MIGRATION IN BRAZILIAN AFRO-DESCENDANTS COMMUNITIES: A NEW APPROACH TO ILLUSTRATE THE MEANING OF CONTEMPORARY QUILOMBO

Resumo: African-descendant populations in Latin America are heterogeneous in their history, culture


INTRODUCTION
The proportion of African-derived people in different Latin-American countries varies widely. A total population ranging from 80 to 150 million people was estimated from data from different sources, including United Nations Children's Fund (UNICEF), The World Bank Group, the Economic Commission for Latin America and the Caribbean, among others (Antón et al., 2009).
During the Latin American period of slavery (1490-1888), communities were founded by run-away or by abandoned former slaves in rural or jungle areas. These populations are tied to slavery resistance and are known by different names in different countries, such as quilombos in Brazil, cumbes in Venezuela, palenques in Cuba and Colombia, marrons in Jamaica, bush negroes in Suriname, and cimaronaje in Cuba and Puerto Rico. These terms may be equivalent to quilombo, meaning the pursuit of freedom, rebellion and resistance against oppression or run-away slaves (Gomes, 2015;Florentino and Amantino, 2012). In Brazil, there are many names for these communities besides quilombos, namely mocambos, comunidades negras (Black communities), comunidades negras rurais (Rural Black communities), comunidades Afro-descendentes (Africandescendant communities), Terra de pretos (Black's land), and Afro-derived communities.
The term quilombo appears initially associated with Quilombo de Palmares (a historical African refugee settlement in Alagoas, Brazil), possibly an allusion to its political organization, similar to Angolan quilombos in Central Africa (Lara, 2010;Filho and Germani, 2014). The classic definition of quilombo dates back to the 18th century and describes these communities as formed by run-away slaves who escaped from their slavers and hid in difficult-to-reach areas, which in Brazil are referred to as "the territory" (Chagas, 2001;Torre, 2013).
While quilombos have existed for hundreds of years, only in recent decades have they have intensified their efforts to make legal claims and gain titles to their ancestral lands. Unfortunately, they have encountered numerous obstacles, many of which stem from disagreements over the definition of a quilombo, as well as from their socially and economically marginalized position within Brazil (Center, 2008). Furthermore, they have encountered difficulties in finding records or other evidence supporting these claims (Fiabani, 2008). Because the definition of what a quilombo is affects the application of laws granting land rights to these communities, as well as access to other affirmative action benefits, quilombo representatives seek an expansive definition that accommodates the historical diversity in these communities, while groups with interest in conflict with their rights seek to restrict this definition. It is noteworthy that in Brazil there are also African-derived communities which do not consider themselves to be quilombos, because they do not have historical or cultural bases to define themselves. These kind of Afro-Brazilian groups may or may not be rural (Anjos, 2011;SEPPIR, 2012). Moreover, there are communities that do not wish to be associated with the quilombo identity because to them it means a return to the slavery system (Silva, 2011;Mattos, 2006).
After decades of historical and anthropological studies and of political and social movements, the definition of quilombo has been expanded, and according to the Brazilian Association of Anthropology, it does not comprise a single social formation, immutable and fixed in time (ABA, 2012). The Black Movement from the 1960's-1970's, together with academic representatives, began a political and ideological struggle to unmask racism in Brazilian society, and to show racial differences in socioeconomic and demographic contexts (Popolo et al., 2011) and to repair a historical debt of Brazilian society with the descendants of slaves. The Black Movement promoted the identification of African-descendants with their historical sense of 'pertencimento' (to belong), or membership, in the African diaspora. In this moment, 'quilombo' was associated to a form of mobilization and struggle for freedom and citizenship, as well as for political, economic and social liberation (Leite, 2015).
Currently, the Brazilian legal system recognizes 'Remnants of quilombos', as ethno-racial groups, according to self-classification criteria, each with their own historical trajectory, associated to their own land, established as part of the resistance to oppression, and with presumed African ancestry related to resistance suffered during historic oppression.
More than 2,600 Afro-derived communities spread around of the country, are recognized by the government as quilombos (Fundação Cultural Palmares, 2016), the majority of which are located in Central Brazil region, where a combination of the ecology, topography and absent army contributed to the establishment of many quilombos (Karash, 1996). This region is known as Chapada dos Negros (Black people highland) and has about seventy seven quilombos communities, the majority of whose inhabitants descend from slaves from mining camps (Lopes, 2009;Fundação Cultural Palmares, 2016).
More than a century after the abolition of slavery in Brazil, the term 'quilombo' continues to evolve new meanings, not all of them associated with its common definition as a runaway slave community (Leite, 2015). This has given rise to the discussion how all such groups fit the contemporary meaning of quilombo and the relation between quilombo definition and the political recognition as well struggle for land rights.
Therefore, the purpose of this study is to illustrate the contemporary meaning of quilombo using the historical circumstances and the diversity in the migration patterns of three quilombos with different levels of urbanization located in Central Brazil region: an urban, a non-isolated rural and an isolated rural. These quilombos are registered by the Brazilian governmental institute (Fundação Cultural Palmares) as a recognition of their full quilombo status.

METHODS
We conducted a cross-sectional population-based study in three quilombos: Cocalinho, Pé do Morro, and Kalunga. The geographic locations of these three quilombos are shown in the Figure 1. Cross-sectional study design is a type of observational study design. In a cross-sectional population-based study, the investigator measures the outcome and the exposures in the study participants at the same time in each study population. The participants in a cross-sectional study are selected based on the inclusion and exclusion criteria set for the research. These studies are used for population-based surveys and can usually be conducted relatively faster and are inexpensive (Setia, 2016).
Cocalinho, a non-isolated rural village, and Pé do Morro, an urban community, have about 700 and 600 inhabitants, respectively. Both are located in the Araguaia River Basin, North region of Brazil. Kalunga, an isolated rural community, has a population of about 5300 habitants and is one of the largest quilombo communities in Brazil. The community covers an area of approximately 258,000 hectares in the countryside of the state in the Middle West region. It is partitioned into several villages, which are scattered throughout the Kalunga territory, without a marked geographical isolation between them. The following variables were collected: age, gender, place of birth, phenotypic self-classification, and number of marriages. We defined native individuals as those who were born into the community. The female/male ratio between natives and immigrants was determined. All participants who reported having a spouse, were additionally asked about the place of birth of the spouse. We also asked about the reason for migrating into their current place of residency. Moreover, we collected these data across three generations to provide evidence about the historical formation. Each individual was asked about the place of birth of their parents and grandparents. The participants who were interviewed were called generation 3 (G3), and participants' parents and grandparents were called generations 2 (G2) and 1 (G1), respectively. We estimated the proportion of couples formed by two native members (named endogamous), by a native member and a migrant (exogamous) and by two migrant members (migrants). The number of each type of marriage was obtained by direct counting. Among exogamous marriages, we estimated the proportion of patrilocal (native men married with migrant women) and matrilocal (native women married with migrant men) marriages. From that, it was possible to determine the ratio of the number of patrilocal to matrilocal unions (P and M index) (Amorim et. al., 2011).
We chose a sampling strategy based on estimates of number of families in each community. Traditionally, quilombo families are large with more than four generations living in the same house. The sample of this study included at least two members of each family, generally the oldest couple of the house.

RESULTS
The distribution of migrants and natives was not the same across the three quilombos (Table 1). Kalunga is the community with the highest frequency of nativeborn individuals. We observed a significant difference in the proportion of men and women between natives and migrants in Cocalinho, with a higher number of migrant men and native women. In contrast, in Pé de Morro there was no difference in the proportion of migrant men and women, while in Kalunga we did not observe any migrant women. We observed a significant proportion of immigrants over 70 years of age in Pé do Morro when compared to native individuals (p=0.001), indicating an association between migration and age group. This relationship was not present in either of the other two populations (data not shown). Differences were observed in the distributions of phenotypic self-classification between Brazilian's census and the three quilombos communities (p ≈ 0.000) ( Table 2). Our data indicated that the proportion of people who consider themselves to descend from African ancestors was higher in the three communities than in urban areas. We observed a higher proportion of Black selfclassification (between 34.92 to 57.97 %) and lower proportion of White selfclassification (between 0.97 to 11.43%) in the three quilombos than in nearby cities and surrounding regions.
The categories of marriages over the past three generations are shown in Table   3. All information about three generations was accessed from generation 3. It was possible to access data from 87 couples for G1, 52 for G2, and 46 for G3 in Cocalinho while in Pé do Morro it was accessed 70 couples for G1, 52 for G2, and 40 for G3. In Kalunga, it was possible to access data from 242 couples for G1, 133 for G3, and 126 for G3.
The highest proportion of endogamous marriages across the three generations was observed in Kalunga, the isolated rural community. The few marriages that took place between migrants were observed in generations 1 (14 -5.78%) and 2 (six -4.51%), while the exogamous marriages across the three generations were mostly matrilocal. In contrast, marriages between migrants were prevalent in all generations in Cocalinho and Pé do Morro, while endogamous marriages were not observed. Exogamous marriages in Cocalinho were observed in generation 3 (19.57%), and they were mostly matrilocal. In Pé do Morro, patrilocal marriages comprised 10% of the total, as indicated by the P and M index shown in Table 3. 197 Revista da ABPN • v.12, nº 32 • março -maio 2020, p.188-208

DISCUSSION
Quilombo, initially regarded as a form of political resistance against slavery and the colonial regime, has become a form of social organization and ethnic category, and a legal category found in public policy (Leite, 2015). Contemporary quilombo communities claim their origin in traditional quilombos describing themselves as part of the movement of resistance to oppression during and after the slavery period and as being descendants of rebel slaves who took up residence on the margins of society. However, it is difficult to find historical documentation of the origin and history of these population and usually their stories are passed down via oral narratives (Funari, 1996;Lopes, 2009). Kalunga quilombo is composed of descendants of enslaved Africans who were taken to work in the "Mines of Tocantins", in an area where today the city of Monte

In this study we documented two examples of demographic composition of
Alegre is located. According to Soares (1995), slaves fled from the mines in the late 18th century and formed the quilombo. However, residents indicated that their ancestors were abandoned by the pioneers when the gold mines were exhausted and, fearing capture again, they remained hidden for decades without contacting the outside world. In 1982 anthropologists drew attention to the construction of a hydroelectric plant in Rio Paranã, which would harm hundreds of Kalunga people. At that point, there was no information 200 about this people, and soon after they started the process of claiming their lands (Baiocchi, 1996). Over time, towns and cities grew near their area and they started trading relations with the inhabitants of urban centers (Meegen- Silva, 1999 (Fiabani, 2008). Since the late 19th century, the population of the Brazilian Northeast region has suffered from intense periods of drought, which triggered migration flows to the Brazilian North region (Costa, 2015).
The first descendants-of-slaves settlers arrived in the north of Tocantins at the beginning of 20th century. More of these landless peasants (campesinos) kept arriving, and were part of the migratory movements in search of land and work that peaked in the 1950's in Brazil (Velho, 2009). According to researchers, the migration of rural Black communities from Maranhão to the North region was not only associated with economic necessity, but also with keeping family groups together (Gomes 2011;.
Therefore, the communities that were formed beginning in the early 20 th Century, but continually through the 20 th Century, were peopled by migrants who were, in many cases, related or who had some cultural cohesion, as they migrated from the same region.

Cocalinho and Pé do Morro were officially recognized by Fundação Cultural
Palmares (2016) as quilombo in 2005 and 2010, respectively. The history and recent formation of these two quilombos do not fit the common sense of quilombo, but do fit the current definition and exemplifies the diversity of quilombo communities. According to Florentino and Amantino (2012), quilombo definition is evident in the absence of taxonomies, which envision quilombos as effectively historical structures, which might 201 have been restricted merely to loose groups or have developed into self-sustaining communities capable of reproducing themselves economically and demographically for long periods in the Americas. These communities are inserted in multicultural societies with a high degree of ethnic-racial diversity and historical economic plurality (Florentino and Amantino, 2012).
In this study we provided evidences about the historical foundation of three quilombos using migration patterns. We were able to reconstruct the marriage structure history during the last three generations in the three quilombos. It is very clear that the level of endogamous marriage observed in Kalunga was much higher than in the other two quilombos both presently and in the past. A predominance of endogamous marriages was observed over the three generations investigated, and the few migrants (3) we sampled were all male. This result suggests that migration is gender-biased so that more men are likely to migrate into the community. The absence of migrants in Kalunga is probably due to Kalunga's location and lack of resources such as roads, electricity, running water and low possibility of irrigation, which make Kalunga a less-than-desirable destination. In Cocalinho, exogamous matrilocal marriages were more common than patrilocal marriages, which in fact is associated with the increased mobility of males in search of work (Oliveira and Jannuzzi, 2005). In Pé do Morro, we observed a predominance of marriages between migrants. Moreover, we observed that there are older migrants there than in other quilombos. In general, the number of marriages between migrants decreases across generations. This is due to the fact that community members take more control of community resources during the process of registration of land (Véran, 2000). In different quilombos from the Northeast of Brazil, it was observed a predominance of endogamous marriages and an intermediate proportion of marriages between migrants (Amorim et al., 2011) when we compared to our quilombos.
Although phenotypic self-classification is an important aspect to recognition as a member of a quilombo, these groups display all or some of the following characteristics: definition of an ethnicity, rituals or religiousness, common origin or ancestry, kinship bonds, relation to slavery, and especially a connection with their territories (Marques and Gomes, 2013). For Afro-derived communities, cultural recognition, a connection to African ancestry, and cultural traditions have become a primary way of obtaining 202 collective land rights and social inclusion (Farfán-Santos, 2015;Brazil, 2006 Affirmative action and public policies intended to promote equal opportunities for people of all ancestries are being implemented in Brazil over the past decade and may have helped promote a sense of self-esteem and, consequently, to an increase in the number of people who classify themselves in the categories "mixed" and "black". Indeed, in the two last Brazilian census, the proportion of people self-classified as black and mixed increased from 6.2% to 7.6%, and from 38.5% to 43.1%, respectively (IBGE, 2011).
The importance of highlighting the ethnic-racial dimension in population studies is, at least in part, due the recognition of the historical discrimination that African peoples and their descendants suffered and that resulted in conditions of marginalization and vulnerability.

203
Revista da ABPN • v.12, nº 32 • março -maio 2020, p.188-208 Although there are differences in the struggle and history of resistance among quilombos, their geographical proximity to urban centers or even overlapping with cities possibly contributes to a new quilombo identity. Accordingly, in order to establish the rights of Afro-derived communities in Brazil, such as the balance between universalist and affirmative action policies, the partial convergence between poverty and racial discrimination and the historical legacy of discrimination in Brazilian society, it is necessary to consider that quilombos may have different beginnings and histories (Ikawa, 2009;Telles and Paixão, 2013 are recognized as such, which have their own forms of social organization, which occupy and use territories and natural resources as condition for its cultural, social, religious, ancestral and economic reproduction, using knowledge, innovations and practices generated and transmitted by tradition." (Author´s free translation). New meanings have been built on the terms remnants of quilombos and quilombos, since there is a diversity of formation histories, representing this term a social category for governmental discussions and public policies (Mattos, 2003;O 'Dwyer, 2007).

CONCLUSION
We have sought to analyze the meaning of quilombo through examples based on our field research carried out in Central Brazil region Afro communities. Our aim in this article has been to illustrate the meaning of quilombo using a different approach such as historical circumstances and diversity in the migration patterns. This approach allow us to loosely sort them into two examples: communities that were formed by either run-away or abandoned slaves (historical quilombos) or communities not directly related to the history of slavery, but associated with the migration of rural land-less Black communities after the abolition of slavery (migrant quilombos). The lack of agreement on the definition of quilombo is a debate that involves economic and social interests of different parts of Brazilian society, as the meaning of the concept is indissociably tied from the historical 'ideal' of what a quilombo "should" be (Center, 2008). Therefore, it is critical to stress that many communities that were historically discriminated may be at risk of losing their land. If there are quilombos, it is because a group achieved political autonomy and social organization to have a territory and fight for better living conditions at a historical moment. However, different groups arrived at their own historical moments in different ways. Therefore, we highlight the quilombo communities represent a struggle for rights and for a consciousness about the African diaspora and its descendants.