge Iván and Rolf Foerster. 2005. Contar a los indígenas en Chile. Autoadscripción étnica en la experiencia censal de 1992 y 2002. Estudios Atacameños, 30: 91-115. Hooker, Juliet. 2005. Indigenous inclusion/black exclusion: race, ethnicity and multicultural citizenship in Latin America. Journal of Latin American Studies, 37: 285-310. INE–Instituto Nacional de Estadística. 2011. XIV Censo Nacional de Población y Vivienda. http://www.ine.gob.ve/censo2011/index/htm. Acessed 12-12-2011. Linárez, Pedro Pablo, Alvarado Carlos and Mendoza, Heiler. 2007. La Loma de los Indios: historia. Document online available at http://enlaceindigenas.gob.ve/doc/PuebloIndigenaGayon.doc. Accessed 12-11-2011. Llorens, José A. 2002. Etnicidad y censos: los conceptos básicos y sus aplicaciones. Bulletin de l’Institut Français de d’Études Andines, 31 (3): 655-680. Marroquín, Alejandro D. 1977 [1972]. Balance del indigenismo: informe sobre la política indigenista en América. Mexico: Instituto Indigenista Interamericano [Ediciones Especiales: 76]. Nava, Julián. 1965. The Illustrious American: the Development of Nationalism in Venezuela under Antonio Guzmán Blanco. The Hispanic American Historical Review, 45(4): 527-543. OCEI–Oficina Central de Estadística e Informática. 1985. Censo Indígena de Venezuela 1982: nomenclador de comunidades y colectividades. Caracas: Taller Gráfico de la OCEI. OCEI–Oficina Central de Estadística e Informática. 1993. Censo Indígena de Venezuela 1992: nomenclador de comunidades y colectividades. Caracas: Taller Gráfico de la OCEI. Pérez, Luis. 2004. Educación superior indígena en Venezuela: una aproximación. Caracas: IESALC/UNESCO. Pueblos indígenas de Venezuela. 2009. Pueblos de varias regiones. N°15. Caracas: Santillana. Samudio, Edda. 2005. La propiedad territorial de las comunidades indígenas en la política agraria de Venezuela a finales de lo siglos XVIII y XIX. In Germán Cardozo and Arlene Urdaneta (ed.)

National Census and Indigeneity in Venezuela

259

Colectivos sociales y participación popular en la independencia hispanoamericana. Maracaibo: Ediciones de la Universidad de Zulia. Van Cott, Donna Lee. 2002. Movimientos indígenas y transformación constitucional en los Andes. Venezuela en perspectiva comparativa. Revista Venezolana de Economía y Ciencias Sociales, 8(3): 41-60 Caracas. Wright, Withrop. 1990. Café con leche: Race, Class, and National Image in Venezuela. Austin: University of Texas Press.

Legislation CRBV–Constitución de la República Bolivariana de Venezuela. 1999. Gaceta Oficial Extraordinaria de la República Bolivariana de Venezuela n° 5.453 del 24 de marzo de 2000. Constitución de la República de Venezuela. 1961. Gaceta Oficial 662 Extraordinario del 23 de enero de 1961. Ley de Demarcación y Garantía del Hábitat y Tierras de los Pueblos Indígenas. 2001. Gaceta Oficial de la República Bolivariana de Venezuela nº 37.118 del 21 de enero. Ley de la Función Pública Estadísitica. 2001. Gaceta Oficial de la República Bolivariana de Venezuela n° 37.321 del 09 de noviembre de 2001. LOPE–Ley Orgánica de Procesos Electorales. 2009. Gaceta Oficial de la República Bolivariana de Venezuela nº 39.240 del 12 de agosto de 2009. LOPCI–Ley Orgánica de Pueblos y Comunidades Indígenas. 2005. Gaceta Oficial de la República Bolivariana de Venezuela nº 38.344 del 27 de diciembre de 2005.

Interview Allais, María Luisa (28 July 2011): transcription of recorded interview (audio). Personal residence of interviewee, Caracas, Venezuela. Recorded by author.

Notes 1 I should like to thank Maria Luis Allais and Michelle Cañizález (INE Caracas) for generously sharing their experiences in and their views on the carrying out of contemporary Venezuelan censuses. I must also thank Sabine Kradolfer for enriching this work with several rounds of comments on earlier drafts.

260

Chapter Eight

2 In this work the noun “indigeneity” refers to “being indigenous”. Understood as a socially constructed category, it is conceptually associated with belonging (or relating) to one of the peoples that inhabited or claim to have inhabited the American continent prior to the European invasions started in the fifteenth century. In this conception, “indigenous people” is tantamount to “originary people” as a people pre-existing the constitution of nation-states in the region. In Latin America today, indigeneity as a social category is central to debates around democratic forms and the existence of collective rights based on cultural difference, but is also pivotal in processes of collective identification at play in spheres as varied as that of social movement activism and statecraft. 3 The difference between these two state organs is essential: the Ministry is an organ of the Executive Power that directly articulates government dispositions and the INE is an autonomous state entity that, albeit legally ascribed to the Ministry of Planning and Development, gives more scope to independent criteria based on notions of technical expertise. The Ley de la Función Pública Estadística (Law of Statistical Public Function [2001]) grants the Institute distinctive juridical personality as well as its own patrimony (Art. 49), which guarantees a minimum degree of institutional autonomy and negotiating strength before other state organs that participate in the production and use of national statistics. 4 These relative percentages were calculated, until the census of 1942, considering the indigenous population as already integrated in the national population numbers. From 1950 to 1992 the indigenous censuses were conducted separately from the general ones, so the calculation of the relative percentage was made by adding the indigenous population to the national one in order to obtain a country’s total population in the census rounds of that period. Given the difference in conceptual definition, geographical scopes and timing of the surveys, these relative percentages are exclusively to be used as an indication of the data available to the state in each period. This table does not (and could not) claim validity for comparative demographical studies. 5 The general census was conducted in 1981 and the indigenous one in 1982. 6 The general census was conducted in 1990 and the indigenous one in 1992. 7 Even nowadays, when the option for self-ascription to the indigenous category is open to every Venezuelan anywhere in the country, there is still a factual distinction between regions that are considered to contain population in “traditional settlements” and others that are not. Only in the former regions do the enumerators apply the so-called “Questionnaires for Indigenous Communities/Populated Centres”, which capture information not gathered by the general census questionnaire even when the respondents self-identify as indigenous. 8 This term has been used by activists and analysts to refer to processes of demographic reduction or elimination of ethnic groups in state records such as censuses and surveys (cf. Cojtí, 2009). The Declaration of the World Conference against Racism, Racial Discrimination, Xenophobia and Related Intolerance (held in Durban, South Africa, in 2001) established the basis for considering these administrative practices as a manifestation of racism. See the Declaration at www.un.org/WCAR/durban.pdf [accessed 12-12-2011].

National Census and Indigeneity in Venezuela 9

261

In the colonial period, lands granted to indigenous communities by the Spanish crown through titles that permitted collective usufruct. 10 See Ficha Técnica del Censo de Comunidades Indígenas of the INE at http://www.ine.gov.ve/index.php?option=com_content&id=352&Itemid=26&view =article%20%20%20%20 [accessed 07-05-2012]. 11 The Instituto Indigenista Interamericano was created in 1940 after the Congress of Pátzcuaro (Mexico). 12 My translation from the Spanish original. I have also translated into English all the subsequent literal quotations from sources in Spanish. 13 With the sole exception of Sucre state, now incorporated into that group. Today, the eight federal states with “traditional occupation” are: Amazonas, Anzoátegui, Apure, Bolívar, Delta Amacuro, Monagas, Sucre and Zulia. In the 1970s Amazonas was not yet a federal state and was administratively regarded as Territorio Amazonas. 14 In Table 8-3 there are 32 denominations, but it is to be noted that the Japreira, Kuiva and Maco were included only as sub-groups of other peoples. See also notes 39 to 44. 15 Venezuela finally ratified this Convention in October 2001. 16 The term guaicaipurismo has been chosen after the figure of Guaicaipuro, the most praised leader of the indigenous resistance against Spanish colonisers in the sixteenth century. 17 Aloha Núñez, vice-minister of the MINPI for Urban Areas, illustratively declared that “[after census 2001] we said that there were around 500 thousand [indigenous] citizens, but we know that they are now many more, approximately a million”. See “Censo 2011 llegará a pueblos indígenas”, in Últimas Noticias. Available at http://www.ultimasnoticias.com.ve/noticias/ciudad/serviciospublicos/censo-2011-llegara-a-pueblos-indigenas.aspx [accessed 02-10-2011]. 18 “Land and habitat” are the terms used in the CRBV and derived legislation to refer to indigenous peoples’ “territories”. 19 In Spanish the question was: “Según sus rasgos físicos, ascendencia familiar, cultura y tradiciones se considera: Negra/Negro – Afrodescendiente – Morena/Moreno – Blanca/Blanco – Other (which one)”. The new question generated some debate, particularly intense around the supporters of the term “afrodescendiente” and the supporters of the term “negro” to capture ethnic selfascription. 20 The wording of the question was grammatically confusing and should have rather been “¿Pertenece a algún pueblo o etnia indígena?” (instead of the existing “¿Pertenece a algún pueblo indígena o etnia?”). Asking if one belongs to an indigenous people or to an ethnie can be misinterpreted: without an adjective that indicates otherwise, an ethnie could also include all those who are not considered “indigenous” ethnies. 21 Sinamaica is the name of a lagoon and a community in municipio Páez, Zulia state, where indigenous Añú population resides. 22 According to the Chilean census of 1992 the indigenous population was 10.33% of the national one, whereas in 2002 it went down to 4.58%.

262

Chapter Eight

23 Those preliminary results have often been published with no critical comments. See, for instance, scientific articles such as Amodio (2007) or Pérez (2004); or, with some commentary, its inclusion in divulgatory publications such as “Pueblos indígenas de Venezuela: pueblos de varias regiones” (2009), which allegedly counted on an “anthropo-linguistic revision”. 24 See, for instance, El Nacional, 21-08-2011, p. 3. 25 Name of an alliance of political actors who agreed to collaborate in order to present a common candidate to challenge Hugo Chávez in the presidential elections of 2012. 26 See Correo del Orinoco, 18-09-2011, p. 8. 27 Venezuela largely relies on the Principles and Recommendations for Population and Housing Censuses of the United Nations. 28 See Últimas Noticias, 04-09-2011, p. 9. 29 Name of an alliance of political actors aligned with Hugo Chávez for the presidential elections of 2012. 30 This type of referendum was one of the pivotal novelties in the model of Participative Democracy that the CRBV promoted. It makes possible the revocation of any elected officer when half of the period for which she/he was elected is completed (Art. 72, CRBV). 31 The shadow of the lista tascón was not only tacitly operating, but also openly mentioned in press and conversation. See for instance articles of opinion such as that entitled “¿Atiendo al censo 2011?” published by the conservative newspaper El Universal. Available at http://www.eluniversal.com/2011/09/02/atiendo-alcenso-2011.shtml [last accessed 16 September 2011]. 32 There are currently 34 peoples with which the user of the service can identify. This number does not coincide with the one listed in the LOPCI nor with the one advanced by the MINPI (see Table 8-3), which illustrates the lack of consensus and coordination among different state organs. 33 See http://www.ine.gov.ve/fichastecnicas/menufichastecnicas.asp [accessed 0210-2011]. 34 The Wapishana dwell in the circum-Roraima region, basically in the south of Guyana and the north of Brazil, so they present a case in which the second filter (that of “nationality”) could have also been applied in the process of clearing the list of the census 2001. 35 See Linárez, Alvarado y Mendoza (2007), available at http://enlaceindigenas.gob.ve/doc/PuebloIndigenaGayon.doc [accessed 14-122011]. 36 The INE has not produced an official publication of results for the census round of 2001. This is the list of results available at the institution after the initial processing of data. 37 According to the MINPI and its list of the “Indigenous peoples in Venezuela”, which for example was circulated, barely a month before the commencement of the 2011 census, among the attendants to the IV Bolivarian Congress Young Indoamerica – Against Capitalism, War and for Socialism, closed on August 9. 38 This term has often been used to denominate the Kariña, but is also a generic name for all those peoples of the Carib linguistic branch.

National Census and Indigeneity in Venezuela 39

263

Included as a sub-group of the Yukpa. Included as a sub-group of the Guajibo. 41 Included as a sub-group of the Piaroa. 42 With this name in the DICA list. 43 Another denomination for the Hodi/Hoti/Chikano. It is separately included with this name in the list of the DICA. 44 The Japreira, Kuiva and Maco were classified as sub-groups of the Yukpa, the Guajibo and the Piaroa respectively. 40

CHAPTER NINE ETHNIC/RACIAL STATISTICS: BRAZIL AND AN OVERVIEW OF THE AMERICAS JOSÉ LUIS PETRUCCELLI

1. Introduction Analysis of ethnic categories in population censuses may be structured around two main pillars: on the one hand, its role in providing a mirror image of political struggles for representation and, on the other, its use as a state tool for population management. In the case of the latter, like all classification carried out by official institutions, the statistical categorisation of the different ethnic/racial groups in effect constitutes a practical intervention in society. This intervention occurs at two levels: 1. by providing the elements that shape a socially legitimated perception of the groups that make up that society; 2. by contributing data and analysis that not only give an accurate picture of social reality but also assemble the basic information that is essential for developing public policy (Solanke, 2007). Applying an ethnic/social classification system therefore entails measuring a phenomenon that poses several different kinds of difficulties of both a theoretical/conceptual and empirical nature and in which the overlapping of political representations, scientific issues and statistical categorisation becomes increasingly apparent. The aim of this chapter is to provide a historical perspective on the evolution of the different forms of population classification in Brazilian statistics, dating back to the nineteenth century, before going on to carry out a comparative analysis between the different methods employed to investigate the ethnic/racial makeup of the population in the censuses of the countries of the Americas, using the case of Brazil as a point of reference. Brazil has an extensive database and in some ways has come to be seen as a role model on the subject for the Latin American region. To conclude, the article will focus on the current situation and expectations

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

265

for the immediate future with regard to the production of ethnic/racial statistics in the country.

2. Some conceptual clarifications The statistical operation of census-taking has systematically presented ethnic/racial classification as an exclusively technical procedure. However, such classification is, in fact, an attempt to naturalise any physical and/or cultural attributes that may define racial or ethnic groups. In providing the categories of composition of the social world that are deemed “legitimate”, the statistical apparatus tends to delimit understanding of it within a preelaborated frame in response to administrative and state intervention concerns (Simon, 1997). Furthermore, the social construction of ethnic/racial identities is in turn informed by the way in which statistical institutions codify the respective groups. This intersection between the definition of supposedly scientific categories–and the resulting classifications–and the interests of specific groups within the population shows the eminently political nature of the census operation, both in the sense of being a means to help manage the polis and the Aristotelian concept of ordering the diverse parts that make up a community (Skerry, 2001). Despite their problematisation and juxtaposition of meanings, the concepts of (skin) colour, race, ethnicity, origin and ancestry tend to be used as categories for analysing social facts. With regard to why censuses may or may not include a question concerning the ethnic/racial categorisation of the population, there is consensus among a wide range of researchers that the reasons are complex and ever changing (Marx, 1998; Nobles, 2000, Bulmer, 2001; Skerry, 2001; Simon, 2001). According to a study by Ann Morning (2006), these reasons do not always point in the same direction and can be summarised as follows: • inclusion: for the purpose of exercising political control over the different groups; • non-inclusion: in the name of national integration of society; • non-inclusion: discourse of national hybridity; • inclusion: so that anti-discrimination policies can be justified. Thus, since there have already been periods in which racial difference within the population has been used to implement a segregationist social order, more recently it has been put at the service of denouncing inequalities and implementing human rights and affirmative action policies.

266

Chapter Nine

However, methodological and substantive problems continue to challenge the assumptions on which the various classification efforts have been based. The objections appear to stem largely from the persistence within the social sciences of conceptual confusion over the notion of race and its often ambiguous relationship with the concept of ethnic group. While in its current academic sense the term has been definitively distanced from the original nineteenth century biological notion, it is still possible to find references to the perceived physical characteristics of individuals as being the defining elements of reality, thereby naturalising such attributes and thus concealing the fact that physical differences that are deemed significant, those that are perceived to be phenotypic variations, are the result of a specific social and historical construct. Indeed, it was not just any visible, perceptible physical characteristic that was racialised. The historical process of European colonial expansion, slavery and the subsequent relations of domination between peoples and nations conferred specific meanings on certain physical features, later associating them with cultural expressions. And so relations between groups identified as racial or ethnic, when integrated within the context of the different countries, led to a range of different proposals for statistically classifying such groups. It should be stressed that the effect that we call racial (in the sense of the social rather than the biological representation of differences) is, in fact, the result of an extremely complex combination of the visible characteristics of different human populations (hair type, eye shape, skin colour, body morphology) and symbolic operations on the skin which are ethnic or national marks. (Carvalho, 2008: 88).

Meanwhile, it is impossible to ignore the tension that exists between finding an acceptable ethnic/racial classification and at the same time ensuring that it will provide the distinction required by state instruments (Stavo-Debauge, 2005). It has to be said that implementing a specific form of statistical categorisation means confronting the dual challenge of, on the one hand, successfully investigating the social reality of a specific ethnic/racial universe while, at the same time, helping, whether one likes it or not, to shape and legitimise the different names assigned to it. The act of categorisation, when it manages to achieve recognition or when it is exercised by a recognised authority, exercises by itself a certain power: “ethnic” or “regional” categories, like categories of kinship, institute a reality by using the power of revelation and construction exercised by objectification in discourse. (Bourdieu, 1991: 223).

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

267

The two issues that arise in relation to this measuring process are: 1. What to measure: namely, identifying the most relevant aspects–the phenotype or physical appearance alongside origin or ancestry, and which groups to include in the categorisation: blacks, Roma (gypsies), indigenous people, etc. 2. How to measure them: if there is total consensus about using selfclassification as a criterion, in some cases supplementary questions on cultural features and language spoken may accompany it. In the Brazil of the second half of the nineteenth century, the debate around constructing a sense of nationality and national identity was particularly important, reaching a peak during the first Republic (18891929). This debate was strongly influenced by the European racial theories and doctrines of the time and manifested itself in a systematic preoccupation with the multi-racial origin of the Brazilian people, which was seen as a source of social contradictions, an obstacle to the country’s development and a hindrance to the definition of a national identity. The notion of race pervaded all those discourses, nevertheless encompassing views that reflected the broadest possible range of intellectual currents and conflicting opinions. These ranged from racist doctrines, according to which other groups were condemned by a ranking system that rendered them inferior to the white European population, to other categories of identification–blacks, Indians, pretos, pardos, caboclos, etc. [see below for an explanation of these terms]–, and more advanced interpretative discourses, which, as well as distancing themselves from a primitively essentialist view, fostered a culturalist conceptualisation that was more suited to the composition of the society. While the concept of race does not in any way correspond–from the genetic viewpoint–to scientific reality, it does represent a social reality that stems from a common perceptual organisation of references to human diversity, with evidence of its persistence as an extremely effective symbolic reality. Its strength lies precisely in the fact that it relies on a “natural” mark, which is immediately visible to the eye and passed down in a hereditary way, thereby making it possible to discursively generate actual social groups or categories that can be called racial. There is now a recognised consensus that the notion of race permeates all social relations, pervades practices and beliefs and determines the place and status of individuals and groups within society. Thus, people can be identified, classified, ranked, protected or subalternised on the basis of the colour/race/ethnic group or origin attributed to them by whoever is observing them. However, when constructing an official classification system, race, as a concept, poses a problem in essentialising the “types”

268

Chapter Nine

being categorised, which can help to legitimise a supposed order or hierarchy by relying on distinctions that are inscribed in the nature of things or beings. In order to avoid this problem and to overcome the essentialist thinking that sees practices as biological or cultural properties inherent to individuals or groups, reality is conceived as being relational, in other words, the result of a symbolic conflict between the occupiers of unequal positions (Bourdieu, 1980). This structural conflict is inherent to any complex society and manifests itself in the different ways in which the empirical world is categorised and represented. Scientific criteria simply record the state of the symbolic struggle for legitimacy between the different representations. In the case of features that go beyond physical appearance or phenotypic similarities, the incorporation of such differences within ideologies and social practices “confers on them a symbolic meaning, be it positive or stigmatised, forming what we can call race, racial groups or race relations” (Paixão, 2009: 2). In this way, a contemporary definition of race aiming at being useful in the practice of social research should also consider the classificatory implications of the ways of thinking associated with the native or popular use of the term, a legacy of former academic conceptualisations. Thus race should be understood as being a category that has been socially constructed throughout history on the basis of one or more signs or features among the overall characteristics of individuals that stand out culturally: a symbolic representation of identities that have been fashioned from physical, cultural or other referents.

3. The importance of ethnic/racial statistics Traditionally the legal apparatus of a country (its constitution, codes of law, etc.) penalises racial discrimination under civil or criminal law. However, cases of “indirect” or institutional discrimination require statistical tools in order to survey and investigate the apparently neutral social processes and practices that produce discriminatory effects as a consequence of their application (Simon, 2005). Here we are in a field which, though falling within the realm of the construction of social indicators for studying inequality, extends to more conceptual spheres of how to think about the production of statistics. The idea is that ethnic/racial characteristics are attributed and considered surreptitiously, without the consent of the victims of discrimination and often in a way that is not totally apparent for the actual people involved in the appointment of candidates to a particular position in the workplace, for example. In this way, current procedures and practices are discriminatory

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

269

simply because of their collective effects on groups. It is therefore on the basis of statistical data that it becomes possible to provide evidence for and help combat subtle and concealed forms of social practice that produce inequalities. In other words, the indicators render the invisible visible (Simon, 2005).

4. A historical perspective of censuses and surveys in Brazil 4.1. Colour and race in population censuses During slavery terminology was developed for describing the appearance or physical features of the racial characteristics of individuals to be applied, for example, in the event that a slave escaped. The owners published “wanted” notices giving the best possible physical description, including details of variations in skin or hair colour and other noticeable features or marks, such as scars, in order to enable and facilitate identification and recovery of the person who had run away. Based on that terminology an official system for the ethnic/racial classification of the country’s inhabitants in public statistics was established and applied in the first national population census carried out in 1872 when slavery was still in force. The categories used in the census were: white, preto, pardo and caboclo, with “preto” corresponding to the colour black and “caboclo” being used to identify the native populations1. Thus, ever since that first census, data on the colour and race of the Brazilian population have been available. Throughout the 140 years in which these characteristics have been investigated, practically the same categories have continued to be used in relation to that important question, albeit employing a wider range of implementation criteria. However, in this first enumeration exercise, which separated the population according to their civil status into “free men” and slaves, while the former selfidentified as far as their race was concerned, the latter were classified by their owners. In 1890, when the second census took place, among the options given for answering the same question, the term “pardo” was changed to “mestizo” (mixed race) in the classification. At a time when racist doctrines held strong sway, distinctive features, the phenotype, the referent for the 1872 census, gave way to the idea of racial mixtures and mestizaje into the subsequent census classification. This pardo/mestizo category showed a significant increase between those first two censuses, influenced in part by that change (see Tables 9-1 and 9-2).

270

Chapter Nine

In the following two censuses, those of 1900 and 1920, racial classification was not included, while the one scheduled for 1910 did not take place. According to some intellectuals of the time, the continued hegemony in Brazil of racist doctrines of European origin characterised this period by a feeling of inferiority, given the composition of Brazilian society with its African and indigenous legacy. Therefore, it seemed better not knowing the actual racial structure of the population. The search for a national identity was taking place at the same time as proposals were being put forward to solve the black and indigenous “problems” by eliminating those racial population groups, either as a result of whitening, mestizaje with the white group–or even by subjecting them to genocide, directly or by systematically failing to ensure the conditions in which they would be able to reproduce. At the same time, the country was pursuing a selective immigration policy, initiated in the second half of the nineteenth century, favouring immigrants of European or “Caucasian” origin. Preference was given to Swiss, German and Nordic immigrants although, some time later, not without some misgivings, Italians and Spaniards, who were less valued but, from the point of view of the time, “at least white” came to be accepted. This plan for whitening the Brazilian population shows the other face of the mestizaje ideology–an ideology that supports the myth of an encounter of the three founding races of the country’s population and results in an “abstract celebration of the interpenetration of cultures” (Carvalho, 2003: 318) but which in practice denies the possibility of legitimately expressing black or indigenous racial identity. Furthermore, in any case, the mestizo was only accepted “as a transitory element that would lead to the establishment of a nation of whites” (Costa, 2006: 63). Population censuses did not get under way again until 1940 as no statistical enumeration took place in 1930 because of political instability. In the new census the reference to native peoples, who had previously been classified as “caboclos”, was removed and the category “yellow” was added in order to take account of the contingent of immigrants from Japan. Starting in 1908, this steady migratory flow essentially continued until the 1930s, having been allowed despite protests from members of Congress opposed to it on the grounds that, given the “Africanisation” to which the country had already been subjected, its “Mongolisation” could not be tolerated. In the 1940 census questionnaire the terms “white”, “preto” and “yellow” were accepted as answers to the racial classification question but, in the event that any other response was given, the census takers were instructed to put a line in the corresponding space, which was later codified as “pardo”. The results of the 1940 census show how significant

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

271

selective European immigration was for the “white” category, given that its relative share rose from 44% in 1890 to over 63% (Table 9-2). The 1950, 1960 and 1980 censuses retained the same ethnic/racial categorisation, explicitly including the term “pardo”, the 1970 census being an exception in that the classification question was omitted. In the 1991 census, after a 101-year absence, “indígena” (“indigenous”) was reintroduced as a category, with the question heading changing to “colour or race” since “indígena” was supposedly a race and not just a “colour”, as the question was headed in the censuses taken between 1940 and 1980. In 2000 we find all five categories, also used in household surveys, in the census questionnaire in the following order: white, preto, yellow, pardo and indígena. They also appeared in the 2010 census.

Chapter Nine

9,930,478 14,334,215 17,438,434 30,635,605 41,236,315 51,944,397 70,191,370 93,139,070 119,011,052 146,815,791 169,799,170

Total

White 3,787,289 6,302,198 26,171,778 32,027,661 42,838,639 64,540,467 75,704,924 90,647,461

Preta 1,954,452 2,097,426 6,035,869 5,692,657 6,116,848 7,046,906 7,335,139 10,402,450

Coulour/race Parda Yellow 3,801,782 4,638,795 8,744,365 242,320 13,786,742 329,082 20,706,431 482,848 46,233,531 672,251 62,316,060 630,659 66,016,783 866,972

(1) “Caboclo” was used instead of “indígena”. (2) “Mestiza” was used instead of “parda” and “caboclo” instead of “indígena”. (3) No information on colour/race was included. Source: IBGE, demographic censuses 1872-2000.

18721 18902 19003 19203 1940 1950 1960 19703 1980 1991 2000

Year

Table 9-1. Resident population by colour/race, Brazilian censuses 1872-2000

272

Indigenous 386,955 1,295,796 294,131 701,462

Ignored 41,983 108,255 46,604 517,897 534,878 1,164,042

100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0

Total

White 38.1 44.0 63.5 61.7 61.0 54.2 51.6 53.4

Preta 19.7 14.6 14.6 11.0 8.7 5.9 5.0 6.1

Coulour/race Parda Yellow 38.3 32.4 21.2 0.6 26.5 0.6 29.5 0.7 38.8 0.6 42.4 0.4 38.9 0.5

(1) “Caboclo” was used instead of “indígena”. (2) “Mestiza” was used instead of “parda” and “caboclo” instead of “indígena”. (3) No information on colour/race was included. Source: IBGE, demographic censuses 1872-2000.

18721 18902 19003 19203 1940 1950 1960 19703 1980 1991 2000

Year

-

Indigenous

Table 9-2. Distribution of resident population by colour/race, Brazilian censuses 1872-2000

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

0.2 0.4

3.9 9.0

Ignored 0.1 0.2 0.1 0.4 0.4 0.7

273

274

Chapter Nine

5. The indigenous population doubles between 1991 and 2000 The 1990s, at least in the Americas, were characterised by a reaffirmation or rescuing of identities which had been historically discriminated against. Just as the organised civil rights movement representing Afrodescendants succeeded in giving new meaning to the term “black” so that it came to denote pride in one’s ethnic origin, indigenous organisations also re-evaluated their identity. As well as being part of a broader continental movement, in Brazil this process also grew out of opposition to the military dictatorship’s geopolitical project during the 1970s to emancipate the Indians. Basically designed to expand the agricultural frontier in the Amazon region, the aim was to de-characterise indigenous groups by means of discriminatory legal differentiation. In 1980, FUNAI, the state body that in theory was supposed to look after indigenous interests in the country, proposed the establishment of supposed “Indianness criteria” (“Criterios de indianidad”) in which indigenous people were to be characterised as people who exhibit “primitive mentality, undesirable biological, psychological and cultural characteristics, the presence of a Mongolian or sacral spot, anthropometric measurements, psychosocial maladjustment, etc.” (CIMI, 2011: 1), a document which was widely rejected by Indians, anthropologists and indigenous associations (Carneiro da Cunha, 1992), resulting in the dropping of the proposal. In fact the effect obtained was the opposite to what was wanted since it can be said that it led to the “reindianisation” of several indigenous communities. But being Indian cannot be summed up in a question merely based on the appearance of constructed stereotypes but rather on a way of being, an “incessant [movement] of differentiation, not a massive state of preexisting stabilised ‘difference’, namely an identity” (Viveiros de Castro, 2006: 2). However, “indigenous” does not refer to an individual attribute but to a collective or relational movement in which “‘indigenous’ identity is ‘relational’ not only ‘in contrast’ to ‘non-indigenous’ identities but relational [...] first and foremost because it comprises intra-referenced and intra-differentiated transindividual collectives” (Viveiros de Castro, 2006: 4). There are indigenous individuals because they are members of indigenous communities, and not the other way round. The components of such communities have begun to reinvent “a culture and way of life–a relational world which, however restricted it may have been by the adverse conditions in which it was cultivated, never stopped being an expression of human life just like any other” (Viveiros de Castro, 2006: 8).

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

275

And in that regard the same author contends that there are no unauthentic cultures, given that there are no authentic ones either. Hence no Indians, whites or Afrodescendants, or anyone else, can be seen as authentic or unauthentic. A visible consequence of these processes of identity affirmation was that, in the period between the 1991 and 2000 censuses, there was an extraordinary increase in the population who self-identified as indigenous, rising from less than 300,000 to over 700,000 (Table 9-1), in other words, a two and a half times increase, with its importance relative to the population as a whole rising from 0.2% to 0.4% (Table 9-2). The only possible explanation for this increase, which by far exceeds the chances of natural growth, is the above-mentioned process of the rescuing of previously relinquished identities or identity revaluation, which, in Brazil as well as in other regions, has also been witnessed in relation to the categories that represent Afrodescendants.

6. Ethnic/racial classification: the most detailed studies In Brazil there have been three data-gathering exercises to try to increase knowledge of the categories used to classify the population by race in surveys and censuses: the first was the 1976 Pesquisa Nacional por Amostra de Domicilios–PNAD (Household Survey), the second was the July 1998 Pesquisa Mensal de Emprego–PME (Monthly Employment Survey) and the third was Pesquisa das características étnico-raciais da população–PCERP (Survey of the Ethnic/Racial Characteristics of the Population) in 2008, which will be looked at in more detail later. The results obtained by spontaneously asking people about their colour or race in the cases mentioned provoked widespread debate about how many classification categories were in fact being used in the country and how many categories would best depict the state of racial diversity in Brazil. The number of answers received, over one hundred in each case, prompted doubts about whether it was possible to construct a representative system of racial classification to be applied in censuses and population surveys, with critics claiming that it would be too complicated and that no one would be able to say who is who in this country. According to this view, supposedly grounded in the recognised mestizo origin of the Brazilian population and the lack of defined boundaries between the racial categories, it would be impossible to properly establish the identity of those who are Indian, black, etc. The spectre of the mestizaje ideology reared its head once again to demonstrate that it was not viable to have a reliable system of ethnic/racial classification and

276

Chapter Nine

develop public affirmative action policies to be targeted at racial groups, given that it would be impossible to identify the beneficiaries–the sham of mestizo equality concealing, in fact, ongoing hegemonic racial prejudice. However, what it is important to stress, and which has not been mentioned in the various criticisms, is that of all the names used to selfdefine race, just over half a dozen accounted for practically all the answers given (95%). In addition to the five standard categories used in censuses and surveys, they consisted of “moreno”2 (literally, “brown”) or some other variant of it, such as “moreno claro” (“light brown”) or “moreno oscuro” (“dark brown”), and had little statistical significance. Given the concentration of responses in this small number of categories, the claim that the country’s racial classification system is extremely complicated therefore evaporates. Table 9-3. Breakdown for the spontaneous question on colour/race in the 1998 employment surveys and 1976 household surveys Categories White Morena Parda Preta Negra (Black identity) Morena clara Yellow Mulatto Light Morena oscura Dark Indigenous Other Total Sources: PME, 1998; PNAD, 1976.

PME/98 54.24 20.89 10.40 4.26 3.14 2.92 1.11 0.81 0.78 0.45 0.38 0.13 0.48 100.00

PNAD/76 49.45 24.80 8.47 5.61 0.10 2.75 1.53 1.24 1.50 0.54 1.08 -2.93 100.00

At any rate, it should be pointed out that the term “pardo” is still employed as a residual category, as it was when it first came into use, thereby generating a certain distrust of the classification, mainly due to its polysemous nature. Several different groups may use this category to selfidentify, including both Afrodescendants or blacks and descendants of indigenous peoples, as well as immigrants from East Asia and their

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

277

descendants. Moreover, in some places, such as the Central Western region of the country, it is also used to refer to historical/geographical origin or social/economic position. To sum up, two points deserve to be highlighted: first, that nearly all Brazilians identify themselves according to a rather restricted set of representations of colour or race; secondly, that both the names given spontaneously and their relation to the standard predetermined categories have remained significantly stable. Thus criticism of the possibility of constructing and using a classification system in surveys and censuses is clearly inappropriate and unfounded (Petruccelli, 2007). The system of ethnic/racial identification used up till now has proved to be not only heuristically interesting but also significantly useful in exposing existing social inequalities.

7. A comparative overview of the countries of the Americas 7.1. The identification of ethnic/racial groups It is abundantly clear that during the 1990s there was a qualitative leap in interest across the countries of the region in knowing the pluri-cultural and multi-ethnic composition of their populations and that this interest has grown significantly in the censuses carried out since 2000. This has resulted in questions on people’s ethnic/racial identity being incorporated into various sources of official data, as well as into population censuses and household surveys. The statistical tools employed have improved and data-gathering methodologies have been refined while respecting the rights of the population, in keeping with the growth in ethnic consciousness in the region. As a result current records show that 25 countries of the Americas included an ethnic/racial identification variable in the census round of 20003. It has therefore been possible to quantify the participation of the different ethnic/racial groups identified within the total population of each country. Indigenous nations were identified in nearly all countries (24) while Afrodescendants4 were found in just 18. However, the devising of mechanisms for gathering data on the ethnic/racial identity of the population is still a sensitive issue that has even caused controversy in some fora and the addressing of which involves several actors. Answers are needed to the following: How should the questions be asked? Who should answer them? How should they answer them? What should they be asked?

278

Chapter Nine

Below is a summary of the progress the countries of the region have made in addressing these conundrums.

7.2. The adoption of basic principles It should be noted that the adoption of “self-adscription” as a fundamental guiding principle of the formulation of the question on ethnic/racial identification reflects an agreed attitude of respect for the human rights of interviewees5. The use of the criterion of self-adscription or self-classification, by which the actual interviewees are the ones who decide how to identify themselves, values the subjective aspect of the answer, resulting in a more active determination of their ethnic/racial identification, since they themselves give “sense” to it. On the other hand, a second criterion, involving other possible methodologies, is aimed at detecting supposedly more “objective” aspects, such as language spoken, clothing used and physical features, to be classified by the interviewer, though with limited opportunities for using them. As already highlighted, most of the censuses in the 2000 round included at least one question on ethnic/racial identification, formulated in accordance with one or other of the two above-mentioned criteria. While there were no problems in implementing the first, in relation to the second it did not always turn out to be appropriate for capturing data on Afrodescendants. For historical reasons, which will not be analysed here, this ethnic/racial category does not, generally speaking, fit the characteristics set out in Convention 169 of the International Labour Organization (ILO), in the sense of having its own name as a community, using a particular language and inhabiting a specific territory, other than in exceptional cases such as those from the community of Palenqueros de San Basilio, in Colombia, or occupying the region before the conquest. A variety of terms were used to refer to the autochthonous populations of the Americas, including “grupos indígenas” (“indigenous groups”) in El Salvador, “grupos étnicos” (“ethnic groups”) in Guatemala and Trinidad and Tobago, “pueblos originarios” (“originary peoples”) in Bolivia and “grupo poblacional” (“population group”) in Honduras. However, the ILO Convention recommends the use of “indigenous peoples”, together with a list of categories giving the actual name of each people present in the country, thereby enabling them to be identified in greater detail. With regard to countries that have been collecting data on Afrodescendants for some time, the questions concerning identification have been found to refer mainly to race and skin colour. Statistical data rely on the existence of races as a historically constructed social

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

279

structuring concept because organised civil society representing Afrodescendants sees “race” and “skin colour” as the pillars around which those groups have constructed their own identities, and partly because in the course of colonisation the languages, traditions and beliefs that shaped their cultural identity became fractured. For example, the Cuban census of 2000 included the question “What is your skin colour?” while the Colombian census of 2005 was unusual in that the question on identity alluded to “physical features” in order to try to obtain information: “According to your culture, people or physical features, are you or do you recognise yourself to be...?” In Brazil the question, in both censuses and surveys, is asked directly: “What is your colour or race?” In other national situations, however, the minorities of African descent have constructed an historical/cultural identity that differentiates them from the rest of the population as a result of using a dialect, living in a geographical enclave or giving a name to their community, for example, the Raizales and Palenqueros in Colombia and the Yungas in Bolivia. In such cases it is possible to include the actual name of the community as a specific answer category. Another challenge to the adoption of the criteria mentioned for formulating the classification question concerns the fact that household data-gathering is not always strictly based on the principle of selfadscription since the questionnaire is usually applied to the person found at home at the time of the interview. That person, sometimes called the “informant”, is the one who provides all the data relating to the residents of the household in question, and not just the data relating to the classification of ethnic/racial identity, as specified in the actual instructions on implementation of the questionnaire, for example, in Brazil6. Identification by another person is also allowed in the case of children and very young people who do not yet have enough experience of socialisation to have a strong ethnic/racial self-perception–even if they are present–and for people who, for a variety of reasons, are unable to provide such information. In the circumstances mentioned, the classification of absent inhabitants and children and young people belonging to the household, whether or not present, is done by means of hetero-classification, either by a member of the household or, in some cases, by the actual interviewer who can add supplementary information to the declaration.

280

Chapter Nine

8. The criteria used in statistical data-gathering tools 8.1. How the question on ethnic/racial identity is formulated Population censuses in most countries currently include a closed-ended or pre-coded question on ethnic/racial identification, the wording of which varies from case to case. In one group of countries in which the main aim was to capture data on the population of indigenous origin, the wording of the question varied as follows: “Do you consider yourself to belong to…?”, “What population group do you belong to?” or “Do you consider yourself to belong to one of the following originary or indigenous peoples?” The 2000 census in Mexico is paradigmatic since it is the only case in which the names of the indigenous peoples appear in the question as follows: “Are you Náhuatl, Maya, Zapotec, Mixtec or from another indigenous group?” However, it does not include the names of the peoples in the answer categories, thereby making it impossible to see the numbers for each people or for individuals belonging to other groups, such as Afro-Mexicans, to selfidentify as such. In Bolivia, on the other hand, the following options for answering the question in the 2001 census were given: Quechua, Aymara, Guaraní, Chiquitano, Mojeño, etc. (see Table 9-4). In another group of countries, in which this time the aim was to include a complex population profile that was not confined to originary peoples but also included Afrodescendants, the question sought to determine ethnic/racial identity on the basis of several different elements. As has already been stated, Brazil, which, together with the United States of America and Canada, has the longest history of ethnic/racial classification of the population in the Americas7, currently uses a pre-coded question in censuses and surveys, asking: “What is your race or skin colour?”. Other countries ask for ethnicity and race separately, as in the case of the USA, or for “your culture, people or physical features…”, as in the case of Colombia, or, as in Costa Rica in 2000, just ask to which culture the person belongs. In the case of the Cuban census, by contrast, as already mentioned, interviewees are only asked to say what their skin colour is. Another approach used is to put two questions, as in the case of the Venezuelan census of 2001. The first asks about self-definition: “Do you belong to an indigenous people?” and if the answer is yes, there is a second open question so that the interviewee can tell the interviewer the name of the group to which they belong. As already mentioned, in other situations minorities of African descent have identified themselves using their own names, such as, for example, the Raizales and Palenqueros in Colombia (2005 census) and the Yungas

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

281

in Bolivia in 2001, as well as the Garífuna in Honduras (2001 census) and Nicaragua (2005 census). Another way of asking is to leave it to the interviewer to decide when to do so, as in the case of the Ecuadorian census of 2001 and the 2007 census in El Salvador in which, in the event that interviewees selfidentified as indigenous, they were asked to which group they belonged. In Belize (2000), Guatemala (2002) and Jamaica (2001), interviewees were asked to which ethnic group or race they belonged. In addition to ethnic group or race, other less specific criteria, such as people or population group, were used in the cases of Colombia and Honduras. Altogether, on examining the various criteria used either separately or together, up to ten different ways have been used to gather data on membership of an ethnic or racial group, focused around two main approaches: historical/cultural characterisation and classification according to physical features. The USA, on the other hand, is characterised by having a centrally established standard system of racial classification for use in federal agencies. Thus, in 1977 the parameters of ethnic/racial classification were determined in a resolution known as Directive 15. As a result of criticisms and changes in legislation, in 1994 a committee was set up to examine different data sources, including the supplementary household surveys of 1996 and 1997. As a result of the work of that committee, the 2000 census contained some significant changes, compared to previous ones, reflecting the need to take account of groups with a multiple racial history. The most significant change made was that the option was given to select one or more categories to indicate racial identity. The idea that the multi-racial “phenomenon” was growing within North American society was therefore institutionalised without providing an official definition of what it was. The efforts of multi-racial organisations were therefore successful in achieving some kind of result, albeit incomplete in that it did not meet their main demand that there should be a separate category for those who self-identified as being of multiple origins (Nobles, 2000: 144; Williams, 2006: 59). The debate between these two alternatives, having a separate identification category for people who consider themselves to be of multiracial origin or the option of ticking more than one of the answer categories provided in the questionnaire, may also influence discussion on the development of a more appropriate classification system in other countries. Another change concerned how Hispanic origin was addressed. Thus “Race” and “Hispanic Origin” were deemed to be two separate and

282

Chapter Nine

distinct concepts. The 2000 census asked, on the one hand, whether the person was Spanish, Hispanic, Latino and, on the other, which race or races they considered they belonged to, allowing them to select “one or more” of the options offered. As a result of these changes, while the data on race from 2000 are not directly comparable with those obtained from the 1990 or earlier censuses, they do make it possible to give a more accurate picture of the state of the debate on ethnicity in the country. The example of the USA and the constant variations in the way in which the question on ethnicity and race is posed in its censuses contrasts significantly with that of Brazil. In this country, persistent resistance to the possibility of changing the configuration of the question on race and the answer categories provided has been grounded on the importance of carrying on with the historical data series to the detriment of the engagement to provide genuinely reliable statistics. As for Canada, its population is the result of a multidimensional process of unequal contributions from the autochthonous populations, the “founding races” and more recent “allogeneic” immigration. Thus, in Canadian censuses from 1871 until 2001, the predominant method adopted for investigating the ethnic origins of the inhabitants was by using the dual criteria of place of birth, or nationality, and race, on the one hand, and mother tongue or language spoken, on the other. In the 2001 census, under the heading of “Socio-cultural Information”, 13 questions were asked, enabling the interviewees to be classified according to place of birth, nationality, citizenship, whether they were immigrants and, if so, date of arrival in Canada, mother tongue and language(s) known and used at home, the ethnic or cultural group(s) to which their ancestors belonged, whether they considered themselves indigenous or aboriginal, their religion, etc. Having evolved over more than a century, the categorisation of ethnic groups has been considerably transformed. Origin is distinguished from language and all reference to paternal lineage and arrival on the continent has been given up. The classification turned towards a more subjective determination, “cultural” identification rather than ancestry. The 2006 census reproduced that whole set of questions exactly as they were in the 2001 census. Canada is an example of a country in which priority has been given to producing statistics that reflect social dynamics, in particular the race question, rather than to safeguarding the historical data series. In both these countries, each of which has a high annual immigration rate, censuses and surveys include a series of questions on place of birth, citizenship and year of entry into the country for the purpose of separating the native population from that born abroad. These data can be used to

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

283

obtain a wide variety of information since they make it possible to distinguish between those who have citizenship as a result of being born in the country and those who have obtained it through naturalisation.

8.2. How the question on ethnic/racial identity is answered As mentioned above, it is the interviewee who establishes his or her ethnic/racial origin by selecting one–and it is usually only one–of the categories included in the list provided in the census questionnaire8. The list usually comprises the names of the main indigenous peoples, Afrodescendants and other ethnic/racial groups who are present in the country. It may also include an open category option of “others” and may or may not ask for further specification. One group of countries– Argentina, Bolivia, Chile and Paraguay–opted for a list of categories based on the names of indigenous peoples. In Brazil where, as already mentioned, the census asks for race or skin colour, five categories are listed: white, preta, parda, yellow and indigenous. In the 2010 census, as will be seen below, if the answer is “indigenous”, the interviewee is then asked also to specify ethnic group and language spoken. This classification system has been systematically criticised for merging two aspects of the phenomenon of ethnicity, namely colour or appearance and ethnic origin, and including categories that are neither exhaustive nor exclusive, while at the same time being subjected to evaluation and possible future change by the country’s official statistics body itself. Other countries (Ecuador, Colombia, Cuba, Guatemala and El Salvador) include certain specific categories for Afrodescendants, such as “black” or “mulatto”. The term “Afro-(nationality in question)”, such as Afro-Ecuadorian, Afro-Colombian, etc., is also usually used. The Colombian census is the only one that includes a possible answer category for Roma people, commonly known as gypsies, thereby acknowledging their presence in the country. In countries in which the population is characterised by the presence of groups identified by the language they speak, usually derived from English, terms such as “black” (Belize and Guyana) or “negro inglés” (“black English”) (Honduras) are used. Other categories found were: “African” (Guyana), “Afro” (Uruguay) and “afrodescendiente” (“Afrodescendant”) (Colombia and Nicaragua). Another example was provided by Uruguay where, in the classification system used in the 2006 Household Survey and the 2011 Census, the ancestry of interviewees was considered.

284

Chapter Nine

Lastly, intermediate categories are used to denote derivatives of the mixing of races, such as “mestizo”, “mixed” and “mulatto”. All of the 20 or so terms listed are also centred around the two main pillars of classification criteria already noted, namely racial markers and historical/cultural characterisation. The combining, or not, of these two criteria, which we can call, respectively, phenotypic and ethnic, or racial and cultural, has led to varying results across the different countries, depending also on which classification categories have been adopted. In Ecuador, for example, the inclusion of the term “mestizo” as one of the possible answers to the question “What do you consider yourself to be?”, while at the same time asking a question about the language spoken, resulted in an apparent underestimation of the country’s indigenous population in 2001. On the other hand, in the case of Colombia, the 2005 census obtained better estimates of both the indigenous and AfroColombian populations than the 1993 one did by broadening the scope of the question to include simultaneously the concepts of people, culture and physical features9 (Del Popolo, 2008: 28) rather than keeping to the notion of belonging to a community, as the previous census had done. El Salvador seems to have had the same experience as Ecuador when, in its 2007 census, it used the category “mestizo”, which was defined explicitly as the result of a “mixture of white with indigenous”, and limited the scope of the racial identification of “black” (“negro”) to just “race”. As a result, over 86% of the country’s inhabitants identified themselves as “mestizos” and almost 13% as “whites”. If we add the criterion of ancestry to these two classification criteria, namely the racial marker or phenotypic and the cultural factor, as Uruguay did in the 2011 census, it is the multidimensionality of the ethnic/racial phenomenon that needs to be given greater attention in the statistics of countries, probably by using more elaborate classification criteria and not limiting census questionnaires to just one or at most two questions on the issue.

8.3. Indirect methods of ethnic/racial identification It has been possible to identify other methods of ethnic/racial identification, depending on the nature and use of the data sources. One of them is identification by surname, which is particularly complex. It requires the drawing up of the fullest possible list of surnames belonging to the target population, preferably with the help of civil society organisations. It has the drawback that, in some cases, there is ambiguity in the classification of individuals with those surnames who may not

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

285

belong to the group in question. However, this approach has been validated to a certain extent by some local studies. Another method of identification includes questions about the language spoken at home, which is deemed important for determining ethnicity, especially in the case of native American peoples, since language operates as a representation of people’s culture. Having a question on mother tongue also makes it possible to obtain the best possible information on the ethnic origin of the members of the household, since more recent generations may have learned, and may speak almost exclusively, the language of the country or region where they live. However, questions on the customary language spoken in the household introduce supplementary elements of ethnic identification, often requiring questionnaires to be translated and the interviewers to be bilingual. The countries that include this method of classification are: Bolivia, Colombia, Ecuador, Guatemala, Mexico, Nicaragua, Paraguay, Peru and Venezuela.

9. Methods of ethnic/racial identification by country Below is a description of the types of questions that have been used by the different countries in their censuses. Self-identification was used as a criterion in the 2000 round of censuses conducted by: Argentina, Brazil, Chile, Costa Rica, Cuba, El Salvador, Honduras, Jamaica, Nicaragua, Panama, Trinidad and Tobago and the USA. The joint criteria of self-identification and language were used by: Belize, Bolivia, Canada, Colombia, Costa Rica, Ecuador, Guatemala, Mexico, Nicaragua, Paraguay and Venezuela. As for the population investigated in the censuses, two countries stand out for having limited the age of interviewees: in Mexico it was addressed to people aged 5 and over and in Bolivia to those aged 15 and over. While recognising that it is desirable for the whole population of each country to be identified in terms of ethnicity and race, it is also clear that children and adolescents generally have had insufficient experience of socialisation to allow them to have constructed an ethnic/racial identity. Also it is usually their parents or guardians who answer the census questions rather than themselves. This being so, the age restrictions mentioned are seen as justified in those countries. Lastly, since 1990 Peruvian censuses have used solely the criterion of the language in which the respondent learned to speak (mother tongue). Of the 25 countries in the region that collect data on the ethnic/racial identity of the population, all except Cuba seek to capture data on the

286

Chapter Nine

“indigenous” population, with 13 of them going on to identify the ethnic group/people to which they belong: Argentina, Belize, Bolivia, Chile, Ecuador, El Salvador, Guatemala, Honduras, Nicaragua, Panama, Paraguay, Peru and Venezuela. By virtue of the characteristics of its population, Cuba seems to be the only country that uses solely the categories of “white”, “mestizo” and “black” in its demographic statistics. A summary of what has been described so far can be found in Table 9-3.

10. An overview of statistics in current Brazil The persistence of structural, institutional and individual racism means that the notion of race, together with its correlates of ethnicity and skin colour, is still accepted as a political and analytical category. The importance of recognising the social inequalities that negatively affect segments of the population, and their connection with discrimination and racism, leads to a questioning of the supposed homogeneity of the population and the adoption, as a consequence, of ideas about the multiethnic, multi-racial and pluri-cultural nature of the society. Analysis of the records resulting from the codification of ethnic/racial identities provides a particularly important picture of how the identity markers of these categories have been formulated and constructed. However, the classification of racial groups does not take place within a neutral setting: it is attended by a marked asymmetry between those doing the classifying and those being classified. A relationship of symbolic domination, interpenetrated by the current use of the categories of colour and race, is clearly present in such classification, in which it can be detected that: The biological appearances and the very real effects that have been produced in bodies and minds by a long collective labour of socialisation of the biological and of biologicisation of the social combine to reverse the relationship between causes and effects and to make a naturalised social construction […] appear as the grounding in nature of the arbitrary division which underlies both reality and the representation of reality and which sometimes imposes itself even on scientific research. (Bourdieu, 2001: 3).

While the different ways of measuring this phenomenon have already been subjected to extensive methodological reflection, the questioning of how the representation of a person’s colour and race is organised and designated only began more recently. The cognitive map of the perception of ethnic/racial variability is built on the socially determined conceptualisation of appearances, since all perception is informed and

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

287

organised on the basis of organising principles that are capable of selecting the relevant data (Magli, 1989). As far as the categorisation of ethnic/racial groups is concerned, it is extremely worthwhile to remember that, already back in 1835, Jean Maurice Rugendas said the following about his trip to Brazil: “Deciding the colour of such and such a person is less down to appearance and physiology than legislation and administration” (Rugendas, 1940: 65).

10.1. PCERP: A special survey on ethnic/racial identification in Brazil Thirty years after classification by colour or race was reintroduced into Brazilian population censuses–as well as incorporated into other monthly or annual surveys–it was deemed essential to conduct a study to assess how the ethnic/racial characterisation of the population was being investigated and the results it was obtaining. Meanwhile, during this period, many authors have turned to this issue and it has also become a focus of attention for academics in other countries in which its importance demands it. The conclusions reached by the different works published in Brazil range from endorsement of the classification system used by the Instituto Brasileiro de Geografia e Estadística–IBGE (Brazilian Institute of Geography and Statistics), to outright rejection. The former include statements such as the following, based on analysis of the results of the 1976 PNAD household survey: Though much criticised, the existing surveys that make it possible to assess some aspects of the classification system used by the Instituto Brasileiro de Geografia e Estadística–IBGE for identifying racial groups suggest that they are appropriate for conducting empirical research into racial inequalities in Brazilian society. (Osório, 2003: 7).

Assertions made by critics of the system include the following: The subtle and fluid character of the subjective identification of race in Brazil raises critical methodological concerns for analysts who use IBGE data to determine the racial composition of Brazil and for those who carry out comparative analyses of socioeconomic differentials by skin colour. (Wood and Carvalho, 1995: 2).

In this context, it should be pointed out that the IBGE is noteworthy for having undertaken, in the second half of 2008, a specific survey in order to investigate the most appropriate methods of capturing data on the

288

Chapter Nine

ethnic/racial groups of the country’s population, which is also intended to serve as a basis for a future improving of the classification system. The survey was called Pesquisa das características étnico-raciais da população–PCERP (Survey of the Ethnic/Racial Characteristics of the Population), and its stated aim was to investigate the construction and use of the terms people currently use to express their identity. In addition, its immediate objectives were to investigate the elements underlying racial classification terminology and design models for representing the different ways in which individuals and groups identify themselves. What is expected is that, through this survey, information will be obtained on the different dimensions that shape the construction of those identities, such as origin, physical features, ancestry, etc. Initial data from the survey in the form of tables of variables (IBGE, 2011) have already been published and another publication will contain studies and analysis which could be used as a basis for possible proposals for improving the ethnic/racial classification system used in the country’s official statistics. The aim of the exercise is to allow ethnic/racial identity to be expressed more reliably through the use of non-exclusive categories, therefore adopting the notion of the multi-ethnic and multi-racial nature of the population and reaffirming the importance of recognising the social differences that negatively affect segments of the population, as well as their connection with discrimination and racism. When broadly compared with censuses and household surveys, the PCERP introduced a methodological innovation with regard to the identification by colour or race of the interviewee. It consisted in selecting just one person from each household, aged 15 or over, who answered exclusively with regard to their own ethnic/racial identity. As already mentioned, this is different from what happens in other statistical operations. The current classification categories, which are presently inadequate for capturing the multiculturalism that characterises Brazilian society, could be turned into contextualised meaningful signifiers, corresponding to the many different elements of reality, based on the results of this survey. In this way, a new classification system, to be designed for future use, will allow the multidimensional expression of the identities of Brazilian society.

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

289

10.2. Ethnic/racial identification in the 2010 Brazilian population census The restoration of racial classification in the 1980 census, despite the opposition of such eminent figures as Gilberto Freyre10, following its omission from that of 1970, coincided with the country’s redemocratisation. The subsequent census scheduled for 1990 was eventually carried out in 1991 with a background of intense campaigning by social organisations representing the Brazilian “black movement”, under the slogan of “Não deixe sua cor passar em branco”11. The preparations for the 2000 census included the testing-out of questions on colour or race and origin that were both open-ended/spontaneous and pre-codified. This test, involving four questions included in the Monthly Employment Survey for July 1998 (Petruccelli, 2007), failed to succeed in changing the established classification system involving the five options already mentioned: white, preta, parda, yellow and indigenous, essentially because there was insufficient time for reflection and the development of a possible alternative. For the 2010 census the earlier campaigning subsided and representatives of social movements invited by the IBGE to attend seminars on evaluation of the results of the PCERP survey were less keen to see possible changes in the classification system, given the relative progress of public policies on inclusion advanced by universities on the basis of the current racial categories. The consolidation, yet again, of these categories, which refer to an “eclectic” reference system (Nascimiento, 2006), fosters the continuity of a mixed classification criterion that includes appearance and ancestry and gives, through the ambiguous idea of “colour”, what is deemed to be the official depiction of national identity. What has changed in this census is that the question on ethnic/racial classification has been reincorporated into the questionnaire applied to the entire population being studied; in other words, every inhabitant in every household in the country was asked about their race and skin colour and not just those who formed part of the census sample, as had happened between 1980 and 2000. This change has brought benefits such as the ability to analyse the data obtained with the desired level of disaggregation, according to both geography and classification categories. Furthermore, in cases where the interviewees identified themselves as indigenous, they were also asked, for the first time in the history of Brazilian censuses, to which ethnic group they belonged and which language they spoke. This innovation, which brings the Brazilian census into line with that of other countries which also identify the ethnic groups

290

Chapter Nine

to which the autochthonous population belongs, is perhaps the most important change made to the census in over a century of the history of statistics in this country.

11. Conclusions Ethnic/racial population classification, which has been found in the majority of population censuses, has not yet spread to all countries of the Americas. However, in the 2010 census round, the number of countries that had used this topic increased, compared to the 2000 round, including, for example, the cases of Uruguay and Argentina, in 2011 and 2010 respectively, for the first time. While current prospects for extending the number of countries are reasonable, the ways in which the data are obtained are characterised by the wide variety of approaches taken. From the conceptual point of view, the focus on measuring ethnicity, race and skin colour or identifying native peoples is juxtaposed with the variety of concepts and terminologies used, some of them discussed here, which stem from the multiple socio-cultural and linguistic reality of the countries of the region. In any case, it is significant that since 1959 the United Nations itself has continuously and repeatedly stressed (United Nations, 1959; 1969) that recommending international classification criteria is impossible: “Since countries collect data on ethnicity in different ways and for different reasons, and because the ethno-cultural composition of a country could vary widely from country to country, no internationally relevant criteria or classification can be recommended” (United Nations, 2008: 140). It therefore needs to be stressed that formulating surveys, censuses or any other type of statistical data collecting tool is not simply a conceptual or technical task but also a political one, in its broadest sense, part of a process in which the interests of states and groups are ideologically involved. It is therefore insufficient to turn to methodological principles for determining statistical enumeration strategies, without also considering the socio-political interpretations that closely accompany the construction of social representations. Continuous dialogue between institutions and social actors must be constantly encouraged and maintained in order to ensure that either one or more questions on ethnic/racial identification are included in the official statistics of countries which still do not have any, or that existing classification systems in those that do are improved. Since identity is expressed in many different ways, the interests in terms of political representations, the academic definition of specific issues and the construction of particular systems of statistical categorisation present, as

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

291

pointed out at the beginning, many difficulties arising from the conflicting interests of groups that have been historically subalternised and the resistance there is to those groups being fully incorporated into the exercise of citizenship. Lastly, while it is true that there have been found a wide variety of approaches and many different categories in measuring ethnic/racial diversity, it is nonetheless important to stress the crucial role the data obtained from surveys and censuses play in detecting the specific social/economic disadvantages and inequalities of the different groups. Despite the variations in approach found in the different countries, the empirical frameworks they share with regard to identifying categories that have historically suffered discrimination make it possible to detect the opposition between dominant ethnic/racial groups and socially disadvantaged minorities, common to the structure of all societies. Nevertheless, conceptual and methodological decisions about which dimensions should be considered when devising the ethnic/racial classification system to be used by statistical measuring reflect not only technical positions but also the socio-political dynamics of the society in question. If the boundaries between phenotypic or racial categories and historical/cultural factors are not well delimited, either in theory or in practice, the application of one or other method of classification will clearly yield a variety of results within the same country. In any case, having such data, despite the noted limitations, makes possible the design and implementation of public policies aiming at compensating and correcting economic and social injustice to which the traditionally disadvantaged population groups in those countries, mainly Afrodescendants or indigenous peoples, are subjected.

Chapter Nine

2. Belize

COUNTRIES 1. Argentina

QUESTIONS, CRITERIA, CATEGORIES

Selfadscription

Selfadscription

Identification criterion 2000 censusround

¿Existe en este hogar alguna persona que se reconozca descendiente o perteneciente a un pueblo indígena? To what ethnic group do you/does....belong?

Census Question

1. Black, 2. White, 3. Chinese, 4. Creole, 5. East Indian, 6. Garifuna, 10. Mennonite 11. Mestizo, 12. Spanish, 13. Other 7. 8. and 9. indigenous groups

17 indigenous peoples plus an “other people” category

Census Categories

Selfadscription

Identification criterion Household Surveys

Table 9-4. Comparative table of ethnic/racial classification questions

292

To what ethnic group do you belong?

Question Household Surveys 2000-2005

1. Black, 2. White, 3. Chinese, 4. Creole, 5. East Indian, 6. Garifuna, 9. Mennonite 10. Mestizo, 11. Spanish, 12. Other 7. and 8. indigenous groups

Categories Household Surveys

6. Chile

Selfadscription

5. Canada

Language spoken Selfadscription

Selfadscription

Language spoken

Selfadscription

Identification criterion 2000 censusround

4. Brazil

COUNTRIES 3. Bolivia

QUESTIONS, CRITERIA, CATEGORIES

À quel(s) groupe(s) ethnique(s) ou culturel(s) les ancêtres de cette personne appartenaientils? ¿Pertenece usted a alguno de los siguientes pueblos originarios o indígenas?

Sua cor ou raça é…

¿Se considera perteneciente a alguno de los siguientes pueblos originarios o indígenas?

Census Question

8 indigenous peoples plus a category of “None of the above”

1. Quechua, 2. Aymará, 3.Guaraní, 4. Chiquitano, 5. Mojeño, 6. Other native, 7. None. 1. Branca, 2. Preta, 3. Amarela, 4. Parda, 5. Indígena Open with 25 examples

Census Categories

Language spoken at home

Selfadscription

Selfadscription

Language spoken

Identification criterion Household Surveys

1. Branca, 2. Preta, 3. Amarela, 4. Parda, 5. Indígena 15 categories, plus “others"

1. Spanish, 2. Quechua, 3. Aymará, 4. Guaraní, 5. Other indigenous

Categories Household Surveys

293

Hay algún miembro de 8 indigenous la familia que hable o languages entienda una de las siguientes lenguas?

Quelles étaient les origines ethniques ou culturelles de vos ancêtres?

A cor ou raça do(a)__é:

¿En que lengua habla habitualmente?

Question Household Surveys 2000-2005

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

10. Ecuador

9. Cuba

8. Costa Rica

COUNTRIES 7. Colombia

QUESTIONS, CRITERIA, CATEGORIES

294

Language spoken

Selfadscription

Language spoken Selfadscription

Selfadscription

Language spoken

Selfadscription

Identification criterion 2000 censusround

Census Categories

¿Cómo se considera...?

¿Cuál es el color de piel?

Pertenece... a la cultura...

1. Indígena, 2. Afro-costarricense o negro, 3. China, 4. None of the above 1. Blanco, 2. Negro, 3. Mestizo o mulato 1. Indígena, 2. Negro (afroecuatoriano), 3. Mestizo, 4. Mulato, 5. Blanco, 6. Other

De acuerdo con su cultura, 1. Indígena, pueblo o rasgos físicos, ... 2. Rom, es o se reconoce como: 3. Raizal, 4. Palenquero, 5. Negro(a), mulato(a), afrocolombiano(a) o afro-descendiente, 6. None of the above

Census Question

Chapter Nine

Selfadscription (for those over the age of 12)

Selfadscription (for those over the age of 10)

Identification criterion Household Surveys

Usted se considera:

¿De cual de los siguientes grupos étnicos se considera usted?

Question Household Surveys 2000-2005

1. Indígena, 2. Mestizo, 3. Blanco, 4. Negro, 5. Mulato, 6. Other, what?

1. Indígena, 2. Rom, 3. Raizal, 4. Palenquero, 5. Negro, mulato, afrodescendiente, 6. None of the above

Categories Household Surveys

13. Guyana

12. Guatemala

COUNTRIES 11. El Salvador

QUESTIONS, CRITERIA, CATEGORIES

Selfadscription

Language spoken

Selfadscription

Selfadscription

Identification criterion 2000 censusround

What ethnic groups do you belong to?

¿A qué grupo étnico (pueblo) pertenece?

b) Si Ud. es indígena, ¿a qué grupo pertenece?

a) Es usted…

Census Question

1. African, Negro, Black, 2. Amerindian, 3. East Indian, 4. Chinese, 5. Mixed, 6. Portuguese, 7. Syrian, Lebanese, 8. White, 9. Don’t know 10. Other

a) 1. Blanco, 2. Mestizo, 3. Indígena, 4. Negro (de raza), 5. Other b) 6. Lenca, 7. Kakawira (Cacaopera), 8. Nahua Pipil, 9. Other (specify) 22 indigenous peoples, Afro-indigenous, Ladino, others

Census Categories

Language spoken

Selfadscription

Selfadscription

Identification criterion Household Surveys 1. Lencas, 2. Maya/ Cacaopera, 3. Nahuat/ Pipil, 4. Don’t know, 5. Others

5 ethnic groups plus Xinca, Garifuna, Nonindigenous, Foreign

¿A qué grupo étnico pertenece?

Categories Household Surveys

295

De acuerdo con sus antepasados y/o costumbres Ud. se considera de origen de algún pueblo indígena? 1. Yes: Which one? 2. No 3. Don’t know

Question Household Surveys 2000-2005

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

Selfadscription

16. Mexico

Language spoken

Selfadscription

Selfadscription

Identification criterion 2000 censusround

15. Jamaica

COUNTRIES 14. Honduras

QUESTIONS, CRITERIA, CATEGORIES

296

¿Es náhuatl, maya, zapoteco, mixteco o de otro grupo indígena?

To which race or ethnic group do you say you/ ... belong?

¿A qué grupo poblacional pertenece?

Census Question

1. Garífuna, 2. Negro inglés, 3. Tolupán, 4. Pech (Paya), 5. Miskito, 6. Lenca, 7. Tawahka (Sumo), 8. Chortí, 9. Other 1. Black, 2. Chinese, 3. Mixed, 4. East Indian, 5. White, 6. Other, 7. Don´t know. 1. Yes 2. No

Census Categories

Chapter Nine Identification criterion Household Surveys

Question Household Surveys 2000-2005

Categories Household Surveys

Selfadscription

19. Paraguay

Language spoken

Selfadscription

Language spoken

Selfadscription

Identification criterion 2000 censusround

18. Panama

COUNTRIES 17. Nicaragua

QUESTIONS, CRITERIA, CATEGORIES

¿Existe en este hogar alguna persona que se considere indígena o perteneciente a una etnia indígena?

¿A qué grupo indígena pertenece?

¿A cuál de los siguientes pueblos indígenas o etnia pertenece?

¿Se considera perteneciente a un pueblo indígena o a una etnia?

Census Question

1. Kuna, 2. Ngöbe, 3. Buglé, 4. Teribe, 5. Bokota, 6. Emberá, 7. Wounaan, 8. Bri Bri, 9. Don’t know. 17 indigenous groups

For those who answer Yes: 13 indigenous groups, Afrodescendants and other ethnic groups

1. Yes 2. No

Census Categories

Selfadscription

Selfadscription

Identification criterion Household Surveys

Es usted indígena?

¿A cuál de los siguientes pueblos indígenas o etnia pertenece?

Question Household Surveys 2000-2005

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

1. Mestizo, 2. Blanco, 3. Criollo, 4. Creole/ Afrodescendiente, 5. Miskito, 6. Mayagna (Sumu), 7. Rama, 8. Other 1. Yes 2. No

Categories Household Surveys

297

21. Suriname

COUNTRIES 20. Peru

QUESTIONS, CRITERIA, CATEGORIES

298

Selfadscription

Language in which the person learned to speak

Identification criterion 2000 censusround

To which populational group does (N) belong?

¿El idioma o lengua en el que aprendió a hablar fue:

Census Question

Identification criterion Household Surveys

1. Indigenous, Amerindian, 2. Maroon, Bushnegro, 3. Creole, 4. Hindostani, 5. Javanese, 6. Chinese, 7. Caucasian, White, 8. Mixed, 9. Other, 10. Don´t know, No answer

1. Quechua, Self2. Aymara, adscription 3. Ashaninka, 4. Other native language, 5. Spanish, 6. A foreign language. 7. You are deaf-mute.

Census Categories

Chapter Nine

¿Por su herencia o cultura se considera perteneciente a algún grupo étnico?

Question Household Surveys 2000-2005 1. Indígena de la Amazonia, 2. Quechua, 3. Aymara, 4. Negro/ Mulato/ Zambo, 5. Mestizo, 6. Blanco, 7. Other

Categories Household Surveys

a) Is this person Spanish/Hispanic/ Latino? b) What is this person’s race: mark x one or more races ¿Pertenece a algún pueblo indígena?

¿Cree tener ascendencia?

To which ethnic group does (N) belong?

Census Question

a) 1. No, 2. Yes, with 3 options plus “Other” b) 13 groups plus “Other” 1. Yes 2. No

1. African, 2. Indian, 3. Chinese, 4. Syrian/Lebanese, 5. Caucasian, 6. Mixed, 7. Other, 8. Don´t know. 1. Afro o negra 2. Amarilla 3. Blanca 4. Indígena 5. Other (specify) 6. Don’t know

Census Categories

Selfadscription

Selfadscription

Identification criterion Household Surveys

¿Cree tener ascendencia?

To which ethnic group does (N) belong?

Question Household Surveys 2000-2005 1. African, 2. Indian, 3. Chinese, 4. White/ Caucasian, 5. Mixed, 6. Other ethnic group 1. Afro o negra 2. Amarilla 3. Blanca 4. Indígena 5. Other (specify) 6. Don’t know

Categories Household Surveys

299

Sources: Compiled by the author from the following sources: Del Popolo, 2008; Paixão, 2009; Antón & Del Popolo, 2008; Population Censuses; Household Surveys.

25. Venezuela

Selfadscription

24. USA

Language spoken Selfadscription Language spoken

Selfadscription

Selfadscription

COUNTRIES 22. Trinidad and Tobago

23. Uruguay

Identification criterion 2000 censusround

QUESTIONS, CRITERIA, CATEGORIES

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

300

Chapter Nine

Bibliographical references Antón, Jhon & Popolo, Fabiana. 2008. Visibilidad estadística de la población afrodescendiente de América Latina: aspectos conceptuales y metodológicos. Santiago de Chile: CEPAL Serie Población y Desarrollo n° 87: 13-38 Bourdieu, Pierre. 1991. Language and Symbolic Power. Cambridge: Polity Press. Bourdieu, Pierre. 2001. Masculine Domination. Cambridge: Polity Press. Bulmer, Martin. 2001. On the Classification of Subnations. Article submitted to the CERI-INED Conference, Paris. CIMI-Conselho Indígena Missionário. 2011. Funai mente sobre Belo Monte e ressuscita critérios racistas de indianidade. http://www.correiocidadania.com.br/index.php?option=com_content& view=article&id=5696:cimi-funai-mente-sobre-belo-monte-eressuscita-critrios-racistas-de-indianidade&catid=33:noticias-emdestaque [accessed in October 2011]. Carneiro da Cunha, Manuela (ed). 1992. História dos índios no Brasil. Rio de Janeiro: Companhia das Letras. Carvalho, José Jorge de. 2003. As ações afirmativas como resposta ao racismo acadêmico e seu impacto nas ciências sociais brasileiras. Teoria e Pesquisa: Revista de Ciência Política, 42-43: 303-340. Carvalho, José Jorge de. 2008. Racismo fenotípico y estéticas de la segunda piel. In Bruno Mazzoldi. (ed) El temblor: las sonrisas. Cátedra Derrida 2005. Bogotá: Pontifícia Universidad Javeriana/Tercer Mundo Editores, pp. 83-114. Costa, Sergio. 2006. O branco como meta: apontamentos sobre a difusão do racismo científico no Brasil pós-escravocrata. Estudos AfroAsiáticos, 28(1/2/3): 47-68. Del Popolo, Fabiana. 2008. Los pueblos indígenas y afrodescendientes en las fuentes de datos: experiencias en América Latina. Santiago de Chile: CEPAL-OPS. IBGE-Instituto Brasileiro de Geografia e Estadística. 2011. Características étnico-raciais da população. Um estudo das categorias de classificação de cor ou raça. Rio de Janeiro: IBGE. Magli, Patrizia. 1989. The Face and the Soul. In Michel Feher et al. (eds) Fragments for a History of the Human Body-Part Two. New York Zone Books, pp. 87-127. Marx, Anthony. 1998. Making Race and Nation: a Comparison of the United States, South Africa and Brazil, Cambridge: Cambridge University Press.

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

301

Morning, Ann. 2006. Ethnic Classification in Global Perspective: a Cross-National Survey of the 2000 Census Round. New York: New York University. Nascimiento, Alexandra Santos. 2006. Classificação oficial e extra-oficial: raça e cor em debate. Perspectivas: Revista de Ciências Sociais, 29:133-148. Nobles, Melissa. 2000. Shades of Citizenship: Race and the Census in Modern Politics. Stanford: Stanford University Press. Osório, Rafael. 2003. O sistema classificatório de “cor ou raça” do IBGE. Texto para Discussão No. 996, Brasilia: IPEA. Paixão, Marcelo. 2009. Realidades da diáspora: presença afrodescendente nas Américas segundo a rodada de censos de 2000. Rio de Janeiro: ANPOCS, Grupo de Trabalho, Relações Raciais e Ações Afirmativas. Petruccelli, José Luis. 2007. A cor denominada. Estudos sobre a classificação racial. Rio de Janeiro: DP&A/LPP-UERJ. Rugendas, Jean Maurice. 1940 [1835]. Viagem pitoresca através do Brasil. São Paulo: Biblioteca Histórica Brasileira, Livraria Martins. Simon, Patrick. 1997. La statistique des origines. “Race” et ethnicité dans les recensements aux États-Unis, Canada et Grande Bretagne. Sociétés Contemporaines, 26: 11-44. Simon, Patrick. 2001. Politiques publiques et catégorisations des populations: l´origine ethnique dans les statistiques françaises. Article submitted to the XXIV Congress of the IUSSP, Salvador. Simon, Patrick. 2005. The Measurement of Racial Discrimination: the Policy Use of Statistics. International Social Science Journal, 57(183): 9-25. Skerry, Peter. 2001. Never Precise, Always Political: Ethnic and Racial Data in the United States. Article submitted to the CERI-INED Conference. Paris. Solanke, Iyiola. 2007. Race, Law and Ethnic Data. Article submitted to the Conference Social Statistics & Ethnic Diversity, Should we count? How should we count and why?, Montreal, Canadá. Stavo-Debauge, Joan. 2005. Mobilising Statistical Powers for Action against Discrimination: the Case of the United Kingdom. International Social Science Journal, 57(183): 43-55. United Nations. 1959. Principles and Recommendations for National Population Censuses, Statistical Papers. New York: Department of Economics and Social Affairs, Statistics Division. United Nations. 1969. Principles and Recommendations for the 1970 Population Censuses, Statistical Papers. New York: Department of Economics and Social Affairs, Statistics Division.

302

Chapter Nine

United Nations. 2008. Principles and Recommendations for Population and Housing Censuses, Statistical Papers. New York: Department of Economics and Social Affairs, Statistics Division. Viveiros de Castro, Eduardo. 2006. No Brasil, todo mundo é índio, exceto quem não é. Povos indígenas no Brasil. Instituto Socio-Ambiental, Rio de Janeiro. Williams, Kim M. 2006. Mark One Or More. Civil Rights in Multiracial America. Ann Arbor: The University of Michigan Press. Wood, Charles & José Alberto M. de Carvalho. 1995. Census Categories and Racial-Ethnic Identity in Brazil. Paper presented at the annual meetings of the Population Association of America (PAA), San Francisco, California.

Notes 1

These terms have developed very specific and often evolving connotations in Brazil. The term “preto”, which was first used in the 1872 census, refers to people of African descent and can be translated as “black”. The term “pardo”, which was also used in the 1872 census and can be loosely translated as “greyish brown”, has had an ambiguous meaning which has changed over the years. More than a colour, it has racial connotations, referring mainly to people of mixed African/white descent but also other people with darker skin, 2 The term “moreno” relates solely to skin or hair colour and has no racial connotations. Furthermore, it can be used as a way of avoiding racial identification. It refers to people whose skin is slightly darker than white and can even include those who are sun-tanned. 3 Argentina, Belize, Bolivia, Brazil, Canada, Chile, Colombia, Costa Rica, Cuba, Ecuador, El Salvador, Guatemala, Guyana, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Suriname, Trinidad and Tobago, Uruguay, USA and Venezuela. 4 Defined as descendants of sub-Saharan Africans who arrived during the period of the slave diaspora, including “groups that were physically and culturally mixed, who mainly identified with this common demographic matrix” (Paixão, 2009: 2). 5 ILO Convention 169 states that “[s]elf-identification as indigenous or tribal shall be regarded as a fundamental criterion for determining the groups to which the provisions of this Convention apply”. 6 http://www.ibge.gov.br/home/presidencia/noticias/noticia_visualiza.php?id_noti cia=1602&id_pagina=1 7 The first Brazilian population census was carried out in 1872 and already included racial classification. In the case of the USA, the first census was carried out in 1790 and also included racial classification. 8 Only the USA allows more than one category to be ticked in the response. 9 Between the two dates, the indigenous population went from 1.6% to 3.4% and the Afro-Colombian population from 1.5% to 10.6%.

Ethnic/Racial Statistics: Brazil and an Overview of the Americas

303

10 Gilberto Freyre. “O Brasileiro-sua cor?”, Folha de São Paulo, 5 December 1979, p. 3, cited by Nobles (2000: 119). 11 This can be loosely translated as “Don’t let your colour be left blank”. However, in Portuguese “branco” can mean both “blank” and “white”. The slogan is therefore a play on words, calling on the black population not to leave the answer to the question on colour blank and thus be counted as white.

CHAPTER TEN INDIGENOUS PEOPLES AND AFRO-DESCENDANTS: THE DIFFICULT ART OF COUNTING FABIANA DEL POPOLO AND SUSANA SCHKOLNIK

1. Why should censuses include questions to identify indigenous people and populations of African descent? The need to identify indigenous peoples and Afro-descendants in the population censuses of Latin America arises from the recognition that these groups experience significant implementation gaps in the exercise of their civil, political, economic, social and cultural rights. This is the result of historic processes of conquest, colonisation and expansion of the Republican States of the region, which determined that both indigenous peoples and Afro-descendants came to occupy a subordinated position in contemporary societies, excluded from advances towards development and welfare. This situation can be addressed by contributing to indigenous peoples and Afro-descendants being truly able to exercise their rights, in equality of conditions. This necessitates having knowledge of their life conditions and the inequalities that affect them. In order to facilitate the design of inclusive public policies it is therefore essential to have information that is relevant, reliable, timely, disaggregated and culturally appropriate. Possession of this data would equate to having a tool, both technical and political in nature, that would be a valuable resource to increase levels of participation and oversight by indigenous peoples and Afro-descendants regarding actions that affect them. For this reason the call for information forms part of the demands made by these groups, and States bear an obligation to respond to it.

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 305

This process is not easy in societies that, historically, have denied that ethnic diversity is one of their constitutive elements, and made its existence invisible statistically–and by other more brutal means. But there have been changes, and the recognition, at least officially, of the multiethnic and pluricultural character of these societies has resulted in an emphasis on the task of collecting information about these peoples and populations. National statistics agencies have begun to address this demand for information about ethnic groups, although they have mainly applied such an “ethnic approach” in population censuses, and principally for indigenous peoples. Its inclusion in household surveys is less frequent, and the lack of attention is evident in continuous registers–at least in statistics on life and health. Given this scenario, there is a need for instruments capable of achieving the application of the ethnic approach to all sources of sociodemographic data, and enhancing it in those cases in which it has already been done. The best means to achieve this–indeed the only way from a human rights perspective–is through the effective participation of members of the region’s various ethnic groups. This document presents conceptual elements for the construction of operational definitions as well as antecedents of experiences involving inclusion of the ethnic approach in Latin American censuses. The special reference to the census reflects the fact that it is the instrument with the most experience of application in the region and that, given its universal character, it is an indispensable source for identifying, quantifying and characterising these groups, especially when they represent a minority within the national population. The census enables reliable information to be obtained for lower-level territorial units and can also serve as a framework for representative surveys. However, it is fundamental that ethnic identification be extended to all data sources.

2. Who’s who: the conceptual definition This section will review some conceptual definitions for selecting relevant variables that can assist in devising questions to identify the indigenous and Afro-descendant population of Latin America and the Caribbean. It will consider notions that are related to their development as social constructions subject to the political and historic context, institutionalised knowledge and social practices. The first of these is the concept of “race”, which has been utilised to justify differences between human groups based on phenotype, and which

306

Chapter Ten

alludes to a taxonomic division founded on biological, morphological and physiological distinctions. Although related to phenotypic features, the term “race” has been utilised in practice “not as a biological referent demonstrated empirically, rather as a social construction that takes certain visible biological traits as classification criteria” (Stavenhagen, 2001). As concepts are not neutral, historically the notion of “race” contained at its root the belief in the existence of “superior races” that would “naturally” have better capacities of adaptation and attributes of domination, while others would be substantially inferior. This thereby opened the door to racism and the “scientific” justification of the exploitation of some social groups by others. In Latin America, racism developed during the processes of conquest and colonisation to legitimate the domination of Europeans over indigenous peoples and enslaved Africans and to perpetuate it for the benefit of the dominant classes (Hopenhayn and Bello, 2001; Bodnar, 2005; Bello and Paixão, 2009). In opposition to the concept based exclusively on phenotypic aspects, this work presents the concept of “ethnic group”, such as defined by Stavenhagen (1991), which implies a shift from biological determination towards cultural and social specification, although it does not exclude it totally. In effect, he defines the ethnic group as a collectivity that identifies itself and is identified by others according to certain common elements such as language, religion, tribe, nationality or race, or a combination of these elements.

This enables him to state that ethnic groups can be considered as peoples, nations, nationalities, minorities, tribe or communities, according to the distinct contexts and circumstances [and indigenous peoples constitute] a special case of ethnic groups [...] given the historic circumstances of their conquest and incorporation into new state structures (Stavenhagen, 1991).

According to this definition, various types of situations can be identified: ethnic groups inside a State that identifies itself as multiethnic or multinational; ethnic groups inside a State that does not formally recognise its own multiethnic composition; national minorities that identify themselves with their ethnicity in a neighbouring State in which they may enjoy majority status; multiple ethnic groups in a State in which none of them enjoys a particular dominant position; ethnic minorities settled on both sides of a border between different States, and that find themselves in

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 307 a minority situation in both States; ethnic emigrants and refugees, the product of significant migrations, particularly from countries of the Third World towards other countries of the Third World or towards industrialised nations; and finally, indigenous and tribal peoples that constitute a special case of ethnic groups, and that are generally considered as minorities, given the historic circumstances of their conquest and incorporation into new state structures, as well as of their attachment to land, territory and of their centuries-long resistance to genocide, ethnocide and assimilation (Stavenhagen, 1991).

This final situation echoes, in relation to indigenous peoples, the definition formulated in 1986 by Martínez Cobo in his “Report to the UN Sub-Commission on the Prevention of Discrimination of Minorities in relation to Indigenous Peoples”: Indigenous communities, peoples and nations are those which, having a historical continuity with pre-invasion and pre-colonial societies that developed on their territories, consider themselves distinct from other sectors of the societies now prevailing on those territories, or parts of them. They form at present non-dominant sectors of society and are determined to preserve, develop and transmit to future generations their ancestral territories, and their ethnic identity, as the basis of their continued existence as peoples, in accordance with their own cultural patterns, social institutions and legal system (2004: point 2).

Consistent with this definition, Convention 169 of the International Labour Organization (ILO) of 1989, titled “Concerning Indigenous and Tribal Peoples in Independent Countries”, states that indigenous peoples are characterised by their descent from the populations which inhabited the country, or a geographical region to which the country belongs, at the time of conquest or colonisation or the establishment of present state boundaries and who, irrespective of their legal status, retain some or all of their own social, economic, cultural and political institutions (ILO, 1989: Article 1).

In this case, consciousness of their indigenous or tribal identity is considered a fundamental criterion for this determination. Over the years, a consensus has consolidated within the field of international organisations in favour of the definition of “indigenous people” articulated by Martínez Cobo (Deruyttere, 2001). This definition has been incorporated into the conventions and other legal instruments developed by the ILO, as already mentioned, and also into those of the Organization of American States (OAS) and the United Nations, as well as

308

Chapter Ten

into documents of indigenous organisations such as the Constitutive Agreement of the Fund for the Development of the Indigenous Peoples of Latin America and the Caribbean (Fondo Indígena), which was ratified by the majority of the countries of the region in the Second Ibero-American Summit of Heads of State and Government in 1992. All these authorities, with the addition of the United Nations General Assembly, which in 2007 approved the Declarations of the Rights of Indigenous Peoples, have sanctioned, if not the “definition”, then the “denomination” of “indigenous people” for these peoples; a denomination that entails the recognition of their rights relating to land, territories and natural resources and represents an important advance in their recognition as subjects of rights, although of course different countries and organisations within those peoples may vary in their designations. Beyond the characteristics enunciated in his definition, self-recognition as indigenous was already considered by Martínez Cobo to be a fundamental element given that he considered that On an individual basis, an indigenous person is one who belongs to these indigenous populations through self-identification as indigenous (group consciousness) and is recognised and accepted by these populations as one of its members (acceptance by the group). This preserves for these communities the sovereign right and power to decide who belongs to them, without external interference (2004: point 2).

This is compatible with a rights-based approach and with the stance of the United Nations, which has requested indigenous groups to define themselves as such1. With respect to the attitude of indigenous peoples, every attempt at definition by actors external to the peoples themselves has been considered incorrect or incomplete, and their representatives have also maintained that it is up to indigenous persons, and each people collectively, to decide who belongs to a people. This has been the position unfailingly sustained by indigenous representatives before the various organs of the United Nations, who advocate self-definition of group membership while also highlighting other elements such as descent, collective identity, acceptance by the group, language, and historic links with land (Aguilar Cavallo, 2006; Gamboa, 2006). A study undertaken in Chile to contribute to the development of a census question, drawing on the perspective of the indigenous peoples and that of non-indigenous specialists, demonstrated that the self-definition of group membership was considered a priority, without it invalidating criteria related to kinship relations and cultural characteristics (Oyarce et

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 309

al., 2005). For the United Nations, it is also clear that it is the peoples and individuals who consider themselves indigenous who must self-define as such (Permanent Forum on Indigenous Issues, 2004), respecting the internal definition of the people and its members; this means that indigenous representatives should participate in decisions made in official bodies regarding, inter alia, the way in which information will be gathered about their peoples. While the recognition of the indigenous as peoples is a debate already settled at the international level, the same does not hold in the case of Afro-descendants. For the indigenous this consensus involves a concept that embraces their ancestral, sociocultural and territorial characteristics, whether implicitly or explicitly. However, for Afro-descendants, such a consensus has not been reached regarding their equivalence with a people or community, except in some countries such as Colombia or Ecuador, or for certain particular groups such as the Creoles in Nicaragua or the Garifunas2 in Honduras, Guatemala and Nicaragua. This relates to the fact that in their case the concept of race, and within it skin colour, has been historically privileged, detached from other considerations. There may be a number of reasons for this. Perhaps one is that it has been more difficult to assign the quality of “people” to Afro-descendants because they appear to comprise a very broad group, one whose specificities are unknown. The Afro-descendant ethnic identity was mainly constructed through the complementarity of modern and traditional elements, between processes of continuity and rupture, and with a great capacity for adaption and assimilation of diverse cultural elements. In some cases, based on a re-creation of the African parent cultures accompanied by the formulation of new guidelines, elements of both the dominant culture and of other cultures with which they had had contact were incorporated. In others, as in the culture of the palenques, quilombos 3 and cumbes , the African cultural origin remained present. In general, within Latin America “Afro-descendants” are considered to be all those peoples and persons descended from the African diaspora throughout the world. This term is used to name the population of African origin, descended from those who were brought by force to Latin America during centuries of colonial slavery, as recognised by the World Conference of Durban (Bello and Paixão, 2009: 3).

However, Afro-descendants may still use different forms to denominate themselves. For example,

310

Chapter Ten the expression “black” has been legitimated by certain Afro-descendant social organisations of the continent [as in the cases of certain national organisations in Panama and Ecuador and some of a regional character], while other movements have sought to move beyond this category because it bears a strong historico-racial burden (Bello and Paixão, 2009: 3).

However, another reason why accordance of the status of people to Afro-descendants has remained elusive may be because Afro-descendants themselves have not demanded this attribute with force until recent years, when an advance to a more comprehensive and proactive ethnic consciousness has been observed. Since the year 2000 this has been reflected in the formation of new Afro-descendant organisations and networks, which have become social and political actors promoting the implementation of public policies, the recognition of rights, and political participation and representation (Rangel, 2009).

3. How to capture ethnic identity: towards an operational definition The indigenous peoples and Afro-descendants of the region constitute heterogeneous ethnic groups. It therefore seems obvious to ask if it is possible to define a set of meaningful characteristics that could enable acquisition of knowledge about their identities through instruments of data collection. This section attempts to define some basic variables that are considered closely tied to the phenomenon of identity. The previous section aimed to show some aspects or characteristics that were associated with the definition of ethnic groups; this prior conceptual exploration has permitted four dimensions to be distinguished that group together the constitutive elements of the definitions reviewed. These would make it possible to establish operational criteria to identify appropriate questions for capturing information on ethnic groups (indigenous peoples, Afro-descendants, and others) in the sources of data (CELADE/CEPAL, 2009; CEPAL, 2007; Schkolnik, 2000; Schkolnik and Del Popolo, 2005). These dimensions are: consciousness of membership, common origin, culture, and territoriality. It needs to be made clear that not all the dimensions have the same rank. From a rights-based approach, consciousness of membership is the necessary (although not sufficient) dimension for capturing details about ethnic groups through sources of statistical data. The other dimensions, for their part, make reference to aspects that are considered part of the definition of “ethnic group” and that, although not currently considered valid for estimating population size, permit–whether individually or in

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 311

association with other variables–the characterising of their internal heterogeneity as well as inequities in relation to other populations. a) Consciousness of membership This refers to the effective exercise of the right to self-identify with a people and self-define as belonging to that people, the development of an individual consciousness of that group membership, and the acceptance of that membership on the part of the same people. This dimension derives from a rights-based approach, and hence has pre-eminence over the other dimensions, regardless of whatever it is that people rely on in making that self-definition. Although the level of self-recognition with a people depends on the level of consciousness, which can oscillate from an assimilationist position to one of self-affirmation, self-identification respects the right of each to decide his or her belonging. b) Common origin This alludes to the existence of common ancestors and of a historic trunk of common origin which underlies the descent of a people up to the present. Among other factors, this considers the social and collective memory of peoples, their relation with their history and the endurance of the past as a permanent re-creation and renewing. As it deals with an ascribed characteristic and not an acquired one, it cannot cease to be considered nor can it be lost with time. c) Culture This involves the connection of individuals with the material and symbolic cultural expressions and manifestations of each people. This dimension embraces the attachment to the culture of origin, to social and political organisation, language, art, religious practices, lifestyle, and form of relationship with the environment, among other elements. It is associated with characteristics that are generally acquired at a very early age and that can be weakened by the effects of acculturation, globalisation, migration, relation with other cultures, or the impact of racism and discrimination. d) Territoriality To understand this dimension it is important to establish a distinction between land, as a factor of production, and territory, which contemplates not just the material aspect but also the collection of social and cultural experiences of the communities that interact with it (González Pazos, 2007).

312

Chapter Ten

The notion of territoriality is linked to the idea of territory, which speaks of an ancestral inheritance and the collective memory of peoples, the occupation of ancestral lands and the exercise of this territoriality based on the people’s belief system. It is for this reason that the indigenous struggle over territory is not solely a controversy regarding its commercial and individual value, but also a demand for a collective right (Mindiola, 2006). These basic dimensions relate to various facets of the concept of ethnic group. The utilisation of various features of ethnic identity has important potentialities, among them the possibility of examining not just quantitative aspects, but also the qualitative aspects of this social reality, the heterogeneity of each people or group, and the changes that can occur within them across time. Before undertaking the operational translation of these dimensions of identity there is at least one pair of general elements that needs to be considered. One of them is that not all ethnic groups necessarily share all these dimensions or all of them at the same time. The other is that these basic dimensions maintain a certain independence between each other. Although in the first instance it could be supposed that all of them would be influenced by the process of globalisation and the role played by the mass media within it, some are definitely more sensitive than others, or could even take different directions. For example, it is possible that the external manifestations of a culture are the first to be abandoned, weakened or replaced by those which the hegemonic culture offers (such as language and customs), but that does not imply that the same shift takes place, or occurs at the same time, with the consciousness of membership, given that the awareness of ethnic self-identification could pass through stages of decay followed by processes of ethnic revitalisation and vice versa.

3.1. Self-recognition, the key step Although a “desirable measurement” that would permit greater knowledge of the characteristics of indigenous peoples and Afrodescendant populations should include indicators for each of the identified dimensions, that of self-recognition has, as already stated, pre-eminence above the others because: a) it is the criterion adopted in international legal instruments and it reflects the position sustained by indigenous peoples before the various organs of the United Nations, which affirms the right of

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 313

people to self-identify, thereby respecting the consciousness that persons have of their identity; b) it considers the right that individuals have as members of a people to be consulted on matters that affect that people; c) it endeavours to avoid definitions proposed by persons from outside of the peoples who may commit errors deriving from a lack of knowledge or to serve their own interests; d) it aims to contribute to the empowerment of peoples via the recognition of the free declaration of identity as a member of one of those peoples; e) it is an explicit element of the official definition adopted by various countries in their legislations, which means it is adopted by public policies to determine their target populations. Thus, from a rights-based approach, the dimension of self-recognition is the most adequate means for capturing the size of the indigenous peoples and Afro-descendant populations of the region’s countries, using the criterion of self-identification. However, although its potential lies above all in quantification, it also allows a good approximation for measuring the social indicators that reflect their life conditions and the inequalities that affect them. It is worth noting that when contrasting the results of the indigenous population count when more than one variable is available–for example, self-definition and language–the size of the population may differ, in some cases significantly, given that in some countries the loss of the original language is considerable, especially among the younger generation. However, in general the differences between the social indicators calculated according to two different criteria were not marked, and both variables made visible the deep gaps between the indigenous and non-indigenous populations in matters such as fertility, infant mortality and others (Schkolnik y Del Popolo, 2005). Although it is the top priority, the criterion of self-identification is not free of bias. In contexts of strong ethnic discrimination it is possible that its utilisation leads to significant underestimates of certain groups. In contrast, in scenarios in which processes of cultural revitalisation are occurring or where indigenous and Afro-descendants have greater prominence, it is probable that this criterion gives rise to the identification of persons who feel socially or politically close to these groups, communities or peoples. However, in Latin America the bias of exclusion and the consequent underestimation of the indigenous and Afrodescendant populations would appear to have a greater impact than that of “false” inclusion, owing principally to discrimination and also to the strong influence of the dominant culture, especially in urban settings.

314

Chapter Ten

The validity of the criterion of self-identification can also be seen to be affected by the extent to which indigenous and Afro-descendants consider the census to be a legitimate instrument and by their degree of commitment to the operation. For this reason it is vitally important to ensure participation by the organisations in the whole process, from the design of the instruments, through the data gathering and awarenessraising campaigns, to the analysis and use of the information generated. Nevertheless, the potential biases do not invalidate a priori the use of the self-identification criterion; rather, they should drive the search for and implementation of strategies that lead to its success, in both the technical and political realms.

3.2. A multidimensional “who” The other dimensions of ethnic identity allow the acquisition of knowledge about persons who self-define as ethnic group members: how they live, the relationship with their common ancestors, the survival of their link with their territories, and their different degrees of attachment to the culture of origin. Hence its potential lies in the possibility for discerning their characteristics, their internal heterogeneity, the reach of the process of assimilation, and the gaps between those who identify as such and those who do not. This potential depends, of course, on the selection of the most adequate indicators for each dimension within the context of each country, and the formulation of questions capable of understanding the multidimensionality of ethnic identity. The national experiences reveal a lack of qualitative studies conducted with indigenous and Afro-descendant participation that would help to properly address these issues. Similarly, the fact that someone is not a bearer of certain characteristics related to these other dimensions–such as indigenous language, the practice of certain rituals or living in their territories–does not invalidate the right to self-identify as indigenous or Afro-descendant. For example, the use of a native language was considered to be a privileged indicator of indigenous identification, above all for the high value that the peoples themselves give it for its capacity to preserve identity through transmission across generations. However, owing to migration to the cities, increased integration of the populations into the global economy and greater schooling, there has been a loss of native languages. This has led language to become an insufficient indicator for quantitative purposes, although it retains its potential for visualising the heterogeneity within and between the indigenous peoples; for example, by

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 315

using monolingualism to identify the most “isolated” sub-groups and analyse the way in which this has a bearing on their conditions of life.

3.3. The crux of the matter In this section, the question being considered is how to capture ethnic identity from the sources of sociodemographic data, and specifically the census. Before answering this, it is useful to review the way in which this has been done until now. In this regard there has been a change in the method of including the ethnic approach in the censuses. Whereas between 1980 and 1990 it involved formulating questions chiefly related to language, in the censuses of 2000 the questions mainly utilised the criterion of self-identification (Table 10-1). Implicitly, this change also reveals a turn from a conception of indigenous peoples as beneficiaries of policies who can be identified indirectly through external features or cultural expressions, toward a perspective that conceives of them as subjects of rights. In the case of Afro-descendants, the criterion of identification in those censuses that did inquire about the topic has been systematically that of self-definition, except in Cuba, where the census-taker registers it according to his or her perception. However, this presents an ambiguous situation, for the classification constructed to gather this information responds to racial categories–in the sense of the social construction based on the phenotype. Brazil and Cuba, countries with a long statistical trajectory in these matters (since 1879 in the case of Brazilian censuses and for over 200 years in the Cuban case), have tended to equate the concept of ethnic group with race in their censuses, reducing it to skin colour. Although it is clear that the exercise of discrimination is based above all on these characteristics, it is also certain that, to the extent to which “racial groups” adopt an ethnic identity and proclaim it collectively, it is possible to consider the four dimensions proposed–that is, selfrecognition, common origin, culture and territoriality–to achieve a more comprehensive measure of Afro-descendant identity.

Chapter Ten

Census in indigenous territories

Cuba: Afro-descendants

El Salvador: Afro-descendants and indigenous

Ecuador: Afro-descendants and indigenous

Colombia: Afro-descendants, indigenous, Roma (Gypsies)

Costa Rica: Afro-descendants, indigenous, Chinese

Census of population

External identification

Self-identification Language spoken

Self-identification

Brazil: Afro-descendants and indigenous

Chile: indigenous

Language spoken Home language

Complementary Survey

Census of population

1980

Bolivia (Plurinational State of): indigenous

Argentina: indigenous

Country/ Ethnic groups included

Home language

Self-identification Language spoken

Self-identification

Self-identification

Language spoken

Census Round / Criteria 1990

External identification

Self-identification

Self-identification Language spoken

Self-identification Self-identification Language spoken Mother tongue Self-identification Language spoken

Self-identification

2000 Self-identification at household level Self-identification Indigenous descent Self-identification Language spoken Mother tongue Self-identification

Table 10-1. Latin America: identification criteria for Afro-descendant and indigenous populations in the 19802000 censuses

316

Census of population

Venezuela (Rep. Bol. de): indigenous

Indigenous census

Census of population

Indigenous census

Peru: indigenous

Paraguay: indigenous

Census of population

Nicaragua: Afro-descendants, indigenous, mestizos of the Caribbean Coast Panama: indigenous

Mother tongue Self-identification, certain areas

Language spoken or heard spoken by mother or grandmother, certain areas

Home language

Self-identification

Mother tongue

Language spoken

Language spoken Home language

Language spoken

Language spoken

Mexico: indigenous

Language spoken

Language spoken

Self-identification Language spoken Mother tongue Indigenous dress

Census Round / Criteria

Honduras: Afro-descendants and indigenous

Guatemala: Afro-descendants and indigenous

Self-identification Indigenous dress Indigenous footwear Home language

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting

Self-identification Language spoken

Self-identification Language spoken

Mother tongue

Self-identification Language spoken Self-identification Language spoken Self-identification Self-identification Language spoken Home language Self-identification Language spoken Territory

Self-identification

Self-identification Language spoken Mother tongue

317

318

Chapter Ten

Note to Table 10-1: In the case of the 1990 Ecuador census, language only permitted identification of indigenous persons. In Brazil, the category of indigenous was included in the 1991 census; in the case of Guatemala, the Afrodescendant population only included the Garifunas. Each country’s census dates can be consulted on the web page of the Latin American and Caribbean Demographic Centre-CELADE-Population Division of ECLAC that deals with the year 2000 Censuses: http://www.cepal.org/cgibin/getprod.asp?xml=/celade/noticias/paginas/6/15116/P15116.xml&xsl=/celade/t pl/p18f.xsl&base=/celade/tpl/top-bottom_cen.xsl. Source: Updated from Schkolnik and Del Popolo (2005) and Del Popolo (2008).

There has also been a significant increase in the number of countries in the region that inquire about the indigenous and Afro-descendant population in their censuses. While in 1970 and 1980 only isolated census enumerations were undertaken, analysis of the two latest rounds (1990 and 2000) reveals growing awareness of the need to give statistical visibility to these groups–although as indicated earlier, this has occurred in a less generalised way in the case of Afro-descendants. Examination of census questionnaires and household surveys shows that queries to identify indigenous peoples have been introduced with increasing frequency–specifically, in 16 of the 19 countries that conducted their census in the year 2000. For Afro-descendants the panorama has been less encouraging, as identification of them was attempted in only around half of those censuses: nine of the 19 countries (and in the case of Guatemala only the Garifuna were identified). However, behind the apparent homogeneity suggested by the adoption of the criterion of self-identification, there is great heterogeneity, both in the formulation of questions and in the categories of response that are included (Table 10-2).

Chile (2002)

Cuba (2000)

Costa Rica (2000)

Colombia (2005)

Categories 17 indigenous peoples are listed plus the category “other people”

1. Quechua, 2. Aymara, 3. Guaraní, 4. Chiquitano, 5. Mojeño, 6. Other native, 7. None 1. White, 2. Black, 3. Brown, 4. Yellow, 5. Indigenous In accordance with your culture, people 1. Indigenous, 2. Roma, 3. Raizal from or physical features, ... are you or do Archipelago of San Andrés & you recognise yourself as: Providencia, 4. Palenquero of San Basilio, 5. Black, mulatto, AfroColombian or Afro-descendant, 6. None of the above Do you belong to the … culture: 1. Indigenous, 2. Afro-Costa Rican or black, 3. Chinese, 4. None of the above What is the colour of your skin? 1. White, 2. Black, 3. Mestizo or mulatto Do you belong to one of the following 1. Alacalufe (Kawashkar), 2. original or indigenous peoples? Atacameño, 3. Aymara, 4. Colla, 5. Mapuche, 6. Quechua, 7. Rapa Nui, 8. Yámana (Yagán), 9. None of the above

Country and census date Question Argentina (2001)a Is there a person in this household who recognises him or herself as descended from or belonging to an indigenous people? Bolivia (Plurinational State Do you consider yourself a member of of) (2001) one of the following original or indigenous peoples? Brazil (2000) Your colour or race is…

319

Yes. Closed question

No

No

If answered 1, “To which indigenous people do you belong?” (open question)

Yes. Option to specify people of “other native” No

Identifies people? Yes. Closed question at level of household

Table 10-2. Latin America: questions associated with the criterion of self-identification in the censuses of 2000

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting

Mexico (2000)b

Honduras (2001)

Guatemala (2002)

El Salvador (2007)

Categories Identifies people? 1. Indigenous, 2. Black (AfroIf answered 1, Ecuadorian), 3. Mestizo, 4. Mulatto, 5. “To which indigenous White, 6. Other people or indigenous nationality do you belong?” a) Are you….? a) 1. White, 2. Mestizo (mix of white with indigenous), 3. Indigenous (answer b), 4. Black (of race), 5. Other b) If you are indigenous, to what group b) 6. Lenca, 7. Kakawira (Cacaopera), Yes. 8. Nahua Pipil, 9. Other (specify) Open option to specify do you belong? people in “other” category To which ethnic group do you belong? 22 indigenous peoples listed, plus Yes. categories Afro, indigenous, ladino, Closed question others To which population group do you 1. Garifuna, 2.Black English, 3. Tolupán, Yes. belong? 4. Pech (Paya), 5. Miskito, 6. Lenca, 7. Closed question Tawahka (Sumo), 8. Chortí, 9. Other Are you Nahuatl, Maya, Zapotec, 1. Yes; 2. No No Mixtec or from another indigenous group?

Chapter Ten

Country and census date Question Ecuador (2001) How do you consider yourself...?

320

Only at level of household, persons are not identified. Argentina utilised this question to define the survey in depth for indigenous peoples. b Only in the extended questionnaire. c Filter question used: “indigenous” in the list of house dwellers, among other variables. d Question included in the questionnaire of the general census, at the end of the form. Source: Del Popolo (2008).

a

Venezuela (Bolivarian Republic of) (2001)

Paraguay (2002)d

Panama (2000)c

321

Categories Identifies people? Q6. 1. Yes; 2. No Yes. For those who answer Yes: Closed question Q7. 13 categories of indigenous peoples, Afro-descendants and other ethnic groups are listed To which indigenous group do you 1. Kuna, 2. Ngöbe, 3. Buglé, 4. Yes. belong? Teribe, 5. Bokota, 6. Emberá, 7. Closed question Wounaan, 8. Bri Bri, 9. None Is there a person in this household who 17 indigenous groups are listed Yes. considers him or herself to be Persons are listed and indigenous or to belong to an the ethnicity recorded. indigenous ethnicity? Do you belong to an indigenous 1. Yes; 2. No In affirmative case, people? specify the people

Country and census date Question Nicaragua (2005) Do you consider yourself to belong to an indigenous people or an ethnicity? To which of the following indigenous peoples or ethnicities do you belong?

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting

322

Chapter Ten

The analysis of the 17 population censuses of the 2000s decade that asked about ethnic origin reveal that 16 of them included a question directed at capturing indigenous peoples, except Cuba, which only counted Afro-descendants. In turn, in eight of these 16 countries there was simultaneous incorporation of categories to identify indigenous and Afrodescendants (Brazil, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Honduras and Nicaragua). In Colombia and Costa Rica other ethnic groups were also considered–Roma people in the former and Chinese in the latter. If the conceptual differences in the drafting of the question are considered, it can be noted that in the Brazilian case there is a direct mention of skin colour or race. This criterion has been employed in censuses in that country since the end of the nineteenth century (heteroclassification was used in the first censuses, and self-identification was implemented from 1950), with the exception of the census of 1970. Although in principle this classification system has been functional for its socio-historic context, characterised by structural racial discrimination, it is not exempt from criticisms. Race, as a perceived phenotypic category, in combination with other social factors such as education and income, configures a racial identity closely tied to the concept of social class (Hasembalg, 2006; Magno de Carvalho and Wood, 2005). Some studies have shown that with social mobility a reclassification of racial ascription is produced: the higher a person is located in the social hierarchy, the more he or she will tend to be “whitened”, whether by self-definition or through assignation on the part of the interviewer (who would tend to locate the interviewee in lighter categories of colour). This introduces biases in the analysis of the black population’s social progress through time, for example (Magno de Carvalho and Wood, 2005). Moreover, this criterion limits the identification of indigenous peoples, insofar as it deals with peoples recognised by the State and ethnically differentiated. In the seven countries that identified the indigenous population solely by means of self-identification, it was asked explicitly–and in a quite direct way–whether the person belonged to an indigenous people, indigenous group or indigenous ethnicity (the term changed in each country, with predominance of the use of the category “people”). Mexico constituted an exception by asking directly if the person “is” indigenous. In the formulation of the question in Argentina there was also incorporation of other elements referring to the dimension of descent or common origin, by inquiring if the person recognised him or herself as the descendant of an indigenous people.

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 323

In the remaining eight countries that simultaneously considered indigenous and Afro-descendants there was a great diversity in the formulation of questions and in response categories. The first aspect to highlight is that in all of them, except Nicaragua, categories of recognition of ethnic membership and racial categories were included together, in some cases even in the wording of the question as well. The consequences this has on the quantification of ethnic groups are difficult to evaluate because the answers may depend strongly on the pressure of each country’s sociocultural and political context and on the level of ethnic consciousness of individuals (Hernández, 1994), plus the fact that confusions can be produced upon combining the criterion of peoples or ethnic groups with phenotypic criteria. Thus, the biases derived from the operational decisions adopted in the census regarding the formulation of questions for the various groups can be multiple and difficult to predict; hence it has been suggested that “empirically it is not consistent to equate a racial factor with an ethnic one and then expect the same result from self-identification” (Antón, 2010: 20). By way of example, in Ecuador the principal problem seems to have been underestimation of the population of the indigenous peoples by also including the category of “mestizo”, which ended up being highly selected, and thereby defeated the expectations indigenous organisations held of verifying a much higher indigenous proportion of the population (Guerrero, 2005). However, other types of problems related to the selfexclusion of the indigenous population regarding census operations cannot be ruled out, in some cases possibly due to discrepancy between the wording selected by the National Statistics and Census Institute (INEC) for the identification question and the proposal of certain indigenous organisations in 2001. In the 2007 census in El Salvador two consecutive questions were used: membership of an indigenous people and self-identification with a racial group. However, the order in which the categories were listed can be questioned, as can the wording of some of those categories, particularly given that this is a country where possession of black or indigenous physical features cannot be excluded as a source of social stigma. Further, this was the first time El Salvador had attempted this type of measurement in a census. The changes in the question in the Colombian census of 2005 from that of 1993 appears to have yielded better results: an indigenous population figure of 1.6% in the first census became 3.4% in the second; in the case of the Afro-Colombians, there was a figure of 1.5% in the census of 1993 but 10.6% in that of 2005. The principal changes affirmed the inclusion of

324

Chapter Ten

the concept of people, culture and physical features, all in the same question, which was in keeping with the diversity of ethno-racial identities in that country. Similarly, the categories appear to have captured the various local terms by which identities are referred to. In the case of Nicaragua there was a change in the criterion of classification: mother tongue was used in 1995 but the 2005 census relied on self-definition. Further, in the formulation of the question there was direct reference to belonging to an indigenous people or ethnicity, offering response categories that had meaning for the communities themselves. There was also an improvement in the level of participation of indigenous peoples and Afro-descendants in the census operation. Finally, the 2005 census concluded that 8.9% of the population belonged to ethnic communities or peoples, compared with 2.4% in the previous census. Analysis of the population under consideration reveals that in all the countries a question was applied to all persons, except in the Plurinational State of Bolivia, where it was asked of persons aged 15 and above, and Mexico, where it applied to persons aged five and above. A lack of information about persons under these ages prevents sociodemographic characterisation of the population of children and adolescents, the ones who must be the targets for provision of services with cultural specificity to close existing gaps; this leads to subsequent recourse to indirect and approximate methods to estimate the missing segment of population. Finally, the review shows that not all countries identified the distinct peoples in the census form. Brazil, Mexico and Costa Rica censused the indigenous population as a whole, and in the latter case only identified the peoples resident in indigenous territories. In general, in countries where indigenous peoples are numerous it was decided to incorporate an open category for their specification, as in the Plurinational State of Bolivia, Ecuador and the Bolivarian Republic of Venezuela. The results of the Ecuadorian census of 2001 were not completely satisfactory, given that there was approximately a 50% lack of response to the open question to identify nationality or indigenous people. The differences in the way the question is dealt with in each country may result from divergences in the appreciation of the ethnic world, to the diverse uses that are attempted to be given to the information obtained, or to the heterogeneity that is verified between and within ethnic groups. This scenario therefore makes it difficult to suggest a similar question for all the countries of Latin America and the Caribbean with a view to having rigorous comparability of data. However, these national experiences do permit some guidelines to be set out, at least in relation to approaches that are not recommended.

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 325

The lack of comparability could be linked, in some cases, to semantic divergences, and Chile exemplifies this situation. In 1992 the question asked about belonging to a “culture” while in 2002 the question referred to “original or indigenous peoples”, and the results showed a significant reduction in this population between the two censuses in both absolute and relative terms–from 10.4% to 4.5%. One of the explanations for this contrast relates precisely to the form in which the questions were formulated in the questionnaires, as they alluded to different subjective requirements of membership commitment: in the case of Chile it has been observed that individuals identify themselves with the Mapuche “culture”, which does not mean that they will affirm membership of the Mapuche “people” when the question is direct (Gunderman, Vergara and Foerster, 2005). Similar situations have been produced in Colombia and Costa Rica. Nonetheless, it should be borne in mind that changes between one instrument and another within the same source and country can also arise from the necessary revisions that tend to be made and that are eminently valid, above all when they respond to a broadening in the participation of the people themselves in the process of definition and testing. Other limitations relate to the fact that in some cases there is identification only of the indigenous or Afro-descendant (or other) population but not of the specific peoples or communities, or else peoples are identified with different levels of detail, or by means of closed questions in some cases and open in others. The categories of response that are offered can differ to a great extent: in some cases, for example, “other native” is utilised without the interviewee having to specify what he or she refers to with this broad concept, while in others it is necessary to detail the uncategorized indigenous or Afro-descendant peoples explicitly. Another difficulty evidenced by prior experience relates to the lack of clarity about the purpose of the ethnic identification query, which does not aim at classifying the whole population, but instead at identifying particular ethnic groups. Thus it is not advisable to include categories of response such as “mestizo” or “mulatto”, which can draw responses from other categories, or ones that are not what they purport to capture, such as “white”. The ideal would be to mention specifically the ethnic groups that are desired to be identified and for which focused policies can be proposed. If there are difficult-to-access zones where certain ethnic groups reside, this may affect the census operation’s estimations of population. Other factors of potential impact are inadequate training of census takers, communication difficulties in multilingual areas, or insufficient participation by the actors involved.

326

Chapter Ten

It is therefore evident that the incorporation of the ethnic approach in the sources of data, and specifically the operationalisation of the selfidentification criterion, is complex and delicate, and requires prior qualitative studies, debates and reflections, and the implementation of pilot surveys. It involves a sensitive issue that will be crucial in terms of the results that are obtained and, in an indirect manner, of the potential for these results to be translated into policies with a bearing on the inequitable reality lived by indigenous peoples and Afro-descendant populations.

4. Hands to the wheel: some recommendations regarding the “how” The step from discourse to action is not generally straightforward. Nonetheless, it is possible to synthesise some recommendations for applying the ethnic approach to a source of sociodemographic data, or else broaden or modify the way in which it has been done. These recommendations are focused on the identification of indigenous peoples and Afro-descendant populations. However, each country should determine if it is appropriate to include other ethnic groups that are also in situations of vulnerability and exclusion. Although special emphasis is placed on censuses, there is also reference to other sources of data.

4.1. General recommendations about the ethnic approach a)

Include identification of indigenous peoples and Afro-descendants in population and housing censuses by means of a criterion of selfidentification in accordance with the prevalent standards of rights. b) Guarantee the full and effective participation of indigenous peoples and Afro-descendants in all stages of the census process, including the formulation and application of the questionnaire, pilot surveys, compilation of data, systematisation, analysis and dissemination of information. c) Apply pilot surveys as a way to evaluate every alternative for exploring the optimal incorporation of the ethnic approach in population censuses. d) To ensure good coverage it is recommended that particular care be taken in zones that are difficult to access. In spite of this, States must respect the rights of indigenous peoples in voluntary isolation and in process of initial contact. e) Consider the different socioterritorial contexts, identifying the monolingual and multilingual areas in order to anticipate the need

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 327

for translation of questionnaires; the training of interviewers and supervisors, who should belong to the respective indigenous and Afro-descendant communities; the adequate design of communication campaigns and dissemination of results, with appropriateness in terms of cultural, linguistic and other characteristics. f) Design campaigns of digital communication directed at indigenous peoples and Afro-descendant populations in order to strengthen processes of ethnic revitalisation and enable them to assume their identity in the census process, particularly in urban settings. Also develop campaigns directed at society in general to promote recognition and respect for cultural diversity. g) Strengthen the capacity to utilise and diffuse census-derived information about indigenous peoples and Afro-descendant populations in a bidirectional way: democratising technical knowledge among indigenous and Afro-descendant professionals, experts and leaders, while also educating information producers and decision makers about the beliefs, culture and practices of these social groups.

4.2. Specific recommendations about the approach for indigenous peoples a)

In the next census round, include a self-identification question in the census forms applicable to all persons, thereby continuing with the approach followed by the majority of the region’s countries in the censuses of 2000. b) Those countries that are still utilising a question about language to ascertain the population of indigenous peoples should also include self-identification, thereby embracing a rights-based approach. Equally, it is relevant to ask about language, as this provides a deeper understanding and facilitates the construction of indicators regarding collective rights. c) With respect to the formulation of self-identification questions and the categories utilised in them, the key recommendation is to do so in collaboration with the indigenous peoples. With this as a guiding principle, we suggest following the approaches supported by prior experiences. d) If the country also inquires about other ethnic groups, consider the necessity of separately asking the question regarding indigenous peoples, in order to avoid a classification that involves categories that are ambiguous or not very specific (for example, “mestizos”).

328

Chapter Ten

The experience of certain countries has demonstrated that this leads to problems in assessing the indigenous population. e) In accordance with the context of each country, the introductory clause of the question should be as direct as possible, avoiding lengthy wording that would lead to varied interpretations by the interviewer and/or be difficult for him or her to read. f) Although the term “indigenous people” is adopted at the international level, each country should use the terms that its own peoples recognise: in Ecuador, for example, they recognise themselves as “peoples” and “nationalities”; in Argentina as “aboriginals”. g) It is fundamental to include not only self-recognition as indigenous but also to identify to which specific people the person belongs to, and to do so for the entire national territory. h) If changes are made to the self-identification question, it is suggested that pilot surveys be undertaken (for example by means of household surveys), bearing in mind the various problems that may arise in different social and geographic contexts. i) The ethnic approach involves more than the inclusion of questions about self-identification and indigenous languages; it also embraces the cultural appropriateness of the contents of the census form (or other instrument). Revision of the census form entails reviewing the dimensions, variables and categories used, and incorporating new elements to construct an instrument that is more inclusive and has greater consideration of cultural diversity. This task should be undertaken in conjunction with the indigenous peoples, using the relevant recommendations of the Permanent Forum on Indigenous Issues and the guidance in the United Nations census manual as a frame of reference. A specific example: in the question regarding the type of school attendance, an option about intercultural bilingual education could be included. j) In relation to the possibility of considering new instruments, one strategy could be to carry out an “indigenous census” in parallel with the general census (as in Costa Rica, Paraguay and the Bolivarian Republic of Venezuela). If this option is selected, the recommendation is to design a complementary form that constitutes part of the census process, applicable in the indigenous territories– and not the undertaking of an independent census. In this sense, the self-definition question must form part of the general questionnaire administered through the country. The form designed for the indigenous territories should allow inclusion of a minimum set of

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 329

questions that are comparable with those at the national level, adapt other variables to the territorial context (as the Bolivarian Republic of Venezuela does regarding household items), and incorporate new dimensions, such as the recognition of traditional indigenous activities; there could also be a community questionnaire, as applied in Paraguay and Peru. These experiences could serve as a basis for including information in the next round that would permit monitoring of collective rights. Other examples are Argentina, which executed a complementary survey of indigenous peoples as part of the census process, or the post-census surveys undertaken by Canada and Australia. It is therefore recommended that for the next census round, countries explore these or other alternatives in line with their realities. k) To address the call for cultural adaptation of the indicators in statistical instruments, it is recommended that at least one question be included that relates to indigenous language. To make this possible with a single question, it is suggested that it be asked of those persons who declare membership of an indigenous people, for example: “Do you speak the language of your people?” This would permit the construction of indicators for the rights of indigenous peoples, such as the percentage that speak their traditional language. l) In countries that have recognised indigenous territories and lands (which do not necessarily coincide with the political-administrative boundaries of the State), the corresponding identification should be included in the census form. All countries should evaluate the possibility of defining the territorial spaces and limits of their indigenous peoples. This task should be accompanied by development of the respective census cartography and undertaken in conjunction with the peoples. In this sense, it is recommended that there be consideration of elements that would permit census microdata to be linked with indigenous territorial units in the cartography, in order to carry out a valid georeferencing of the information. m) A perspective of territorial rights would require the statistics to provide information about aspects of the indigenous territories (social, demographic, biotic, physiographic, among others), including the location of human settlements and their spatial distribution. However, political-administrative units do not generally coincide with these territories, even at lower levels, and it can be difficult to reconstruct these territorial units from census

330

Chapter Ten

segments, which reduces the likelihood of meeting these requirements. Hence it is important to determine the geographic limits of the territory considering the real limits of the sociocultural interaction of each people. This task is complex when, for example, various ethnicities reside in the same territorial space, or when a people’s members are spread through a territorial continuum crossing extensive geographic areas of a country (as in Mexico and Guatemala). However, there are countries that have achieved advances in this regard, such as Paraguay and the Bolivarian Republic of Venezuela.

4.3. Specific recommendations about the approach for Afro-descendants a)

It is recommended that in the next census round an increased number of countries consider gathering information on Afrodescendants from a rights-based perspective, in order to overcome the imbalance between the identification of indigenous peoples and that of Afro-descendants (in countries that already ask questions on ethnic identification), with self-identification as the criterion of classification. Given the particular relation of the Afro-descendant population with the notion of people, commented upon previously, it will be necessary for each country to make an evaluation, in conjunction with Afro-descendant organisations, regarding the best method for capturing them. b) It is important to give attention to the terms utilised in the framing of the question, given that Afro-descendant corresponds to an ethno-racial distinction, even when the local term to designate it is a phenotypic category such as “black”. It is also necessary to consider the various local meanings of the categories used and their social and territorial variations. For example, as well as the opposed attitudes generated by the denomination “black”, which is stigmatising for some Afro-descendants while representing an assertion of socio-racial pride for others, there are also some groups that do not recognise themselves with this label nor with skin colour, but instead with their own terms, such as the Raizales of the archipelago of San Andrés and Providencia or the Palenqueros of San Basilio, in the case of Colombia. Some countries utilise a denomination that is associated with nationality, such as AfroEcuadorians or Afro-Colombians. In this case it is recommended that there be consideration of an additional category referring to

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 331

c)

Afro-descendants who are not native to the country in which they are being censused. In some countries of the region there are Afro-descendant groups who conserve their own language, and even a strong ethnoterritorial identity. The recommendations made for indigenous peoples in relation to language and territoriality are therefore also applicable to these cases, and in consequence it is suggested that these realities be identified and considered in each country.

Bibliographical references Aguilar Cavallo, Gonzalo. 2006. La aspiración indígena a la propia identidad. Universum, 21 (1): 106-119. http://www.scielo.cl/scielo.php?script=sci_arttext&pid=S071823762006000100007 Antón, Jhon. 2010. La experiencia afrodescendiente y la visibilidad estadística en el Ecuador. Santiago de Chile: Naciones Unidas (CEPAL). http://www.eclac.org/publicaciones/xml/0/42700/W368afrodescendien te.pdf Bello, Álvaro and Marcelo Paixão. 2009. Una mirada a la situación de los derechos de los afrodescendientes en América Latina. In Jhon Antón et al. (ed.), Afrodescendientes en América Latina y el Caribe: del reconocimiento estadístico a la realización de derechos. Santiago: Naciones Unidas (CEPAL), pp. 39-84. Bodnar, Yolanda. 2005. Colombia: apuntes sobre la diversidad cultural y la información sociodemográfica disponible en los pueblos indígenas. Notas de Población, 79: 231-262. CELADE/CEPAL–Centro Latinoamericano y Caribeño de Demografía/Comisión Económica para América Latina y el Caribe. 2009. Censos 2010 y la inclusión del enfoque étnico: hacia una construcción participativa con pueblos indígenas y afrodescendientes de América Latina. Santiago de Chile: Naciones Unidas (CEPAL) CEPAL–Comisión Económica para América Latina y el Caribe. 2007. Panorama social de América Latina 2006. Santiago de Chile: Naciones Unidas (CEPAL). Del Popolo, Fabiana. 2008. Los pueblos indígenas y afrodescendientes en las fuentes de datos: experiencias en América Latina. Santiago de Chile: Naciones Unidas (CEPAL).

332

Chapter Ten

Deruyttere, Anne. 2001. Pueblos indígenas, globalización y desarrollo con identidad: algunas reflexiones de estrategia. Banco Interamericano de Desarrollo. http://www.rimisp.org/fida_old/documentos/docs/pdf/0040-002317pueblosindiacutegenasglobalizacioacuten.pdf Gamboa, Juan Carlos. 2006. Notas sobre el concepto "pueblos indígenas". www.utopiaverde.org/modules.php?name=News&file=article&sid=34 4 González Pazos, Jesús. 2007. Bolivia: la construcción de un país indígena. Barcelona: Editorial Icaria. Guerrero, Fernando. 2005. Población indígena y afroecuatoriana en Ecuador: diagnóstico sociodemográfico a partir del censo de 2001. Santiago de Chile: Naciones Unidas (CEPAL-BID). Gundermann, Hans, Jorge Iván Vergara and Rolf Foerster. 2005. Contar a los indígenas en Chile: autoadscripción étnica en la experiencia censal de 1992 y 2002. Estudios Atacameños, 30: 91-115. http://www.scielo.cl/pdf/eatacam/n30/art06.pdf Hasembalg, Carlos. 2006. Ciclos de vida y desigualdades raciales en Brasil. In CEPAL (ed) Pueblos indígenas y afrodescendientes en América Latina y el Caribe: información sociodemográfica para políticas y programas. Santiago de Chile: Naciones Unidas (CEPAL), pp. 167-196. Hernández, Isabel (1994). Población y cultura: el caso de los pueblos indígenas en Bolivia. In Estudios sociodemográficos de los poblaciones indígenas contemporáneas, Santiago de Chile: CELADE-ICI-FNUAPCIDOB. Hopenhayn, Martín and Álvaro Bello. 2001. Discriminación étnico-racial y xenofobia en América Latina y el Caribe. Santiago de Chile: Naciones Unidas (CEPAL). ILO–International Labour Organization. 1989. Convention Nº 169 on Indigenous and Tribal Peoples in Independent Countries. Geneva. [On line, consulted on May 4, 2012 http://www.ilo.org/dyn/normlex/en/f?p=NORMLEXPUB:12100:16582 11106972734::NO:12100:P12100_ILO_CODE:C169:NO] Magno de Carvalho, José and Charles Wood. 2005. Estimating the Stability of Census-Based Racial/Ethnic Classifications: the Case of Brazil. Paper presented at the XXV International Population Conference, Tours (France). http://iussp2005.princeton.edu/download.aspx?submissionId=50412 Martínez Cobo, José. 2004. The Concept of Indigenous Peoples. Background Paper Prepared by the Secretariat of the Permanent

Indigenous Peoples and Afro-Descendants: The Difficult Art of Counting 333

Forum on Indigenous Issues. Workshop on data collection and disaggregation for indigenous peoples (New York, 19-21 January 2004). New York: United Nations. [ref PFII/2004/WS.1/3]. Mindiola, Omaira. 2006. Gobernabilidad y consulta previa a los pueblos indígenas. Fundación Canadiense para las Américas. http://dspace.cigilibrary.org/jspui/bitstream/123456789/11009/1/Gober nabilidad%20Y%20Consulta%20Previa%20A%20Los%20Pueblos%2 0Indigenas.pdf?1 Oyarce, Ana María, Malva Marina Pedrero and Gabriela Pérez. 2005. Criterios étnicos y culturales de ocho pueblos indígenas de Chile. Notas de Población, 79: 133-170. Permanent Forum on Indigenous Issues. 2004. Report of the Workshop on Data Collection and Disaggregation for Indigenous Peoples to the Permanent Forum on Indigenous Issues, Third Session. New York. [Document N° E/C.19/2004/2]. Rangel, Marta. 2009. Una panorámica de las articulaciones y organizaciones de los afrodescendientes en América Latina y el Caribe. In Jhon Antón et al. (ed.), Afrodescendientes en América Latina y el Caribe: del reconocimiento estadístico a la realización de derechos. Santiago: Naciones Unidas (CEPAL), pp. 87-103. Schkolnik, Susana. 2000. Algunos interrogantes sobre las preguntas censales para identificar población indígena en América Latina. Paper presented at the Seminario “Todos contamos”: los grupos étnicos en los censos, I Encuentro Internacional, Cartagena de Indias, Colombia. Schkolnik, Susana and Fabiana Del Popolo. 2005. Los censos y los pueblos indígenas en América Latina: Una metodología regional. Notas de Población, 79: 101-132 Stavenhagen, Rodolfo (1991). Los conflictos étnicos y sus repercusiones en la sociedad internacional. Revista Internacional Ciencias Sociales, 127(1): 125-140. http://www.cholonautas.edu.pe/modulo/upload/CONFLICTOS%20ET NICOS.pdf Stavenhagen, Rodolfo. 2001. El derecho a la sobrevivencia: la lucha de los pueblos indígenas en América Latina contra el racismo y la discriminación. Santiago de Chile: CEPAL-BID, mimeo. http://www.iidh.ed.cr/comunidades/diversidades/docs/div_enlinea/stav enhagen-derecho%20sobrevivencia.htm

334

Chapter Ten

Notes 1

http://www.cinu.org.mx/prensa/especiales/indigenas/ [accessed 22.03.2012] See http://www.garifunaresearch.com for general information on the Garifuna people. 3 In all the Spanish and Portuguese colonies in the Americas the slave population found zones of flight and resistance. In Brazil and the La Plata river region the communities established by escaped slaves were called quilombos, while in other parts of the continent they were known as palenques. Mambices, cumbes, ladeiras and other denominations were also adopted by the rebel slaves. 2

CONTRIBUTORS

Luis Fernando Angosto Ferrández is a Lecturer in Anthropology and Spanish and Latin American Studies at the University of Sydney. He graduated from the University of Granada with a degree in political science and sociology and then completed a Master of Philosophy in folklore and ethnology at University College Cork. His PhD in social anthropology at Queen’s University Belfast examined the relationships between the Venezuelan state and indigenous peoples within the context of the Bolivarian political process. After completing his PhD he worked and researched in Venezuela for five years. His research interests include the relationships between indigenous peoples and states in Latin America; ethnicity and race; multiculturalism and plurinationality; and electoral processes, political agency and organisation. Pilar Barrientos trained as a primary school teacher (1986) and holds a degree in Education Sciences from the Faculty of Philosophy and Letters of the University of Buenos Aires (1995). She is currently completing a Master’s degree in Social Anthropology at the same university with a thesis that examines originary peoples–state relations. She is a member of two research groups at the Faculty of Philosophy and Letters of the University of Buenos Aires within the programme on “Political economy and the social formation of frontiers: Ethnicities and territories redefined”, headed by Dr. Sebastián Valverde from the Institute of Anthropological Sciences at the same university. She has acted as a job training consultant at the National Institute of Statistics and Censuses since 1996. She has advised other statistical institutes within Latin America (Bolivia, Peru and Honduras). She is a member of the Ibero-American Network of Experts on Public Administration Management (National Institute of Public Administration – Government of Spain – CEDDET Foundation). Gemma Celigueta Comerma is an Associate Professor in Latin American History at the University of Barcelona. She received her doctorate in social anthropology from the École des hautes études en sciences sociales, Paris, in 2009. She has carried out ethnographic research among the K’iche’ Indians of Guatemala. Her research expertise includes

336

Contributors

political anthropology, ethnography and regional ethnology (Mesoamerica; Guatemala). She is joint editor with Gemma Orobitg of the book Autoctonía, poder local y espacio global frente a la noción de ciudadanía. La experiencia indígena y afroamericana desde la interacción política, social y étnica, Publicacions de la Universitat de Barcelona, 2011. Fabiana Del Popolo. After undergraduate studies in Statistics (National University of Rosario, Argentina) and a Postgraduate Diploma in Population Dynamics and Development Policies and Programs at the Latin American and Caribbean Demographic Centre (CELADE), 1990, she is currently undertaking a PhD in Demography at the National University of Cordoba (Argentina). In 1991 she started her professional activities at CELADE, working on issues of childhood health and mortality. Between 1993 and 1999 she worked as a researcher at Foundation Bariloche (Argentina), investigating demographic topics at the sub-national level, poverty and local development, as well as integrated systems of indicators. Between 1993 and 2000 she taught statistics and demography in several degree programmes at FASTA University (Mar del Plata, Argentina). From 2001 she worked in the Demographic Analysis and Information Area of CELADE, developing and coordinating activities of the Regular Program, most recently regarding the situation of indigenous peoples and Afro-descendants in Latin America from a rights perspective, and systems of indicators for monitoring international agreements. She has numerous publications on these issues, most recently linked to the inclusion of the ethnic approach in data sources, indigenous and Afro-descendant youth, and migration of indigenous peoples. She is a co-representative of the Economic Commission for Latin America (ECLAC) at the United Nations Permanent Forum on Indigenous Issues and a representative to the Interagency Group on Indigenous Peoples in Chile. Sabine Kradolfer is an anthropologist and sociologist. She is currently a postdoctoral fellow of the Swiss National Foundation at the Universitat Autònoma de Barcelona (Spain) and at the Universidad Nacional de Río Negro (Argentina). In 2005 she obtained a joint doctoral degree from the Universities of Lausanne (Switzerland) and Paris III - Sorbonne Nouvelle. Her dissertation is based on research on social organisation of the Mapuche communities in the province of Neuquén (Argentina). Alongside her investigations on indigenous peoples in social and cultural anthropology, she was co-director of a research project on women’s academic careers (2006-2008). Between 2000 and 2009 she worked at the University of

Everlasting Countdowns

337

Lausanne as a teaching and research assistant, and as a substitute professor. Gloria Patricia Lopera Mesa has an MA in Anthropology from the University of Antioquia (Medellín, Colombia) and a PhD in Law from the University of Castilla – La Mancha (Spain). She is a Professor at the University of EAFIT Law School and a member of the research group “Justice and Conflict” of the same university, where she has carried out research projects on Theory of Law and Legal Argumentation. Her current work focuses on the relationship between Indigenous Peoples and the Law. Mònica Martínez Mauri is a postdoctoral researcher at the University of Barcelona in the Study Group on Indigenous and Afro-American Cultures (CINAF). She obtained her PhD in Social Anthropology in 2007 from the Autonomous University of Barcelona (UAB) and the École des hautes études en sciences sociales in Paris, after postgraduate studies at the Graduate Institute of Development Studies in Geneva. Since 1999, she has carried out fieldwork in Panama in the Comarca Kuna Yala. Her research and publication focuses on cultural mediation, in particular on international and cross-cultural collaboration in development, environmental declarations, cultural tourism and indigenous rights. She has recently published two monographs: La autonomía indígena en Panamá: la experiencia del pueblo kuna, siglos XVI-XXI) and Kuna Yala, tierra de mar. Ecología y territorio indígena en Panamá (both at Abya Yala, 2011). She is a member of the research group at the UAB on the Anthropology and History of the Social and Political Construction of Identity (AHCISP) and the Interdisciplinary Group on Development and Multicultural Studies (GIEDEM) at the University of Lleida (UdL). José Luis Petruccelli has a doctorate in Demography and Social Sciences from the École des hautes études en sciences sociales, Paris. He is Senior Researcher at the Department of Population and Social Indicators Studies, Instituto Brasileiro de Geografia e Estatística–IBGE, Rio de Janeiro, Brazil. His most recent areas of work are racial inequalities, ethnic/racial identification, public affirmative action policy and ethnic/racial classification categories. As a specialist in race relations, he has conducted research, delivered courses, organised seminars and participated in conferences in Brazil, Uruguay and France. He is the author of two books and has contributed chapters to several other books and

338

Contributors

written dozens of articles for scientific journals, magazines and various international publications. Pablo Regalsky is an anthropologist and has a PhD in Human Geography from Newcastle University. Born in Buenos Aires (1949), he has been permanently resident in Bolivia since 1976. His main interests are in Globalisation and Andean Indigenous Territoriality, Andean Indigenous Peasants’ Livelihood Strategies, and Climate Change. He has conducted more than two decades of continuous in-the-field research action and ethnodevelopment with Quechua communities in Cochabamba (Bolivia) with CENDA (Centro de Comunicación y Desarrollo Andino). He coordinates a Master’s degree in Territoriality and Interculturality at the University of Cochabamba (CESU-UMSS). He has participated in DFID/British Council Higher Education links with the University of Newcastle and currently at DARN/University of Newcastle. He is advisor to the Indigenous Organisations at the Constituent Assembly on issues of Indigenous Territorial Autonomy. He has lectured in Cochabamba, Arica, Buenos Aires, Newcastle, Vienna, and other universities. His most recent books are: Etnodesarrollo, Tierra y Vida. Una alternativa a la crisis alimentaria y energética. CENDA-O’DAM (2011); Las Paradojas del Proceso Constituyente Boliviano (2010); Ethnicity and Class: The Bolivian State and Andean Space Management (2009). Susana Schkolnik holds an undergraduate degree in Sociology from the University of Buenos Aires, Argentina, followed by postgraduate studies in Demography in the Latin American and Caribbean Demographic Centre (CELADE) Santiago (Chile) and a Master of Science in Medical Demography at the London School of Hygiene and Tropical Medicine of the University of London. Her professional activities were developed in the National Institute of Statistics and Censuses (INDEC) of Argentina, at the International Institute for Environment and Development (IIED), London, at the Regional Centre for Higher Education for Latin America and the Caribbean (CRESALC) of UNESCO, Caracas (Venezuela) and, principally, in CELADE, where she retired as Head of the Demographic Area in 2004. Since then she has been a consultant for ECLAC/CELADE, the United Nations Population Fund (UNFPA), the Pan American Health Organization (PAHO) and the Fund for the Development of Indigenous Peoples of Latin America and the Caribbean (Indigenous Fund). Her main interests and her publications have been related to the sociodemographic study of fertility, infant mortality, population trends, population aging and indigenous populations. Most

Everlasting Countdowns

339

recently she has been working on the inclusion of ethnic and gender approaches in censuses of population. David Sulmont is currently Principal Professor at the Social Science Department and coordinator of the Sociology Program of the Social Sciences Faculty at the Pontificia Universidad Católica del Perú (PUCP). He is also a member of the PUCP’s Public Opinion Institute. Between 2001 and 2003 he was head of the data analysis unit and a member of the final report editorial board of the Peruvian Truth and Reconciliation Commission (2001-2003). His current research focuses on ethnicity, race and inequality in Peru and Latin America; and on electoral behaviour and political institutions in the same regions. He is a collaborator of the Project on Ethnicity and Race in Latin America (PERLA) of the University of Princeton and the Comparative Studies of Electoral Systems (CSES). As a consultant he has been an advisor to Truth Commissions, Human Rights NGOs and Human Rights projects in Latin America, Africa, Europe, Indonesia and the Solomon Islands. Néstor Valdivia holds a Licenciatura in Sociology from the Pontificia Universidad Católica del Perú and is a doctoral candidate in Sociology at El Colegio de México. Currently, he is an Associate Researcher in the Development Analysis Group (GRADE), in the areas of Education and Labour Markets, and Poverty, Equity and Ethnicity. He has carried out research projects on small enterprise and the employment prospects and career paths of higher education graduates. He has been a consultant on studies related to processes of decentralisation and regional and local educational management. In recent years, he has developed studies in the fields of social exclusion and poverty of indigenous and afrodescendent people. He has also carried out research to analyse the use of ethnic categories in censuses and surveys in Peru.