5. After affirmative action:
changing racial formations

Graziella Moraes Silva

Over the past few decades in Brazil, conversations about race have become ubiquitous. Prior to the 1980s, researchers commonly portrayed Brazil as a country in which race was a forbidden word. In contrast, today we see what Calvo-González and Ventura Santos (2018, p. 254) have called an ‘explosion’ of race in Brazil, accompanied by ‘a complex process of sedimentation, in which new (or not so new) narratives and practices about race overlap and/or intermingle with those of old “strata”’. In this chapter, I hope to better understand this ‘sedimentation’ or what can be called, following Omi and Winant’s (1986) classical concept of racial formation, the socio-historical process by which Brazilian racial categories have been created, inhabited, transformed and destroyed over the past few decades.

Brazil is a particularly good case for looking at how race is socially constructed through continuous and changing processes. Presented and studied as an example of racial harmony during most of the first half of the twentieth century (e.g. Freyre, 1933; Pierson, 1942; Tannenbaum, 1946), the country was increasingly described as a case of hidden or cordial racism by the century’s end (e.g. Hasenbalg and Valle e Silva, 1988; Guimarães, 2001; Twine, 1998) and, ultimately, praised as a state committed to racial redress by the beginning of the twenty-first century (Htun, 2004; Lima, 2010; Paschel, 2016). The conservative turn marked by the election of Jair Bolsonaro in 2018 has largely constrained the possibilities of racial inclusion but, as argued later in this chapter, the horizon of possibilities of anti-racism is shaped also by bottom-up mobilisations that have their roots in previous decades.

In order to understand the (im)possibilities of Brazilian racial formations, it is important to keep in mind that they have always unfolded in a broader context of changes in global debates about race. Especially after World War II, and largely following a European lead, the hegemonic goal (and ideal) was to abolish race. The idea was that if we stopped talking about race, racisms were bound to disappear. Accordingly, race was deemed a ‘fiction’ due to the lack of biological basis for its use, and most countries around the world abolished racial classification as official categories, including in their censuses (Morning, 2008). A broader modernisation framework predicted a world in which we all would be colour-blind, ‘raceless’ and, though it remained unspoken, Western and Eurocentric.

The persistence of racial tensions without de jure racial discrimination made evident that it was racism that created race and not the other way around. In addition, the recognition that diverse paths of development were possible, and that difference and equality could be understood as compatible and complementary in the pursuit of justice, brought ‘race’ back as an anti-racism weapon (Reis and Moraes Silva, 2015). Many argued that, through affirmative action and other policies that recognised and valued diversity, we would find a new progressive path towards equality in a multicultural world. In the United States, the election of Barack Obama, proudly portrayed as the first African American president of the country, was presented as the ultimate evidence that we had finally reached a post-racial world (Tesler and Sears, 2010). The rise of Donald Trump and his project to Make America Great (and white) Again came as a reminder that history rarely follows a linear path, and the same has been true for anti-racism progress.

As with most Latin American countries, Brazil was caught in those global movements and was even described as exemplary of the multicultural turn towards equality (Paschel, 2016; Loveman, 2014). In spite of the heated debates around the merits of multiculturalism (Hale, 2002), by the end of the first decade of the twenty-first century it seemed that Brazil had chosen ‘the prism of race’ as a path to social inclusion (Lehmann, 2018). Similar to the US case, Bolsonaro’s 2018 election, supported by a campaign with overt expressions of anti-Blackness and anti-indigenous racisms, was a harsh reminder that racialisation remains an important force of social exclusion (Silva and Larkins, 2019).

In the next pages, I take seriously the idea that racial formation happens through ongoing and open-ended processes that create a dynamic horizon of (im)possibilities. Building on the work of Saperstein, Penner and Light (2013), I analyse these (im)possibilities of Brazilian racial formations by focusing on the intersections of macro, meso and micro levels, or on how ‘contested categories at the macro level and fluid and complex identities and performance at the micro level coexist with persistent racial inequality in the present’ (Saperstein, Penner and Light, 2013, p. 371). I start with the macro-historical narratives, in particular the interactions with global debates on race that, in the words of Htun (2004), allowed Brazil to go from racial democracy to affirmative action. In the second section, I focus on the institutional changes that have permitted Brazil to implement top-down policies (i.e. from the state to society) to tackle structural racial inequalities (arguably successfully). In the third section, I analyse how these global and institutional changes created new tensions and disputes at the micro level – in particular around racial classification – that have also transformed Brazilian racial formations from the bottom up. In conclusion, I attempt to address this volume’s question ‘where are we going?’ by mapping the ongoing disputes and tensions that have emerged from the current conservative turn that brought to power a government that openly flirts with anti-Blackness and embraces a colour-blind project. I explore how this new project interacts with recent dynamics at the global and local levels and new horizons of (im)possibilities.

Global and local dynamics: the rise and fall of Brazilian racial democracy

Racialisation is at the basis of what we understand today as ‘the Americas’, a continent constituted through European colonial invasions that led to the genocide, slavery and oppression of indigenous and African peoples. Race was also at the forefront of most wars for independence and civil wars of the eighteenth and nineteenth centuries which constituted the American nation states. But while the United States (and Canada in its policies towards the native population) insisted on open and formal racialised policies through most of the mid-twentieth century (e.g. Jim Crow segregation), most Latin American countries followed a different path by embracing, at least discursively, the idea of racial mixture and mestizaje.

Mestizaje is commonly presented as the hegemonic building block of Latin American racial formation (Telles and PERLA, 2014), or, to use Omi and Winant’s (1986) concepts, its main ‘racial project’ during most of the late nineteenth and twentieth centuries. Mestizaje often implied a discursive rejection of white supremacy and biological racism, but, in practice, was accompanied by continuing subsidies to European migration, formal and informal privileged access to land and newly created industrial jobs for these migrants, forced assimilation policies for indigenous groups, and the absence of policies to redress centuries of Black slavery (Graham et al., 1990).

In Brazil, mestizaje, later celebrated as racial democracy, had a particular and ironic character. As a colony, the country was by far the largest importer of slaves, which, in contrast to other countries in the region, made the ‘Black problem’ more important than the indigenous one (Wade, 1997).1 In addition, although it was one of the last countries in the region to abolish slavery (partly due to the lack of an independence war, as argued by Andrews, 2004), Brazil was one of the first to embrace an image of racial democracy. Due to the inevitable contradictions between this image of racial harmony and a harsh reality of racial inequalities, racial democracy is today largely referred to as a myth (Guimarães, 2001).

Nevertheless, as social scientists know too well, discourses (and myths) have practical consequences. The idea of Brazil as a racial democracy was the basis for the creation of laws that punished blatant racism in the early 1950s, even if they downplayed Black movements’ socio-political demands and stressed the exceptionality of racist acts (Dávila, 2017). Concomitantly, the nationalistic policies of the Estado Novo relied on national ideas of a colour-blind Brazilianness in labour laws. Limiting the hiring of (mostly white) immigrants to 30 per cent of the workforce, these labour policies created possibilities of upward mobility for the urban lower middle class, especially brown men – part of what Degler (1971) called the ‘mulatto escape hatch’. Finally, questioning the idea of race as a biological category allowed racial boundaries to be contextually negotiated in the interface of region, skin colour, cultural habits and socio-economic status (Moraes Silva, 2016). In practice, this meant that more people were allowed to ‘become white’.

As discussed by a number of authors (e.g. Hofbauer, 2006; Schwarcz, 2011), the defence of racial mixing meant, in practice, the whitening of the Brazilian population. As shown by the historical series of the census, between 1890 and 1960, the number of people in Brazil who were identified (or self-identified, since 1950) as white grew from 45 to 60 per cent.2 Evidencing persistent racial hierarchies, this was celebrated as key to the modernisation of the country – as whiteness was a necessary condition of modernity (Schwarcz, 2011). Although this was partly due to the subsidised European migration, it was also accomplished through individual reclassification. During that same period, the number of people who identified their colour as Black went from 15 per cent to 8 per cent, in what Abdias Nascimento (1989), among others, has labelled the Black ‘statistical genocide’.3

Much has been written on the hegemony of the Brazilian racial democracy narrative and its sins in twentieth-century Brazil, and it would be impossible to summarise all the debates here (for a good review, see Guimarães, 2001). Instead, I want to focus on another particularity of the country in relation to other nations that proved to be key to the success of the Brazilian so-called multicultural turn: together with the United States, South Africa and Cuba, Brazil continued to ‘count race’ through most of the twentieth century (Loveman, 2014; Powell and Moraes Silva, 2018). Despite the fact that in 1970, the question about ‘colour’ was excluded from the Brazilian census following a broader Latin American movement (Loveman, 2014), in 1976, at the height of political repression, Brazil conducted a pioneer survey on racial classification. Partly as a result of this survey, the ‘colour’ question was added back to the census in 1980, asking Brazilians to identify according to four categories: branca (white), preta (Black), parda (brown) and amarela (yellow), allowing researchers to measure trends in racial inequalities. As argued elsewhere, this was possible due to an alliance between academics, social movements and census bureaucrats that questioned the project of racial democracy by defining race as a social indicator (Powell and Moraes Silva, 2018). More importantly, this allowed a production of statistical data on racial inequalities that played a key role when the global discourses on race changed at the beginning of the twenty-first century.

The 2001 World Conference against Racism in Durban is an important landmark in the transformation of global discourses on race and racism. Despite many institutional failures and the early exit of the US and Israel from the meeting, Durban pushed the agenda on sensitive topics such as compensation for colonialism and slavery. It also allowed more space for social movements and NGOs to question countries’ official narratives about racism and discrimination. In the case of Brazil, it marked the official acknowledgement of the persistence of racial inequalities in the country.4 In fact, the Brazilian anti-racism agenda was more visible in the conference than that of other countries of the region, partly because other Latin American Black movements and NGOs arrived with strong political agendas but only anecdotal data on the persistence of racism and discrimination in their countries. Brazilian organisations, in contrast, had more than a century of statistical data to support their arguments.5

Paschel (2016) sees Durban as exemplary of a global multicultural alignment, or the strengthening of a transnational anti-racism agenda within local contexts, with strong impacts on Latin America. The global move towards multiculturalism cannot be naively celebrated as a synonym of redistribution and recognition, as thoroughly discussed by Charles Hale (2002), among others. But the multicultural alignment between the global multicultural agenda and a Brazilian local anti-racist project did open space for important transformations within the Brazilian state and concrete anti-racist policies such as racial quotas, as discussed in the next section.

Changing race in the state: constructing multiracial Brazil top-down

If the implementation of racial quotas in Brazil might have come as a surprise internationally, domestically the narrative of Brazil as a racist country had been gaining ground throughout the second half of the twentieth century. In this section, I describe different dimensions of this state change, some of the public policies implemented to (un)make race and the institutional limits to changing a racial project from the top, through public policies that were sometimes at odds with beliefs about race entrenched at the bottom.

By the 1990s, a decade before the Durban meeting, referring to racial democracy as a ‘myth’ had become typical in intellectual and more progressive circles. Nearly all studies in social sciences had strong evidence of racial inequalities and racial discrimination (e.g. Hasenbalg and Valle e Silva, 1988; Valle e Silva and Hasenbalg, 1992; Hasenbalg, Valle e Silva and Lima, 1999). In addition, national polls showed that if most Brazilians did not see themselves as racists, they at least acknowledged the widespread existence of racism in the country (Turra, Venturi and Datafolha, 1995). The 1995 Zumbi march, a call for Black conscientisation with an estimated participation of thirty thousand people, showed the strong organisation of Brazilian Black movements and received widespread coverage in the media, which was unusual for ‘racial’ issues at that time (Rios, 2012).

In that same year, Fernando Henrique Cardoso, a sociologist who had participated in the 1950s UNESCO race study, became Brazil’s president. Although a few scholars stress the importance of the Cardoso administration in initiating federal actions targeting the Black population (e.g. Htun, 2004), others emphasise the limits of such initiatives. As Lima (2010, p. 81) puts it: ‘The discursive and political strategy of this [Cardoso] government was to promote recognition with little investment in redistributive aspects.’ Regardless of the role played by the Cardoso administration, during that period Black movements and civil society organisations mobilised for more concrete actions and arrived well organised at the Durban meeting.

As mentioned, racial statistics played an important role in this preparation. They were also at the core of many of the demands of the Black movements, who were pushing for changes in official ethnoracial categories. In the lead-up to the 2000 census, the Black movement defended the inclusion of a more political category, negro, in place of the colour categories, preta and parda. The census technical committee, dominated by social scientists, argued for the need to keep the historical terms and, relying on survey studies, showed that negro was not a category that resonated with most of the population (Schwartzman, 1999). A balance was achieved in which the official categories were kept but increasingly merged in official reports (Powell and Moraes Silva, 2018).6 This practice, which can be traced back to IBGE reports from the 1980s, allowed pretos and pardos to be presented as negros, which added up to 46 per cent of the population in 2000 and allowed Brazil to present itself as the country with the largest Black population outside Africa.

With the election of Luiz Inácio Lula da Silva in 2003, race debates were further developed and institutionalised within state institutions. As argued by Lima (2010), the Workers’ Party (PT) administration implemented a number of educational, health and labour policies and laws that directly and indirectly benefited the Black population.7 More importantly, the PT administration opened space to the Black movements within the state. The creation of the Special Secretary for Public Policies to Promote Racial Equality (SEPPIR), an office with ministerial status, created what Paschel (2016) termed an ‘ethnoracial state apparatus’.

The entrenchment of the Black movement within the state has been analysed as co-optation or part of a broader tradition of state corporativism (as insightfully discussed by Lehmann, 2018). As argued by Paschel (2016), it also marks a shift in the approach of many Black militants, as the goal of becoming a mass movement was sidelined and the focus became the implementation of policies. In practice, this allowed individuals and organisations broadly identified as affiliated to Black movements to successfully push forward policies that addressed racial inequalities in Brazil. Even if the PT’s track record on indigenous rights is debatable (e.g. the decision to construct the Belo Monte hydroelectric dam), for most anti-racism activists, Lula’s administration was a turning point in the Brazilian state’s approach to racial inequalities. As Silva and Larkins (2019, p. 18) put it, ‘Even though we believe that PT did not go far enough in addressing the brutal consequences of antiBlackness in Brazil, their race-conscious policies did begin to transform Brazilian society in important ways.’

This widespread and constitutionally backed implementation of race-based policies and laws can be read as the final evidence that the racial democracy or mestizaje racial project had been left behind. Affirmative action has allowed Brazil to enrol a record number of students who identified as pretas/os and pardas/os in higher education, not only due to racial quotas but also with the support of University-for-All, or ProUni, a programme that gave scholarships to private universities and also took race into account. Given the importance of educational credentials in the country, this has the potential to reduce racial inequalities in wages and, arguably, create a Black middle class in the country.

Nevertheless, as exhaustively argued by sociologists and anthropologists, the state is not a unidimensional and homogeneous entity (e.g. Gupta, 2012). A number of scholars have analysed the difficulties of different state bureaucracies in dealing with race after decades of silencing it, as one of the effects of the complex process of sedimentation discussed by Calvo-González and Ventura Santos (2018). This is partly because the racial inequality agenda did not evenly penetrate the state. For example, the centrality of race issues in debates about the school curriculum, health and higher education is in stark contrast to its near absence in debates on access to credit and wealth, spheres in which racial inequalities are striking. In addition, it cannot be overlooked how much successful policy implementation depends on interaction and negotiation with other bureaucratic levels (Pires, 2019). For example, since the creation of Law 10.639/2003, which included the history of Africa in the official curriculum of Brazilian schools, evaluations indicate that there is resistance from educators to discussing the subject in the classroom. Conducting fieldwork in six schools in the northern region of Brazil, Coelho and Coelho (2012) identify this resistance: i) by the reproduction of stereotypes about Africa; ii) by the delimitation of the theme to discrimination only; iii) by the allocation of the debate to fairs and during the month of Black consciousness; and iv) by the great difficulty of teachers in dealing with the content outlined in law. Similarly, studies in health policy have analysed how the unified health system (Sistema Único de Saúde – SUS) bureaucracies reacted to the enactment of anti-racist policies in public health. They found that the absence of clear guidelines for handling racial discrimination allowed passive resistance from staff, which in turn reproduced the invisibility of racism in these bureaucracies (Milanezi and Moraes Silva, 2019).

It is also unclear how much the state was successful in legitimising its new approach to race among the general public. Although national surveys have evidenced public support for racial quotas across racial categories, they have also shown strong opposition, particularly among those with university degrees (e.g. Datafolha, 2008). In addition, surveys and in-depth interviews demonstrate the persistent fear of many Brazilians that racial quotas and other racialised discourses might create racism and stronger racial boundaries (Lamont et al., 2016; Moraes Silva, 2016). This raises the question: how have ordinary people engaged with the changing Brazilian racial project at the micro level?

Changing identities and reclassification: negotiating a multiracial Brazil bottom-up

In this section, I turn to changes in the Brazilian racial project at the micro level. A number of authors have discussed the mismatch between the top-down categories mobilised by Brazilian racial policies – in particular racial quotas – and the way Brazilians see race (e.g. Bailey, 2008; Schwartzman, 2009). The focus of this section is on the consequences of the public policy implementation that mobilises the language of race for racial identification. In particular, I focus on multiple explanations for why people increasingly identify as negra/os, preta/os or parda/os and on the role of these changes in the transformation of Brazilian racial formation.

As discussed previously, during the period when the racial democracy project became hegemonic, there was a clear change in the way Brazilians identified themselves: a preference for identifying as branca/o and a rejection of identifying as preta/o. Between 1960 and 1991, the number of people who identified their colour as preta was declining, reaching its lowest point in 1991 (5.15 per cent). Since 1991, however, there has been an increase in the number of people identifying their colour as preta: first to 6.20 per cent in 2000 and to 7.52 per cent in 2010. In the 2019 National Household Survey (PNAD), it was 9 per cent, the highest percentage since 1950. Throughout this period the growth of people who identified their colour as parda, meant to include those who saw their skin colour as in between Black and white, or mixed-race, has been the most remarkable: from 21.32 per cent in 1940, it continued increasing to 43.3 per cent in 2010. In the 2019 PNAD it was 47 per cent. In 2010, the sum of those who identified as preta/os and those who identified as parda/os was already larger than those who identified as whites. The prediction for the next census was that those who identify as parda will be the largest group and the sum of preta/os and parda/os will account for nearly two-thirds of the Brazilian population.8

If we look at the growth of people identifying as parda/o and preta/o from the perspective of the Black activists’ campaign launched prior to the 1991 census, this is a huge success. A poster showing the bare torsos of three people with varying skin tones became the iconic image of this campaign (Oliveira, 2001, p. 85). It urged Brazilians: Não deixa sua cor passar em branco. Responda com bom (C)senso (‘Don’t let your colour pass into white [an expression that also means ‘passing unnoticed’ in Brazil]. Respond with good sense.’) By overwriting the ‘s’ with a ‘C’, the poster cleverly urged Brazilians to use good sense, on the census. (Senso and censo have the same sound in Portuguese, the latter meaning ‘census’ and the former ‘sense’.) The goal was to encourage Brazilians with African ancestry not to identify as white, rendering the Blackness of Brazil statistically visible.

As previously discussed, joining pretos/as and pardos/as in the category of negra/os has been common practice in Brazilian social sciences and in many state institutions, dating at least to the 1980s. A similar strategy was commonly used when defining beneficiaries of affirmative action. The 2012 federal decree that made racial quotas mandatory in federal universities defined the size of reserved quotas based on the state percentage of people who identified as preta, parda and indigena according to the ‘what is your colour or race?’ question on the 2010 census. (These quotas are known by the acronym ‘PPI’.)9

It must be noted that the first multicultural policies in Latin America aimed mostly at the protection of cultural difference, leading some scholars to label affirmative action ‘indigenous inclusion/Black exclusion’, since they saw the need to protect the indigenous but largely ignored the Afro population, which was perceived as insufficiently ‘ethnic’ (Hooker, 2005). By employing self-identification with the census categories, affirmative action policy designers could broaden the policy’s scope. Beneficiaries were not defined in terms of cultural difference or racial identity, but simply by their identification with census colour categories that had been roughly the same since 1872. Mobilising simultaneously social and racial quotas was also instrumental in guaranteeing political support. Since the association between colour and disadvantage in Brazil was supported by a plethora of statistical evidence, affirmative action policies gained broad public legitimacy and support. In short, using the IBGE census categories (coupled with socio-economic categories) paved the way for the creation of quotas for urban pretos and pardos – most of whom (at least until recently) did not possess a distinct political, ethnic or cultural identity (Paschel, 2016), nor did they clearly identify with the more political negro category mobilised by the Black movements.

This is particularly important because, despite its widespread use in policy reports and academic papers, until recently the term negra/o was not a widespread racial identification for most of the population. The 2003 Brazilian Social Survey (Pesquisa Social Brasileiro, PESB), one of the few surveys to force interviewees to choose between Black and white (i.e. not allowing a mixed category), found that interviewees who identified as pardos according to the census categories were split – 50 per cent chose white and 50 per cent Black (Bailey, 2008). Based on a 2010 survey, Telles and PERLA (2014) found that the percentage of negros in the country can vary from 6 per cent (when we rely on spontaneous self-identification as negro) to 59.3 per cent (when we classify as negros those identified by survey interviewers as pretos and pardos).

More recent studies have shown that, as with the preta category, identification with the term negro has been growing. Awareness campaigns, coupled with the availability of transnational repertoires for talking about race and valuing Blackness, have certainly contributed to the increasing number of Brazilians who identify as preto, pardo and negro (Moraes Silva, 2016). Within universities, there has been an expansion of Black collectives and organisations that have pressured universities to include these topics in their course syllabus and public debates, encouraging students to ‘become’ Black. Studies have also shown that identification as negro is also correlated with higher levels of education and income (even after controlling for skin colour tone); therefore, the general upward mobility of non-white sectors of the population may have contributed to this change (Telles and Paschel, 2014).

It is also possible, and non-contradictory, to argue that access to affirmative action and racial quotas may have encouraged people to identify as pretas/os, pardas/os or negras/os. Based on a survey experiment, Bailey (2008) found evidence that mentioning quotas before asking people to identify as Black or white nearly doubles the percentage of respondents who choose to identify as Black. Analysing changes in ethno-racial identification after graduation among university students who were enrolled before and after the implementation of quotas, Francis-Tan and Tannuri-Pianto (2015) found that students increasingly identified as negras/os, especially after the implementation of quotas.

What do these changes in identification tell us about the broader changes in Brazilian racial formation? On one hand, the increase in people willing to identify as pretas/os, pardas/os and negras/os can be seen as a successful consequence of the policy itself, particularly as the policy has the objective of creating growing awareness about race. As Silva and Larkins (2019) point out, for the first time, there were advantages to being Black in Brazil, and people were willing to embrace that identification. On the other hand, this growth was also received with scepticism and accusations of fraud and ‘afro-convenience’ – or the use of a Black ancestor to claim the rights to quotas (Rosa, 2016). Because self-identification was the only requirement to be included in racial quotas, anyone who identified as indígena, pretas/os or pardas/os could benefit from quotas without the need for any documentation (as was the case for quotas for alumni of public schools or low-income students). But in a country that had for a long time argued that ‘we are all mixed’ – or, as Sovik (2009) puts it, ‘Here [in Brazil], nobody is white (Aqui ninguém é branco)’ – who was not entitled to claim being parda/o? Does that mean that all or most Brazilians were somehow entitled to quotas? Or was a certain degree of Blackness necessary for that?

These questions became even more salient when racial quotas expanded to prestigious civil servant selections, for posts as judges, state prosecutors and university professors. Silva and Larkins (2019, p. 911, n. 6) note: ‘Some Brazilians claimed Blackness based on ancestry rather than phenotype to further their careers through the quota system.’ The visibility of these positions also raised eyebrows about the lack of ‘Black faces’. As Frei David, a Black leader and founder of Educafro, denounced in a recruitment process for the Public Prosecutor’s Office, ‘[Eight] out of 10 selected candidates [through quotas] could not be considered Black under any circumstances’ (Conselho Nacional do Ministério Público, 2015). Similarly, Black collectives within the universities started to actively denounce cases of fraud in racial self-identification, in particular in highly selective programmes such as medicine (Rosa, 2016).

These demands have resonated with the state, and in 2016 the Ministry of Education published a directive (Orientação Normativa no. 13, 1 August 2016) mandating procedures for checking the truthfulness (veracidade) of the racial self-identification (auto-declaração racial) of candidates for civil servant positions. This same document made the establishment of verification commissions (comissões de verificação da veracidade da auto-declaração racial) mandatory. Self-identification is still the basis for any inclusion in race-based policies (i.e. no racial identification is imposed on anyone), but because the state’s responsibility is to avoid fraud and misuse of public policies, these committees may reject access to racial quotas if candidates for university and civil servant positions do not provide enough evidence that they are Black. Evidence is both embodied in phenotype and argued through narratives about experiences of discrimination and other forms of racial identification.

Verification committees, praised by some and feared by others, signal a radical break with old narratives of racial democracy and mixed racial boundaries. Nevertheless, as these committees were being discussed, President Dilma was being ousted from office, initiating a radical change in the Brazilian federal administration, in particular for the ethno-racial status apparatus. In a reverse from the multiracial narratives of the PSDB and PT eras, in 2019 the Bolsonaro government came to power with a discourse much closer to the old narrative of racial democracy and a campaign fuelled by openly racist statements (Silva and Larkins, 2019).

Recognitions, polarisations and backlashes: what comes next?

Recent changes and their open-ended consequences evidence how much Brazilian racial formations remain under construction. As argued by Omi and Winant (1986), racial formations are continuous and historical projects. They accumulate previous formations while responding to contemporary dilemmas and power relations. In addition, as argued by Saperstein, Penner and Light (2013), these changes are shaped by the interactions between macro, institutional and micro levels. In this inconclusive conclusion, I return to these three dimensions to identify some of the horizon of (im)possibilities of the Brazilian changing racial formation after its conservative turn.

It is clear that what Calvo-González and Ventura Santos (2018) called ‘the explosion of race’ in Brazil has links to transnational changes. Brazilian Black and indigenous movements, explicitly mobilising an ethno-racial language, gained more space and formed important alliances in global anti-racist movements. Transnational repertoires about being Black are increasingly visible in the country, not only in aesthetic signals and cultural products consumed in everyday interactions, but also in political mobilisation and narratives. At the same time, the global rise of extreme right-wing political groups, in particular in the form of nationalist and anti-immigration sentiments, has involved more overt expressions of anti-Blackness, identified by Silva and Larkins (2019) in their analysis of the 2018 Brazilian presidential campaign.

The implementation of verification commissions illustrates how the racial boundaries in Brazilian society are under negotiation also through state institutional practices. The consequences of these commissions are open ended.10 Supporters claim that they may serve to curb fraud and help to further debates about race and white privilege in the country. Because access to public funds is at stake, they argue, some form of control is not only important but necessary. Critics claim that these commissions can become race trials and, by relying on fixed notions of race, discourage people entitled to racial quotas from applying out of fear that they are not Black enough. Perhaps more important is the fear that debates about these commissions may serve as opportunities for an unsympathetic government to question affirmative action policies.

Bolsonaro is openly opposed to affirmative action policies. He also has a long record of racist statements, often using old repertoires of racial democracy (Alfonso, 2020). Although Bolsonaro is not the first to employ these narratives (Cardoso famously mobilised his Black ancestry by claiming he had ‘a foot in the kitchen’), his rhetoric is not empty but rather a frame for concrete policies that reproduce racial inequalities and privileges. His public security policies are particularly harmful to Black youth, whose deaths by police reached record highs during his first year in office. His environment policies portray indigenous people either as potential enemies (manipulated by international NGOs who want to ‘steal’ the Amazon) or as victims of underdevelopment who wish to be assimilated and become ‘just like any other Brazilian’. More broadly, his economic liberal policies leave little space for social inclusion and may threaten affirmative action policies not only through the dismantling of the policies themselves but also through the undermining of public universities and civil service positions. After all, affirmative action only makes sense if there are selective positions to be occupied.

On a more hopeful note, racial formation is also being transformed from the bottom up, partly as a consequence of previous global and state transformations at the macro and meso levels. At the same time as Bolsonaro was elected president, more women, more Black men and more Black women were elected to legislative power (Mazza, 2018). Of course, the under-representation is still striking (and the chance of a Black woman being elected is still more than five times less than that of a white man), but many were elected with an anti-racist agenda that is much more salient in media and public debates today than it was a few decades ago. On the individual level, surveys show that people are also more aware of experiences of racial discrimination and will respond to them more often, including with legal action.

It is clear that in contrast to the optimism of the era of the Brazilian multicultural alignment (Paschel, 2016), there are multiple ‘disalignments’ between global, institutional and everyday debates. In addition, the strengthening and growing visibility of Black movements – at the global level with Black Lives Matter initiatives, at the micro level and within state institutions – are in stark contrast to the government’s conservative turn towards narratives of benevolent miscegenation and towards policies that reinforce racial exclusion. But as the optimism of the previous period was too naive and did not notice the growing power of old and new forms of racism, let’s hope the pessimism of the current moment is underestimating the horizon of possibilities of emerging Brazilian anti-racist racial projects.

Acknowledgements

I thank Veronica Toste Daflon for her comments on an earlier draft and Sarah Sachs for her careful editing, as well as the editors for their suggestions and continuous support. Of course, all mistakes and imprecisions are my sole responsibility.

References

Alfonso, D.A. (2020) ‘Bolsonaro’s take on the “absence of racism” in Brazil’, Race & Class 61(3): 33–49.

Andrews, G.R. (2004) Afro-Latin America, 1800–2000 (Oxford and New York: Oxford University Press).

Bailey, S. (2008) ‘Unmixing for race making in Brazil’, American Journal of Sociology 114(8): 577–614.

Calvo-González, E. and R. Ventura Santos (2018) ‘Problematizing miscegenation: the fact/fiction of race in contemporary Brazil’, Journal of Anthropological Sciences = Rivista Di Antropologia: JASS 96: 247–54.

Carvalho, J.A.M. de, C.H. Wood and F.C. Drumond Andrade (2004) ‘Estimating the stability of census-based racial/ethnic classifications: the case of Brazil’, Population Studies 58(3): 331–43.

Coelho, W.N.B. and M.C. Coelho (2012) ‘Por linhas tortas – a educação para a diversidade e a questão étnico-racial em escolas da região Norte: entre virtudes e vícios’, Revista da ABPN 4(8): 137–55.

Conselho Nacional do Ministério Público (2015) ‘Fraudes no sistema de cotas e mecanismos de fiscalização – O papel do Ministério Público’, Ata da Audiência Pública, 11 March.

Daflon, V.T., J. Feres Júnior and L.A. Campos (2013) ‘Ações afirmativas raciais no ensino superior público brasileiro: um panorama analítico’, Cadernos de Pesquisa 43(148): 302–27.

Datafolha (2008) ‘O racismo confrontado’, <https://www1.folha.uol.com.br/fsp/especial/fj2311200801.htm> (accessed 30 April 2021).

Datafolha (2019) ‘Cresce percepção de preconceito entre Brasileiros na última década’, <http://datafolha.folha.uol.com.br/opiniaopublica/2019/01/1986041-preconceito-por-genero-e-cor-dobra-em-uma-decada.shtml> (accessed 28 January 2020).

Dávila, J. (2017) ‘Desafiando o racismo no Brasil. Ações judiciais no contexto da “Lei Anti-Discriminação” de 1951’, Varia Historia 33(61): 163–85.

Degler, C.N. (1971) Neither Black nor White: Slavery and Race Relations in Brazil and the United States (New York: Macmillan).

Francis-Tan, A. and M. Tannuri-Pianto (2015) ‘Inside the black box: affirmative action and the social construction of race in Brazil’, Ethnic and Racial Studies 38(15): 2771–90.

Freyre, G. (1933)The Masters and the Slaves (Casa-Grande & Senzala): A Study in the Development of Brazilian Civilization, 2nd edn (New York: Knopf).

Graham, R., T.E. Skidmore, A. Helg and A. Knight (1990) The Idea of Race in Latin America, 1870–1940 (Austin: University of Texas Press).

Guimarães, A.S. (2001) ‘Democracia racial: o ideal, o pacto, e o mito’, Novos Estudos do CEBRAP 61: 147–62.

Gupta, A. (2012) Red Tape: Bureaucracy, Structural Violence, and Poverty in India (Durham, NC: Duke University Press).

Hale, C.R. (2002) ‘Does multiculturalism menace? Governance, cultural rights and the politics of identity in Guatemala’, Journal of Latin American Studies 34(3): 485–524.

Hasenbalg, C. and N. Valle e Silva (1988) Estrutura Social, Mobilidade e Raça (São Paulo: Vértice).

Hasenbalg, C.A., N. Valle e Silva and M. Lima (1999) Cor e estratificação social (Rio de Janeiro: Contra Capa Livraria).

Hofbauer, A. (2006) Uma história de branqueamento ou o negro em questão (São Paulo: Editora UNESP).

Hooker, J. (2005) ‘Indigenous inclusion/black exclusion: race, ethnicity and multicultural citizenship in Latin America’, Journal of Latin American Studies 37(2): 285–310.

Htun, M. (2004) ‘From “racial democracy” to affirmative action: changing on state policy on race in Brazil’, Latin American Research Review 39(1): 60–89.

Lamont, M., G. Moraes Silva, J. Welburn, J. Guetzkow, N. Mizrachi, H. Herzog and E.P. Reis (2016) Getting Respect: Responding to Stigma and Discrimination in the United States, Brazil, and Israel (Princeton, NJ: Princeton University Press).

Lehmann, D. (2018) The Prism of Race: The Politics and Ideology of Affirmative Action in Brazil (Ann Arbor: University of Michigan Press).

Lima, M. (2010) ‘Desigualdades raciais e políticas públicas: ações afirmativas no governo Lula’, Novos Estudos CEBRAP 87: 77–95.

Loveman, M. (2014) National Colors: Racial Classification and the State in Latin America (Oxford: Oxford University Press).

Mazza, L. (2018) ‘Mulher negra (não tão) presente’, revista piauí (blog), 12 November, <https://piaui.folha.uol.com.br/mulher-negra-nao-tao-presente/> (accessed 28 January 2020).

Milanezi, J. and G. Moraes Silva (2019) ‘Silêncio – reagindo à saúde da população negra em burocracias do SUS’, in R. Pires (ed.), Implementando Desigualdades: Reprodução de Desigualdades Na Implementação de Políticas Públicas (Brasília: IPEA).

Moraes Silva, G. (2016) ‘After racial democracy: contemporary puzzles in race relations in Brazil, Latin America and beyond from a boundaries perspective’, Current Sociology 64(5): 794–812.

Moraes Silva, G., V. Toste and C. Giraut (2018) ‘Affirmative action in Brazil: emerging dispute over racial categories in “validation commissions”’, Working paper.

Morning, A. (2008) ‘Ethnic classification in global perspective: a cross-national survey of the 2000 census round’, Population Research and Policy Review 27(2): 239–72.

Nascimento, A. (1989) Brazil, Mixture or Massacre?: Essays in the Genocide of a Black People (Dover, MA: Majority Press).

Oliveira, J.P. (1999) ‘Entrando e saindo da “mistura”: os indígenas nos censos nacionais’, in J.P. de Oliveira (ed.), Ensaios em Antropologia Histórica (Rio de Janeiro: UFRJ).

Oliveira, N.F.I. de (2001) Consciência negra em cartaz (Brasília: Editora Universidade de Brasília).

Omi, M. and H. Winant (1986) Racial Formation in the United States: From the 1960s to the 1980s (New York: Routledge & Kegan Paul).

Paixão, M. and L. Carvano, eds (2008) Relatório Anual das Desigualdades Raciaisno Brasil; 2007–2008 (Rio de Janeiro: Garamond).

Paixão, M., I. Rosseto, F. Montovanele and L. Carvano (2011) Relatório Anual das Desigualdades Raciais No Brasil; 2009–2010 (Rio de Janeiro: Garamond).

Paschel, T.S. (2016) Becoming Black Political Subjects: Movements, Alignments and Ethno-Racial Rights in Brazil and Colombia (Princeton, NJ: Princeton University Press).

Pierson, D. (1942) ‘Negroes in Brazil: a study of race contact at Bahia’ (unpublished thesis, University of Chicago).

Pires, R., ed. (2019) Implementando Desigualdades: Reprodução de Desigualdades na Implementação de Políticas Públicas (Brasília: IPEA).

Powell, B.M. and G. Moraes Silva (2018) ‘Technocrats’ compromises: defining race and the struggle for equality in Brazil, 1970–2010’, Journal of Latin American Studies 50(1): 87–115.

Reis, E.P. and G. Moraes Silva (2015) ‘Processes and national dilemmas: the interplay of old and new repertoires of social identity and inclusion’, Revista Colombiana de Sociología 38(2): 15–38.

Rios, F. (2012) ‘O protesto negro no Brasil contemporâneo (1978–2010)’, Lua Nova: Revista de Cultura e Política 85: 41–79.

Rosa, A.B. (2016) ‘Coletivos universitários denunciam #AfroConveniência em possíveis fraudes de cotas’, Huffington Post Brasil, 5 January.

Saperstein, A., A.M. Penner and R. Light (2013) ‘Racial formation in perspective: connecting individuals, institutions, and power relations’, Annual Review of Sociology 39(1): 359–78.

Schwarcz, L.M. (2011) ‘Previsões são sempre traiçoeiras: João Baptista de Lacerda e seu Brasil branco’, História, Ciências, Saúde-Manguinhos 18(1): 225–42.

Schwartzman, L.F. (2009) ‘Seeing like citizens: unofficial understandings of official racial categories in a Brazilian university’, Journal of Latin American Studies 41(2): 221–50.

Schwartzman, S. (1999) ‘Fora de foco: diversidade e identidades etnicas no Brasil’, Novos Estudos CEBRAP 55: 83–96.

Silva, A.J. Bacelar da and E.R. Larkins (2019) ‘The Bolsonaro election, antiblackness, and changing race relations in Brazil’, Journal of Latin American and Caribbean Anthropology 24(4): 893–913.

Sovik, L.R. (2009) Aqui ninguém é branco (Rio de Janeiro: Aeroplano).

Tannenbaum, F. (1946) Slave and Citizen: The Negro in the Americas (New York: Knopf).

Telles, E. and PERLA (2014) ‘Who is black, white, or mixed race? How skin color, status, and nation shape racial classification in Latin America’, American Journal of Sociology 120(3): 864–907.

Tesler, M. and D.O. Sears (2010) Obama’s Race: The 2008 Election and the Dream of a Post-Racial America (Chicago: University of Chicago Press).

Turra, C., G. Venturi and Datafolha (1995) Racismo Cordial: A Mais Completa Análise Sobre o Preconceito de Cor No Brasil (São Paulo: Editora Atica).

Twine, F.W. (1998) Racism in a Racial Democracy: The Maintenance of White Supremacy in Brazil (New Brunswick, NJ: Rutgers University Press).

Valle e Silva, N. and C. Hasenbalg (1992) Relações Raciais no Brasil Contemporâneo (Rio de Janeiro: IUPERJ).

Wade, P. (1997) Race and Ethnicity in Latin America (Chicago: Pluto Press).