Before they were largely decimated and dispersed by the effects of European colonization, Arawak-speaking peoples were the most widespread language family in Latin America and the Caribbean, and they were the first people Columbus encountered in the Americas. Comparative Arawakan Histories, in paperback for the first time, examines social structures, political hierarchies, rituals, religious movements, gender relations, and linguistic variations through historical perspectives to document sociocultural diversity across the diffused Arawakan diaspora.
Comparative Arawakan Histories
Rethinking Language Family and Culture Area in AmazoniaBy JONATHAN D. HILL FERNANDO SANTOS-GRANEROUniversity of Illinois Press
Copyright © 2002 Board of Trustees of the University of Illinois
All right reserved.ISBN: 978-0-252-02758-1Contents
Acknowledgments..............................................................................................................................................................................................viiIntroduction JONATHAN D. HILL AND FERNANDO SANTOS-GRANERO....................................................................................................................................................11. The Arawakan Matrix: Ethos, Language, and History in Native South America FERNANDO SANTOS-GRANERO........................................................................................................252. Arawak Linguistic and Cultural Identity through Time: Contact, Colonialism, and Creolization NEIL L. WHITEHEAD...........................................................................................513. Historical Linguistics and Its Contribution to Improving the Knowledge of Arawak SIDNEY DA SILVA FACUNDES................................................................................................744. Rethinking the Arawakan Diaspora: Hierarchy, Regionality, and the Amazonian Formative MICHAEL J. HECKENBERGER............................................................................................995. Social Forms and Regressive History: From the Campa Cluster to the Mojos and from the Mojos to the Landscaping Terrace-Builders of the Bolivian Savanna FRANCE-MARIE RENARD-CASEVITZ.....................1236. Piro, Apurinã, and Campa: Social Dissimilation and Assimilation as Historical Processes in Southwestern Amazonia PETER GOW...........................................................................1477. Both Omphalos and Margin: On How the Pa'ikwené (Palikur) See Themselves to Be at the Center and on the Edge at the Same Time ALAN PASSES.............................................................1718. A New Model of the Northern Arawakan Expansion ALBERTA ZUCCHI............................................................................................................................................1999. Shamanism, Colonialism, and the Wild Woman: Fertility Cultism and Historical Dynamics in the Upper Rio Negro Region JONATHAN D. HILL......................................................................22310. Secret Religious Cults and Political Leadership: Multiethnic Confederacies from Northwestern Amazonia SILVIA M. VIDAL....................................................................................24811. Prophetic Traditions among the Baniwa and Other Arawakan Peoples of the Northwest Amazon ROBIN M. WRIGHT.................................................................................................269References Cited.............................................................................................................................................................................................295Contributors.................................................................................................................................................................................................327Index........................................................................................................................................................................................................331
Chapter One
The Arawakan Matrix: Ethos, Language, and History in Native South America
FERNANDO SANTOS-GRANERO
The relationship between language and culture has been the subject of much speculation in Western philosophy and social sciences. In the recent past, the tendency has been to contest the one language–one culture hypothesis implicit in the writings of eighteenth-century German philosopher Johann G. Herder, an idea that, under several guises, dominated early anthropology and linguistics. In his 1769 essay "On the Origin of Language," Herder asserted that polities are unified neither by the acceptance of a common sovereign power, as proposed by Hobbes, nor by a social contract based on the general will, as advocated by Rousseau (Barnard 1969). Instead, he indicated, first, that the basis for the sense of collective political identity was the sharing of a common culture and, second, that the emergence and reproduction of a group's culture are based on the use of a common language. Herder referred to units possessing a common culture and language by the term Volk, or nationality. Members of such communities are united by the collective consciousness of a common cultural heritage. This consciousness, which distinguishes members of a collective from those of similar communities, is what Herder calls "national character."
Herder's propositions found their way into historical linguistics and through it to modernist ethnology. The detection of connections between Sanskrit, Persian, and European languages by Sir William Jones in 1786 and the discovery by Jacob Grimm of the existence of regularities of sound change between different but related languages established the framework for the emergence of comparative linguistics. The Neogrammarians, a group of scholars working in Leipzig, formulated the principles and methods of the new discipline in the late nineteenth century. Its basic premise is that languages that have a large number of cognate words descend from a common ancestral language and thus belong to the same family. Implicit in this proposition is the assumption that speakers of the protolanguage constituted a culturally unified community and that its descendants share at least some aspects of that culture.
This was Brinton's (1891) approach in The American Race, which provides the first modern linguistic classification of Native South American languages. In addition to associating language and culture—in Herder's tradition—the author posited a linkage between language and race. Brinton (1891, 57) suggested, "Similarity of idioms proves to some extent similarity of descent and similarity of psychic endowments." In an intellectual environment in which evolutionism was becoming a dominant mode of thought, it was only a short step from here to assert that certain combinations of language, culture, and race were superior or inferior to others. Anthropology was quick to reject this argument by refuting the assumption that there existed an immanent connection between language and culture or race and culture.
In The American Indian, Wissler (1917) asserted that there are "no correlations between culture and linguistic type." According to him, the fact that speakers of the same language stock are represented in several cultural areas demonstrates that "language can travel independently of culture" (Wissler 1917, 332). Edward Sapir (1921/1931, 143) went a step further, declaring that "races, languages and cultures are not distributed in parallel fashion, that their areas of distribution intercross in the most bewildering fashion, and that the history of each is apt to follow a distinctive course." Franz Boas (1928) was equally definitive. From then on it became almost anathema in anthropology to propose a connection between language and culture, except as formulated in the milder versions of the Sapir-Whorf hypothesis, which postulate that language influences but does not determine the way we think and perceive the world (see Sherzer 1992, 274).
The notion of language family, seen by Brinton (1891, 57) "as the only one of any scientific value," became outdated among anthropologists and was replaced in subsequent classifications of South American indigenous peoples by that of culture area. Julian H. Steward's (1946–59) Handbook of South American Indians is the most outstanding example of this new approach. He and Louis C. Faron (1959, 26) argued that although linguistic affiliation and distribution have certain implications for the cultural history of South American Indians, "there is no direct relationship between language groups and culture." The principal value of linguistic classifications and the comparative method was to supply, through the method known as glottochronology, information on the dates of divergence of languages belonging to the same family and thereby on possible migration routes.
However, even the scholars who most firmly opposed the idea that people belonging to the same language family shared a common cultural tradition had trouble accounting for the numerous ethnographic cases that did not fit their model (e.g., Wissler 1917, 333). In fact, to explain the relative cultural homogeneity that characterizes members of the most important Amazonian language stocks, despite their widespread geographic distribution, Steward and Faron (1959, 26) suggested that they must have separated quite recently and in fairly rapid waves of migration.
A certain ambiguity with respect to the sociological value of the notion of language family is also found in two collections devoted to the study of Carib-speaking societies. In the first, Ellen B. Basso (1977, 17) identifies eight material and nonmaterial cultural traits of what she considers to be a typical Carib complex. However, Basso warns the reader that these traits are not uniquely Carib and concludes that most Amazonian indigenous societies fall "into general social and cultural units that often encompass local groups of different language affiliation and history." In short, after a painstaking reconstruction of Carib culture, we are advised that it is more productive to think in terms of culture area than of language family.
This same stance is defended by Audrey Butt-Colson (1984a, 11) and, even more strongly, by Simone Dreyfus (1983–84, 39–40), who argues that what is perceived as "Carib unity" appears to be based more on the characteristics of culture areas—geographic, ecological, historical, and sociological—than on linguistic affiliation. From a slightly different perspective, it is also supported by Whitehead (1994,34; chapter 2),and by Urban and Sherzer (1988,297),who assert that to have a better understanding of the history of Amerindian cultures, researchers should pay more attention to cases of actual multilingualism and to the more-than-one-language-per-culture hypothesis.
In this chapter I assess the sociological significance of the notion of language family by analyzing "Arawak unity," which has been the object of little attention from specialists. Based on the premise that ethnohistorical studies can be meaningful only if one adopts an interethnic perspective (Santos-Granero 1995), I examine the issue of "Arawakanness" by comparing Arawakan cultural and social organization with that of their most meaningful counterparts in three tropical regions: eastern Peru (Panos), northwestern Amazonia (Tukanos), and northeastern South America (Caribs). I argue that there is a connection between language and culture expressed in the fact that peoples belonging to the same language family share a common cultural matrix and a certain ethos.
In so doing, it is not my intention to perpetuate early colonial stereotypes or subsequent ethnographic accounts depicting the Arawak as "gentle, hospitable and cultured" pioneers, in contrast with other peoples, such as the Carib, characterized as bellicose "adventurers, cannibals and colonists" (Radin 1946, 23, 25, 49). It is not my intention to uphold the notion that such a connection is in any way immanent or essential. The dialectical relationship between language and culture is historical and can be understood only through history. Thus, together with determining what are the central features of the Arawakan ethos, I examine situations of interethnic cultural influence and exchange. I pay particular attention to the emergence of what I call transethnic identities, that is, groups that adopt the cultural ethos of another language stock but retain their language or, conversely, groups that adopt a different language but retain their ethos. It is in the boundaries between peoples of different linguistic affiliation that we may observe the intricate ways in which ethos, language, and history combine in the negotiation of ethnic identity, sometimes reinforcing existing ones and sometimes promoting the formation of new identities.
The Arawak-Pano Cluster of Eastern Peru
When the Spanish entered into the tropical forest region east of the Peruvian central Andes in the second half of the sixteenth century, they encountered two different types of indigenous societies. Along the Upper Ucayali and Lower Urubamba rivers they found two bellicose peoples—probably the Pano-speaking Conibo and the Arawak-speaking Piro—living in large settlements under the rule of powerful war leaders (Alès 1981, 93, 95). In contrast, the Arawak-speaking peoples living along the eastern slopes of the Andes lived in small, scattered settlements led by headmen with little political power (Santos-Granero 1980, 29; Renard-Casevitz 1981, 130). Because of the high uniformity of their cultural practices, these hinterland peoples came to be known collectively as Chunchos or Antis in the sixteenth century and as Campa in the seventeenth century. In the early 1900s, linguists began to refer to them as the pre-Andine Arawak (Rivet and Tastevin 1919–24).
In the second half of the seventeenth century, when Franciscan and Jesuit missionaries started working in the area, the Conibo and Piro monopolized trade along the Ucayali-Urubamba axis by means of large armed fluvial expeditions. These two dominant peoples are described as engaged in constant war with each other and with their weaker semiriverine or interfluvial neighbors: the Pano-speaking Shipibo, Setebo, Cashibo, Amahuaca, Mochobo, Comabo, Remo, and Sensi, who lived along the tributaries of the Lower Urubamba and the Ucayali River, and the Arawak-speaking peoples who inhabited the vast area along the Andean piedmont (see map 5, p. 10). The Spanish divided these latter peoples into two large nations: the Campa (the present-day Asháninka, Ashéninka, Matsiguenga, and Nomatsiguenga) and the Amage (the Amuesha, known at present as Yanesha). For reasons of simplicity, from here onward I shall refer to all of these peoples as Campa, except when it is necessary to distinguish them from the Yanesha.
Pano warfare against the Arawak had a long tradition. Archaeological evidence and oral tradition suggest that the Arawak were gradually pushed against the Andean range by successive waves of Pano-speaking peoples who settled in the Ucayali River (Lathrap 1970a; Santos-Granero 1998). The Piro (Yine), who had reached the Urubamba River from the east (see chapter 6), also pushed the Arawak in a westward direction.
In addition, Pano-speaking groups were in constant war with each other. In general terms, the Conibo raided the semiriverine Pano groups, while all of them assailed the more interior, headwater groups; these, in turn, were continually raiding each other (Frank 1994; Erikson 1994; Townsley 1994; Dole 2000; Morin 2000). Pano endo-warfare, under the guise of blood feuds, slave raids, and pillaging, increased in colonial times as each group attempted to monopolize trade in tools with the Spanish (Frank 1994, 148).
Conibo raiders killed adult men and elderly people who could not work as servants and took with them their women and children (Biedma 1981, 95; Huerta 1983, 123–24). They also plundered their villages, stealing all valuable goods, such as spun cotton, cotton textiles, feather ornaments, and salt. Enemies were beheaded, their heads ritually insulted, and their blood drunk, mixed with manioc beer (Amich 1975, 108). As a sign of their courage, victorious warriors hung the heads and dried hearts of their enemies from the rafters of their houses (Biedma 1981, 95). Captive women were taken as co-wives or servants. As a result, most Conibo men had two or more wives (Beraún 1981, 181). Captured children had to work for their captors but were raised as Conibo and eventually married into the group (Amich 1975, 93). The Conibo came to depend so much on pillage that their women no longer spun or wove cotton. Conibo pillaging and raiding created many enemies for them. To enhance their capacity for defense, allied Conibo groups gathered in large settlements of up to 2,000 people under the joint rule of their respective leaders (Amich 1975, 93).
The interfluvial Arawak presented a stark contrast with their Pano neighbors. As France-Marie Renard-Casevitz (1985, 1993) has pointed out, one of the outstanding features of pre-Andine Arawak social organization was the absence of endo-warfare. The only exceptions were the Piro, described in more detail later in this chapter. This in no way means that conflicts, hostility, and violence had been eradicated from Campa society but, rather, that aggression was canalized through intrafamily or intracommunity individual violence or through exo-warfare (Renard-Casevitz 1985, 90). When conflicts at the local level could not be resolved peacefully, they usually led to sorcery accusations and even to murder. However, these killings involved individuals rather than the social groups to which they belonged and never gave rise to intratribal cycles of vendetta. The only context in which intratribal attacks were admitted was in case of betrayal of common "tribal" interests, such as support of foreign invaders when this was not a collectively approved political strategy (see Biedma 1981, 162; Amich 1975, 73).
The rejection of endo-warfare among the Campa must not be taken as an expression of pacifism or a peaceful disposition. With the exceptions of the Yanesha and the Matsiguenga (Rosengren 1987), Arawak leaders derived much of their authority from their capacity as warriors. This was especially true of the Campa peoples of the Apurimac, Ene, Tambo, Ucayali, and Gran Pajonal areas, whose territories bordered on those of the Piro and Conibo. Although weaker in military capacity than their riverine enemies, Campa leaders not only defended their communities from enemy attacks but also waged an active war against them (Steward 1946–59, 3:537). However, the Campa shared none of the ritual war practices characteristic of the endo-warring Pano, and there is no evidence that they took war captives to transform them into wives or servants. This took place much later on, during the late nineteenth century, as a result of the demands of powerful rubber extractors.
Renard-Casevitz (1993, 32) explains Arawak suppression of endo-warfare as resulting from the need to present a common front vis-à-vis constant attempts by Andean state societies to subjugate them. However, the fact that other peoples facing similar threats—such as the Jivaro of eastern Peru and Ecuador—did not follow the same strategy suggests that the suppression of endo-warfare must be attributed to other factors. Risk of revenge feuding and slave raids was avoided among Campa groups by promoting regular dialogue between local leaders, intersettlement quarantine in case of conflicts, or, in extreme cases, geographic distancing (Renard-Casevitz 1993, 33). A concentric model of social organization, based on ever-increasing spheres of solidarity in which each local settlement created its own social network through marriage ties, residence rules, ritual gatherings, commerce, and political alliances, generated a dense web of relations of exchange, stimulating greater social cohesion.
Particularly important in this context were salt extraction and trade centered on the Cerro de la Sal, which each year, during the dry season, promoted the peaceful communication of hundreds of peoples belonging to the diverse groups that composed the Arawakan cluster (Tibesar 1950, 108). Central features of this kind of trade were formal trading friendships, which could be established only between Arawak speakers and were based on deferred exchange (Bodley 1984, 49; Schäfer 1991). After the expulsion of the Spanish in 1742, trade between Arawak-speaking groups was further enhanced by the emergence of numerous temples that functioned simultaneously as forges in the Upper Perené region (Santos-Granero 1987). Yanesha and Campa ironworks became the center of an active trade in iron tools that overlapped with the more traditional salt trading network.
(Continues...)
Excerpted from Comparative Arawakan Historiesby JONATHAN D. HILL FERNANDO SANTOS-GRANERO Copyright © 2002 by Board of Trustees of the University of Illinois. Excerpted by permission of University of Illinois Press. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
