Skip to main content

Endangered languages in town: The Urbanization Of Indigenous Languages In The Brazilian Amazon

de Oliveira
Gilvan Muller

Brazil's territory covers an area of 8,500,000 square kilometers (3,286,170 square miles), and is home to a population of about 169,500,000 inhabitants, only 16 million of whom live in rural areas -- the number has not changed since 1950. (IBGE,(1) 1991 census) The country's remaining 154 million live in 5,507 urban nuclei, the seats of municipalities (the country's smallest administrative units). Brazil is a federal republic composed of 26 states that have the power to mandate and enforce their own public policies, including those concerning education.

Brazil's indigenous population constitutes a minority of about 350,000 people, amounting to 0.2 percent of the overall population. These are the survivors of the four million native Brazilian Indians estimated to have lived during the time of European arrival in what is today the Brazilian territory. The continuous wars against the Indians, which persisted until the second half of the 20(th) century (Leonardi, 1996), reduced their numbers to a small fraction(2) of the original population.

Even with such small populations, the Brazilian indigenous peoples speak 170 different languages, which are derived from various linguistic branches and families. A great number of these languages belong genetically to the Tupi, Jê, Karib, and Aruak branches. All Brazilian indigenous languages may be considered endangered, especially if one takes into account the reduced number of people who speak each language and the country's sociolinguistic situation. Brazil thus serves as yet another confirmation of Crystal's (2000) claim that a mere four percent of the world's six billion people speak 96 percent of the world's languages.

The most populous indigenous minorities -- the Guarani, in the southern and central regions of the country; the Tikuna, along the River in the central Amazon region; and the Yanomami, in the mountains neighboring Venezuela -- have about 30,000 native members. At the opposite end of the continuum are peoples like the Umutina or the Bará, with only a single remaining speaker of their original languages. Most indigenous languages, however, have a few hundred speakers. Under these demographic and sociolinguistic conditions, the disappearance of many languages can be predicted for the next decades or even years, repeating the trend that, according to anthropologist Darcy Ribeiro, led to the extinction of 54 indigenous languages between 1904 and 1957.

The assimilationist language policy steering the country toward Portuguese monolingualism changed officially with the 1988 Constitution. The new Constitution granted native Brazilian Indians rights not only over their lands, but also over their cultures and languages. (Brazilian Federal Constitution, Article 231) This change of direction in the country's policy allowed the development of language policies which, in a worst-case scenario, could slow down the rhythm of language extinction. At best, they might well reverse it, at least in some regions.

The political change now underway in many Latin American countries is accompanied by a slow change of mentality among the socalled "national" (or non-indigenous) population, and calls for a reconsideration of the status of the urban native Brazilian Indian population. Brazilian Indians have not, after all, ceased to be indigenous because they now live in the cities. We must, therefore, take stock of the development of indigenous languages in urban settings.

The Urban Native Brazilian Indian Population

The concepts used by the Brazilian National Indian Foundation (FUNAI) to describe native Brazilian Indians outside of their territory -- de-tribalized or de-villaged Indians -- are no longer true to reality. Today we find "urban villages," places where members of particular Indian nations succeed in maintaining their networks of sociability, even though they are immersed within a non-Indian majority and are experiencing enormous economic hardship.

There is no reliable data on the size of urban native Brazilian Indian populations. Estimated figures vary from 26,000 (official FUNAI estimate data) to 50,000 (according to sources at the same institution's Assistance Department). In the absence of reliable information, we estimate the urban native Brazilian Indian population to be around 100,000 people -- a quarter of the total Indian population.(3)

Whereas the Brazilian government formulates public policies (with the participation of Indian communities) in the interest of Indians living in native territories, no policies have been formulated concerning the urban native Brazilian Indian population, as these are not officially recognized as Indian. Current legislation in the educational domain, for example, allows for the emergence of a language policy formulated for each particular native Brazilian Indian nation, according to its own project for the future. Such policy might be concerned, for instance, with maintaining and developing the Indian language (e.g., the lexicon), teaching Portuguese, or teaching a foreign language. The Brazilian federal states, which are legally responsible for the formal education of native Brazilian Indians, must therefore recognize curricular propositions coming from the communities, and must help them put these into practice with the support of funds from the states' budget allocations.

This system paves the way for the development of an interesting model for the management of ethnic, cultural, and linguistic diversity; a model in which the government does not formulate a program for all ethnic groups, but instead supports each group's own formulation. The policy has been implemented in the native Brazilian Indian territories -- formerly known as Indian reservations -- but not in towns and cities.

The recent approval by the National Council of Education of Resolution No. 3 (October 10, 1999), however, introduces new possibilities for native Brazilian Indian schooling wherever the population may be -- in the traditional Indian territories, or in towns and cities. The second article of Resolution No. 3 reads:

Basic elements for the organization, structure and functioning of native Brazilian Indian schools:

Their location in territories inhabited by Indian communities, even if these are spread over territories belonging to several contiguous states or municipalities.

We know that having a place of status in the city -- being visible in the city -- is important for the maintenance of Indian languages in rural areas, since the prestige emanating from cities is transferred to the languages. Their renewed prestige within traditional territories may help slow down the process of language shift to Portuguese. Supporting the urbanization of Indian languages and their use in urban institutional spaces may therefore constitute a decisive linguistic policy for Indian peoples interested in maintaining their languages and, consequently, their ethnic identity.

S...o Gabriel do Cachoeira: A Multilingual Town

S...o Gabriel da Cachoeira is a town situated at the heart of Brazil's most multilingual region, the upper Rio Negro river region in the state of Amazonas, which is near the country's borders with Colombia and Venezuela. The urban nucleus, with nearly 8,000 inhabitants (according to the 1991 census), today has about 10,000 inhabitants and is the main hub for a 112,000 square kilometer area. Spread over this area are 409 Indian villages and 165 bilingual native Brazilian Indian elementary schools (offering the first four years of schooling). This situation is atypical for Brazil, but it presents a concrete case of urban multilingualism, along with its political and linguistic aspects.

The city of S...o Gabriel da Cachoeira receives a large amount of migrants who come from Indian villages all over the region, mostly Indians from the Tukano linguistic group, or speakers of other languages of the eastern Tukano family. Brandhuber (1999) has studied the causes of the migrations (and urbanization) of the Tukano. She stresses internal conflict in the villages as one of the most important factors in understanding the phenomenon. The search for basic services, such as formal education and health, is also important. The Indians form the majority (more than 90 percent) of S...o Gabriel da Cachoeira's population.

At present, there is no precise sociolinguistic description of S...o Gabriel da Cachoeira. A study by Renault-Lescure (1990) focuses on the analysis of a particular urban group -- namely, students in an urban school -- and their linguistic practices at home. The group studied was comprised of 457 young students enrolled in classes from the 5(th) grade of elementary school to the end of secondary school, mostly youngsters between 11 and 18 years of age (though some are older than 20).

At the time of the study, 66 students (14.4 percent) spoke an indigenous language as their first language, and 125 students (27.37 percent) spoke a regional language as a second language. Two hundred and forty-one students (52.73 percent) had a passive knowledge of one of these languages. Considering the multilingual structure of the Rio Negro region, a portion of the sample may have been monolingual in Portuguese; another portion either spoke or understood two or three other languages.

Nheengatu was once the language of wider communication in S...o Gabriel da Cachoeira (and of the whole Brazilian Amazon); it gave way to Portuguese in its expansion inland from the coast. Tukano is expanding either over the traditional territories of other languages of the Rio Negro region, especially in the upper rivers (the Tiquié, for example), or in the town itself.(5)

In her analyses of the conditions for the reproduction of the two most widely spoken languages (Nheengatu and Tukano) in the group of 57 students she studied, Renault-Lescure (1990) concludes:

"Thus, there is a discontinuity in the transmission of languages: if the mothers prove to be more connected to the indigenous languages than the fathers, only 28 percent of the General Language (Nheengatu) speaker-mothers and 47 percent of the Tukano speaker-mothers use these languages when communicating with their children. Regarding the fathers, only 13 percent and 27 percent, respectively, use these languages when communicating with their children.

"The use of Portuguese imposed by the parents corresponds to a wish for integration to the national society, a parent's wish not to see their children being the target of ethnic discrimination.... The Portuguese language comes up as the key to integration, while mastery of an indigenous language entails a mark binding them to the world of the `non-civilized', from which they have to escape. At the Missions schools, for example, the idea that speaking an indigenous language is an obstacle to the learning of Portuguese was also implanted."

Her conclusion points to the linguistic shift to Portuguese and away from indigenous languages. Given the feeble intergenerational linguistic reproduction from adults to children, we could also expect a similar conclusion from the analysis of the macro political and linguistic situation of the country, which makes even the most widely spoken of Indian languages endangered. Three points, however, deserve our attention.

First, one piece of information provided by the author herself must be underscored: children speak Tukano and Nheengatu (the two indigenous languages most widely spoken by the students in the area) more frequently to their parents than their parents speak to them in those languages.

Even if in certain cases the difference is very small (three percent), in other cases it reaches seven to eight percent, which can indicate the need to analyze pressure factors for linguistic reproduction other than those having to do with interaction within the family. We know, according to Calvet (1994), that street languages may exert stronger pressure on the linguistic repertoire of speakers than the family itself. In this sense, the intra-familial sociolinguistic study reveals only one of the signs of linguistic maintenance or shift in a given community.

Second, Renault-Lescure's sample comes from schools in the town center, which is the traditional territory of the "non-Indian" and of the imposition of Portuguese monolingualism. It is therefore questionable whether her conclusion can be generalized for the town as a whole. Also questionable is whether the phenomena of language shift are taking place in the same direction and with comparable intensity in the peripheral areas of town. More importantly yet, the numerous new Indian contingents arriving in town after 1995 largely compensate, so to speak, for the quantitative losses of Indian language speakers, notably of Tukano speakers, a dynamic comparable to that of Spanish in the United States. Inter-generational loss is an important problem, but monolingual newcomers nurture the "threat" of "hispanicization" and add political weight to the minority language. Language shift in Indian areas may be produced by the shift from other languages to Tukano, either in the form of a "bilingual stage" in the transition toward Portuguese -- as has been the case with Nheengatu-Portuguese bilingualism since 1870 throughout the Brazilian Amazon. (Bessa Freire, 1982) Language shift may also be produced as the stabilization of an urban situation in circumstances where a central language (i.e., Portuguese) and a peripheral language (i.e., Tukano) co-exist.

The third and last point concerns the important changes in the relations of power since 1990 at the municipal level. A clear empowerment of native Brazilian Indians has taken place, especially after the creation, in 1984, of the Federation of Indian Organizations of the Rio Negro (FOIRN), an entity comprising 36 Indian organizations which has the support of various national and international NGOs. At the same time, the role of missionaries has lost much of its importance, particularly those belonging to the Society of St. Francis de Sales, who dominated the political scene along the Rio Negro for most of the 20(th) century.

The municipality's educational policy, especially the creation of a preparatory course for 200 bilingual Indian teachers for the municipal schools, has favored the positioning of this group. This is particularly the case in terms of political and linguistic matters connected to pedagogical practices. For example, teaching reading and writing not in Portuguese but in Indian languages is currently underway in various communities. The 165 municipal schools, formerly known as "rural schools," have been officially recognized as Indian schools after the creation, in 1999, of the Indian educational subsystem (in the same year of the creation of the Rio Negro Indian Teachers Association). In January 2000, a FOIRN general assembly convening 196 representatives from 36 organizations released a call for legal, political, and linguistic consultation. Their aim was to assess the chances of a movement demanding that the three Indian languages of wider regional communication -- Nheengatu, Tukano, and Baniwa -- be made official at the municipal level, an unprecedented move in Brazilian history.

These developments and similar initiatives suggest new possibilities for the maintenance of indigenous languages in the town of S...o Gabriel da Cachoeira. They also signal the emergence of new urbanization models for minority languages in Brazil, at least in those municipalities with a predominantly native Brazilian Indian population.

In other cases of small minorities living in the cities (these cases constitute the great majority), the possibility exists for the development, to use De Certeau's terms, of tactical rather than strategic language policies. (Conditionally, these minorities have a small chance of constituting their own circumscribed area within the town/city/municipality limits).

By recognizing the right of Indian nations to their languages in 1988, and by granting native Brazilian Indians the right to formulate their own educational policies, the Brazilian state has, within certain limits, established teachers, communities, associations and other entities as agents of language policies. It has, therefore, transferred to them responsibilities which are, in other models, entirely governmental obligations. The visibility of urban native Brazilian Indians and their capacity for articulation with regional public forces and with the population of the villages will allow them to take over planning tasks which, in turn, may enable them to expand their small numbers. This process will help them maintain their ethnic borders, an important strategy for the political and economic survival of native Indians in their relationship with the "national society." Town and city become new Indian territory. The occupation and maintenance of this territory may prove to be a decisive factor in the survival of Indian languages in Brazil.

(1). Brazil's official census bureau.

(2). For the sake of comparison, 62 million Brazilians are of African descent (household sample census, 1996). This is the largest population of those with African backgrounds outside of Africa. Although these 62 million African Brazilian people speak some variety of Portuguese (i.e., they do not speak any African languages), the native Brazilian Indians speak about 170 different languages.

(3). According to IBGE census data, in the state of Goiás alone there are 2,400 native Brazilian Indians living in towns, in contrast with only 203 living in three Indian villages. The urban Indian population is tenfold the rural Indian population in that state -- a situation usually inverted in other states. In Manaus, the capital of the state of Amazonas, the figures vary from 10,000 to 30,000, according to the Catholic Church's Indian Missionary Council (CIMI). In Campo Grande, the capital of the state of Mato Grosso do Sul, there are about 5,000 urban Terena Indians. Boa Vista, the state capital of Roraima, has an Indian population of about 12,000, who are mostly Macuxi and Wapixana. In the city of S...o Paulo, there are Guarani Indians, but also Pankararu Indians from the Brazilian northeast, who are thousands of miles away from their original communities. Apart from these populations living in large cities, it is important to consider the Indian communities in small and medium-sized towns throughout the country.

(4). Arapaso, Baniwa, Baré, Dow, Desana, Hupda, Kubeo, Kuripako, Nadöb, Piratapuia, Siriano, Tariana, Tukano, Tuyuka, Wanana, Werekena, Yanomami, Yuhupda and Nheengatu. FOIRN/ISA Mapa-livro Povos Indígenas do alto e médio Rio Negro. Brasília, MEC/SEF, 1998.

(5). The theory that Tukano is expanding in the town itself is corroborated by a brief survey carried out in a new neighborhood of S...o Gabriel da Cachoeira in July, 2000, ten years after Renault-Lescure's research.

References & further reading

Bessa Freire, J. (1983). Da "fala boa" ao português na Amazínia Brasileira. Amerôndia 8.

Calvet, L.-J. (1994). Les voix de la ville. Introduction à la sociolinguistique urbaine. Paris: Éditions Payot e Rivages.

Calvet, L.-J. (1999). Pour une écologie des langues du monde. Paris: Plon.

Crystal, D. (2000). Language death. Cambridge: Cambridge University Press.

FUNAI quer terra do INCRA para nova casa do índio (1979, Aug 16). A Notícia. Manaus.

Leonardi, V. (1996). Entre árvores e esquecimentos: história social no sert...o do Brasil. Brasília: UnB/Paralelo 15.

Renault-Lescure, O. (1990). As línguas faladas pelas crianças do Rio Negro (Amazonas): descontinuidade na transmiss...o familiar das línguas. In Crianças na Amazônia: um futuro ameaçado. Franco, H.B.F., Leal, M.F.M., Eds. Belém: Universidade Federal do Pará/UNICEF. Pp 315-324.

Article copyright Cultural Survival, Inc.

Our website houses close to five decades of content and publishing. Any content older than 10 years is archival and Cultural Survival does not necessarily agree with the content and word choice today.