Racism in Brazil: From hypothesis to premise, without ever becoming an object

Abstract

The area study of race relations is one of the oldest and most consolidated in Brazilian Social Sciences. Yet paradoxically, there is still little research focused on understanding and explaining Brazil’s racism, its mechanisms and characteristics. In addition to illustrating these claims, this text aims to explain why the mechanisms of racial discrimination still manage to escape the radar of sociology and, above all, to outline the general lines of a research agenda that takes discrimination as an object of investigation. It seems that racial discrimination stopped being a research hypothesis, reformulated and tested in different ways between the 1970s and 1990s, to become a premise in the area from the 2000s onwards. However, this came to be without racism being properly constructed as an object of sociological research itself.

Keywords:
Discrimination; Racism; Race; Social Sciences; Brazil

Resumo

A área de estudos sobre relações raciais é uma das mais antigas e consolidadas das Ciências Sociais brasileiras. Mas paradoxalmente, ainda são poucas as pesquisas focadas em compreender e explicar o nosso racismo, seus mecanismos e características. Além de qualificar essas afirmações, este texto pretende explicar por que os mecanismos de discriminação racial ainda escapam do radar de nossa sociologia e, sobretudo, delinear as linhas gerais de uma agenda de pesquisa que tome a discriminação como objeto de investigação. Ao que parece, a discriminação racial deixou de ser uma hipótese de pesquisa, reformulada e testada de diferentes modos entre os anos 1970 a 1990, para se tornar uma premissa dos estudos da área a partir dos anos 2000. Contudo, isso se deu sem que o racismo fosse construído enquanto objeto em si da pesquisa sociológica.

Palavras-chave:
Discriminação; Racismo; Raça; Ciências Sociais; Brasil

Introduction

The racial issue has always occupied, and still occupies, a central place in the Brazilian Social Sciences. Despite the controversies about the foundation of the discipline in Brazil, the theme of race was part of the thought of all the players within our canon. Nowadays, the sociology¹ of race has not only been institutionally consolidated as a thematic area of the main camps of the discipline (Campos, Lima and Guimarães, 2018), but has also rapidly expanded the number of published articles, theses, and dissertations (Artes and Mena-Chalco, 2017).

But if the “studies of race relations” form a consolidated field in the Brazilian social sciences, the same cannot be said about racism studies. Despite similar labels, these two fields are far from the same (Miles, 2003; Steinberg, 2007). In the Anglo-Saxon academia, where research on racism was first consolidated, the studies of race relations were more concerned with investigating the modes of interaction between racialized groups (Black and White people, for example), their artistic and cultural expressions, their comparative inequalities, forms of struggle and activism, etc. While such approaches may presuppose the existence of racism, they do not necessarily promote investigations into how discriminatory practices function (Miles, 2003).

Despite the multiplicity of definitions for the concept, studies on racial discrimination are more focused on the social practices that lead to racial distinctions and hierarchies, whether they are seen as intersubjective relations or processes with structural effects. The consideration of racial discrimination as an subject of investigation indeed faces numerous conceptual, epistemological, and methodological obstacles. Firstly, by the semantic inflation suffered by terms such as “racism” and “discrimination” (Idem). Currently used to refer to several doctrines, structures, processes, practices, and inequalities (Campos, 2017), these terms have always had a strong political charge, which imposes constraints on attempts to define them conceptually. Secondly, because the analysis of racism almost always involves a complex interaction between ideologies, practices, and social structures (Idem), which requires complex theoretical and methodological designs for its consequent constitution as a subject of investigation. Thirdly, the very moral and legal condemnation of racism around the world has transformed it into an phenomenon insistently concealed or denied by social actors. For this very reason, its elucidation generally depends on indirect or experimental methodological strategies of limited scope (Pager, 2006).

We understand racial discrimination as a social practice, the result of direct or mediated interactions, in which hierarchies and disadvantages are imposed on individuals in a repeated and systematic way - although not necessarily deliberate or conscious - based on a conception of race or ethnicity attributed to them. There exists a debate about whether the concept of discrimination should include in its definition the presence of structural disadvantages and whether the mere existence of differential treatment would be sufficient to justify its use regardless of the presence of any ideological conception of race (Pager and Shepherd 2008, p. 182). However, since our aim here is to review the studies on the subject in an ecumenical way, we do not believe it is necessary to deal with this controversy but rather to operate with a general definition for the term.

More than delimiting the contours of racial discrimination in a specific context, we want to call attention to how the practical expression of racism, its mechanisms and effects, have been marginalized in the Brazilian Social Sciences. As pointed out by Antônio Sérgio Guimarães, racism is a “very specific way of naturalizing social life, that is, of explaining personal, social and cultural differences based on differences taken as natural”, and, therefore, “each racism can only be understood from its own history” (Guimarães, 1999, pp. 11-12). This does not imply, however, that it is not possible to employ a broader definition capable of abstracting particularities to indicate the object across different sociological studies.

We develop this general goal in the four following sections. First, we defend that the classical reflection about race in Brazil, developed in the greater part of the 20^th century, did not properly address racism or discrimination as we understand those terms today. Racism tended to be taken as synonymous with “doctrine” and discrimination as synonymous with “segregation”, making its semantic application very limited. Secondly, we will try to demonstrate that racial discrimination became a research hypothesis from the work of Carlos Hasenbalg and Nelson do Valle Silva in the 1970s without, however, becoming an empirical object of investigation. In the third section, we show that the success of this approach has made racial discrimination no longer a research hypothesis to be treated as an uncontested theoretical premise without, however, being taken as an object of sociological research in and of itself. In the fourth section, we start from the few works that do not comply with this rule, in order to outline the general methodological lines of a research agenda capable of taking racism, and not only racial relations, as the focus of analysis. Finally, we return to the original argument to summarize several conclusions.

Studies of race relations or the non-place of discrimination

The racial debate in Brazil long predates the institutionalization of the Social Sciences in the country. In fact, many of the self-proclaimed founders of sociological thought held the theme as one of their main focuses, if not the main one. It is also worth noting that the rupture with biological views of race, typical of modern social thought, occurred in a much slower and more complex way in the Brazilian Social Sciences than in other national contexts. Oliveira Vianna, for example, considered one of the forerunners of Brazilian social thought, espoused a strongly Arianist notion of race, as some of his works make explicit (Oliveira Viana, 2005).

Traditionally, Gilberto Freyre’s influential work is considered to have had a more sociological and culturalist notion of race in comparison with the naturalizing concepts that were more popular during his time (Freyre, 2006). However, this reputation has already been strongly contested by analyses that call into question its pioneering spirit - Manoel Bonfim, for example, challenged the strictly naturalistic and hierarchical concepts of race before Freyre (Bomfim, 2017) - including its constructivism and its rupture with racializing discourses. In his analysis of Freyre’s work, Benzaquen de Araújo (1994) draws attention to the permanence of a Lamarckian notion of race, a concept much more malleable and susceptible to environmental and cultural influences, but still profoundly biological. But beyond these controversies, the fact is that the notion of race is much earlier in Brazilian social reflection than the concept of racism or concepts such as prejudice or discrimination, whose meanings are much more recent².

It is within the scope of the so-called “Unesco Project”, a set of research on race relations sponsored between the 1940s and 1950s by the international agency of the UN, in which different authors tried to put to the test the hypothesis of the supposed Brazilian racial harmony (May 1999). In works such as those by Donald Pierson (1967) and Charles Wagley (1963), the idea was to show how the record of discourses indicative of prejudice was sparse and, most of the time, confused with class prejudices. Therefore, racial discrimination was not an object itself or was treated as synonymous with spatial segregation, which was seen as non-existent in Brazil.

It was only in the 1940s that the notion of “racial prejudice” began to be seen as the basis for the reproduction of inequalities between groups and linked to discrimination, especially beginning with the second generation of the Unesco Project (May 1999). Despite their particularities, authors such as Florestan Fernandes, Roger Bastide, Oracy Nogueira and Virgínia Bicudo seemed more comfortable than their predecessors in pointing out racial prejudice as an object of research itself.

However, prejudice here was not yet seen as an autonomous cause of social inequalities between whites and Blacks, but as an anachronistic survival of the slavery past. Everything happened as if post-abolition Brazil had inherited extreme social inequality, in which Black people left unprepared for competition in the job market and at a disadvantage compared to more competent European immigrants. In addition to this was the survival of anachronistic racial prejudices, typical of the slave order:

The process [of industrialization] took place so quickly that the survivals of the slave society and the innovations of the capitalist society still coexist side by side. Color prejudice, whose function was to justify the servile work of Africans, will now serve to justify a class society, but this will not change the old stereotypes; they will only change their purpose (Bastide, 1951).

This perspective involves the initial articulation of the grammar underlying our prejudices. The remarkable analytical distinction between “mark prejudice” and “origin prejudice” coined by Oracy Nogueira (2007), as well as the studies of prejudiced attitudes in children by Virgínia Bicudo (1951) and Aniela Ginsberg (1951) are just examples of how prejudice gained the status of a sociological object of reflection. This was not accompanied, however, by a focus on discrimination as a practice that emerges from prejudice.

It is not necessary here to summarize the complex theories about prejudice from this period, but rather to draw attention to the contextual meaning of this category. If today the notion of prejudice was almost equated with concepts such as racism and discrimination, in the 1930s and 1940s it was something different. As Antônio Sérgio Guimarães notes, terms such as racism and discrimination were restricted to segregated realities such as the USA and South Africa, while the notion of prejudice was used to indicate personal tendencies that were not very effective in the production of a segregated social structure. In this sense, the notion of “color prejudice” was placed in opposition to the concepts of racism and discrimination which, even when used, denoted inorganic phenomena in relation to Brazilian social constitution:

For the generation of Pierson, Wagley and Harris, in the United States, racial class inequalities between Blacks and whites were perpetuated thanks to racial prejudice, discrimination and segregation. Because Brazil had the same inequalities, yet the causal factors mentioned above were relatively weak, the American authors concluded that such inequalities were only due to differences in starting points, and should disappear in the future (i.e., Blacks came from subordinate castes). For Florestan and his generation, however, prejudice not only existed, but in a sense prevented the new competitive order from developing to its fullest. It was, however, a matter of prejudice and discrimination out of place, a kind of alienated consciousness of social agents (Guimarães, 2004, p. 32).

Barreto et al. summarize this narrow interpretation of the concept of discrimination: “the discussion about whether or not there was racial prejudice and discrimination occurred in a context marked by the belief in the non-existence of racism in Brazil” (Barreto et al., 2017). In summary, the idea of racial discrimination as an interactional practice engendered by prejudices, and which leads to hierarchies and inequalities that are at the basis of the processes of racialization itself, emerges late in the Brazilian Social Sciences. Both terms have been popular in the Anglophone social sciences since at least the 1930s. The semantic expansion of both began, however, in the 1940s, after the horrors of the Holocaust were revealed. But it was only in the 1960s and 1970s that terms such as “racism” and “racial discrimination” began to be systematically used in a bibliography focused on their own dynamics.

Studies of racial stratification or racism as a hypothesis

In the 1970s, the existence of racial discrimination was converted into a systematically testable hypothesis by the analyzes of Carlos Hasenbalg (2005) and Nelson do Valle Silva (1978), both interested in collecting empirical evidence of its centrality via research on social mobility. Using statistical models taken from the sociology of stratification, Hasenbalg and Silva sought to compare the chances of upward and downward social mobility of white and non-white individuals with similar conditions and class backgrounds. Despite the increasing sophistication of them models, their work indicated systematic disadvantages for self-declared black and brown³ people compared to white people.

Therefore, an inequality of opportunities emerged from the data that (1) was not reducible to class inequality and (2) continued to operate even after the country’s intense industrialization between the 1930s and 1970s. This indicated that, contrary to previous literature, racial inequalities in Brazil could not be entirely explained by the cursed legacy of slavery, nor by strict class inequality. As they were already analyzing data from post-industrialization Brazil, they also challenged the hypothesis that existing discrimination in the country would be naturally mitigated by the modernization process.

Another empirical derivation of Hasenbalg and Silva’s theoretical model is the research on marital selectivity, that is, investigations into the way in which racial inequalities are reflected in the patterns of marital choice that individuals make (Ribeiro and Silva, 2009; Silva, 1991).

These studies aim to test the hypothesis that there is racial discrimination based on data that crosses racial stratification with social patterns. They indicate that there is a high and growing rate of interracial marriages in Brazil, but a hierarchical order of marital preferences, in which “pardos” (brown people) have higher rates of intermarriage with “brancos” (whites) than with “pretos” (blacks), particularly with “pretas” (black women). More than corroborating the existence of systematic racial biases, this data helps to show how social mobility rates bring mixed-race and Black people closer together, while sociability patterns distance them (Silva and Leão 2012). We will return to this point in the last section.

The works of Hasenbalg and Silva remain landmarks for having collected robust evidence of the existence and effects of racial discrimination in Brazil. However, both authors repeatedly highlighted the limits of their methodological strategies, focused more on measuring the socioeconomic effects of racial discrimination than on investigating the concrete dynamics of its operation. They therefore emphasized the importance of research that sought to elucidate the ways in which discriminatory practices exist and the connection with their consequences for social stratification. In an interview with Antonio Sérgio Guimarães, Hasenbalg summarized this point:

[…] in that 1979 book, inequalities between color or racial groups are attributed to racial discrimination and racism. It is worth clarifying that, in this book and in subsequent works with Nelson, discrimination is not directly observed. It is inferred from the analysis of the disparity in social outcomes of groups of color, controlled for the relevant variables. […] Differences in the performance of whites and non-whites are always observed using the relevant control variables, that is, the equality of other conditions, such as social origin, family income and educational level. These controls allow conclusions to be drawn about differences in the appropriation of social opportunities by color or racial groups. In all topics analyzed for more than twenty years, non-whites end up at a disadvantage (Hasenbalg apudGuimarães, 2006, p. 260).

The relevance of the theoretical-methodological framework of Hasenbalg and Silva’s works for subsequent research, both quantitative and qualitative, is undeniable. Despite increasing methodological refinement and complexity, the most recent quantitative studies on the socioeconomic consequences of racism maintain the focus on the reproduction of Brazilian racial stratification, especially on the comparison of the relative rates of social mobility of whites and non-whites (Henriques, 2001; Ribeiro, 2014). But it is possible to notice a great influence of this approach in more qualitative investigations, almost always concerned with the production of racialized identities in concrete contexts, particularly through cultural and artistic movements (for analyzes in this field, see Hofbauer, 2006). Little progress has been made, however, in terms of a sociology of racism, that is, a set of investigations into the contextualized dynamics of discrimination processes, as well as their consequences in and beyond Brazilian socioeconomic stratification.

Race studies today or racism as a premise

The second millennium began with an inflection point in the way that Brazil politically dealt with racism. A complex articulation between academia, the Black movement, state administrators and international organizations managed to include debates and, subsequently, policies with racial criteria on the government agenda (Htun, 2004). Social Sciences focused on race had a significant role in this process, especially with the production of quantitative research on systematic inequalities in racial opportunities and qualitative studies on Black cultural expressions and political articulations.

During this period, academic views on the racial issue were divided, however, with the advent of affirmative action policies in the 2000’s and the great public controversy surrounding them (Hofbauer, 2006). Once close, sociologists of racial stratification began to oppose anthropologists of Black identities: while the former tended to consider affirmative action an advance, the latter tended to be more critical about the way these policies would impact the national imagination around processes of racialization:

Among the countless and different existing theoretical approaches, it is possible to discern two opposing poles of the argument. On the one hand, we can see a basically sociological tradition, which focuses on the analysis of “relations” between “Blacks” and “whites”, and more specifically on the aspect of social inequality between these “racial groups”. On the other end of the spectrum, we have a series of studies that depart from classic concerns and conceptions of Social and Cultural Anthropology (Hofbauer, 2006, p. 11).

For our purposes, however, this dichotomy matters more regarding the matter of discrimination for each of these epistemic groups. On the one hand, sociologists of racial stratification resorted to increasingly sophisticated models for measuring the effects of discrimination, which reproduced the premises and objectives proposed by Hasenbalg and Silva. The intention was to refine the calculations of mobility rates, either to test the original hypothesis that discrimination in Brazil is strong, or to identify the class transitions with which this discrimination interferes.

By taking the issue of race/color as an independent variable, a set of studies from this group also focused specifically on validating the relevance of the official IBGE categorization. Despite methodological advances, these studies have two major limitations, one external and one internal. The main external limitation is that they remain focused on the observable effects of presupposed discrimination. The main internal limitation refers to the reiteration of traditional ways of measuring the race/color variable, which can sometimes underestimate, sometimes overestimate, racial inequalities. This is because the border between racial groups in Brazil is more continuous than discrete, which also seems to affect the incidence of discrimination itself. Carlos Costa Ribeiro even suggests that inequalities of opportunities are affected by a racial continuum, which impacts all measurements based on the dichotomous variable “white” versus “non-white” (Ribeiro, 2017).

On the other hand, studies on the anthropology of Black identities focus precisely on the complexity of the process of formation of racial identities, taking as a variable to be explained what sociologists of stratification use as an explicative variable. The emphasis here more so concerns the dynamics and construction of symbolic borders, almost always seen as porous and complex spaces, than on the possible power relations that produce statistically measured inequalities (Hofbauer, 2006, p. 31). In addition to not integrating their ethnographic findings with statistically measured inequalities, most of these studies gave little space to racial discrimination in the process of racialization of identities. With exceptions (Sansone, 2004), everything happens as if the polysemy of racial classifications emerged from the unique historical constitution of Brazilian racial formation.

In both approaches, racial discrimination is transformed as a premise. None of these studies deny the existence of racial discrimination, but they disagree regarding its function and scope. Paradoxically, few of them focus specifically on identifying the concrete dynamics of discrimination, its fundamental mechanisms and modes of expression.

Although the division mentioned by Hofbauer no longer reflects the recent bibliographical development in the field, data on specialized literature from the last two decades suggest that racism and discrimination remain marginal in studies. The chart below, ⁴ taken from Campos et al (2018) shows how the divide between “racial stratification studies” versus “Black culture studies” in fact dominated the field in the 1994-1998 period but became diluted in subsequent years until dispersed themes emerged in the years of 2009-2013. Another important fact is the relative marginality of studies on racism and discrimination, indicated by the last bar of each series with parallel lines. Between 1994 and 1998, no text was included in this category, but in 1999 and 2003, 9% of all production in the area of race in human sciences articles on the Scielo platform dealt with the issue centrally. In the last four years, however, this percentage fell to less than half, indicating not only the fluctuation of attention given to the topic, but also its marginalization over time.

Just to illustrate, we separated all articles on the Scielo platform, published between 2014 and 2022 with terms such as “race,” “Black,” “racism,” “discrimination,” etc., as well as their correlates, in the abstract, title or keywords, in the same research journals as mentioned above. Of the 1,145 articles on racial issues, only 97 (8.4%) mentioned the terms “discrimination” or “racism” in their abstracts. But not all texts in this short list, however, focus on empirical research on the mechanisms of discrimination in Brazil. A still exploratory categorization of this material indicates that the term racism tends to appear more in texts about anti-racist movements, about the racial thinking of a particular author, about national and international racial norms, conceptual reflections about racism, discussions in the public sphere about racism, etc. Of these 97 articles, only 26 sought in some way to analyze concrete practices of discrimination, either through their effects on social stratification (inequalities) or based on the perceptions of specific individuals about their experiences with discrimination (perceptions). This data indicates that racism is no longer a hypothesis to be researched (as occurs in the work of Hasenbalg and Silva) but has become a widespread theoretical premise. In other words, specialized sociological studies today begin from a diffuse diagnosis according to which racism is a fact and then investigate its history, its presumed effects and political reactions to it. Although such research has great academic and political value, it rarely attempts to understand how racism empirically operates in practice.

Chart 1
Proportion of academic articles accordingt to the most frequent themes in each period

Discrimination as an object: an agenda

The marginalization of studies on racial discrimination in Brazil does not imply, however, its non-existence. Since the 1990s, some investigations into other dimensions of our race relations have made contributions to a sociology of racism, albeit lateral. We present some of these contributions below. Reconstructing the typology of methodologies available for investigating racial discrimination proposed by Pager and Shepherd (2008), we are able to highlight four strategies: research based on participant observation, judiciary-flux research, research based in perception, and experimental studies. To illustrate this, we point out some research that contributed to the understanding of racial discrimination in Brazil. It is not the intention to exhaustively review the work of each line of research, but rather to highlight examples of work that establish methodological paths of exploration.

Given the high degree of development in the area of racial stratification studies and its focus more on the effects of discrimination than on its dynamics, we decided to disregard this area. Given our focus on empirical research on the current mechanisms of racism, we also ignore more theoretical discussions on the topic and studies in historical sociology⁵.

Participant observation studies

Hasenbalg and Silva already pointed out that the discriminatory mechanisms attributed by their statistical models require qualitative approaches for their understanding. Despite this, ethnographic studies of Brazilian race relations focused more on the processes of identity formation, along with Black political and cultural activism, than on investigating the patterns of discrimination that prevail in the country.

An exception is the work of Lívio Sansone (2004) on racial relations in Bahia. Although also interested in the processes of identity formation, Sansone paid special attention to the way his research subjects circulated in social space and perceived discriminatory experiences, hence the division of racial interactions into what he calls “hard areas” and “soft areas”. The hard areas would contain the social spaces of work relations, in which racial friction would become more common, in opposition to the soft areas, such as family and party spaces, in which racial discrimination would be alleviated (Sansone, 2004, p. 248).

But despite this, he highlights that racial discrimination remains is not much noticed by his informants and, even when it is identified, this does not necessarily imply the adoption of a strong ethnic identity: “[…] more is needed than African ancestry or the experience of discrimination to make people become ‘Black’ or Afro-Brazilian on their own. This is a case in which ‘one swallow doesn’t make a summer’” (Sansone, 2004, p. 248). Despite this, he foreshadowed his field, which in the 1990s would see the emergence of a new Black generation more sensitive to discrimination and attentive to Black cultural symbols.

More recently, the division between soft and hard areas has become more complex. Elizabeth Hordge-Freeman observed how Bahian families give different treatments to their “lighter” members compared to their “darker” ones (Hordge-Freeman, 2021, 2022). Using interviews and participant observation of families with different racial configurations, she shows that discrimination is established in the simplest dimensions of everyday life, from the distribution of domestic tasks to the judgment of the romantic relationships of young people and adolescents according to their color/race of their partners.

Studies based on participant observation have the advantage of accessing the experiences and perceptions of racial discrimination in all its complexity with few mediations. However, they face problems of generalization and systematization of their general traits. We cannot confirm, for example, whether Hordge-Freeman’s findings contradict the model deduced from the field by Sansone, or merely point to more recent or particular dynamics. Furthermore, they provide little information about the ways in which these discriminations become systematic and institutionalized. Hence the importance of studies more focused on institutions.

Institutional studies of the legal-judicial process

The concept of institutional racism has a long and complex history. Its most recognized point of origin is the book Black Power by Stokely Carmichael and Charles Hamilton in which they define the expression by its consequences:

When white terrorists bomb a black church and kill black children, this is an act of individual racism, widely deployed by most segments of society. But when in the same city - Birmingham, Alabama - 500 black babies die each year for lack of adequate food, clothing, sleeping and medical facilities, and thousands more are destroyed or physically, emotionally, or intellectually maimed because of conditions of poverty and discrimination of the Black community, this is due to institutional racism (Carmichael and Hamilton, 1969, p. 6).

Despite pointing to the complex mediation and potential role that institutions can assume in racist dynamics, this definition is more political than sociological. Carmichael and Hamilton highlight that, from a political perspective, it is of little importance to elucidate the mechanisms that lead to systemic inequalities in results between whites and Blacks, as in the example of motherhood. From a sociological perspective, however, the greatest difficulty is the opposite: determining the functioning of discriminatory practices and connecting them to racially based prejudices.

In a critique of the concept of institutional racism, Robert Miles argues that racism will always depend on the actions of concrete people mobilized by racist ideologies and that, therefore, institutionalist notions of discrimination are of restricted use (Miles, 2003, p. 71). However, he recognizes that institutions may be capable of reproducing discriminatory patterns in two circumstances:

We propose that the concept of institutional racism refers to two sets of circumstances: first, circumstances in which exclusionary practices arise from, and therefore embody, racist discourse, but which may no longer be explicitly justified by such discourse; and second, circumstances in which explicitly racist speech is modified in such a way that the explicitly racist content is eliminated but other words carry the original meaning (Miles, 2003, pp. 109-10).

There is no need to advance this discussion here, but at least to point to the importance of studies focused on the ways in which macro-institutions can reproduce or enhance discriminatory practices. And again, Brazil provides a heuristically rich case in how racial inequalities can be reproduced precisely because of the denial of racism.

In his qualitative analysis of racism reports registered at police stations and processed by the justice system, Antonio Sérgio Guimarães highlights the importance of the naturalization of the denial of racism as the basis for the invisibility of discrimination. Although these spaces may admit that there is prejudice in the country, they assume a legal definition of discrimination that is so demanding that it makes it impossible to classify disrespect against Black people as a crime. These occasions are, therefore, treated as prejudiced speeches and, therefore, classified as dishonorable crimes rather than as racism (Guimarães, 2004).

This strand of studies on racism and the judicial process does not focus exclusively on qualitative studies, but also encompasses quantitative methodologies (Adorno, 1995; Ribeiro, 1995; Vargas, 1999). Each in its own way, these studies count the convictions of defendants for different crimes according to their race/color. Although almost all of them detect racial biases against “pretos” and “pardos”, the precariousness of judicial data limits more systematic accounts of racist tendencies in law enforcement processes.

In this dossier, the works of Juliana Vinuto (2024) and Luiz Henrique Marques and Luiz Cláudio Lourenço (2024) help to complexify this area with a common provocation. Although they start from different researches, the two articles show how agents of the Brazilian justice and punishment system tend to have a strongly essentialized view of those accused or sentenced. From “seeds of evil” (Vinuto, 2004) to “disturbers of social peace,” offenders of the most different types are classified as naturally inferior sub-humans. Despite this, these same agents dispense with explicitly racist perspectives in their communications and, when provoked, vehemently deny the racial nature of their essentializations. An intricate dilemma thus arises, in which judicial systems of punishment end up naturalizing the misconduct attributed to people who are majority Black and mixed-race, but which do not require an explicitly racist vocabulary to do so.

Studies of perception

Most studies of discrimination are dependent on the perceptions of its victims and/or the prejudices attributable to its agents. To some extent, ethnographic research also uses people’s perceptions, although this is not the only source of evidence that can be mobilized in these cases. However, we define perception surveys here as those focused exclusively on perceptions captured through structured or semi-structured interviews. In general, studies of perception can be divided into different subtypes according to three analytical axes:

• Quanti versus quali: in one of the axes, there are quantitative versus qualitative research, the first being based on structured interviews of the “survey” type and the second on semi-structured interviews.

• Direct perception versus indirect perception: on another axis, there is research based on direct perception of racial discrimination, in which interviewees are directly questioned about them, and studies of indirect perception, in which interviewees are asked about their views on discrimination in general (Pager and Shepherd, 2008).

• Discriminated against versus discriminators: it is possible to focus studies of perception on potential discriminated groups, as well as potential discriminating groups. Although in this case the interview scripts require different designs, nothing prevents both from being included in the same questionnaire.

On the other hand, research based on perception tends to be limited precisely by its dependence on the perspective of the interviewees, whether they are framed as possible victims or agents of discrimination. To some extent, its results are influenced by the degree of politicization or concealment that racist practices suffer within a given racial formation. This limitation becomes even more prominent in a country like Brazil, whose racism tends to be veiled and the borders of identity relatively fluid. This is because the perception of discrimination itself tends to be weak, as does the formation of a political opinion based on these experiences.

The best solution in this sense is the precedence of indirect questions in relation to direct ones and surveys that combine quantitative and qualitative designs. That is, instead of directly questioning respondents about their views on racial discrimination, indirect surveys focus on generic sensations and experiences with discrimination, concerned with measuring how badly individuals from different racial groups feel or are treated well in different spheres of society. The attribution of a racist character to the different forms of treatment occurs as a posteriori, after the responses have been computed.

An example of an indirect survey of this type applied to the Brazilian reality can be found in the article by Daflon, Carvalhaes e Feres Júnior (2017). Using the Social Dimensions of Inequality Survey (PDSD), carried out in 2008 by the Center for the Study of Wealth and Social Stratification of the Institute of Social and Political Studies of the State University of Rio de Janeiro (Ceres/Iesp-Uerj), the authors analyzed how different racial groups responded to questions about different situations of discrimination without classifying them, however, as racial in nature. The questions were purposefully generic, such as “Do you feel you are treated with less respect than other people?” or “do you feel that people act as if they are better than you”, or “do you feel that building doormen treat you with distrust?” Among the most significant results is the fact that Black and mixed-race people express similar perceptions of discrimination, but not uniformly. The greatest similarity emerges from the comparison between self-declared Black people in general and self-declared low-class mixed-race. Self-declared upper-class mixed-race people, on the other hand, tend to report discrimination rates closer to white people when other variables are controlled (Daflon et al., 2017).

This data can be interpreted in multiple ways, and we will mention three of them here. It could either mean that mixed-race people who have risen socially do not perceive discrimination like Black people in general, or that the former do not actually suffer discrimination like the latter. A third possibility is that among the self-declared mixed-race middle and upper class, many are perceived as white, either due to access to whitening body signs or due to problems related to self-declaration. All of this shows how more studies on the indirect perception of discrimination must be carried out in Brazil, but how they must be combined with other types of studies so that we can unravel part of its interpretative ambivalence.

Marcelle Félix’s (2024) research on the processes of racialization of Black and mixed-race people in Rio de Janeiro also drew thought-provoking conclusions about the plural role that discriminatory experiences have in the process of identity transformation or what she calls self-racialization. Generally speaking, discriminatory experiences do not always lead to a strong process of self-racialization, not even when their racist nature is explicitly recognized. Although his study does not allow us to extrapolate its conclusions beyond the cases studied, it opens a line of research that can connect the sociology of racism with the sociology of the processes of formation of racial identities.

Experimental studies

Despite constituting methodological borders within contemporary social sciences, different types of experiments still receive little attention in the sociology of race in Brazil. This has understandable reasons: in addition to being complex, social experiments tend to be expensive and have uncertain results. It is possible to distinguish three types of experiments applicable to social sciences: the so-called laboratory experiments, field experiments, and natural experiments.

Laboratory experiments seek to test a hypothesis by artificially producing highly controlled environments capable of isolating the intervening variables from the real world as much as possible. Instead of isolating the intervening variables, field experiments seek to control them by comparing random samples, which would allow measuring the effect of the focus variable in real situations. In an example cited by Devah Pager: “instead of asking college students to rank hypothetical job candidates on a laboratory test, a field experiment would present two equally qualified candidates to real employers in a real job search context” (Pager, 2006, p. 75). In addition, there are natural experiments that use the occurrence in delimited spaces of natural events that, for this very reason, can have their effects on other variables.

Nathalia Bueno and Thad Dunning (2017) organized an important laboratory experiment to determine the extent to which Brazilian voters racially discriminated against election candidates. They subjected different people to completely fictitious television election programs in which candidates with different profiles and speeches were presented. The fictitious applications were organized in pairs, in which only the colors of the hired actors varied. After multiple rounds, potential voters were asked to express their voting intentions. At the end, the authors did not detect systematic pro-white or anti-Black biases: “in short, regardless of how we tried to characterize potential compilers or stratify the sample to focus on subgroups in which we might expect race-related effects, we were unable to reject the null hypothesis as absent of effect” (Bueno and Dunning 2017, p. 20)⁶.

In the first known field experiment on racism in Brazil, Antonio Sérgio Guimarães and Nadya Guimarães (2000) organized a quantitative and qualitative research based on sending equivalent CVs of white and Black candidates for job vacancies advertised in the biggest newspapers in the city of São Paulo. After a careful separation of occupations according to the discriminatory potential of the selectors, the research trained 11 students to introduce themselves to the vacancies for which they were recruited and then report their impressions. Despite the sophistication of the research design, the authors highlight some problems that affected the results, such as the low availability of students to search for vacancies, their overqualification in the selected advertisements, as well as the contamination of the experiment by variables such as age and “personality” of those selected. For this reason, they recognize the inconclusive nature of the experiment.

More recently, Felipe Dias (2020) repeated a similar field experiment. Equivalent fictitious CVs were randomly distributed to selection companies in two cities in Brazil. Photos of men and women attached to them were manipulated with the help of Artificial Intelligence to generate white and Black profiles. The statistical differences in the chances of being called up by companies for the white and mixed-race profiles were similar. However, Black women have much lower chances of being drafted, while the status of the intended occupation appears to reduce the effect of race.

In these cases, it never hurts to remember the epistemological adage that the absence of evidence of a phenomenon is not evidence of the absence of that phenomenon. That is, the fact that these experiments do not detect strong general discriminatory tendencies between whites and non-whites does not imply that discrimination does not exist. That is, the fact that these experiments do not detect strong general discriminatory tendencies between whites and non-whites does not imply that discrimination does not exist.

It is never known whether the results express the non-existence of the phenomenon or whether the characteristics of its methodological design prevented its detection. Again, the particularities of a country so marked by a tradition of denying racism require more complex experimental designs, while their results have high potential for the Brazilian context.

Conclusion

Although the field of race relations is one of the most traditional and consolidated in Brazilian sociology, discrimination studies have escaped the discipline’s radar. Until the middle of the 20th century, this was explained by the traditional denial of racism that marked Brazilian history. Between the 1940s and 1970s, however, studies on our racial prejudices tended to euphemize their impacts on discriminatory behaviors, almost always seen as residual or anachronistic.

This scenario began to change in the 1970s, when different sociologists transformed the existence of racism in Brazil into a research hypothesis to be statistically tested. Combining official statistics with analytical models from the area of racial stratification, these approaches pointed to the systematic disadvantages of social mobility for Black and mixed-race people when compared to white people. In a pioneering way, the existence of racial discrimination was transformed into a research hypothesis, tested by applying different statistical models and using various surveys.

The analytical and political success of this approach provided the necessary support for social movements and public administrators to justify demands for racial affirmative actions. Another consequence was the theoretical framing of racism as an uncontested premise of most studies in the area. However, this happened without the dissemination of research concerned with establishing discrimination as an object of sociological study.

This does not imply, however, the total absence of attempts to outline the striking features of our racism. Some works, reviewed in the fourth section, have been exploring methodological strategies toward this aim. Roughly speaking, we can divide them into four approaches: research based on participant observation, institutional research, research based in perception, and experimental studies.

Each of these approaches presents specific potentials and limitations for a greater understanding of the phenomenon. Once seen as more fluid, spaces of sociability and affective relationships seem to have more rigid discriminatory standards than previously thought according to discrimination studies (Hordge-Freeman, 2021). On the other hand, Brazilian discriminatory patterns seem to affect Black people more strongly than mixed-race people (Daflon et al., 2017; Dias, 2020; Ribeiro and Silva, 2009), which contradicts some of the studies on the effects of discrimination on stratification, in which these groups appear closer. What remains, however, is to integrate these efforts into a denser critical mass of studies focused on a common objective: to describe and analyze the explanatory mechanisms of racial discrimination and its consequences. This objective is fundamental so that we can better guide our political actions against racism and, thus, help to mitigate it.

Referências Bibliográficas

• ADORNO, Sérgio. (nov. 1995), “Discriminação racial e justiça criminal em São Paulo”. Novos Estudos Cebrap, 43: 45-63.
• ARAUJO, Ricardo Benzaquen. (1994), Guerra e paz. Rio de Janeiro, Editora 34.
• ARTES, Amélia & MENA-CHALCO, Jesús. (2017), “Expansão da temática relações raciais no banco de dados de teses e dissertações da Capes I”. Educação e Pesquisa, 43 (4): 1221-38. DOI: 10.1590/S1517-9702201702152528.
  » https://doi.org/10.1590/S1517-9702201702152528.
• BARRETO, Paula; LIMA, Márcia; LOPES, Andrea & SOTERO, Edilza. (2017), “Entre o isolamento e a dispersão: a temática racial nos estudos sociológicos no Brasil”. Revista Brasileira de Sociologia - RBS, 5 (11). DOI: 10.20336/Rbs.223.
  » https://doi.org/10.20336/Rbs.223
• BASTIDE, Roger. (1951), “Introdução”. In: BASTIDE, Roger & FERNANDES, Florestan (orgs.). Relações raciais entre negros e brancos em São Paulo. São Paulo, Anhembi/Unesco.
• BICUDO, Virginia. (1951), “Atitudes dos alunos dos grupos escolares em relação com a cor dos seus colegas”. In: BASTIDE, Roger & FERNANDES, Florestan (orgs.). Relações raciais entre negros e brancos em São Paulo. São Paulo, Anhembi/Unesco.
• BOMFIM, Manoel. (2017), A América Latina: Males de origem. Rio de Janeiro, Topbooks Editora.
• BUENO, Natália S. & DUNNING, Thad. (2017), “Race, resources, and representation: evidence from Brazilian politicians”. World Politics, 69 (2): 327-65.
• CAMPOS, Luiz Augusto. (2017), “Racismo em três dimensões: Uma abordagem realista-crítica”. Revista Brasileira de Ciências Sociais, 32 (95): 329507. DOI: 10.17666/329507/2017.
  » https://doi.org/10.17666/329507/2017.
• CAMPOS, Luiz Augusto; LIMA, Marcia & GUIMARÃES, Ingrid. (2018), “Os estudos sobre relações raciais no Brasil: Uma Análise da Produção Recente (1994-2013)”. In: Sociologia brasileira hoje. São Paulo, Ateliê Editorial, vol. 2, pp. 199-234.
• CARMICHAEL, Stokley & HAMILTON, Charles. (1969), black power. Nova York, Pequin.
• DAFLON, Verônica Toste; CARVALHAES, Flávio & FERES, João. (2017), “Sentindo na pele: Percepções de discriminação cotidiana de pretos e pardos no Brasil”. Dados, 60 (2): 293-330. DOI: 10.1590/001152582017121.
  » https://doi.org/10.1590/001152582017121.
• DIAS, Felipe A. (2020), “How skin color, class status, and gender intersect in the labor market: Evidence from a field experiment”. Research in Social Stratification and Mobility, 65: 100477. DOI: 10.1016/j.rssm.2020.100477.
  » https://doi.org/10.1016/j.rssm.2020.100477.
• FÉLIX, Marcelle. (2024), Quando eles nos veem: percepções de processos de racialização entre pretos e pardos no Rio de Janeiro. Rio de Janeiro, tese de doutorado, Universidade do Estado do Rio de Janeiro.
• FREYRE, Gilberto. (2006), Casa-grande & senzala. São Paulo, Global Editora.
• GINSBERG, Aniela. (1951), “Pesquisas sobre as atitudes de um grupo de escolares de São Paulo em relação com as crianças de cor”. In: BASTIDE, Roger & FERNANDES, Florestan (orgs.). Relações raciais entre negros e brancos em São Paulo. São Paulo, Anhembi/Unesco.
• GUIMARÃES, Antonio Sérgio Alfredo. (1999), Racismo e Anti-Racismo no Brasil. Rio de Janeiro, Ed. 34.
• GUIMARÃES, Antonio Sérgio Alfredo. (2004), “Preconceito de cor e racismo no Brasil”. Revista de Antropologia, 47 (1): 9-43. DOI: 10.1590/S0034-77012004000100001.
  » https://doi.org/10.1590/S0034-77012004000100001.
• GUIMARÃES, Antonio Sérgio Alfredo. (2006), “Entrevista com Carlos Hasenbalg”. Tempo Social, 18 (2): 259-68. DOI: 10.1590/S0103-20702006000200013.
  » https://doi.org/10.1590/S0103-20702006000200013.
• GUIMARÃES, Antonio Sérgio Alfredo & GUIMARÃES, Nadya. (2000), “Documento de Trabalho n. 3 do Projeto Ford/USP, ‘Tratamento desigual de brancos e negros na procura de empregos na Região Metropolitana de São Paulo’”.
• HASENBALG, Carlos A. (2005), Discriminação e desigualdades raciais no Brasil. 2. ed. Rio de Janeiro, Iuperj.
• HENRIQUES, Ricardo. (2001), Desigualdade racial no Brasil: evolução das condições de vida na década de 90. Brasília, Ipea.
• HOFBAUER, Andreas. (2006), “Ações afirmativas e o debate sobre racismo no Brasil”. Lua Nova: Revista de Cultura e Política, (68): 9-56. DOI: 10.1590/S0102-64452006000300002.
  » https://doi.org/10.1590/S0102-64452006000300002.
• HORDGE-FREEMAN, Elizabeth. (2021), A cor do amor: Características raciais, estigma e socialização em famílias negras brasileiras. São Carlos, EdUFSCar.
• HORDGE-FREEMAN, Elizabeth. (2022), Second-class daughters: Black Brazilian women and informal adoption as modern slavery. Nova York, Cambridge University Press.
• HTUN, Mala. (2004), “From ‘racial democracy’ to affirmative action: Changing state policy on race in Brazil”. Latin American Research Review, 39 (1): 60-89.
• MAIO, Marcos Chor. (1999), “O projeto Unesco e a agenda das ciências sociais no Brasil dos anos 40 e 50”. Revista Brasileira de Ciências Sociais, 14 (41): 141-58. DOI: 10.1590/S0102-69091999000300009.
  » https://doi.org/10.1590/S0102-69091999000300009.
• MILES, Robert. (2003), Racism. 2nd edition. Londres; Nova York, Routledge.
• NOGUEIRA, Oracy. (2007), “Preconceito racial de marca e preconceito racial de origem: sugestão de um quadro de referência para a interpretação do material sobre relações raciais no Brasil”. Tempo Social, 19 (1): 287-308. DOI: 10.1590/S0103-20702007000100015.
  » https://doi.org/10.1590/S0103-20702007000100015.
• OLIVEIRA VIANA, Francisco José. (2005), Populações meridionais do Brasil. Brasília, Senado Federal.
• PAGER, Devah. (2006), “Medir a discriminação”. Tempo Social, 18 (2): 65-88. DOI: 10.1590/S0103-20702006000200004.
  » https://doi.org/10.1590/S0103-20702006000200004.
• PAGER, Devah & SHEPHERD, HANA. (2008), “The sociology of discrimination: Racial discrimination in employment, housing, credit, and consumer markets”. Annual Review of Sociology, 34 (1): 181-209. DOI: 10.1146/annurev.soc.33.040406.131740.
  » https://doi.org/10.1146/annurev.soc.33.040406.131740.
• PIERSON, Donald. (1967), Negroes in Brazil: a study of race contact at Bahia. Illinois, Feffer & Simons.
• RIBEIRO, Carlos Antonio Costa. (1995), Cor e criminalidade: Estudo e análise da justiça no Rio de Janeiro. Rio de Janeiro, Editora da UFRJ.
• RIBEIRO, Carlos Antonio Costa. (2014), “Mobilidade e estrutura de classes no Brasil Contemporâneo”. Sociologias, 16 (37): 178-217. DOI: 10.1590/15174522-016003711.
  » https://doi.org/10.1590/15174522-016003711.
• RIBEIRO, Carlos Antonio Costa. (2017), “Contínuo racial, mobilidade social e ‘embranquecimento’”. Revista Brasileira de Ciências Sociais, 32 (95): e329503. DOI: 10.17666/329503/2017.
  » https://doi.org/10.17666/329503/2017.
• RIBEIRO, Carlos Antonio Costa & SILVA, Nelson do Valle. (2009), “Cor, educação e casamento: tendências da seletividade marital no Brasil, 1960 a 2000”. Dados, 52 (1): 7-51. DOI: 10.1590/S0011-52582009000100001.
  » https://doi.org/10.1590/S0011-52582009000100001.
• SANSONE, Lívio. (2004), Negritude sem etnicidade: o local e o global nas relações raciais e na produção cultural negra do Brasil. São Paulo, Salvador, SciELO - EDUFBA.
• SILVA, Graziella Moraes & LEÃO, Luciana T. de Souza. (2012), “O paradoxo da mistura: identidades, desigualdades e percepção de discriminação entre brasileiros pardos”. Revista Brasileira de Ciências Sociais, 27 (80): 117-33. DOI: 10.1590/S0102-69092012000300007.
  » https://doi.org/10.1590/S0102-69092012000300007.
• SILVA, Nelson do Valle. (1978), “Black-white income differentials: Brazil 1960, Ano de obtenção”. Ann Harbor, Universidade de Michigan.
• SILVA, Nelson do Valle. (1991), “Estabilidade temporal e diferenças regionais no casamento interracial”. Estudos Afro-Asiáticos, 21: 49-60.
• STEINBERG, Stephen. (2007), Race relations: A critique. Stanford, Stanford University Press.
• VARGAS, Joana Domingues. (1999), “Indivíduos sob suspeita: a cor dos acusados de estupro no fluxo do sistema de justiça criminal”. Dados, 42 (4): 729-60. DOI: 10.1590/S0011-52581999000400004.
  » https://doi.org/10.1590/S0011-52581999000400004.
• WAGLEY, Charles. (1963), An introduction to Brazil. Nova York, Columbia University.

• 1
  . We use the term “sociology” in lowercase here to refer to the studies of social relations in general, rather than Sociology as a specific discipline.
• 2
  . For an analysis of the historical relationships between the concepts of race and racism in Brazilian sociology and elsewhere, see Guimarães, 2024.
• 3
  . It’s always hard to translate the particular race classifications in Brazil, but our censuses tended to use the current five race-and-color categories: “branco” (white), “preto” (black), “pardo” (brown), “indígena” (natives) and “amarelo” (yellow or Asians). Historically, “pardo” was used to denote mixed-race people. However, after the 1980’s, some social movements started to struggle against the official imaginary that portrayed Brazil as a “racial democracy” made by mixed people without clear race borders. Based on statistical and sociological data, the activists claimed that “pardos” (browns) are seen and discriminated as “pretos” (blacks) and, because of that, should be consider part of the same afro-descendent aggregate politically called “negros”. In despite of many countries where the N-word has a pejorative meaning, in Brazil “negro” is the politically correct term to define people from Afro descent. Nonetheless, there are still many social disputes around the actual meaning of being “pardo”. The category can be adopted not only by light-skin black people, but also by people from indigenous descent or even by people that don’t like to fit in race categories.
• 4
  . It is based on a thematic classification of academic articles published in the 18 most prestigious academic journals in the Social Sciences according to the Qualis-Capes platform and indexed on the SciELO platform.
• 5
  . Readers will find in this dossier important discussions on these three themes in texts by Danilo França, Antonio Sérgio Guimarães and Matheus Gato de Jesus, respectively.
• 6
  . Rosário and Aguilar also carried out similar electoral experiments, but not exactly with the aim of detecting discriminatory biases and more with the aim of identifying the preference for candidates of the same race as the voter.

Publication Dates

• Publication in this collection
  23 Sept 2024
• Date of issue
  May-Aug 2024

History

• Received
  29 Feb 2024
• Accepted
  22 Apr 2024