French Journal for Media Research

Nuno Oliveira

Talking about boundaries: Symbolic boundaries and ethno-racial repertoires in contemporary Brazil.


The article argues that there is a shift in core conceptions of identity and belonging in Brazil’s national identity narrative. Analysing the discourse on affirmative action in favour of black people in contemporary Brazil, we discuss three differentiated repertoires with implications for the construction of collective boundaries of an ethnoracial nature, which the author calls dilutionnegotiation and salience. These frameworks are ways of talking about boundaries between groups and therefore make it possible to organise the perception of ethnic and racial diversity. Furthermore, the article highlights the importance of narratives of national identity in the ways these frameworks are structured. The article also posits the need to consider the role of the state and its agents in the establishment of the legitimate codifications of belonging.

Texte intégral


1This article seeks to reflect on the cognitive ways of approaching ethno-racial classifications and, ultimately, the processes involved in constructing and adhering to certain collective identities. To that end, the concepts of repertoire, as formulated by Lamont (1995), and of boundaries – specifically, the construction or making of those boundaries – are used. Consequently, the focus will be on ethno-racial boundaries in contemporary Brazil, inquiring into the systems of meaning mobilised for such a social operation.

2The implementation of affirmative action policies in Brazil – particularly, the adoption of quotas for black people in public universities – triggered a heated debate on the boundaries of ethno-racial classification. Controversy over race-based affirmative action divided Brazilian society between those who argued that in a mixed society like Brazil (also genetically mixed) those divisions were artificial and others who argued that, since racial discrimination was a widespread social phenomenon in Brazilian society, race was relevant for whoever felt being a victim of racism. At this juncture, it seems timely to question boundary making between groups, or to look into how individuals position themselves in relation to those boundaries. The purpose here is not to question the way in which these individuals include themselves within those boundaries, but rather to attempt to understand what they think of their legitimacy. In other words, it is to understand the ways of structuring political and symbolic expressions that define and redefine group boundaries by means of legitimated public classifications.      

3In Brazil, central statistical offices classify individuals according to a gradation of colours — white, brown, black, and yellow — that does not fully coincide with the “categories of practice” used in everyday social relations (Loveman, 1999). This incompatibility has been interpreted in opposite ways: on the one hand, the idea that systems of ethno-racial classification are fluid and interrelated with other variables of a different nature (Sansone, 1996); on the other hand, the idea that, although fluid, these systems have a racialised translation and the categories of colour express everyday racialised divisions (Guimarães, 1999).    

4Recently, a line of research has developed tools to study the “making and unmaking” of ethno-racial boundaries in Brazil (Silva, 2016; Silva and Reis, 2012). Although it has contributed with meaningful perceptions on how people in different social positions understand ethno-racial belonging and its current changes, it seems to ignore the institutional dynamics that retroactively reinforce the role of those changes. This article argues that the approach to boundaries (to group boundaries) should not be solely based on social categorisation. Therefore, its starting point is a combination of cognitive elements (translated into the language of cultural repertoires) and the political dynamics of institutionalising the above-mentioned categories, with the aim of understanding the current change in ethno-racial boundaries in Brazil. Just as that relationship has been somewhat neglected, the role of the state and its social penetration have been insufficiently explored (but see Loveman, Muniz, and Bailey, 2011; Bailey, Fialho, and Peria, 2015; Silva, 2016). Closely linked to the role of the state are the national identity narratives underpinning a certain vision of the construction and cohesion of Brazilian society. Therefore, this article seeks to show the relationship established between these narratives and the frameworks people mobilise, highlighting the conditioning role of the latter in boundary-making strategies.

Group boundaries, institutions, and repertoires

5The current theoretical standpoints on ethnicity and race assume they have overcome the primordialist views; i.e. race and ethnicity are no longer considered a second nature, but rather social constructions (Barth, 1969; Wallman, 1978). The constructivist approach, in turn, has been successfully applied to understand the ways in which groups negotiate their ethnic classifications and boundaries (see, for example, Song, 2003; Alba, 1990; Waters, 1990). Recent strands in the theory of ethnic boundaries offer an alternative approach that links constructivism to cognitive and moral patterns, understood within specific institutional configurations. Cultural traditions and institutional/structural conditions restrict the use of specific boundaries, as well as the principles that govern their construction. For example, Lamont (2000b) insisted that members of different national communities are not liable to equally mobilise the same cultural tools to construct and evaluate their social worlds. Therefore, there are national (and group) differences in the availability of symbolic social systems, which result in certain classification principles being more prominent in some contexts than in others. These systems are designated as cultural repertoires (Lamont, 1995; Swidler, 1986). According to this conception, cultural repertoires are systems of meaning employed to categorise groups and establish boundaries between them.

6Lamont and Wimmer argue convincingly in favour of establishing links between symbolic boundaries, cultural repertoires, and institutions (Lamont, 1995, 2000a; Wimmer, 2013). This analytical angle explicitly connects cultural repertoires to historical institutional configurations. Understanding those patterns involves considering mechanisms and resources that influence the process of ethnic or racial boundary making. Wimmer (2013) emphasises elements such as power control in a given social field, institutional arrangements that provide incentives for specific symbolic boundary configurations, modes of categorising groups, cultural markers defining social belonging, as well as types of political organisation and mobilisation that reinforce ethnic and racial categories.   

7These analytical specifications coincide with the analysis of Brubaker (2004) of the cognitive dimension of group construction or groupness. Both approaches emphasise the strategic nature of categorisation and association practices and how these are constitutive of the struggles for the legitimate divisions of the social world. Swidler, in turn, suggests there is an intrinsic instability in the repertoires mobilised by individuals. Therefore, the relationship between repertoires and social change is dynamic, that is: when in settled situations, culture is integrated with action and responds to a uniform model; when in unsettled situations, competing models for organising action emerge.   

8This relationship between the intersubjective and structural levels highlights the historical, processual, and relational dimension of ethnic boundaries and calls for an analysis of social phenomena such as political speeches, national identity narratives, or the public repertoires of social movements. Depending on that affinity, it is possible to distinguish between those historical times when ethnicity does in fact become “ethnic politics” and other situations in which ethnic boundaries are irrelevant. The way in which collective identifications are reinforced derives from the network of interdependencies – such as the state where these “formations” (Tilly, 1978) are included and the perceptions its members have of those configurations – in a retroactive dynamic that can both strengthen or weaken the definition of group boundaries and the repertoires of its members.   

9Repertoires are important because they mitigate the analytical effects of unified conceptions of culture, such as those underpinning national cultural models. Cultural repertoires, as systems of meaning endowed with principles of classification, are inevitably tied to institutional contexts and these, in turn, are constructed by power relations. In other words, to analyse the modes of recognising ethnic and racial boundaries, it is necessary to understand the role played by institutions and by struggles within institutional fields (Wimmer, 2008).

10Contrary to Lamont, however, this article does not focus on the conceptions and frameworks of reference of “ordinary people” (Lamont and Mizrachi, 2011). Power is an important dimension of agency. Considering that macro-actors are those assumed to have greater structuring capacities, this implies reassessing the frameworks mobilised by “ordinary people”, since “unequal access to the means of social construction” (Mouzelis, 2008) involves ranking discourses, strategies, and legitimate visions of the social world. Hence, in an effort to complement the programme of Lamont, this article shifts its focus to the promulgators “of distinct visions of collective identity” (Eisenstadt, 2002: 40), such as policy-makers, community leaders, academics, and actors with public responsibilities.

11In what follows, the article emphasises the process of institutionalising the category “black” in the sphere of the state and how that created a tension in the national narratives of belonging, since the former traditional national narrative is being contested and can no longer claim hegemony.

12This article is based on qualitative fieldwork involving the collection of data in Brazil over three distinct periods: 2010, 2011, and 2015. During those three periods, semi-structured interviews were conducted with 25 key actors and different documents were gathered from a variety of sources. The group of interviewees resulted from a previous selection guided by the way people position themselves in power configurations. In this way, the interviews targeted key actors, such as state officials responsible for institutionalising rules and codes, academic actors with a role in discursively defining and structuring cognitive and moral models, as well as social movement leaders.  

13All interviews were coded and analysed using MAXQDA (v. 12), which allowed to inductively identify patterns of response based on primary-code frequency. Besides the interviews, this article draws on a corpus of texts published in websites, documents from civil society and public authority institutions, a collection of legal texts, as well as other written sources, such as blogs and newspaper articles. Although this corpus was not analysed with MAXQDA, it also served as the basis for reflections and inferences that strengthened the conclusions. Additionally, participation in debates and conferences on affirmative action and the new legal framework for granting collective rights also played a crucial role.

14In the following section, the policy-making process that led to the consolidation of an affirmative action framework and to the institutionalisation of an ethno-racial semantics in Brazil is outlined. The plural modes of constructing ethno-racial boundaries, as well as their interaction with collective narratives, is then highlighted through the characterisation of repertoires taken from the discursive corpus. Finally, the idea of distinct forms of constructing group boundaries is suggested, and the different ways in which they turn to the symbolic elements of the national narrative are described.

Institutionalising a framework for ethno-racial claims

15The history of Brazil is part of a wider civilizational process that approached ethnic and racial mixture as a narrative for building the nation-state (Gross, 2012). The celebration of ethnic and racial mixture, under different labels, was crucial in these processes that took place across the whole of Latin America during the beginning of the 20th century: raza cósmica in Mexico, modern criolismo in Peru, racial democracy in Brazil, Afro-Latin non-racialism in Cuba, and national mestizaje in Colombia. These ideologies were part of the nation-building projects of these countries and were used as a symbolic element of the nationalist unification processes aimed at overcoming sharp class differences through narratives of miscegenation. This was a national-populist consensus rooted in a corporate society and laden with a traditionalist paternalism whose pattern was repeated in several countries (Touraine, 1988; Linz and Stepan, 1996).

16In Brazil, that national consensus disintegrated after the 1970s. Social movements, supported by evidence provided by scholars of race, began to denounce the persistence of racial inequality as the most convincing proof that, despite the ideology of mixture, asymmetries based on ethnic and racial classifications continued (Hasenbalg, 1979; Valle Silva, 1985; Wood and Carvalho, 1988). In order to understand the reinvention of the collective project that animates contemporary Brazilian society, it is necessary to understand the institutional and political changes that led to the consolidation of policies against racial discrimination. This institutionalisation process marks a turning point in the paradigm of hybridity upon which racial democracy was based. The new model offers the black population a space of representation based on a symbolic and practical recognition of the importance of group belonging as a criterion for gaining rights. At the same time, it reformulates the traditional national identity narrative, criticising the premise of social harmony. This is part of what Loveman (2014) identifies in all of Latin America as a dominant ideological change based on the institutionalisation and recognition of clear ethno-racial boundaries, contrary to their traditional vagueness. In Brazil, the start of this trend can be traced, at least, to the end of the 20th century.        

17In fact, during the 1990s, the political claims made by social movements denouncing white hegemony and racial discrimination gained greater impact within the national political sphere. For the 1991 census, for example, activists and academics organised a campaign on the recognition of the category “black”, under the motto “Do not let your colour go unnoticed”,1 calling for Brazilians to choose a darker colour in the census (Nobles, 2004 [2001]). That inquiry was followed by the official recognition of racial discrimination, which led President Fernando Henriques Cardoso (FHC) – former sociologist of race relations – to receive a delegation of activists and union leaders on the occasion of the march of Zumbi dos Palmares (a fugitive slave and iconic figure of the black movement) against racism. In this circumstance, the delegation presented a National Plan Against Racism, which outlined a number of measures for the black population (Souza, 1996; Fry, 2000). In support of this initiative, the President appointed an Inter-ministerial Working Group for Improving the Status of the Black Population, charged with developing public policies for its inclusion.   

18The first Human Rights National Programme, launched in 1996, defines a set of such measures, as well as the consequent need to institutionally circumscribe the target of those measures.2 This is how the suggestion of the black activist Dora Bertúlio3 to aggregate all non-white categories into a single black category to be used in the official statistics produced by the IBGE (Brazilian Institute of Geography and Statistics) is adopted within the scope of the programme.

19The recognition of racial inequality as a structural characteristic of Brazilian society thus complemented the logic of criminalising racism, previously enshrined in the 1988 Constitution (article 5, XLII).4 To its framework on the traditional liberal principles of individualism and universalism in the application of justice, it added the promotion of a certain social category under-represented in some institutional space or role (Feres Júnior and Campos, 2013). Its redistributive nature, as opposed to a merely judicial logic, marks a turning point in the previous paradigm concerning the direction of affirmative action measurestaken by the state, even though the Constitution itself already contained the possibility of legislating in those terms.

20It is important to highlight the growing organisation of black activists and the appearance of the MNU (United Black Movement) as an expression of this recognition. The idea of reversing inequalities based on ethnicity or race had gained force at least since 1978 when certain civil liberties had been restored. A new generation of activists and university graduates radically changed the former perception based on the culturalist and assimilationist matrix.5 In 1983, the Deputy Abdias do Nascimento, one of the leading figures of the black movement, presented Draft Law 1332/1983 to Congress, which established compensatory proposals aimed at promoting equity between blacks and whites in fields as diverse as education, access to work, remuneration, and treatment by the police; the document was shelved in 1989 (Jacoud, 2009).  

21As a result of the increasingly audible claims of the black movement, some regional governments set up councils to support the black population, such as the Council for the Participation and Development of the Black Community of the State of São Paulo (1984), whose structure involved the leaders of the largest black organisations in the region. This first experiment was followed by others in Bahia (1987), Rio Grande do Sul (1988), and Rio de Janeiro (1991), later expanding to smaller regions.    

22As noted by Guimarães (1999), from 1985 onwards, with the return of democracy, following 21 years of dictatorship, the state sought to redefine its relationship with social movements, including the recently formed MNU. Several activists and militants emerged from community activism (in the favelas, for example), supported by non-governmental organisations. This is the case of Benedita da Silva, federal representative of the PT (Workers Party) and governor of the state of Rio de Janeiro, whose origins are connected to the grassroots community movements and whose line of political action merged the central themes of the PT with the political agenda of the black movement.6 The agenda of racial equality continued to be pursued by black deputies such as Carlos Alberto Caó, Maria Benedita Pais, and Paulo Paim, who proposed the Statute of Racial Equality in 2006 (approved in 2010) and who defended the need to implement compensatory measures for the population classified as black.    

23In 2001, at the World Conference Against Racism, in Durban, the Brazilian delegation was authorised by the president (FHC) to defend the adoption of racial quotas with a view to reducing inequalities.7 At the countless preparatory meetings – both national and international – a network of black activism began to take shape and was discursively prepared for international litigation on racism and discrimination (Machado, 2004). Edna Roland, the Brazilian representative, took this opportunity to propose that measures from the inter-ministerial working group be adopted, with special emphasis on the implementation of positive measures in the employment and health sectors and on fine-tuning the methods of data collection concerning Afro-Latin populations (Ribeiro & Nascimento, 2003).  

24The period between 1991 and 2003 was of intense legislative activity on affirmative action: the enactment of laws in this field increased by over 90% and 24 affirmative action measureswere recorded during those years (Jacoud, 2009). In 2002 a decree established a programme of affirmative action in the Federal Public Service, to be observed by all direct and indirect administration bodies.8 The programme is clear on the achievement of target percentages concerning the participation of Afro-descendants in federal bodies (article 2, I) and in companies hired by federal public administration bodies (article 2, IV).9

25However, it was under the presidency of Lula da Silva (in office between 2003 and 2011) that the measures considered during the government of FHC began to be put in place. The most significant were the creation of SEPPIR (Special Secretariat for Policies Promoting Racial Equality) in 2003 and the establishment of the National System for Promoting Racial Equality (SINAPIR) – a network of public organisations dedicated to eliminating racial discrimination and inequality.10 Those bodies, ranging from municipal offices to councils for racial equality, did in fact draw black activism close to the state. The secretariat was led for the first time by Matilde Ribeiro, an activist and feminist of the black movement, and later, in 2011, by the minister Luiza Bairros,11 a former militant of the United Black Movement, following the tradition of nominating people affiliated with the black movement.

26The mission of SEPPIR was the national coordination of racial equality measures. Three priority areas were selected: education, health, and the distribution of quilombo lands.12 Measures and policies were implemented in each of these areas, consolidating the logic of categories, together with a concern to reverse structural inequalities. The changes in education focused on introducing the subject of African and Afro-Brazilian History and Culture in all primary and secondary schools (Law 10639 of 09/01/2003) and on incorporating the programme ProUni, granting public-school black students scholarships to study at private universities. In 2004 the first quotas in federal higher education started being applied. However, between the first experiments with quotas for black people in public higher education and their nationwide implementation, a number of legal initiatives concerning their constitutionality took place. On the one hand, the opposition of countless white students who claimed they were being discriminated against; on the other hand, the suit filed in 2009 by the DEM (Democrats Party) against the University of Brasilia for alleged violation of several constitutional principles by admitting race as a selection criterion.

27In 2012, the Federal Supreme Court (STF) decided on the constitutionality of the policy of “racial quotas”, although it defended its transitional application. The president of the Supreme Court justified this decision on the grounds of the historical nature of prejudice and on the fact that “those who are not affected by the prejudice of colour are already at a huge advantage”.13 The quota system for black people quickly spread across national public universities and, by 2015, 128 federal education institutions had implemented Law 12711, also known as “Quota Law”. Finally, the Statute of Racial Equality, approved through Law 12228 of 20 July 2010,14 provides an operational definition of the black population (article 1, IV), namely “the group of people who claim to be black or brown when answering the question of colour or race” made by the IBGE. In this way, it defines an ethno-racial group that emerges as the target of rights. A national policy for the health of the black population, for example, is therefore created by the Statute.

28More recently, in 2014, President Dilma Rousseff approved Law 12990, which establishes quotas for black people in the public service. 20% of the vacancies offered in federal public tenders are now reserved for the self-declared “blacks” or “browns”. According to data from SEPPIR, in the first year of its application 638 black individuals entered the public service. Finally, in 2017, colour records in the public health system became compulsory as a result of the publication of Order 344 of the Ministry of Health.

29These examples show that new collective understandings of group boundaries are emerging, backed by institutional actors. However, had the perception of the cultural frameworks defining the different places of social categories not changed, that institutionalisation would have not occurred. The politicisation of ethno-racial categories, driven by claims of black activists and the strategy of social appropriation15 resulting from their incorporation into state structures, successfully repositioned ethno-racial boundaries. At the same time, it disturbed the previous consensus on the meaning and place of those boundaries.  

Plural repertoires and identity resignification

30The frail nature of the agreement on group boundaries was revealed during the controversy surrounding racial quotas in the education system. More than describing a disagreement on the merits or demerits of the quota system, this debate highlighted the divides on the definition of legitimate ethno-racial boundaries.

31Following interviews with people involved in this discussion, as well as key actors in the strategy of implementing the new measures, these discourses were thematically organised. Based on the justifications and views of these actors on the ethno-racial categorisations, the most structuring themes were then highlighted. It is this evaluative dimension, tied to cultural understandings of group belonging that is inherent in the idea of repertoire. The discourse of these actors reveals several repertoires they employ to culturally define which categorisations should be recognised and which categorisations they adhere to.

32Following an inductive strategy, three repertoires are highlighted. They emerge demarcated by the thematic content analysis and are followed by a brief explanation of their structure. Classification of the repertoires results from the analysis of the set of interviews collected during fieldwork in Rio de Janeiro, Bahia de Todos os Santos, and Florianópolis. Other accessed documents were also taken into account, namely: legislation, journalistic pieces, and texts in the blogosphere. Although these were not considered for analysis with MAXQDA, they strengthened the construction of the main features of the repertoires. A methodological aspect that is important to mention is the choice not to take self-identification as a principle for understanding social positions. That work, mainly carried out by Silva and Reis (2012), is not the one developed here. In this sense, the interview guide focused on the understandings interviewees showed of the meanings of ethno-racial categorisations used socially, both at everyday level and at the institutional level. Therefore, the interview was divided into three parts that correspond to three analytical vectors. Specifically: the institutional, which takes into account organisational and power configurations, as well as legal frameworks; the symbolic, where issues linked to identities and identifications are included, such as the spaces of (national or group) identifications and the reformulation or institutionalisation of identification categories; and the political, which accommodated questions on models for managing diversity, as well as the meaning of (and adherence to) the ideas of multiraciality and differentiated policies.

33The first aspect to be noted is that the result of the analyses made with MAXQDA clearly shows three distinct repertoires (see figure 1). These three combinations are obtained through the code matrix that shows how many segments are attributed to certain documents. The corpus is made up of 1629 thematic units extracted from 25 documents (semi-directive interviews). This is a mere graphical representation that does not show any statistical association. It does, however, allow to see what groups can be established, in the sense that certain interviewees share clusters of themes.

Figure 1:Code matrix


34A first repertoire, designated here as dilution (following Alba's notion of blurring and its reformulation by Wimmer), aggregates the themes of mixture, tied to a perception of a particular national ethos – Brazilian exceptionalism. These interviewees insist on the permanence and effectiveness of such an ethos, ascribing this theme a preponderant place in the organisation of their systems of meaning. Miscegenation, Brazil as the country of mixture, and the notion of a “Brazilian type” of racism diluted in everyday encounters intersperse the adherence to this ethos.

35This ethos expresses the hegemonic repertoire up until now, within the symbolic and institutional restrictions imposed on the national narrative of miscegenation. This preponderance of the national identity is mentioned by one of the interviewees in the following way:

(…) you can say, for example, that it is a country of fusion, of various populations, there is no need for "race". This idea of fusion is the old idea that is being strongly rejected. But that was part of citizens’ mentality. [Int. 13, female, academic]  

36This discourse postulates the importance of the dilution of ethno-racial boundaries as a social dynamic inherent in Brazilian society. Insistence on the hybrid nature of Brazilian society – whether on the level of the representations of national cohesion and its importance, or on the level of everyday interactions (where proximity, not social distance, is integral to it) – marks to a large extent the priorities of this repertoire. Even if interviewees accept the existence of discrimination, the danger of a disruption or even a fragmentation of this social harmony becomes imminent if space is given to identity claims that clash with this matrix. Opposition to affirmative action is grounded in the idea that racialisation is a foreign phenomenon to Brazilian society. Social and cultural mixing is the unquestionable nucleus in the cognitive form of adherence to the national ethos, an association well represented in the anti-quota manifesto endorsed by several public figures in 2008:  

The Brazilian nation has constructed an identity anchored in the anti-racist idea of miscegenation and has created laws that criminalise racism. […] The subterranean condition of prejudice proves there is something very positive in the Brazilian national identity, it does not prove our historical failure. […] The fabrication of official races and the selective distribution of privileges according to labels of race […] would represent a radical revision of our national identity and a renunciation of the possible utopia of universal effective citizenship. By judging racial quotas, the STF will not be ruling on a method to enrol in universities, but on the meaning of the nation.16   

37This repertoire establishes the continuity between ethnic and racial mixture and the principle of citizenship. In this way, hybridity is also a symbolic space for social integration and, ultimately, is linked to an essentialist sharing of the same ethos, which makes all ethno-racial differentiation ineffective or dangerous. Contrary to the salience of this type of categories, these interviewees think citizenship is based on suppressing those identities from the state and public spheres. The excerpt below reflects the belief in a mixed society and the fear of that anthropological constant being disturbed in case ethno-racial identities are incorporated into those spheres.    

[…] what that Brazilian population is like, how it perceives itself, how it sees itself, etc. I don't really know how that will change... And with that one thing is lost […] which some people think is a myth, but it is something very real, which is miscegenation in this country. Get it? It has all the difficulties you can think of, a number of cons... some mixture has to be considered in this country: people here are mixed, they really are, get it? [Int. 12, female, academic]

38For these interviewees, race is not a prominent factor in explaining inequality. As in researches that revealed the difference between individualistic and structural explanations of racial inequality (Bobo and Kluegel, 1993), the relationship between structure and race is minimised, while class-based factors are underlined. Poverty becomes the preferred explanation as a social proxy for the differences between ethno-cultural categories. However, this is framed in terms far beyond the traditional considerations on social mixing. A former state official refers to the perversity of the quota system in the following way:

And actually there you don't need a protection system... In fact, there is a whole population out there that has already achieved quite a good position because they managed to overcome a number of barriers that most of the population has not. But that population, I think it should fight for its place in society like the others, it should not be favoured as a special group. [Int. 15, male, former state official]

39Although it does not entirely reproduce an individualistic type of explanation, the coincidence between racial category and structural position is not referred by these actors. Their discourses do not ascribe a central role to racial inequality or discrimination, preferring instead to highlight factors such as poverty and individual mobility.

40For these actors, racial mixture is the meaningful framework through which ethno-racial relationships are conceived. The framework assumes that sociocultural and biological features are interwoven. In this sense, it uses the old ideological script of “racial democracy” as a symbolic resource. As is known, the idea of “racial democracy” was structured by Freyre's social anthropology of Brazilian society (Freyre, 1945, 1960). Through it, Freyre postulates a socio-anthropological continuum between racial mixture – “miscegenation” – and social stability. Unlike the reality of racial division in the United States, in Brazil assimilation was something that emerged naturally without the support of the state. Brazil “invented” a cultural and political paradigm of ethno-racial interpenetration, where discrimination – even if existing – could not culminate in social conflict. However, even if the people interviewed here do not fully endorse this narrative – recognising in some cases the distance between what it presumes and social reality – they argue in accordance with the same moral and anthropological principles that structure it. Consequently, the interviewees who are strongly based on that symbolic construction are those who most emphatically opposed affirmative action.

41The second repertoire is that of negotiation. In this case, the most structuring themes from the system of meanings used by individuals to understand the creation of ethno-racial boundaries are: optional ethnicity, malleability of ethno-racial boundaries, and racial discrimination/inequality. Although not as representative as in the previous repertoire, the national ethos as a cultural resource continues to be considered operational. In this sense, despite the recognition of structures of racial inequality, the myth of mixture as a cultural and biological value still holds normative implications, as one of the interviewees argues:

Construction of the nationality continues to be based on miscegenation. […] In a way, phenomena such as indigenous law, recognition of the quilombo, etc. challenge the idea of a homogenising national identity. And racial discrimination too, of course, that myth is questioned. But I think it still holds value; it continues to serve as a regulatory framework for social relations. [Int. 14, male, academic]

42Recognising the importance of the national ethos is consequential for the fluidity of ethno-racial boundaries – something these interviewees stress. From that point of view, considering the negotiable nature of identities implies both rejecting the essentialist version of national identity and supporting a non-essentialist vision of wide-ranging belongings to specific groups. This reasoning tallies with the definition of optional ethnicity (Waters, 1990) and is linked to the ontological transition from identities to identifications. In contrast with the next repertoire, interviewees consider that ethno-racial categorisations are not fixed and individuals can choose the aspects that are more appealing to them.

For me, black is a racial identity category... That means the person does not necessarily have to have physical features, the person might not be perceived as black, but can identify as black, from a political-ideological standpoint, in view of rescuing a real or mythical ancestry, right, but racial identity is constructed as a constellation of elements that each person constructs... [Int. 8, male, state official]

43It is part of this framework to insist on the malleability of identifications (which cease to be identities), even if – or because – they are conditioned by historical and social phenomena. To claim to be “black” no longer holds ontological value, as will be seen in the case of salience, and is perceived as an option within a game of options:

Well, the thing is the discourse is a bit constructed through polarities. If in fact a group in society constructs itself as white, another has to construct itself as black. And there, yes, there is a risk of racialisation. Nevertheless, I think that if we don't take isolated discourses, but take instead the dynamic of constructing the public space, what we notice is there is a public sphere, a very dynamic public debate that reconstructs and deconstructs racialisations. [Int. 14, male, academic]

44Unlike the previous repertoire, discrimination here is systematically highlighted. Racial inequality is, for all purposes, a structural variable that imposes a hierarchy on the social order of identifications; or rather, it crystallises certain categories in social places that are predetermined, even before the everyday possibility of negotiation resulting from the various social contexts. It is in this sense that the actors who share this repertoire accept the inscription of racialised categories in the public and state spheres.  

45In sum, although boundaries can be negotiated and ethno-racial adherences have a margin of choice, the racialised structure of Brazilian society makes certain categories be socially imposed. However, at a more empirical level, the intersection with other categories might reinforce or mitigate that phenomenon. This discourse emphasises precisely the plural adherences. Unlike the previous one, no adherence is preponderant over the others, nor does it have an ontological reflection providing meaning to the place occupied in the social world. For example, the category “black” is understood as being political and symbolic and, as such, is mobilised as a metaphor for public claims.    

46Finally, the third repertoire, designated as salience, is organised around the following themes: group identity (interviewees emphasise their African roots or origins and the need for the black group to be considered, while also defending the critical reformulation of the national ethos); racial discrimination (the discourse stresses the several instances of discrimination, but mainly the structural inequality between blacks and whites); recognition (interviewees insist on the importance of the phenotype as a categorisation mark, they claim an ethnic consciousness, and allude to a stigmatised racial identity that needs to be made positive); social movement (positive contributions from the social movement, or affinity with the social movement). Concerning this last aspect, it is important to note that interviewees emphasise the role of civil society, organised towards changing identification processes. This becomes particularly relevant when contrasted with the idea of  instrumentalisation on behalf of the state and social movements (noted by interviewees falling into the repertoire of dilution and connoted with classification opportunism). As this aspect does in fact differentiate between two stances on the ethical righteousness of differentialism, it signals a moral judgement on identity politics.    

47Central in the discourse of these interviewees is the claim for an ethnic consciousness. This is tied to the criticism of the national paradigm of hybridity, mixture, and its foundations in the ideology of “racial democracy”. As main principles of organisation, this repertoire emphasises the operationality and prominence of ethno-racial boundaries, valuing the assumption of an “ethnic consciousness”. In this repertoire, the identity of a specific group (the black one) takes up a central place in the discourse of interviewees. Although most of these interviewees were – more or less intensely – involved in the social movement, some identified as white. The actors reject the traditional definitions of the exceptionalism of Brazilian culture. This being the case, mixture no longer frames the social understandings of race relations or national unity. On the contrary, racism, racial inequality, and the historical subordination of black people make up the central narrative that marginalises the culturalist interpretation: i.e. the cultural attributes imputed to the nation no longer ensure symbolic unity. Since “the place of the black” becomes crucial, in order for this national collective narrative to be appropriated on an equal basis, it has to be reformulated. The link between the resignification of national history and black identity claims is approached in the following passage:    

We who came here as slaves, have lost that memory... so what keeps us alive is precisely to imagine and acknowledge that we came from there, otherwise we would go mad, because it was very violent […] we are Africans outside of Africa, because we left there in an extremely violent situation, we remained in that violent situation and... how can I put this, for us those values are what nurture us. [Int. 11, female, academic and activist]

48Therefore, the reinforcement of boundaries takes on a symbolic collective shape that overcomes the national bond. This shape is that which incorporates a very specific social position, mediated by ethnicity, which in practical terms translates into the need for racial self-identification, i.e. for gaining “ethnic consciousness”. As Nilma Bentes from the United Black Movement says, it is not enough to be conscious of your colour, “you need to have an acute perception of belonging to a subjugated and exploited people, both in the past and in the present”.17 An interviewee formulates that conception in the following way:

Right, there is something very important in that question of black consciousness, which is essential, which is that you assume you are black. Because it is so difficult to be black in Brazil... […] And it is one of the things we call black consciousness, which is to acknowledge you are black, to not feel ashamed of being black, to not try to become whiter, which is what happened to me during my teenage years. [Int. 18, male, state official]

49For these individuals, society is structured according to racial lines that affect the social status of individuals and their trajectories. In this sense, for these actors, structural conditions are not overcome by one (even if imagined) wide-ranging cultural condition that postulates mixture as “a [cultural] filter through which individuals evaluate their racial identity” (Silva and Reis, 2012: 396). In contrast, the understanding of ethno-racial boundaries emphasises the overlapping of racial categorisation and social structure. It is because of this coincidence that racial identification becomes salient, unlike what happens in a framework (of racial democracy) that emphasises everyday social relations as non-problematic social interactions. A conception of this nature implies the denial (or reformulation) of a hegemonic version of national identity. Even if based on a political motivation, for these actors, boundaries between ethno-racial groups should be reinforced: “to become black”, as an interviewee mentions, is equivalent to “not letting yourself go white”. Paradoxically, against the reinterpretation of the cultural and historical continuum symbolised in the national ethos, most interviewees who mobilise this repertoire insist on a transhistorical condition that binds a group together through a shared experience of subjugation.

50Figure 2 shows a graphical representation of the relations between the different codes (themes) and the associations they make. It distinguishes the three repertoires analysed before in their thematic relation with each other. It is a graphical representation that does not have a statistical interpretation.  

Figure 2:Code co-occurrence model


Concluding remarks

51It became clear that the repertoires analysed, well beyond a discussion on the merits or demerits of affirmative action reveal three ways of thinking ethno-racial categorisations and, consequently, the boundaries of collective identities. Therefore, this article intended to consider a controversial social issue to, through it, understand the modes of constructing the ethno-racial boundaries admitted by the actors as legitimate. The accentuation and dissemination of ethnic or racial semantics in the sphere of the state – in which ethno-racial classifications emerge as principles for the distribution of material and symbolic resources – has implied a reformulation of the Brazilian national paradigm. Bouchard (2011) referred to the Brazilian collective imagination as “mixity” (mixité) and Silva and Reis (2012) sustain that mixture is the “cultural filter” that evaluates race relations. The mere assumption of the need for affirmative action to emancipate a given social category whose contours are ethno-racial challenges the simplistic acceptance of such a paradigm. This view thus appears as far too faultless and clashes with what clearly emerges as a reconfiguration of cultural interpretations and of the adherences they arouse.      

52As suggested by Swidler (1986), a situation of disagreement between culture and agency causes conflicting or concurrent narratives to emerge. The repertoires analysed here indicate three ways of understanding ethno-racial boundaries, which are explained by the relative influence of cultural and institutional changes. They also indicate the fragmentation of the collective imagination and its traditional foundations. From the dilution promoted by the national ethos, where the boundaries of social solidarity are circumscribed by the specific history of the Brazilian nation, to the salience of new boundaries in the strategic options of social movements and the corresponding critical revision of that same ethos, moving through the negotiation of ethno-racial identifications and the widening of their space – these three cognitive frameworks exemplify ways of thinking categorisations and the meaning structures that construct them. In spite of the cosmopolitan or transnational criticism of national feelings and belongings, it continues to be important to recognise the structuring power of these narratives in understanding the modes in which ethnic and racial diversity is approached. The political and institutional changes briefly described should be understood against the backdrop of fundamental changes in the frameworks for collective boundaries and symbolic identifications. The relevance of a black identity is partially explained by the political process that led to the incorporation of a black category into the scripts of state organisations. These dynamics are interdependent with the views of the elite on the boundaries of the social organisation of difference and the codifications of belonging. The incorporation of black elites into state structures and the concerted policy the state has followed with a view to greater racial justice embody a strategy of social appropriation, which simultaneously brings about a redefinition of the hegemonic classifications of identity.  

53Alba, R. (1990), Ethnicity in America. The Transformation of White Ethnicity, New Haven, Yale University Press.

54Alberti, Verena, and A. Pereira (2007), História do Movimento Negro, Rio de Janeiro, Fundação Getulio Vargas.

55Bailey, Stanley R., F. Fialho, and M. Peria (2015), “Support for race-targeted affirmative action in Brazil”, Ethnicities, 0 (0) pp. 1-34.

56Barth, Fredrik (1969), “Introduction”, in Fredrik Barth (org.), Ethnic Groups and Boundaries. The Social Organization of Cultural Difference, London, Allen & Unwin, pp. 9-38.

57Bobo, Lawrence, and J. Kluegel (1993), “Opposition to race-targeting: self-interest, stratification ideology, or racial attitudes?”, American Sociological Review, 58, pp. 443-464.

58Bouchard, Gerard (2011), “What is interculturalism?”, McGill Law Journal, 56 (2), pp. 435-468. Brubaker, Rogers (2004), Ethnicity without Groups, Cambridge, MA, Harvard University Press.

59Campbell, John L. (2004), Institutional Change and Globalization, Princeton, Princeton University Press.

60Costa, Sérgio (2006), Os Dois Atlânticos. Teoria Social, Antirracismo, Cosmopolitismo, Belo Horizonte, Editora UFMG.

61Eisenstadt, S. N. (2002), “The construction of collective identities and the continual reconstruction of primordiality”, in S. Malesevic and Mark Haugard (orgs.), Making Sense of Collectivity. Ethnicity, Nationalism and Globalisation, London, Pluto Press, pp. 33-87.

62Feres Júnior, João, and Luiz Augusto Campos (2013), “Liberalismo igualitário e acção afirmativa: da teoria moral à política pública”, Revista de Sociologia e Política, 21 (48), pp. 85-99, DOI:10.1590/S0104-44782013000400005

63Freyre, Gilberto (1945), Brasil. An Interpretation, New York, Alfred A. Knoff.

64Freyre, Gilberto (1960), Brasis, Brasil, Brasília, Lisboa, Livros do Brasil.

65Fry, Peter (2000), “Politics, nationality and the meanings of ‘race’ in Brazil”, Daedalus, 129 (2), pp. 83-118.

66Gross, C. (2012), Políticas de la Etnicidad. Identidad, Estado y Modernidad, Bogotá, Instituto Colombiano de Antropologia y Historia.

67Guimarães, A. S. (1999), Racismo e Antirracismo no Brasil, São Paulo, Editora 34.

68Hasenbalg, Carlos (1979), Discriminação e Desigualdades Raciais no Brasil, Rio de Janeiro, Graal.

69Jaccoud, L. (2009), A Construção de Uma Política de Promoção da Igualdade Racial. Uma Análise dos Últimos 20 Anos, Brasília, IPEA.

70Lamont, Michèle (1995), “National identity and national boundary patterns in France and the United States”, French Historical Studies, 19 (2), pp. 349-365.

71Lamont, Michèle (2000a), The Dignity of Working Men. Morality and the Boundaries of Race, Class, and Citizenship, Cambridge, MA, Harvard University Press, and New York, Russell Sage Foundation.

72Lamont, Michèle (2000b), “The rhetorics of racism and anti-racism in France and the United States”, in Michèle Lamont and Laurent Thévenot (orgs.), Rethinking Comparative Cultural Sociology. Repertoires of Evaluation in France and the United States, Cambridge, UK, Cambridge University Press, pp. 25-55.

73Lamont, Michèle, and Nizam Mizrachi (2011), “Ordinary people doing extraordinary

74things: responses to stigmatization in comparative perspective”, Ethnic and Racial Studies, 35 (3), pp. 36-381.

75Linz, Juan, and A. Stepan (1996), Problems of Democratic Transition and Consolidation. Southern

76Europe, South America and Post-Communist Europe, Baltimore, Johns Hopkins University Press.

77Loveman, Mara (1999), “Comment: is ‘race’ essential?”, American Sociological Review, 64, pp. 891-


79Loveman, Mara (2014), National Colors. Racial Classification and the State in Latin

80America, New York, Oxford University Press.

81Loveman, Mara, J. O. Muniz, and S. R. Bailey (2011), “Brazil in black and white? Race

82categories, the census, and the study of inequality”, Ethnic and Racial Studies, 35 (8), pp. 1466-1483.

83Machado, E. P. (2004), Desigualdades Raciais e Ensino Superior. Um Estudo sobre a Introdução das

84“Leis de Reserva de Vagas para Egressos de Escolas Públicas e Cotas para Negros, Pardos e Carentes” na Universidade do Estado do Rio de Janeiro (2000-2004), Rio de Janeiro, PhD thesis in Anthropology and Sociology from the Institute of Philosophy and Social Sciences (IFCS) of the Federal University of Rio de Janeiro (UFRJ).

85Mattos, Hebbe (2005), “Repairing the past: confronting the legacies of slavery, genocide & caste”, in

86Proceedings of the Seventh Annual Gilder Lehrman Center International Conference, New Haven, Yale University.

87McAdam, Douglas, Sidney Tarrow, and Charles Tilly (2004), Dynamics of Contention, Cambridge, UK,

88Cambridge University Press.

89Mouzelis, Nicos (2008), Modern and Postmodern Social Theorizing. Bridging the Divide, Cambridge,

90UK, Cambridge University Press.

91Nobles, Melissa (2004 [2001]), “Racial categorization and censuses”, in David I. Kertzer and

92Dominique Arel (orgs.), Census and Identity, Cambridge, UK, Cambridge University Press, pp. 43-70.

93Pereira, Amauri (2003), “Um raio em céu azul: reflexões sobre a política de cotas e a identidade

94nacional brasileira”, Estudos Afro-Asiáticos, 25 (3), pp.463-482.

95Ribeiro, Omar Thomaz and Nascimento, Sebastião (2003), Entre a Intenção e o Gesto. A

96Conferência de Durban e a Elaboração de Uma Pauta de Demandas de Políticas Compensatórias no Brasil, São Paulo, NUPES, Universidade de São Paulo.

97Sansone, Livio (1996), “Nem somente preto ou negro: o sistema de classificação da cor no Brasil que

98muda”, Afro-Asia, 18, pp. 165-188.

99Silva, Benedita (1999), “The Black Movement and Political Parties”, in Michael Hanchard (org.),

100Racial Politics in Contemporary Brazil, Durham and London, Duke University Press, pp. 179-187.

101Silva, G. M. (2016), “After racial democracy: contemporary puzzles in race relations in

102Brazil, Latin America and beyond from a boundaries perspective”, Current Sociology, 64 (5), pp. 794-812, DOI: 10.1177/0011392115590488.

103Silva, G. Moraes, and Elisa P. Reis (2012) “The multiple dimensions of racial mixture in Rio de Janeiro, Brazil: from whitening to Brazilian negritude”, Ethnic and Racial Studies, 35 (3), pp. 382-399, DOI: 10.1080/01419870.2011.589524.

104Song, Miri (2003), Choosing Ethnic Identity, Cambridge, UK, Polity Press.

105Souza, Jezé (org.) (1996), Multiculturalismo e Racismo. O Papel da Ação Afirmativa nos

106Estados Democráticos Contemporâneos, Brasília, Ministério da Justiça.

107Swidler, Ann (1986), “Culture in action: symbols and strategies”, American Sociological

108Review, 51 (2), pp. 273-286.

109Tilly, Charles (1978), From Mobilization to Revolution, Reading, MA, Addison-Wesley.

110Touraine, Alain (1988), La Parole et le Sang. Politique et Societé en Amérique Latine, Paris,

111Odile Jacob.

112Valle Silva, Nelson do (1985), “Updating the cost of not being white in Brazil”, in P. M. Fontaine (org.),

113Race, Class and Power in Brazil, Los Angeles, UCLA Center for Afro-American Studies, pp. 42-55.

114Wallman, Sandra (1978), “Boundaries of ‘race’: processes of ethnicity in England”, Man, 13, pp. 200-


116Waters, Mary C. (1990), Ethnic Options. Choosing Identities in America, Berkeley, University of

117California Press.

118Wimmer, Andreas (2008), “The making and unmaking of ethnic boundaries: a multilevel process

119theory”, AJS, 113 (4), pp. 970-1022.

120Wimmer, Andreas (2013), Ethnic Boundary Making. Institutions, Power, Networks, Oxford, Oxford

121University Press.

122Wood, Charles, and J. A. M. Carvalho (1988), The Demography of Inequality in Brazil, Cambridge,

123UK, and New York, Cambridge University Press.


1 A literal translation of the Portuguese (“Não deixe sua cor passar em branco”) would be something like “Do not let your colour pass as white” (a play on words with the expression “passar em branco”, meaning “to go unnoticed”). T.N.  

2 Programa Nacional de Direitos Humanos (PNDH-1), Decree-Law 1904, 13 May 1996, available at: (accessed on 21-06-2017).

3 This suggestion was put forth at the International Seminar on Multiculturalism and Racism: The Role of “Affirmative Action” in Contemporary Democratic States, Brasília, 1996. The seminar aimed at gathering guidelines for the formulation of policies specifically targeting the black population and is considered by many as a milestone in affirmative action policies (see, inter alia, Costa, 2006).

4 The Constitution states that “the practice of racism is a felony and indefeasible crime, subject to imprisonment, under the terms of the law”.

5 On this issue, see the autobiographical stories told by these individuals and reproduced in Verena Alberti and A. Pereira, História do Movimento Negro (2007).

6 As she herself describes in: Benedita Silva, “The Black Movement and Political Parties” (1999: 179-187).

7 Mentioned in: Amauri Pereira, “Um raio em céu azul: reflexões sobre a política de cotas e a identidade nacional brasileira” (2003: 463-482).

8 Programa Nacional de Ações Afirmativas, Decree 4229 of 13/05/2002.

9 IPEA, Diretoria de Estudos Sociais, Instrumentos Normativos Federais Relacionados ao Preconceito e às Desigualdades Raciais – 1950-2003, Projeto BRA/01/013 PNUD, 2012.

10 In 2015 the number of PIR bodies (Racial Equality Promotion bodies) rose to approximately over 300. See: Levantamento de Órgãos e Conselhos de Igualdade Racial para o SINAPIR, Secretaria de Políticas de Promoção de Igualdade Racial, Governo Federal de Brasil, May 2015.

11 Luiza Bairros took office in 2011, during the government of Dilma Rousseff.

12  Lands owned by fugitive slaves during the imperial period.

13  Taken from (accessed on 12/05/2016).

14 Law 12288 of 20 July 2010 establishes the Statute of Racial Equality and amends Laws 7716 of 5 January 1989, 9029 of 13 April 1995, 7347 of 24 July 1985, and 10778 of 24 November 2003.

15   The term is taken from McAdam, Tarrow, and Tilly (2004) and refers to mechanisms that enable oppressed or resource-poor populations to overcome their organisational shortcomings.

16 Quoted from “113 Anti-racist Citizens Against Racial Laws”, Folha de São Paulo, 14-05-2008.

17 Quoted in Verena Alberti and A. Pereira, História do Movimento Negro (2007: 526).

Pour citer ce document

Nuno Oliveira, «Talking about boundaries: Symbolic boundaries and ethno-racial repertoires in contemporary Brazil. », French Journal for Media Research [en ligne], Conference Proceedings/Actes de colloques, Minorities/Minorités, mis à jour le : 22/12/2018, URL :

Quelques mots à propos de :  Nuno Oliveira

Instituto Universitário de Lisboa (ISCTE-IUL), Centro de Investigação e Estudos
de Sociologia (CIES-IUL), Lisboa, Portugal



Licence Creative Commons
Ce(tte) œuvre est mise à disposition selon les termes de la Licence Creative Commons Attribution 4.0 International.