Quotas for blacks in higher education and forms of racial classification

Abstracts

This article presents and discusses data obtained with the application of a questionnaire focused on variables for racial classification and opinion about the policy of quotas for blacks; the questionnaire was applied to a sample of 476 pupils from the last year of secondary education of the public school system of a peripheral town in the Metropolitan Area of Rio de Janeiro. We have tried to understand the elements that shape the classifications of color or race, as well as the stance these pupils were taking before a policy of quotas that could help them in their attempts to have access to a public university. It must be noted that the pupils interviewed would soon be facing the possibility of competing for a place in higher education via an entry exam with racial quotas to a public university that keeps a campus in the same town where they live and study. This problem and this kind of investigation seem to us fundamental nowadays, because quotas for blacks have been put in place since 2003 at several institutions of higher education, and have been subjected to criticism and undergone juridical dispute, as a result, among other things, of the forms of classification proposed. In the study conducted here it was possible to advance in the discussion of how the options of racial classification used so far in these policies are related with the forms of self-identification and identification of the other commonly present in the daily lives of the schools researched, and also to observe how the idea of a racial quota is evaluated by its potential beneficiaries.

Higher education; Quotas; Racial classifications

Este artigo apresenta e discute dados referentes à aplicação de um questionário voltado para variáveis de classificação racial e opinião sobre a política de cotas para negros em uma amostra de 476 alunos do último ano do Ensino Médio da rede pública de um município periférico da região metropolitana do Rio de Janeiro. Buscamos compreender os elementos que informam as classificações de cor ou raça, bem como o posicionamento que esses alunos tomavam frente à política de cotas que poderia beneficiá-los no acesso a uma universidade pública. Deve-se ressaltar que os alunos entrevistados estariam em breve frente à possibilidade de disputar uma vaga no Ensino Superior em um vestibular com cotas raciais numa universidade pública que mantém um campus no próprio município onde estudam e residem. Essa problemática e esse tipo de investigação nos parecem fundamentais na atualidade, pois as cotas para negros que vêm sendo implantadas desde 2003 em várias instituições de Ensino Superior têm sofrido críticas e atravessado controvérsias jurídicas também por conta das formas de classificação propostas. Na pesquisa realizada, foi possível avançar na discussão de como as opções de classificação racial até o momento utilizadas nessas políticas se relacionam com os formatos de auto-identificação e de identificação do outro, comumente presentes no cotidiano das escolas pesquisadas, bem como verificar como a idéia de cota racial é avaliada pelos seus possíveis beneficiários.

Ensino Superior; Cotas; Classificações raciais

ARTICLES

Quotas for blacks in higher education and forms of racial classification

André Augusto Brandão^I; Mani Tebet A. de Marins^II

^IUniversidade Federal Fluminense

^IIUniversidade Federal do Rio de Janeiro

^Contact

ABSTRACT

This article presents and discusses data obtained with the application of a questionnaire focused on variables for racial classification and opinion about the policy of quotas for blacks; the questionnaire was applied to a sample of 470 pupils from the last year of secondary education of the public school system of a peripheral town in the Metropolitan Area of Rio de Janeiro.

We have tried to understand the elements that shape the classifications of color or race, as well as the stance these pupils were taking before a policy of quotas that could help them in their attempts to have access to a public university. It must be noted that the pupils interviewed would soon be facing the possibility of competing for a place in higher education via an entry exam with racial quotas to a public university that keeps a campus in the same town where they live and study. This problem and this kind of investigation seem to us fundamental nowadays, because quotas for blacks have been put in place since 2003 at several institutions of higher education, and have been subjected to criticism and undergone juridical dispute, as a result, among other things, of the forms of classification proposed.

In the study conducted here it was possible to advance in the discussion of how the options of racial classification used so far in these policies are related with the forms of self-identification and identification of the other commonly present in the daily lives of the schools researched, and also to observe how the idea of a racial quota is evaluated by its potential beneficiaries.

Keywords: Higher education – Quotas – Racial classifications.

This article results from a research which, making use of a specific sample, tried to understand the elements that inform the classifications of color or race in Brazil. This problem becomes pressing as the affirmative action policies for entrance into higher education which have been put in place since 2003 have suffered criticism and undergone juridical dispute precisely on account of the forms of classification they employ.

We have therefore taken as our object pupils in secondary education at public schools located in the town of São Gonçalo, a peripheral area of the Metropolitan Region of Rio de Janeiro. These students would be shortly facing the possibility of competing for a place in higher education through an entrance exam marked by racial quotas.

Our more specific objectives were:

a) To map out and understand objectively – based on sampling – the forms of racial self-classification and of racial classification of the other in society which were employed by pupils from secondary public schools;

b) To map out the opinion of this segment about the policies of quotas for blacks in higher education, as well as the relationship between such opinions and the perspectives of racial classification mobilized.

The research was conducted based on a methodological orientation focused on the production, organization and analysis of quantitative data through the creation of a database. In this context, we applied 476 questionnaires to pupils at the last year of secondary education in five state schools located near poverty pockets of the town. Thus, the pupils were not chosen randomly, but they did not have to respond to the questionnaire. Although we have not covered all pupils at the last year of schooling, the number of those who did respond represents more than 80% of all pupils.

The questionnaires applied attempted to map out the classification categories used by the individuals. We worked with only one open question related to the color or race of the respondent. All other questions were closed and related to: color or race according to the IBGE classification; color or race according to a bi-racial classification; afro-descent or otherwise; criteria for self-classification; criteria for the classification of the other in society; opinion about the existence of racism in Brazilian society; self-definition of racism and opinion about the policy of racial quotas in universities.

From the crossing of variables it was possible to establish a quantitative evaluation which at many points made reference to other analyses that touched upon the issue investigated here and at other points was supported by theoretical constructions pertinent to the area and theme.

Data analysis

We present below a detailed analysis of the data, collected and organized in simple and crossed tables.

We can see from the quantitative data that the female presence in our sample is more significant, having approximately 11 points above the male presence. This number is related to the fact already pointed out in recent studies that we have observed during the last decade a considerable increase in female¹ schooling.

When studying a preparatory course in the same town, Brandão (2004) found that women constituted the majority, both among students beginning the preparatory school and among those finishing it, and also among the few that were approved in the entrance exam.

In another research, the same author (Brandão, 2004a) detected among the low-income population of São Gonçalo statements and representations that pointed to different patterns of investment and expectation of the families with regard to their sons and their daughters. According to Brandão (2004a), there was indeed a tendency of the families studied to expect from their young sons a drive towards the job market, whereas their school trajectory came in second place. But their daughters could, in some cases, remain longer under family wings (albeit largely tied to house chores) and, consequently, away from paid occupations, thereby making it easier for them to stay at school.

This gender division – founded on an essentialist representation of men and women – appears as a phenomenon belonging to a "natural order of things", because it develops in two ways in the social world: in the material objectivity itself, and in the subjectivity of the individuals, that is, "in the bodies and in the habituses ..., working as systems of perception schema, of thought and action" (Bourdieu, 1999: 17). Thus, the divisions and criteria for divisions between the sexes, which are always arbitrary and taken as "natural", can explain the female advancement in the school trajectory.

The secondary schools investigated in this study display a racial configuration different from that exhibited by the whole of the town in the 2000 Demographic Census (IBGE, 2000). In that year, São Gonçalo had in its population 53.07% of whites, 10.40% of blacks, and 35.27% of browns.

The existing differences can probably be explained by the fact that the schools chosen for the application of the questionnaire are situated in poverty pockets within the town. In a study about São Gonçalo, Brandão (2004a) has shown that this town reveals a continuum of racial concentration in which the presence of blacks and browns increases as one moves towards the periphery, although the town itself is already located at the periphery of the Metropolitan Region.

Analyzing the table for open classification of color or race we find eleven classification terms employed by the 476 interviewees of our sample.

From among these eleven terms, five are the ones used by the IBGE in their assessments (corresponding to the so-called "closed question about color or race": white, black, brown, yellow and Indian). Indeed, if we totalize those using one of these five categories we find 63.71% of the sample. Thus, the apparent dispersion of classification so often pointed out as characteristic of the Brazilian way of thinking about color or race is not so pronounced. Beyond the classification limits used by the IBGE, the dispersion is restricted to approximately 37% of the respondents. Nevertheless, among these 37% no less than 28.05%¨are limited to just two categories: Negro (16.24) and dark (11.81%). The other four categories (fair, blonde, mixed-race and mulatto) represent only around 6% of the interviewees, since 2.74% of the respondents did not declare their color or race openly.

This configuration somehow resembles that encountered in the National Survey of open color or race (a supplement to the National Survey of Household Sampling – PNAD) conducted by the IBGE in 1976 (Telles, 2003), and that found by Brandão (2004a) on a periphery plot in São Gonçalo.

In the national survey of 1976, the IBGE identified in a sample of 82,577 individuals 135 classification terms, however around 95% of the interviewees limited themselves to six terms: white, black, brown, negro, dark, and light dark.

Brandão (2004a) found in a simple of 691 interviewees 14 open classifications; among these 60.93% corresponded to those used by the IBGE. Also, 24.60% of the respondents considered themselves as dark, and 7.38% as negroes. Considering the 1.16% of non-respondents, only 5.97% of the interviewees were dispersed among the other categories.

The Ethnic-racial Census of UFF (Brandão e Teixeira, 2003) found 20 open categories of color or race among approximately 11,000 interviewees. However, 62.8% of those categories coincided with one of the five IBGE categories.

Keeping these trends in mind, we can say that the much publicized profusion of categories of racial classification in Brazil is in fact just relative. As we have seen, despite the fact that in each study the number of terms employed increases with the size of the sample or of the universe of respondents, more than 60% of the answers were always associated to one of the categories white, black, brown, yellow or Indian.

In the case studied here the categories white and brown are respectively the first (with 33.97% of answers) and the second (with 23.42% of answers) categories more frequently chosen. The interesting fact here is that the self-declared negroes are third in frequency (16.24%), above the category dark (11.81%), which came in fourth place.

It should be noted that we have detected a pattern of preference for the category negro over the (apparently related) category black. This same configuration can be identified both in the study by Brandão and Teixeira (2003), which interviewed students at the UFF, and in Brandão (2004a), which took as its object the inhabitants of a periphery plot.

Thus, in the open question, those who would belong to the group more phenotypically associated with an African origin prefer the classification Negro to black, even if the former is not part of the list presented by census or sampling assessments carried out by the official body designated for the task. A possible explanation for that can be found in the idea that if the category Negro is more related to identity, differently from the category black (more phenotypical), we could be witnessing a process of identitary intensification among Brazilian Afro-descendants. This process has been pointed out by Sansone (2004), who identified it only among the younger population. It is no accident, therefore, that in the sample of the periphery plot that appears in the study by Brandão (2004a), where the interviewees were largely adults, the difference between self-declared Negroes and blacks is less than 2 percent in favor of the former, whereas in the sample analyzed here, which selected almost exclusively youngsters, this gap is around 13 percent.

Lastly, is seems probable that the relative fit of the open self-classification to the IBGE classification principles is related to a "molding" that the national census produces of the way in which society classifies itself (contrary to assuming that the IBGE made at the outset adequate choices of classification terminology). This hypothesis finds support in the fact that, despite the open question being asked first, a significant part of the interviewees seem to know the official categories and match with them.

At any rate, the color continuum is far less significant (either in national or in local samples) than we were led to imagine by reading the list of open classifications declared by interviewees.

On the table above, we continued to follow possibilities of classification of color or race among the 476 interviewees. We tested a bi-racial model of classification. What stands immediately out is the large percentage of non-declared answers. While these were around 6% in the closed question with IBGE categories, and 2% in the open color or race question, they represent 20.68% of answers in the bi-racial classification. There seems to be, therefore, a significant rejection to the format of this classification.

Apart from that, it is significant that whites, who were around 34% at the two other classifications, now jump to 44.73%. On the other hand, blacks and browns, which were close to 50% at the IBGE classification, go way above the percentage of Negro (34.60%) in the bi-racial classification.

Thus, besides the large number of non-declared answers, this format of classification provokes an increase in the number of self-declared whites.

On Table 5 we see that no less than 72.78% of interviewees consider themselves as Afro-descendants or of Negro origin, despite the fact that only 34.60% declare as Negroes in a bi-racial classification. On the other hand, it calls attention the small percentage of non-declared answers, pointing to the non-rejection of the question above (Do you consider yourself as an Afro-descendant or of Negro origin?) when compared to the question that originated Table 4 (What is your color or race? (...) white or (...) Negro). It would seem that the interviewees are telling us that having Negro origin is not the same as being Negro; that is, that despite the origin being recognized, it does not translate a priori into a classification principle (which at this level of analysis seems to confirm the indications of Nogueira, 1985 about the importance of the phenotypical "mark" as a fundamental element for the attribution of color or race in Brazil).

The UFF Ethnical-racial Census (Brandão & Teixeira, 2003) also found more self-declared Afro-descendants (42.90%) than blacks and browns (31.10%). However, in our sample this gap is much wider, since we have 50.63% of blacks and browns and 72.78% of Afro-descendants.

We can venture some hypotheses for such scenario. The first would be that the ideological strength of the so-called "foundation myth of the three races" would explain this configuration. So, Brazilians would be led to declare themselves, according to their subjective convictions, as: Afro-descendants, Indo-descendants or European-descendants.

Going back to the analysis of our sample of secondary school students, we could ask ourselves if this trend in the answers about Afro-descent is not also related to the issue of the policy of quotas for Negroes in higher education. The University of the State of Rio de Janeiro (UERJ), which at the 2002-2003 entrance exams was the first public university to adopt the system of racial quotas, keeps a middle-sized campus in São Gonçalo, where they offer courses related to the so-called licentiateships, which in general have the largest number of black and brown students, and are among the least wanted courses of the universities. Thus, it is likely that the reflection about their racial origin drifted amidst these secondary school students, since they were close to taking the entrance exams.

On Table 6 we see that more than 50% of the interviewees declared themselves against the policy of quotas for blacks in public universities, whereas only 21.94% were strongly in favor of it. Considering that a significant share of the interviewees could benefit directly from this policy, it seems that the values and representations are here more important than their personal interests. Further along this work we shall analyze this rejection to the policy of quotas in more detail, by crossing it with other variables.

Tables 7 and 8 should be analyzed jointly. We see that only 8.86% of the interviewees declare themselves as racists. However, they recognize that there is racism in Brazil (82.07%) or that there is "more or less" racism (13.92%); only 2.95% of the interviewees state that there is no racism in our society.

The reason for such a contradiction might well be that there would be the perspective that racism is always in the others or in the abstract concept of "society" and not in the individual himself. Thus, the practice of racism is transferred from the citizen into society, that is, from the micro-social into the macro-social level. It is as though there could be a divide between society as a collective entity and its individuals, individuals who generate, through gestures, representations and everyday actions, the racist practices.

In that sense, this may be a situation in which the interviewees opt for the "politically correct" claim that they are not racists, as if, by doing so, they lived up to a specific "ethics" about the matter without overlooking, however, the racism which lies within the relationships between whites and blacks.

On Tables 9 and 10 above, the interviewees stated their opinion about the most important criteria to define the color or the race of another individual as well as their own. With regard to the other's racial definition, we notice that the most significant criterion is the color of skin (40.08%). Second comes the family's origin (35.86%). If we add the two most closely phenotypical options (color of skin and physical aspects), we will have a total of 57.17% of interviewees.

For self-definition, though, the interviewees are rather based on the family's origin. However, if once again we add up those people whose answer was color of skin and those who answered physical aspects we will come up with very significant figures (48.47%).

The important conclusion that can be drawn from the analysis of such tables is that despite the weight attached to the phenotype (and not exclusively to color), the interviewees do not fail to refer, to a significant extent, to the family's origin. This discovery is interesting, for the structure of Brazilian racism had been defined by Nogueira (1985 and 1998) as "of mark". Thus, for this author, the criteria behind discrimination would refer to the individual's phenotypical traits rather than to his origin. He who gets phenotypically closer to a black group would be more likely to be classified as a non-white and, therefore, to be discriminated.

However, as it has been shown, although the physical aspects and the color of skin, when taken together, may be regarded as a key element to self-classification and that of the others, the interviewees do not fail to take the family's origin into account. Which proves to us that, to a large extent, common sense also regards the origin as a racial identifier in our society.

On the tables above we can notice that the criteria for self-definition of color or race are differently used by racial groups. For blacks, the color of skin is the main characteristic, adding up to 55.38% of the answers, followed by the family's origin, with 33.71%. The color of skin added to the physical aspects amount to over 75% of the answers in this group.

For browns, the most important element is the family's origin (50.29%), followed by the color of skin. The two most phenotypical criteria together amount to 55.71% of the interviewees.

Among whites we also find that the family's origin is predominant, but featuring a small difference with regard to the color of skin (4.21%). Once again, the two phenotypical criteria added amount to approximately 48% of answers.

In brief, the preference for the phenotype as an identifier of self-classification is visible in the three groups, although it outstands among the blacks. Considering particularly the item color of skin, we notice that this tends to be given less relevance than the family's origin, both for whites and for browns (with a more important weight for the latter). Among the blacks, though, the family's origin has the smallest percentage of choices in comparison to whites and browns.

Such results seem to tell us that for blacks the family's origin is not very relevant when compared to the reality of phenotype. For the browns, the fact that they are situated on a phenotypical plan which may be considered as less definite in Brazil makes them less attached to the phenotype and enables them, to a larger extent, to use the family's origin as a criterion of self-classification. The whites, though, are roughly divided between two criteria (for the very reason that they are not victims of discrimination on grounds of color or race).

This pattern of answers, however, cannot be isolated from the understanding of what the interviewees consider to be the way they classify the others. Thus, the same trend found in each racial group for its self-definition is reproduced in the way these groups attach importance to the classification criteria for another individual: both the color of skin and the physical aspects add up to 53.14% of the browns, approximately 60% of whites and 70.77% of blacks interviewed.

Once more, it is worth stressing that even though the data confirm the predominance of criteria of "mark" (Nogueira, 1988 and 1995), the family's origin is always referred to in significant numbers and, therefore, is not forgotten or dismissed by the interviewees.

In a recently published article, Brandão and Marins (2005) analyzed in depth the characteristics of UFF pupils who, in the Ethnic-Racial Census conducted by that institution in 2003, claimed to be Afro-descents. The conclusions of such an essay problematize to some extent the hypotheses presented in the analysis of Table 5 above. For the authors found some regularity of socioeconomic characteristics which, in that environment, distinguished each color or race sub-group, that is, blacks/afro and blacks/non-afro; browns/afro and browns/non-afro; whites/afro and whites/non-afro.

Within this context, the claim of Afro-descent was not randomly spread among the interviewees. In fact, by crossing the variables available, the analysis from the UFF Census pointed out that the poorer the individual of any of these three racial groups is, the closer he will be to claiming his Afro-descent. Also, the better-off he is, the further he will be from stating an African origin. And this applied not only to blacks, but also to browns and whites.

Thus, Brandão and Marins (2005) pointed out that it would be possible to suppose that the whites who claim their Afro-descent have a family origin which is closer to blacks and, therefore, partially inherit the cumulative socioeconomic disadvantages (Hasenbalg, 1979) produced by the racism that is inflicted upon the Brazilian black population. But what would make the few blacks and browns who have their origins in higher-income families to consider themselves non-Afro-descents? The study indicated that by reaching a higher status (considering income, the access to private education and their parents' formal education, elements which lead to larger cultural assets incorporated), these blacks and browns would be more likely to get closer to the "white pole" which would be socially more valuable.

The authors take the discussion further, by showing that the acceptance of Afro-descent is processed differently in each of the three groups. Blacks and browns would, generally speaking, accept it due to the phenotype, whereas the whites would accept it because of their origin, and so it is not a coincidence that the latter were poorer than the non-Afro-descent whites. Even though it is not possible to "round out" an explanation for the refusal or acceptance of Afro-descent on the level of quantitative analysis developed by the authors, what seems to matter is the evidence that such choices are not at random, that is to say, theu are connected to economic and social factors not only with regard to the differences between the groups of color or race but also within each of these groups.

On Table 4, we had already identified the existence of some rejection from the interviewees to the bi-racial classification that had been proposed to them in the questionnaire, which is expressed in the high percentage of non-respondents. Considering Tables 14 and 15 together we may observe the way the latter define themselves in the question about color or race under IBGE categories.

We notice, thus, that it was mainly the browns who rejected the bi-racial classification, while it was only to a small extent that whites and blacks refused to be placed under the categories of whites and Negroes.

When we consider Table 15 in particular, we see that the largest "consistency"² in the choices, once crossed, lies in the group "whites" (97.58% of those define themselves as whites under IBGE classification and keep the same answer in the bi-racial classification). The blacks' "consistency" is also meaningful (90.77% of them define themselves as blacks under IBGE classification and as Negroes in the bi-racial classification). The browns, however, are characterized by their "inconsistency" (20.00% identify themselves as whites and 48.57% as Negroes). But they also tend to reject the question (31.43% of non-respondents).

We may conclude that the operational impossibility in the bi-racial classification lies mainly in the fact that such classification is not able to offer a place for those who define themselves as browns under IBGE's logics and, therefore, call for an alternative option to choose on their own.

When we cross the two variables above, we notice that the largest consistency lies among those who define themselves as whites in the closed question, for these people self-declare as whites in 93.94% of the open questions (the remaining being dispersed in three more categories: fair, dark and brown). The blacks are concentrated in approximately 92% in the open categories of Negro and black (the remainder defining themselves as darks). Among the browns, on the other hand, there is a larger dispersion, for although 60.00% of those confirm the choice for this category when asked an open question, the remainder are dispersed in five other categories (white, mixed-race, dark, black and dark).

Once again we notice that the category of browns characterizes itself for bringing together individuals who in the open questions are spread over a larger number of possible self-classifications, even though they predominantly opt for terms which refer to a concept of racial or phenotypical mixture (migrating to a small extent into the open category of whites). Thus, if we now look at Table 17, we shall notice that all the open categories of classification which refer to phenotypical mixtures come up with a higher frequency in the closed category of "browns" (with the exception of the open category "mulatto"), to which no respondents refer in the closed answer.

On Table 6 we had verified that the majority of secondary education pupils interviewed objected to the policy of quotas for blacks. We can now verify in what way such rejection is expressed in the three groups of color or race we are looking into. What immediately draws our attention is the fact that the blacks are the ones who most categorically oppose this policy, the whites coming second. The browns are the ones who reject it the least.

Initially, it appears that these pupils might be included within the context of an "ideological farce", largely spread in the bosom of society, which does not recognize the policy of quotas as capable of solving the Brazilian educational problem, and finds it more efficient to invest in primary and secondary education.

Furthermore, two other elements pointed out by Brandão (2004b) seem to guide these results. In that study, the author analyzes field interviews conducted with pupils of a course oriented to low-income black students who would take a university entrance exam ("Preparatory course oriented to blacks and underprivileged – PVNC³), also based in the town of São Gonçalo – RJ. All interviewees declared themselves blacks or browns, and they were going to use the policy of quotas in the UERJ exam later in that year, but demonstrated sometimes to disagree with the policy of racial quotas, or to be somewhat reluctant to resort to it. For the interviewees, the problems concerning this policy could be summed up as follows.

a) the racial quota would circumvent the "normal" or "typical" rules of access to university;

b) the black individual who entered university via a policy of quotas could become a target of discrimination or verbal abuse by white pupils;

c) this discrimination would take place because such "quota pupils" would not be considered by whites as pupils who would have the same "right" to be there;

d) quotas for "the poor" would be fairer and would not stimulate discrimination.

Brandão (2004b) reaches the conclusion that the pupils who have been interviewed would be entangled in the "ideological mesh" that pervades Brazilian racial order and asserts the existence of a "racial democracy" in Brazil. On the other hand, these interviewees are also steeped in the "merit ideology", which corroborates the operation of the capitalist societies based on liberal perspectives. In that sense,

The fear of subverting this abstract and inconsistent concept of merit is associated, among these youths, to their fear of being once again victims of the discrimination that has marred their lives. (Brandão, 2004b, p.155)

It would also appear that among the secondary education pupils who answered our questionnaire, the abstract concept of merit becomes an element that conceals the social differences and the historically-accumulated injustices. The dissemination of this idea in society affects the individuals and the groups as a whole, including those victimized by the same racism which ultimately produces differences and reproduces injustices.

For the sake of comparison, we may cite the survey conducted by the DataFolha Institute in 1995, which presented a nationwide sample. In this survey, the interviewees were asked: "What is your opinion on quotas for Negroes in schools and in jobs? A percentage as significant as 40% of the Negro respondents supported the policy wholeheartedly and 15% of them gave it partial support. We may wonder what will have changed in this period of nearly 10 years. The effective implementation of the quotas system in public universities is very likely to have placed this discussion on the press agenda and, due to the criticism systematically expressed, by large-circulation newspapers in particular, a significant pressure and a new "ideological curtain" may have been created, which has had an impact even on the segments which benefit from the quotas system.

When we verify the question of criteria for self-definition of color or race among those who declare themselves as Afro-descents (or not), we notice that the Afro-descents are slightly more prone to choose the family's origin as the most relevant criterion. Nevertheless, if we add up the two criteria which refer to the phenotype (color of skin and physical aspects) we will notice that both groups are nearly equaled, even if the item color of skin alone is approximately 8% more important for the non-Afro-descents.

On the table above, by crossing the self-definition of color in the bi-racial pattern with the declaration of Afro-descent, we see that the percentage of Afro-descent Negroes is far more significant than the percentage of whites (although among the latter Afro-descent goes way above 50%). The most interesting aspect, though, seems to be the realization that out of the significant number of non-respondents whose color is bi-racial, the majority will identify themselves with the Afro-descent.

The crossing of open color or race with the declaration of Afro-descent leads us to notice an interesting scenario: in a higher percentage than the blacks, the self-declared Negroes define themselves as Afro-descents (and the percentage of blacks is almost as high as the browns' in this item) This data may reflect to some extent, in the open declaration of color or race, a more meaningful sense of identity among Negroes than among blacks.

Furthermore, it is worth noting that dark, mixed-raced and mulatto interviewees (the latter having a not very significant weight in this sample) are predominantly prone to state their Afro-descent.

When we analyze the percentage rates of acceptance of the quotas policy for Negroes in higher education compared to the declaration of Afro-descent, we verify that there is no significant difference between the two groups. Even if there are, among Afro-descents, a larger number of interviewees who are totally in favor of such a policy (24.64% against 15.45% of the non Afro-descents), when we add those "in favor" to those with "no strong opinion", we end up with approximately 37% among the Afro-descents and 34% among the non Afro-descents. In the same way, we notice that both groups are quite close to one another with regard to the rejection of racial quotas for the public university.

Conclusion

The social differences associated to color or race are widely known and officially documented in Brazil. Federal agencies such as IPEA and IBGE, alongside with researchers and research groups acting in Brazilian universities, have pointed out that among whites and Negroes (considering the self-declared blacks and browns altogether) one may find a wide range of social performances, regarding life expectancy, child mortality rates, housing, income, job status, unemployment etc.

Particularly with regard to education we also find huge differences between the two groups. According to the study by Henriques (2001) – which is based on the data generated by PNAD of IBGE – towards the end of the 1990's the average difference in years of schooling between a Negro and a white person, both aged 25, was 2.3 years, which corresponds to a high inequality, insofar as the average adult schooling does not exceed 6 years. The most significant aspect, though, is that such a pattern of inequality, concerning the average years of schooling, has been stable for decades now.

The higher the schooling level, the larger the number of such differences will be, reaching a critical point in higher education. This level is attained by only 7.1% of Brazilians aged between 18 and 25. However, among the whites within this age bracket the access rates rise to 11.2%, whereas among Negroes it does not exceed 2.3%.

Insofar as the fundamental item in the ascending social mobility in Brazil lies in education (Pastore and Silva, 2000), these inequalities have been the propelling force and the justification for the implementation of policies of racial quotas in public universities. These policies, which fall within the field of "affirmative actions", require, to be implemented, a definition of categories and racial classifications.

If the Brazilian way of classifying is so ambiguous as to allow one single individual to be classified differently from the way he classifies himself, to implement affirmative actions requires some rather specific knowledge of our complex logic of classifications. In other words, it requires us to understand the objective and subjective elements used to connect an individual to a color or race group, making him the target of discriminations which interfere with the course of his social life. This is exactly what we aimed at in this research.

Thus, the problem resides in the fact that the racial classifications used by Brazilians are not as clearly defined as the ones that have been used in other societies, such as the American one, for instance. This ambiguity in our classification is visible at once in the concept of "color" itself. National common sense tends to use the term "color" instead of "race" to indicate phenotypical differences between the individuals. The concept of "race" has been disseminated for approximately three decades, by academic researchers in the areas of social sciences and applied social sciences, as well as by militants in the Black movement, but in fact it is not widely used in society as a whole.

Our concept of color refers, as it has been said, to the phenotype. This term encompasses combined elements such as the color of skin itself, hair texture, the shape of lips and nose. The color, thus, does not refer to origin, but to the phenotypical mark the individual bears.

To date, most of Brazilian publications on this matter have accepted that the structure of Brazilian racism was defined by Nogueira (1998)⁴. According to this author, our racial prejudice could be named as "of mark". Thus, the criteria which are behind discrimination would not be connected to the ethnic or racial origin of the individual (as in the American "differentialist" racism, in which prejudice seems to be "of origin" and therefore relatively dissociated from physical appearance), but to the phenotypical aspects. The closer the individual is to the African phenotype, the more likely to be discriminated he is. In this system, the "native" concept of color has a privileged place in the demarcation of differences and substitutes the concept of race, which becomes, therefore, implicit.

Brazilian racial order, in that sense, dissimulates existing racism and dilutes it within the concept of "racial democracy" and in the assertion of a "color continuum". However, neither the former – as an ideology – nor the latter prevent both the self-declared blacks and the self-declared browns from holding underprivileged socioeconomic positions in society, positions which are far below the ones held by whites. This takes place because racism (even if transmuted into prejudice and discrimination on grounds of "color") is deep-rooted in common sense and in national culture itself; it permeates not only the most generic sociability relationships but also the relationships with the Negroes that are established by public organizations and by the private sector. For this very reason, the differences in social performance between whites and Negroes in Brazil may be found in all existing indicators.

Given the fact that, in Brazilian racial order, the shape of prejudice is the "physical mark", the essentialism that binds individuals to an immutable racial group is less visible. Children born by the same inter-racial union who bear opposed phenotypical characteristics tend to face very distinct problems in their future social lives.

As we have mentioned, the PNAD of 1976 has openly collected the interviewees' self-classification of color or race and found approximately 100 categories. At the origin of this rather fluid classification lies the very architecture of our racial order. In our society, inequality between whites and Negroes has been maintained and reproduced since the abolition of slavery without the need to produce a discriminatory and segregating legislation; not even the concept of race had to be used. In the United States and in South Africa, on the other hand, the unequal racial order has been built in a legal fashion, which required the making of specific systems of racial classification, systems which would not leave room for doubts with regard to the division between the two racial groups, (Marx, 1998).

In the American case it was a rule that a drop of Negro blood defined an individual as a Negro regardless of his physical aspects. Racial origin determined the bond between the individual and the group. The legislation that maintained official segregation in the United States was abolished in the 1960s; however, this logic of racial classification by origin has infiltrated in the common sense and continues to operate in that society (Telles, 1993).

In the Brazilian case, we verify the opposite logic. We may say that among us a drop of white blood produces an individual who may try to include himself in a myriad of categories and try to escape the classification of Negro or black that is, in general, an indicator of an underprivileged status, objectively and subjectively speaking, in our society.

As we have seen, the categories used by IBGE accommodate the common sense, insofar as they present the term "brown", also entirely ambiguous and which may refer both to the "mulattos" and the "mixed-race" people from several different origins. However, these "censitary" categories used to "count" the population also produce ways of classification which end up being adopted by the population as a whole.

We believe that in complex societies social classifications are not processed in a sort of sociological vacuum. In that sense, also the racial (or color) classifications are the product not only of cultural varieties of differentiation, but also of the way men establish relations with the institutions around them, including the market and the State.

Thus, the myriad of racial classifications existing in the country are not only the result of a cultural aspect, but also the product of very objective political and economical interactions which have imposed their characteristics onto subjectivity. As we know, right after the abolition of slavery, the Brazilian State bet on a strategy of "whitening" of the population, resorting to a policy of mass immigration from Europe. If the target to be achieved was a white population and if the situation should be transformed precisely because of the population of Negroes, it was left to the latter (stigmatized by the stereotype of backwardness and inferiority) the attempt to escape such a negative classification.

We believe that the creation of a "color continuum" represents the Negroes' necessity to escape the degrading treatment that both the State and the society offer them. In that sense, any level of miscegenation creates the possibility of an intermediate classification, which could, to some extent, mean less discrimination.

In fact, we notice a non-polarized classification between two categories. But this non-polarization is only "virtual", for as even the most recent social indicators have been demonstrating (see, for instance, Henriques, 2001), the self-declared browns in the IBGE surveys, who would represent a synthesis of the color continuum, are not, from the point of view of social indicators, halfway between whites and Negroes. The browns are slightly better situated than the blacks but a long way behind the whites.

On the other hand, the adoption of policies of quotas for Negroes should have an impact on Brazilian racial classification by encouraging the acceptance of blackness or, at least, of the "color" brown or Negro.

Returning to the main point for us, the dispersion in the racial categories used by Brazilians for self-representation in social life, we see that, although such dispersion sets limits to our differences with regard to other societies, where the racial order is "differentialist", it does not prevent blacks, Negroes, browns, mixed-race people and others from being socioeconomically far behind whites and from being the target of everyday discrimination.

Thus, our evaluation about a merely virtual character of this racial non-polarization is strengthened, for there are certainly very practical mechanisms to identify color or race, which create the possibility for the Negroes (and here we refer to the continuum that goes from black to all categories which indicate the mixture of this group with others) to be identified, to receive discriminatory treatment (even if dissimulated in many cases and in many different moments in the course of a life) and to suffer the consequences of such treatment.

If we devote ourselves once again to the data presented above, we will be able to show the validity of the indications we have just built.

Analyzing the open answers about color or race, we verify that, despite the fact that we have a varied listing of classifications, most of such classifications were concentrated on those used by IBGE (63.71%).

When we compare the declarations of open color or race, we notice that the strongest consistence in both classifications comes from the whites. In second comes the consistence from blacks in the closed question and the Negroes in the open question. The browns of the closed question, though, present a large dispersion in the open question, but this occurs in the majority of classifications which indicate a racial mixture or race-crossing.

The weight of IBGE categories in the open questions of color or race may indicate that the official data obtained end up "regulating" the social ways of classification. IBGE has been using a pattern of classification which, being official, has been assimilated by the majority of population as valid or accurate, creating, thus, an adequate image of racial classification. Therefore, some sort of regulation of the official vocabulary has been forced upon common sense, a regulation which was eventually built up through everyday usage.

In that sense, the value attributed by part of the literature on this subject to the dispersion and lack of consistence of our racial classification can be clarified by the evaluations above.

In the analyses we have verified two problems in the bi-racial (white/Negro) classification. The first one concerns this category by the three groups of color or race, but especially by the blacks and the browns (which can be verified in the percentage of non-respondents). The second problem is that when such a classification is compared to the one presented by IBGE, or even to the open one, the amount of whites grows significantly, due to the fact that the browns tend, to a large extent, to migrate to the group of whites.

In that sense, the official adoption of a bi-racial classification, as it is supported by part of the National Negro Movement, would end up generating a situation in which the demographic weight of this group (considering the blacks and browns in IBGE classification) would be smaller. We conclude, therefore, that the category of "browns" is necessary in order to settle the multitude of classifications that comprise the so-called "color continuum".

The analysis we have done demonstrate that if there is some rejection to bi-racial categorization, there is no rejection to the question: "Do you consider yourself an Afro-descent or of a Negro origin?" In fact, the percentage of Afro-descents is far larger than the number of blacks and browns taken together precisely because many whites also identify themselves this way. The interesting aspect is that a large percentage of browns and blacks oppose to the phenotypical (bi-racial) polarization, but do not oppose to the polarization of origin (Afro or non-Afro).

We can conclude that the self-declaration of Afro-descent does not seem to be an appropriate parameter of inclusion in affirmative action policies (be that in the scope of higher education or even in the scope of the other social policies) which aim at giving privileges, through positive discrimination, to groups that have been kept in the lower strata of social hierarchy, due to lengthy processes connected to the building and maintenance of our racial order (Brandão and Marins, 2005).

We have seen that the secondary school pupils interviewed are, to a large extent, opposed to the quotas policy for Negroes in university, despite the fact that many of them might benefit from such a policy. Only qualitative research can go deeper in explanations for such a configuration. We can, however, anticipate two fundamental elements. The first one is the abstract conception of merit, which is structural in capitalist societies oriented by liberal principles. The second problem lies in the effects of the ideology of racial democracy which cloaks in lies the toughness of inter-racial relations in Brazil.

We have verified that pupils point to the racist nature of Brazilian society, but do not declare themselves as racists. In other words, they refer to racism as something which is beyond their individuality, as if it were possible to separate the society level from the individual practices which ultimately move society and reproduce its relations.

When we investigate the criteria used for self-classification of color or race and for classifying other individuals, we verify that in fact the phenotypical aspects stands out; however, origin is not dismissed. Quite the contrary, such demarcation on color or race arises as the second most important aspect, always given a significant weight in percentage. This conclusion is fundamental for it puts in check the argument by Nogueira (1985 and 1988) concerning "mark" as a basic classificatory element in Brazilian racial relations, an argument that is widely accepted by racial studies conducted in Brazil.

We can say that, given the limitations of our sample, we have found a pattern of racial classification which corresponds to: a) the definition of a white pole and a black/Negro pole between which are distributed classifications of color that are reunited within the category of browns; b) an open dispersion which is simply relative and; c) a high value given to phenotype and to the family origin, with a more important weight given to the former.

In that sense, we conclude that the criticism to affirmative action policies based on the impossibility of a definition about the potential beneficiaries of such policies is groundless, for even if we do not have an absolutely defined classification of color or race, we have a classificatory logic which, in spite of its flexibility, creates a pattern of identification socially used.

To conclude, we may refer to the classical essay Primitive Classification, by Émile Durkheim and Marcel Mauss (Durkheim and Mauss, 1984), in which the authors insist on proving what they call an absolutely social nature of form through which men create their classifications. In that sense,

Society was not simply a model according to which the classificatory thinking would have worked; the frameworks of society were themselves used as frameworks by the system (Durkheim and Mauss, 1984, p. 198)

In order to elaborate their arguments, Émile Durkheim and Marcel Mauss made use of reflections on classification models used by tribal societies ("primitive" ones, in their view) from several regions in the world. Within the context of the early 20^th century, when social science was still being established as an independent and valid field for studies, the authors, when concluding their essay, emphatically pointed out that the principle they had discovered could be used for the comprehension not only of classifications, but also of a number of other ways of understanding (such as the social concepts of time, space, substance etc).

Generalizing this classical argument in order to understand Brazilian racial relations may lead us to the conclusion that the seemingly unclear way of our logic of racial classification would result from an also unclear way of differentiating the racial groups in society. Two reflections may stem from this.

The first one depends on our accepting the notion that arises in studies which indicate the ineffable existence of a color continuum irreducible to a small number of categories of racial classification. From this principle, we would actually be under the aegis of the "myth of racial democracy" which, being a myth, is the founder of a type of social relationship between individuals phenotypically distinct and of the way of classifying the differences between them. The continuum "would be useful" to reassert a gradation of color which could be relativized because of a non-racialization.

The second reflection arises from a point beyond the first one. Looking into this continuum and considering it as a background to other classificatory indications existing in common sense, we realize that this continuum, even if it does not actually point to a bipolar model, perfectly fits within a "triadic" model. More specifically, there would be definite polar categories: on one side, the white category, on the other side, the homothetic categories of black and Negro. The so-called continuum, in fact, brings together everything that can be thought between these two categories, that is, all the categories which bear the signals of mixture and are likely to be represented and grouped within the classification of browns.

It is not by chance, therefore, that when we divide whites, blacks and browns in socioeconomic categories, we verify a definite demarcation between them, which displays the whites on the top of social hierarchy, way above the browns, who are on the base, but still a little above the blacks.

Thus, if Durkheim and Mauss (1984) were correct in their opinion, our classification addresses a specific form of society in which the differences are recognized and are expressed not only in the "taxonomy" of color or race, but also in the distribution of individuals in the social structure.

Bibliographic references

Contact:

André Augusto Brandão

Universidade Federal Fluminense

Campus do Gragoatá, Bl.E, s. 322

24210-350 – Niterói – RJ

e-mail: [e-mail]

Received 07.12.2005

Accepted 22.06.2006

André Augusto Brandão has a doctorate in Social Sciences, and is a Professor at Fluminense Federal University, working in the Program of Post-Graduate Studies in Social Policies. He coordinates national projects on the evaluation of public social policies.

Mani Tebet A. de Marins is an undergraduate student at Fluminense Federal University and has won a scholarship from CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico). She has received the First Award Building Equality in Gender, organized by CNPq and the Special Secretariat of Policies for Women ([e-mail])

• BOURDIEU, P. A dominação masculina Rio de Janeiro: Bertrand Brasil, 1999.
• BRANDÃO, A. A. Avaliação de um pré-vestibular voluntário para negros pobres. In: REUNIÃO ANUAL DA ANPED, 27., 2004, Caxambu. Anais... Caxambu-MG, 2004.
• ______. Miséria da periferia Rio de Janeiro: Pallas, 2004a.
• ______. Discursos sobre o mérito entre alunos do PVNC. Cadernos PENESB, n. 5, Niterói: EDUFF, 2004b.
• BRANDÃO, A. A.; MARINS, M. T. A. Quem são os alunos afro-descendentes da UFF?. In: Reunião da Sociedade Brasileira de Sociologia, 12., 2005, Belo Horizonte, Anais eletrônicos.. Belo Horizonte MG, 2005. Disponível em: <www.sbsociologia.com.br>.
• BRANDÃO, A. A.; TEIXEIRA, M. P. Censo Étnico-racial da UFF e da UFMT Niterói: EDUFF, 2003.
• DURKHEIM, E.; MAUSS, M. Algumas formas de primitivas de classificação. In: DURKHEIM, E. Sociologia São Paulo, Ática, 1984. p.183-203.
• FOLHA DE SÃO PAULO. Caderno Especial, 25/06/1995.
• HASENBALG, C. Discriminação e desigualdade racial no Brasil Rio de Janeiro: Graal, 1979.
• HENRIQUES, R. Desigualdade racial no Brasil: evolução das condições de vida na década de 90. IPEA: Textos para discussão, n. 807, 2001.
• IBGE. Censo demográfico brasileiro: 2000. Rio de Janeiro: IBGE, 2001.
• MARX, A. Make race and nations London: Cabridge Press, 1998.
• NOGUEIRA, O. Tanto preto quanto negro: estudo de relações raciais em São Paulo. São Paulo: T. A. Queiroz, 1985.
• _______. Preconceito de marca: relações raciais em Itapetinga. São Paulo: Edusp, 1998.
• PASTORE, J.; SILVA, N. do V. Mobilidade social no Brasil São Paulo: Macron Books, 2000.
• SANSONE, L. Negritude sem etnicidade Rio de Janeiro: Pallas, 2004.
• TELLES, E. Racismo à brasileira Rio de Janeiro: Relume Dumara, 2003.

1

. Brandão and Teixeira (2003) have already showed that in the courses at the Fluminense Federal University there were more women enrolled (55.57%) than men (44.43%).

2

. We call "consistency" the matching of two classifications made by the same interviewee.

3

. The PVNC actually consists in a networked movement currently congregating dozens of preparatory courses – denominated nuclei – throughout the Metropolitan Region of Rio de Janeiro. Each of those nuclei has between 40 and 100 students.

4

. It is worth noting that this work was originally published in 1955.

5. As shown b the studies carried out from the late 1970s to the beginning of this century (for example, Hasenbalg, 1979 and Henriques, 2001).

Publication Dates

• Publication in this collection
  25 July 2007
• Date of issue
  Apr 2007

History

• Received
  07 Dec 2005
• Accepted
  22 June 2006