Columbia University’s Bahian Legacy

by Conrad Phillip Kottak

Professor and Chair, Department of Anthropology

The University of Michigan

Ann Arbor

 

            With Thales de Azevedo, Charles Wagley directed the Bahia-Columbia University Community Study Project in 1951-52.  This pioneering comparative research effort culminated in Wagley's edited book Race and Class in Rural Brazil (1952, 1964), the first modern anthropological look at race relations in Brazil.  Other participants in the project included Marvin Harris (Minas Velhas–Rio de Contas), Harry W. (“Bill”) Hutchinson (Sao Francisco do Conde), and Ben Zimmerman (Monte Santo).

            Wagley's longstanding interest in race, class, and ethnicity is also illustrated by Minorities in the New World: Six Case Studies (1958), which he wrote with Harris.  The Latin American Tradition: Essays on the Unity and Diversity of Latin American Culture (1968) brought together several of Wagley's papers on traditional and modern Latin America, including an influential paper on social race. In that paper, “The Concept of Social Race in the Americas,” Wagley (1959/1968) argued convincingly that races are culturally constructed categories that may have little to do with actual biological differences. In Wagley’s terms social races are groups assumed to have a biological basis but actually defined in a culturally arbitrary, rather than a scientific, manner.

            Wagley (who died in 1991) built on a legacy of research on race, ethnicity, and social change pioneered by his professors at Columbia (from which he received his Ph.D. in 1941), especially Franz Boas and Ruth Benedict.  After working as an instructor at Columbia for a year (1940-41), Wagley traveled to Brazil to do research and, eventually, applied anthropology (in public health and sanitation--for the Brazilian and American governments during World War II). ^[1]

            In 1946 Wagley returned to Columbia, where he taught from 1946 to 1971.  He directed Columbia's Institute of Latin American Studies between 1961 and 1969 and served as the Franz Boas Professor of Anthropology from 1965 to 1971, when he moved to Gainesville, Florida as a Graduate Research Professor.  Wagley retired from the University of Florida in 1983.

            Wagley's study of Itá, a community of peasant farmers and rubber tappers on the Amazon, began in 1948 and produced two editions of his book Amazon Town: a Study of Man in the Tropics (1953/1964).  It also provided a fourth case study (along with the three Bahian towns studied as part of the UNESCO project) for Race and Class in Rural Brazil.

            Wagley chaired more than fifty doctoral dissertations at Columbia and Florida, nurturing, guiding, and inspiring some of today's most prominent anthropologists--American and Brazilian, including Marvin Harris.  Several of his students, and their students (such as this author), have benefited from the legacy of Wagley’s interest in race, ethnicity, and social change.

 

Marvin Harris, Minas Velhas, and Patterns of Race

            In the United States and internationally, Marvin Harris, who died in October, 2001, is mainly known as a theoretician.  However, students of Brazil are familiar with his ethnographic study of Minas Velhas in Bahia and his studies of Brazilian racial classification.  Harris’s field work in Minas Velhas (part of the UNESCO project) is the basis of his Town and Country in Brazil (1956) and of his chapter in Race and Class in Rural Brazil.  Wagley was Harris’s dissertation chair and long-time collaborator and associate at Columbia and Florida, where Wagley was also Graduate Research Professor.  Together they wrote Minorities in the New World (1958).

            Harris’s work on race receives its fullest treatment in Patterns of Race in the Americas (1964), a systematic comparison, using a cultural materialist framework, of the divergent racial patterns that emerged in Brazil, the United States, the Caribbean, and highland Latin America.  In that book Harris took particular issue with “cultural heritage” and national character explanations of racial patterns, particularly those advanced by the historian Frank Tannenbaum for the Caribbean and by the Brazilian social theorist Gilberto Freyre for Brazil.  Freyre had stressed the role of Portuguese national character in forming Brazilian race relations, and indeed in creating a “new world in the tropics,” based on a penchant for racial tolerance and mixture–what Freyre called mestiçagem.  In Patterns of Race in the Americas Harris argued persuasively for the role of material conditions in forming the patterns of race in different parts of the Americas.  He also took issue with Freyre’s contention that slaves received more humane treatment in Brazil than in the United States, supposedly because of differences in Portuguese versus English national character and attitudes toward non-Europeans.  Both in Town and Country in Brazil and in Patterns of Race in the Americas Harris confronted the harsher dimensions of Brazilian race relations.  In Town and Country and Race and Class in Rural Brazil, while providing a vivid description of racial prejudice in Minas Velhas, Harris also showed that prejudice did not necessarily translate into systematic discrimination.  Again he took issue with the notion that attitudes and temperament are the best predictors of behavior.

            Harris is also well known for his work on Brazilian racial classification, especially his research on the multiple racial categories in use throughout Brazil and their relation to the categories used in the Brazilian census.  The last fieldwork he conducted, in the early 1990s, took him back to Minas Velhas, where he worked with the Brazilian social anthropologist Josildeth Consorte, who had assisted him with his previous fieldwork there.  Harris took issue with the four official racial terms used in the Brazilian census: branco (“white”), pardo (“grey”), preto (“black”), and amarelo (“yellow”).  Harris recognized that the term “pardo” was not in common use; its use in the census, he claimed and demonstrated, led to an overestimation of the number of whites in the Brazilian population and an underestimation of the number of mixed Brazilians.  Harris noted the much more common use among ordinary Brazilians of the terms mulato and especially moreno.  In his last field work in Minas Velhas, Harris and his associates conducted an experiment, operating like census takers and using random samples of residents (see Harris, Consorte, Lang, and Byrne 1993).  One sample of Minas Velhas residents was asked to self identify with reference to the four terms used in the official census.  For the other sample, the common term moreno replaced the official term pardo.  Harris found that when given the option of choosing moreno rather than pardo, many more Brazilians classified themselves as mixed race (moreno), and the number of self-identified whites (brancos) fell.  Harris hoped to convince sociologists and others who routinely make use of Brazilian census data of its serious overestimation of the white, and underestimation of the mixed, segments of the national population. 

            Writing with his student Conrad Kottak (1963) Harris coined the term hypodescent to contrast American and Brazilian racial classification.  With hypodescent, mixed children (e.g., those from a union between an African American and a Euro-American) are always assigned to the minority category.  Hypodescent did not operate in Brazil, where racial classification was based more on phenotype and social perceptions, and where full siblings could be classified as members of different social races.

            In the United States, according to the hypodescent rule, one acquires his or her racial identity at birth, but race isn’t based on biology or on simple ancestry. In the case of the child of a “racially mixed” marriage involving one black and one white parent, even though it is known that 50 percent of the child’s genes come from one parent and 50 percent from the other, American culture overlooks heredity and arbitrarily classifies the child as black. American rules for assigning racial status can be even more arbitrary. In some states, anyone known to have any black ancestor, no matter how remote, is classified as a member of the black race.

            The system Brazilians use to classify biological differences contrasts with the American system, but recalls classificatory systems of other Latin American countries.  Brazilians use many more racial labels (over 500 have been reported [Harris 1970]) than North Americans do. In northeastern Brazil, I found 40 different racial terms in use in Arembepe, then a village of only 750 people (Kottak 1999). Through their classification system, Brazilians recognize and attempt to describe the physical variation that exists in their population. The system used in the United States, by recognizing so few races, blinds North Americans to an equivalent range of evident physical contrasts. Although there is evidence for recent reduction and simplification in Brazilian racial terminology, the hypodescent rule remains mainly an American peculiarity. 

 

The Story of Arembepe: The Legacy of the Columbia-Cornell-Harvard-Illinois Summer Field Studies Program, in Anthropology

 

            Conrad Kottak and Isabel (Betty) Wagley Kottak began field work in Arembepe, an Atlantic coastal community in Bahia, in 1962.  We married in 1963 and returned to Arembepe several times: in 1964, 1965, 1973, 1980, annually between 1982 and 1987, again in 1991, 1992, 1994, and most recently in October-November 2003. 

            Conrad Kottak’s (my) field work in Arembepe began not as problem-focused ethnography but as an undergraduate experience in the now defunct Columbia-Cornell-Harvard-Illinois Summer Field Studies Program in Anthropology. ^[2]   During that time (June-August 1962), under the guidance of Marvin Harris, I investigated racial classification in Arembepe.  One of the fruits of that experience was the hypodescent paper mentioned previously.  My research in Arembepe has continued because of the fascinating process of change that has taken place there.  Arembepe's radical social and economic transformation serve as subject matter for my book Assault on Paradise: Social Change in a Brazilian Village (3^rd ed., 1999).  (I summarize the key elements of that transformation below.)

            Publication of the first edition of Assault on Paradise followed an 1980 revisit to Arembepe after a seven year absence.   By that time, through the mass media and direct contact with outsiders, Arembepe clearly participated in a global process of cultural exchange.  Mules and donkeys had become rarer than automobiles.  Television antennas bedecked even modest homes (See Kottak 1990).  Telephones were about to arrive.  These developments mirrored what was happening throughout Brazil.  Nationally, the percentage of households with TV sets had increased from seven percent to fifty-one percent between 1964 and 1979.  With the arrival of electricity villagers now enjoyed the advantages of water pumps, refrigerators, and freezers.  Local people were clearly attracted by a newly available inventory of consumer goods.  Future archeologists excavating Arembepe of 1980-81 could uncover hundreds of different products designed and marketed by corporations based thousands of miles away.

            Let me highlight the main features of the change process experienced by Arembepe.  In the 1960s the trip to Arembepe from Salvador took three hours of travel on dirt and sand roads in a vehicle with four-wheel drive.  The local population was 750 people, living in 160 houses.  Fishing was the mainstay of the economy.  Most men fished for subsistence and cash.  Arembepe's most regular visitors were fish buyers from Salvador.  The fleet was not motorized.  Fishermen sailed to the nearby continental slope, where they specialized in migratory species (most notably olho de boi–the horse-eyed bonito).

            In the early 1960s Arembepe’s local economy supported little socioeconomic differentiation.  Besides fishing, villagers grew and sold coconuts, ran small stores and sold low value items from their homes.  Except in storekeeping, women had few opportunities to make money.  Arembepe's demographic profile included more males than females in the younger age categories.  In part this imbalance reflected neglect of female children and girls' lower survival chances.

            Despite an evident gender stratification, an ideology of socioeconomic equality prevailed, reflecting the reality that everyone in Arembepe belonged to the national lower class.  "We're all equal here," said villagers.  "No one is really rich."  Sailboats and fishing equipment were inexpensive and available to any industrious fisherman.  A fully equipped boat cost the equivalent of 400 kilograms of marketed fish.  Since boats rarely lasted a decade, few were inherited.  Land holdings were meager, produced little cash, and were fragmented through inheritance.  Any ambitious villager could find land to plant coconut trees, which supplied Arembepe's second export.

            When I revisited Arembepe in 1973, after an eight year absence, these characteristics were in flux.  By 1980, when I next returned, major and dramatic transformations were evident.  Three economic changes had enmeshed Arembepe much more strongly in the Brazilian nation and the world capitalist economy: 

            (1)       changes in the fishing industry, from wind power to motors;

            (2)       opening of a paved highway and the rise of tourism, attracting people from all over the world;

            (3)       construction of a nearby factory and resulting chemical pollution of Arembepe's waters.

            By 1980, growing wealth disparities were evident.  The poorest villagers were getting poorer; and the rich, dramatically richer.  Nowhere were the changes clearer than in the fishing industry.  Fishermen were getting fewer fish per day's labor than they had in the 1960s, while boat owners were drawing ten times their previous profits.

            Arembepeiros had begun motorizing their boats during the early 1970s, with loans from the government agency (SUDEPE) charged with developing small-scale fishing.  The agency loaned money to successful captains, owners, and land-based entrepreneurs.  However, young industrious fishermen, who in the past would eventually have been able to buy their own boats, lacked sufficient collateral to get a loan.  Nor was it possible (as it had been in the past) to accumulate enough money to buy a (motor) boat through their own fishing efforts.

            Profits from motorized fishing were reinvested in more costly fishing technology, including larger and much more expensive boats.  As the value of property increased, so did the owners' share of the catch. ^[3]   Social relations in the fishing industry grew less social, more economic.  Owners became bosses instead of co-workers.  Given their traditional ideology of equality, Arembepeiros resented these changes.  Many stopped fishing, but a swell of immigrants helped fill the void.

            The paved highway had done much to end Arembepe's isolation.  Its early completion, coinciding with the international hippie diaspora of 1969-1971, was assured by financial assistance from the owners of the chemical factory (Tibrás).  After that, a flood of tourists from Bahia joined the hippies and fueled a rise in property values and rents in Arembepe.

            The end of isolation transformed Arembepe's entire economy, bringing occupational plurality while changing the nature and role of fishing.  During the 1970s, Arembepe's economy diversified.  Fishing declined as the main local occupation. ^[4]   Many young men ^[5] found jobs at the nearby titanium dioxide factory (Tibrás), built by a multinational corporation (Bayer) based in Germany.  By 1980 fourteen per cent of male, and thirty-one per cent of female, cash earners worked in business.  Many of them catered to the weekend and summertime tourist trade that had developed because of the highway.

            The new economy promoted general socioeconomic stratification while reducing gender stratification.  It offered women new chances to make money in sales, services, and rents.  Female status rose as access to resources by women and men became more equal.  Like fishermen, female cash earners acquired pension rights from the government.  Women became less dependent on men for support.

            As the economy grew more complex, so did the local social structure.  Arembepe was now divided by social class, occupation, neighborhood, place of origin, and religion (Catholicism, fundamentalist Protestantism, and Afro-Brazilian candomblé).  Several factors created new divisions in this once fairly homogeneous and egalitarian community.  For example, people who had moved to new satellite villages and neighborhoods were no longer considered to be Arembepeiros.  Occupational diversity also meant different activities and associations for villagers.  And many kinds of outsiders played regular roles in local life.

            By the mid-1970s the stage had been set for the transformation evident in Arembepe in the 1980s—a process that continues today.  The qualitative change had occurred by 1980.  Since then, and through the present, there has been quantitative change--e.g., more people, more outsiders, more Protestants, growth of settlements--but no new qualitative transformation.  Nowadays change in Arembepe is of degree rather than of kind.

 

The Value of Teamwork

 

            Arembepe has become a longitudinal field site, not just for the Kottaks but for several others. Generations of researchers have monitored various aspects of change and development there.  Brazilian and American researchers worked with us on team research projects during the 1980s (on television’s impact–see Kottak 1990) and the 1990s (on ecological awareness and environmental risk perception–see Kottak and Costa 1993; Kottak, Costa, Prado, and Stiles 1995; Kottak, Costa, and Prado 1997).

            Graduate students from the University of Michigan have drawn on our baseline information from the 1960s as they have studied various topics in Arembepe. In 1990, Doug Jones, a Michigan student doing biocultural research, used Arembepe as a field site to investigate standards of physical attractiveness. In 1996–97, Janet Dunn studied family planning and changing female reproductive strategies. Chris O’Leary, who first visited Arembepe in summer 1997, has investigated a striking aspect of religious change in Arembepe—the spread of Protestantism. Later he did a study of changing food preferences. 

            Arembepe, like Bahia more generally, is thus a site where many field workers have worked as members of a longitudinal team. The more recent researchers have built on prior contacts and findings to increase knowledge about how local people meet and manage new circumstances. I learned from Wagley and Harris that scholarship should be a community enterprise. The information we gathered in the past is there for new generations to use. Thus, to monitor changing attitudes and to understand the relation between television and family planning, Janet Dunn reinterviewed many of the women we had interviewed in the 1980s. Similarly, Chris O’Leary, who compared food habits and nutritional status in Arembepe and another Brazilian town, had access to dietary information from our 1964 interviews.

            Charles Wagley and Thales de Azevedo encouraged team research and international collaboration through the UNESCO project.  The need for such a collaborative model is even more apparent today.  Contemporary forces of change are too pervasive and complex to be understood fully by “the lone ethnographer”— a researcher who starts from scratch and works alone, for a limited period of time, and who views his or her field site as relatively discrete and isolated.  No longer can any ethnographer imagine that his or her field site represents some sort of pristine or autonomous entity. Nor should the ethnographer assume that he or she has exclusive (owner’s) rights to the site, or even to the data gathered there. That information, after all, has been produced in friendship, cooperation, and consultation with local people. More and more anthropological field sites have been restudied.  Ideally, later ethnographers collaborate with and build on the work of their predecessors. Compared with the lone ethnographer model, team work across time (as in Arembepe) and space (as in the comparative studies encouraged by UNESCO in various Brazilian towns) produces better understanding of cultural change and transformation. 

 

                                                              References Cited

 

Harris, M.

1952   Race Relations in Minas Velhas.  In Race and Class in Rural Brazil, C. W. Wagley, ed.  Paris: UNESCO. 

1956   Town and Country in Brazil.  New York: Columbia University Press.

1964   Patterns of Race in the Americas.  New York: Walker and Company. 

1970   Referential Ambiguity in the Calculus of Brazilian Racial Identity.  Southwestern Journal of Anthropology 26(1):1-14.

 

Harris, Marvin, J. G. Consorte, J. Lang, and B. Byrne

1993   Who Are the Whites?: Emics and Etics of the Racial Demography of Brazil.  Social Forces 72: 451-62.

 

Harris, M. and C. P. Kottak

1962   The Structural Significance of Brazilian Racial Categories (with C. P. Kottak).  Sociologia 25:203-209,

 

Harris, Marvin, and C. W. Wagley

1958   Minorities in the New World (with C. W. Wagley) New York: Columbia University Press.

 

Kottak, C. P.

1990   Prime-Time Society: An Anthropological Analysis of Television and Culture.  Belmont, CA: Wadsworth.

1999   Assault on Paradise: Social Change in a Brazilian Village, 3^rd ed.  New York: McGraw-Hill.

 

Kottak, C. P., and A. C. G. Costa

1993   "Ecological Awareness, Environmentalist Action, and International Conservation Strategy," Human Organization 52(4):335-343.

 

Kottak, C. P., A. C. G. Costa, and R. M. Prado

1997   "The Sociopolitical Context of Participatory Development in Northeastern Brazil." Human Organization.  (56,2):138-152.

 

Kottak, C. P., A. C. G. Costa, R. M. Prado, and J. Stiles

1995   "Environmental Awareness and Risk Perception in Brazil." Bulletin of the National Association of Practicing Anthropologists.  Issue entitled Global Ecosystems: Creating Options through Anthropological Perspectives, edited by Pamela J.  Puntenney, pp. 71-87.  Washington: American Anthropological Association.

 

Wagley, C. W.

1964   (orig. 1953) Amazon Town: A Study of Man in the Tropics. New York: Alfred A. Knopf.  

1968   (orig. 1959). The Concept of Social Race in the Americas. In The Latin American Tradition, ed. C. Wagley, pp. 155–174. New York: Columbia University Press.

 

Wagley, C. W., ed.

1953   Race and Class in Rural Brazil.  Paris: UNESCO.

1968   The Latin American Tradition.  New York: Columbia University Press.