Mobility and Migration in Indigenous Amazonia: Contemporary Ethnoecological Perspectives

Editor's Preface

MIGUEL N. ALEXIADES

This book focuses on the relationship between mobility and migration and human–environment relations in indigenous and folk Amazonia. Most of the chapters are revised versions of papers presented at the International Society of Ethnobiology's Ninth International Congress, held at the University of Canterbury, Kent, UK, in June 2004. The motivation for putting together the panel – named ‘The Ethnobiology of Mobility, Displacement and Migration in Indigenous Lowland South America’ – and subsequently for compiling this volume grew out of several interests and concerns. The first of these, echoing the broader theme of the Congress – Ethnobiology, Social Change and Displacement – recognises mobility and migration as highly relevant, both theoretically and in more applied contexts, for all of us interested in a better understanding of how human societies perceive, experience and structure their symbolic and material interactions with the environment in an increasingly articulated, changing and contingent world.

A second, more specific, motivation for collating these papers emerged from a personal sense that a disjuncture of sorts exists within the field of ethnobiology and ethnoecology in Amazonia – and perhaps beyond. While human ecologists, ethnobotanists and Amazonianists in related fields have become increasingly sensitive to, and interested in, the historical contingency of nature and of human–environment interactions, and while there has been a simultaneous, generalised and growing interest in the processual aspects of ethnoecology,¹ there remains a strong residual tendency, often implicit, to view these largely as the product of long-term historical emplacement and spatial stasis. Even when spatial stasis is not assumed, it is still not given the same level of attention as other historical processes.

While the consequences of demographic collapse and integration into the state and market economy have drawn considerable attention, the other major impact of European colonisation – the spatial reorganisation of indigenous societies – has not been as extensively or systematically examined (cf. Taylor and Bell 2004: 1). Moreover, while there has been a considerable amount of discussion on the relationship between indigenous migration and changes in mobility and demography and health and subsistence (McNeill 1980; Kroeger and Barbira Freedman 1988; Roberts et al. 1992; Zent 1993; Hill and Hurtado 1995; Coimbra et al. 2002), relatively little attention has been directed at exploring the links with other aspects of human–environment relations, including ethnobotanical and ethnoecological knowledge, the evolution, distribution and abundance of plant resources, or the symbolic representation of nature and landscapes (but see, for example, Cárdenas and Politis 2000; Little 2001; Posey 2002a; Rival 2002; Pieroni and Vandebroek 2007).

The time seems ripe, therefore, to pay closer attention to the links and relationships between movement, mobility, migration and displacement and indigenous and folk ethnoecology. Contributing to the growing literature on the historical and political ecology of Amazonia (Schmink and Wood 1992; Balée 1994; Fisher 2000; Little 2001; Posey 2002b; Heckenberger 2005) and drawing on a wide range of perspectives, theoretical approaches and disciplines – including social anthropology, historical ecology, geography, ethnobotany, botany and evolutionary biology – this collection of case studies from different parts of the Amazon and Orinoco river basins in Colombia, Ecuador, Brazil, Peru, Bolivia and Venezuela seeks to fill some of the existing gaps in our current understanding of the relationship between social change and ethnoecology in lowland South America. Rather than attempt to offer a systematic or comprehensive overview of what is an extremely broad and complex topic, this book aims to illustrate links, identify interesting patterns and trends, demonstrate a range of approaches and flag areas for future study. A substantial introduction is also included at the outset in the hope that it will provide a suitable background and help orient readers not familiar with some of the concepts, ideas or vast literature relevant to indigenous mobility and migration, and their relation to Amazonian history and ethnoecology.

As the title suggests, the outlook of this volume is premised on the logic and coherence of certain geographical, social and epistemological boundaries and categories. Amazonia is used here in its broadest sense, and taken to include the Orinoco basin, as well as adjacent coastal areas (including the Brazilian Mata Atlântica), and the Andean piedmont. While the clear-cut boundary between the Amazonian lowlands and the adjacent highlands is, as Dudley (this volume) underscores, a recent historical construct, and while the two regions have been thoroughly interconnected throughout much of their human history, the Andean highlands and Amazonia have experienced quite different historical trajectories and, though interconnected, are socially and ecologically quite distinct.

The scope of this book is not restricted to people who self-identify as indigenous, but also covers so-called ‘folk’ societies; social collectivities with relatively long histories of occupation in Amazonia. This includes the detribalised descendants of indigenous peoples, many of whom extensively intermarried with Europeans and Africans, forging an Amazonian peasantry widely referred to as ribereños or caboclos. It also includes the descendants of runaway slaves – known as quilombos or cimarrones in Brazil and parts of Latin America, respectively – and more generally Afro-indigenous groups, who form distinct ‘traditional’ communities throughout much of the lowland tropics of South America. Deserving of treatment, but not included in this volume due to necessary limitations in size and scope, are the recent migrant and settler communities, who comprise an important proportion of rural and urban Amazonia, and whose background and reality pose a different, though clearly not mutually exclusive or unrelated, set of questions and problems for ethnoecologists.²

I would like to thank the participants of the panel, including Edmond Dounias, who acted as discussant, as well as those who were not able to attend in person but who subsequently prepared a chapter for the volume. Their contributions and lively discussions, during and after the conference, have been most instructive. I am especially grateful to Roy Ellen for his help and support in putting together the panel and for his subsequent editorial help, advice and patience. Manuel Arroyo-Kalin, Simone Athayde, Rodrigo Bernal, Charles Clement, Oliver Coomes, Meredith Dudley, Phillipe Erikson, Alf Hornborg, Egleé López, Manuel Macía, Helen Newing, Dario Novellino, Manuel Pardo, Elizabeth Reichel, Laura Rival, Daniel Rodriguez and Steven Rubinstein all provided very useful comments on different chapters, including the introduction. I am particularly grateful to Daniela Peluso for her help with many aspects of the book and to Jana Traboulsi for her assistance with several maps. The background research, preparation and editing of this book were made possible thanks to the support of the Nuffield Foundation (Advanced Career Fellowship 2002–2005), the Darrell Posey Foundation (Field Fellowship 2005–2007), the British Academy (Small Grant 35207) and the University of Kent (2003 Faculty Grant). Miguel Pinedo-Vásquez has kindly assisted me over the past years with access to several key sources and journals during this time. Last but not least, I am indebted to the Department of Anthropology at the University of Kent for offering their facilities, and to Michael Fisher for his advice on a wide range of issues relating to digital technology.

References

Balée, W. 1994. Footprints of the Forest: Ka'apor Ethnobotany – The Historical Ecology of Plant Utilization by an Amazonian People. New York: Columbia University Press.

Balée, W. and C.L. Erickson (eds) 2006. Time and Complexity in Historical Ecology. Studies in the Neotropical Lowlands. New York: Columbia University Press.

Barrera-Bassols, N. and V.M. Toledo 2005. ‘Ethnoecology of the Yucatec Maya: Symbolism, Knowledge and Management of Natural Resources’, Journal of Latin American Geography 4 (1): 9–41.

Berkes, F., J. Colding and C. Folke 2000. ‘Rediscovery of Traditional Ecological Knowledge as Adaptive Management’, Ecological Applications 10: 1251–62.

Cárdenas, D. and G. Politis 2000. Territorio, Movilidad, Etnobotánica y Manejo del Bosque de los Nukak Orientales, Amazonía Colombiana. Santafé de Bogota: Universidad de los Andes.

Coimbra, C.E.A.J., N.M. Flowers, F.M. Salzano and R.V. Santos 2002. The Xavánte in Transition: Health, Ecology, and Bioanthropology in Central Brazil. Ann Arbor: University of Michigan Press.

Ellen, R.F., P. Parkes and A. Bicker (eds) 2000. Indigenous Knowledge and its Transformations: Critical Anthropological Perspectives. Amsterdam: Harwood Academic Publishers.

Fisher, W.H. 2000. Rainforest Exchanges: Industry and Community on an Amazonian Frontier. Washington and London: Smithsonian Institution Press.

Goodman, D. and A. Hall (eds) 1990. The Future of Amazonia: Destruction or Sustainable Development? London: Macmillan.

Heckenberger, M.J. 2005. The Ecology of Power: Culture, Place, and Personhood in the Southern Amazon, A.D. 1000–2000. London: Routledge.

Heckler, S. 2002 ‘Traditional Ethnobotanical Knowledge Loss and Gender among the Piraoa’, in J.R. Stepp, F.S. Wyndham and R.K. Zarger (eds), Ethnobiology and Biocultural Diversity: Proceedings of the Seventh International Congress of Ethnobiology. Athens, GA: International Society of Ethnobiology/University of Georgia Press, pp. 532–48.

Hill, K. and A.M. Hurtado 1995. Ache Life History. The Ecology and Demography of a Foraging People. New York: Aldine.

Hunn, E.S. 2002. ‘Evidence for the Precocious Acquisition of Plant Knowledge by Zapotec Children’, in J.R. Stepp, F.S. Wyndham and R.K. Zarger (eds), Ethnobiology and Biocultural Diversity: Proceedings of the Seventh International Congress of Ethnobiology. Athens, GA: University of Georgia Press, pp. 604–13.

Kroeger, A. and F. Barbira Freedman 1988. ‘Cultural Change and Health: The Case of South American Rainforest Indians’, in J.H. Bodley (ed.), Tribal Peoples and Development Issues. A Global Overview. Mountain View, California: Mayfield Publishing Company, pp. 221–36.

Little, P. 2001. Amazonia: Territorial Struggles on Perennial Frontiers. London: Johns Hopkins University Press.

McNeill, W.H. 1980. ‘Migration Patterns and Infection in Traditional Societies’, in N.F. Stanley and R.A. Joske (eds), Changing Disease Patterns and Human Behaviour. London and New York: Academic Press, pp. 27–36.

Nazarea, V.D. 1999. ‘Introduction. A View from a Point: Ethnoecology as Situated Knowledge’, in V.D. Nazarea (ed.), Ethnoecology. Situated Knowledge/Located Lives. Tucson: University of Arizona Press, pp. 3–20.

Painter, M. and W.H. Durham (eds) 1995. The Social Causes of Environmental Destruction in Latin America. Ann Arbor, MI: University of Michigan Press.

Pieroni, A. and I. Vandebroek (eds) 2007. Traveling Cultures and Plants: The Ethnobiology and Ethnopharmacy of Human Migrations. New York and Oxford: Berghahn.

Posey, D.A. 2002a. ‘Environmental and Social Implications of Pre- and Postcontact Situations on Brazilian Indians’, in D.A. Posey, Kayapó Ethnoecology and Culture. Edited by K. Plenderleith. London and New York: Routledge Harwood Anthropology, pp. 25–33.

——— 2002b. Kayapó Ethnoecology and Culture. Edited by K. Plenderleith. London and New York: Routledge Harwood Anthropology.

Reyes-García, V., V. Vadez, E. Byron, L. Apaza, W.R. Leonard, E. Perez and D. Wilkie 2005. ‘Market Economy and the Loss of Folk Knowledge of Plant Uses: Estimates from the Tsimane’ of the Bolivian Amazon’, Current Anthropology 46 (4): 651–56.

Rival, L.M. 2002. Trekking Through History. The Huaorani of Amazonian Ecuador. New York: Columbia University Press.

Roberts, D.F., N. Fujiki and K. Torizuka (eds) 1992. Isolation, Migration and Health: 33rd Symposium Volume of the Society for the Study of Human Biology. New York: Cambridge University Press.

Ross, N. 2002. ‘Lacandon Maya Intergenerational Change and the Erosion of Folk Biological Knowledge’, in J.R. Stepp, F.S. Wyndham and R.K. Zarger (eds), Ethnobiology and Biocultural Diversity: Proceedings of the Seventh International Congress of Ethnobiology. Athens, GA: University of Georgia Press, pp. 585–92.

Schmink, M. and C.H. Wood 1992 (eds) 1984. Frontier Expansion in Amazonia. Gainesville: University of Florida Press.

——— 1992. Contested Frontiers in Amazonia. New York: Columbia University Press.

Schumann, D.A. and W.L. Partridge 1989. The Human Ecology of Tropical Land Settlement in Latin America. Boulder, CO: Westview Press.

Taylor, J. and M. Bell 2004. ‘Introduction. New World Demography’, in J. Taylor and M. Bell (eds), Population Mobility and Indigenous Peoples in Australasia and North America. London and New York: Routledge, pp. 1–10.

Zarger, R.K. 2002. ‘Acquisition and Transmission of Subsistence Knowledge by Q'eqchi’ Maya in Belize’, in J.R. Stepp, F.S. Wyndham and R.K. Zarger (eds), Ethnobiology and Biocultural Diversity: Proceedings of the Seventh International Congress of Ethnobiology. Athens, GA: University of Georgia Press, pp. 593–603.

Zent, S. 1993. ‘Donde no hay médico: Las consecuencias culturales y demográficas de la distribución desigual de los servicios médicos modernos entre los Piaroa’, Antropológica 79: 41–84.

——— 2001. ‘Acculturation and Ethnobotanical Knowledge Loss among the Piaroa of Venezuela: Demonstration of a Quantitative Method for the Empirical Study of Traditional Ecological Knowledge Change’, in L. Maffi (ed.), On Biocultural Diversity. Linking Language, Knowledge, and the Environment. Washington and London: Smithsonian Institution Press, pp. 190–211.

1. This interest in processual issues is exemplified by recent interest in questions relating to the dynamism, variability and transformation of ethnoecological systems. See, for example, Nazarea (1999), Berkes et al. (2000), Ellen et al. (2000), Zent (2001), Heckler (2002), Hunn (2002), Ross (2002), Zarger (2002), Barrera-Bassols and Toledo (2005), Reyes-García et al. (2005) and Balée and Erickson (2006).

2. For case studies discussing environmental and political issues relating to internal migration and colonisation in Amazonia, see, for example, Schmink and Wood (1984), Schumann and Partridge (1989), Goodman and Hall (1990) and Painter and Durham (1995).

CHAPTER 1

Mobility and Migration in Indigenous Amazonia: Contemporary Ethnoecological Perspectives – an Introduction

MIGUEL N. ALEXIADES

Introduction

A diverse group of scholars have, in recent decades, been challenging many of the stereotypes that have for a long time permeated scientific and public representations of Amazonia and its people. These new insights support the notion of Amazonian social and ecological systems as dialectically interrelated, and as more complex, diverse, historically contingent and dynamic across multiple spatial and temporal horizons, than was previously thought.¹

This heightened historical sensitivity, shared by human ecologists, ethnoecologists, ethnobotanists and others interested in human–environment relations, does not extend equally to all domains of socio-environmental experience, however. In particular, there seems to be a widespread residual tendency to view indigenous societies, and their associated ethnecologies, as historically emplaced: that is, as the product of a long history of engagement with particular locales. In other words, while many ethnoecologists and ethnobiologists are now comfortable with notions of social dynamism, and while some have incorporated these into their research agenda, there still exists an assumption, often implicit, of spatial stasis.

Despite important continuities in the spatial, symbolic and material histories of many Amazonian societies (Heckenberger 2005), the historical record points to a long history of movements, exchanges, displacements and changes in the spatial – and not just social – organisation of Amazonian societies. As Paul Little (2001: 4) notes, ‘the history of Amazonia is filled with examples of people in movement: nomadism, group migrations, long distance trade, explorations, forced dislocation, colonisation, labour migration’. Indeed, different kinds of spatio-social processes and movements are associated with each of the major phases in the historical development of Amazonia: the early settlement of Amazonia by Pleistocene hunters, the rise of complex agricultural societies and the development of pre-Columbian regional systems of exchange, their subsequent demise, dispersion and reorganisation following European conquest, and, finally, the late twentieth-century demographic, political and social resurgence of indigenous societies and their increased articulation with global networks of exchange.

Post-conquest demographic and political disruptions had a particularly dramatic effect on patterns of movement and settlement. By the nineteenth century, for example, the major tributaries of the Amazon – core areas of Amerindian social and political growth and environmental transformation in pre-conquest times – were largely emptied of their indigenous inhabitants. In many instances their surviving descendants live in fragmented territories along the margins of the basin (Roosevelt 1994a; Taylor 1999). As a result of post-conquest history, moreover, relatively few indigenous societies today occupy the same areas they did a century or, in some instances, even a few decades ago (de Oliveira 1994).²

The implications of widespread displacement, migration and spatial reorganisation for ethnoecology need to be re-examined in the light of recent advances in our understanding of Amazonian ecological and social systems. Ecological and botanical research, for example, has highlighted the diversity, patchiness and dynamism of lowland Amazonia at different spatial and temporal scales (Foster 1980; Tuomisto et al. 1995), linking these to different forms and scales of environmental disturbance (Pärssinen 2003a).³ The existence of a complex mosaic of Amazonian environments, each offering different sets of opportunities and constraints (Sponsel 1986; Huber and Zent 1995; Moran 1995), underscores the contingency of social and ecological relations following social change, migration and displacement. This in turn has profound and under-examined implications for such ethnoecological processes as the dynamics of incorporation, diffusion, transformation and innovation of environmental knowledge and practices, and their resulting material outcomes.

The relationship between culture, myth, history and place, on the other hand, has recently been theorised in novel and exciting ways in Amazonia and beyond (Hill 1988, 1996; Rappaport 1990; Whitehead 1994; Basso 1995; Santos-Granero 1998; Rumsey and Weiner 2001; Bender 2002; Stewart and Strathern 2003). Weaving together symbolic, political and historical elements, these approaches can generate valuable insights into the social appropriation of ‘nature’, an endeavour central to the ethnoecological approach (Toledo 2002), particularly in the context of migration, displacement and the dynamics of territorialisation.

Drawing from a wide range of disciplinary orientations and geographical contexts, this collection of case studies illustrates how some of these contemporary understandings and theoretical approaches can be used to examine the relationship between spatial mobility – including migration, displacement and dislocation – and a wide range of ethnoecological processes: processes such as the classification, management and domestication of plants and landscapes, and the incorporation and transformation of environmental knowledges, practices, ideologies and identities. In doing so, this volume seeks to contribute to the development of an ethnoecology that is historically informed, process-centred and attuned to the interplay between the symbolic and material forms and outcomes.

The remainder of this chapter is divided into four sections. After providing some general definitions, I present a schematic overview of the history of indigenous mobility and migration in Amazonia, both before and after European conquest. I then discuss some of the common themes and questions raised by the different chapters of the volume, which are presented in two parts. Chapters in Part I, Circulations, explore the links between mobility and environmental knowledge, perceptions and resource management. Part II, Transformations, mostly focuses on how migration and the concomitant circulation of peoples, plants, technology and knowledge have transformed social and ecological systems and, hence, ethnoecologies. I conclude this introduction with some thoughts on emergent themes and questions for future research.

A Typology of Mobility and Migration

As a ‘means to ends in space’ (Bell and Taylor 2004: 266), mobility refers to all forms of territorial movement by people. These movements take place at different spatial and temporal scales and reflect a wide range of underlying factors and motivations. One important distinction is between individual mobility – involving traders or peripatetic shamans (Schäfer 1991; Wrigley 1917), for example – and collective movements. Binford (1980, cited in Kelly 1992), uses the term ‘residential mobility’ to denote group-level movements between different dwelling sites, in contrast with ‘logistical mobility’, which refers to individual or small-group localised movements related to foraging and around a particular dwelling site. Territorial, circular or long-term mobility refers to the cyclic or centripetal movements of a group between different areas (Kelly 1992). Trekking is an often cited and extensively examined example of circulatory – in this case seasonal – mobility, subject to distinct environmental, socio-cultural and historical interpretations (Maybury-Lewis 1971; Turner 1979; Posey 1985; Balée 1994; Good 1995; Rival 2002). Permanent movements or migration, on the other hand, commonly refers to directional movements across larger temporal and spatial scales, and is associated with a change in residence of a minimum specified duration, usually from one to five years (Newbold 2004). Nomadism and sedentism represent extremes in a continuum of collective mobility; whereas the former involves frequent movement, often between permanent base camps, the latter refers to a condition of relatively reduced mobility (Kent 1989; Kelly 1992). Kent (1989) distinguishes between permanent and seasonal sedentism, and suggests at least six months a year of continuous occupation of a site as a defining standard.

These categories are ideal types, delineating points along a dynamic range of movements, in which considerable overlaps in time and space occur (Clarke 1984). As Pinedo-Vasquez and Padoch (this volume) illustrate for ribereños, for example, many individuals rotate between multiple, ‘multi-sited’ or transient households. This, in turn, is consistent with indigenous notions of ‘residence’, which are often defined on the basis of a region, as opposed to a particular locale, with processes of temporary mobility sometimes superimposed over longer-term and larger-scale migratory movements. Moreover, changes in one kind of mobility often presuppose adjustments in other kinds. For example, decreased residential mobility may sometimes entail an increase in logistical mobility.

Case studies in this volume discuss and illustrate some of the many factors involved in shaping migrations and their outcomes. Demographers often refer to the complex interplay of ‘push–pull’ factors in both places of origin and destination as the mechanism triggering the decision to move. Clearly, these push–pull factors operate at different levels of social organisation: state or regional, communal, familial or personal (Kleiner et al. 1986). ‘Displacement’ refers to movements induced largely through environmental, political or economic ‘push’ factors, including crises. Likewise, drivers to migration involve experiential, symbolic, economic, ecological, and political dimensions, as well as the interaction between various indigenous and non-indigenous forms of agency. As the name implies, forced displacement or relocation refers to movements resulting from external coercion, nowadays most often linked to state-led development, colonisation or conservation initiatives (Brand 2001; Chatty and Colchester 2002; Escobar 2003; Geisler 2003; Redford and Fearn 2007) or, as in some Andean states – most notably Colombia and, until recently, Peru – due to war and conflicts relating to armed insurgency and the drugs trade (Segura 2000; Guevara 2004; Navarrete-Frías and Thoumi 2005).

Indigenous Mobility and Migration in Amazonia: a Historical Synopsis

Archaeological, linguistic, ethnohistorical and genetic evidence suggests that Amazonian societies have undergone dramatic changes in their distribution, degree of mobility and spatial and social organisation, in both pre- and post-European conquest time frames. Environmental changes and disturbances, including mega-Niño events, droughts or tectonically induced relocation of entire flood plains (river avulsions), may have also induced and shaped human movements at different points in time (Meggers 1994, 1995; Pärssinen 2003a).

Early (Pre-colonial) Mobility, Migrations and Settlement

Modern humans started their migration out of East Africa between 100,000 and 50,000 years BP (Arsuaga and Martínez 1998; Lahr and Foley 1998). Ideas concerning the peopling of the Americas have been in a rapid state of flux in recent years. The commonly held view that the earliest inhabitants of the Americas – hunters specialised in large game using characteristic Clovis stone tool technology – crossed by land from the Bering Straits at the end of the Pleistocene (c. 11,000 years BP) has become increasingly questioned. Archaeological remains, distinct from those associated with the Clovis tradition and as old, if not older, have been found in several sites across the Americas (Roosevelt et al. 2002; Meggers and Miller 2003; Schurr 2004), suggesting an earlier migration date.⁴ Evidence also suggests that these early migrations probably followed a coastal route, with subsequent sequential expansions into the interior from different points along the west coast of North America (Dillehay 1999; Schurr 2004).⁵ Lithic tools used by foraging palaeo-Indians have been found in the Peruvian highlands and coast, as well as in the Amazon river (Renard-Casevitz et al. 1988; Roosevelt et al. 1996), suggesting a fairly rapid occupation of South America – including Amazonia – by the early Holocene; between 11,000 BP and 9,000 BP (Roosevelt et al. 1996; Dillehay 1999). The diversity of projectile points recovered from different sites in Amazonia suggests the presence of different human populations and overlapping colonisations by the early Holocene (Arroyo-Kalin, 2008).

Archaeological and palaeobotanical research traces the early origins of plant and landscape domestication in South America to between 10,000 and 8,600 BP (Piperno and Pearsall 1998; Denevan 2001), with important modifications of soils and vegetation evident in many areas by 2,000 BP. The oldest known cultivated plants in South America are manioc and maize. Archaeological finds in the coast of Peru dated to 8,000 to 6,000 years BP suggest the domestication of manioc prior to that, most likely in south-western Amazonia (Pearsall 1992; Schaal et al. 2006). These early dates raise the possibility that plant domestication may have preceded pottery making (Arroyo-Kalin, in preparation). Maize, domesticated in Mesoamerica before 6,000 BP (Buckler and Stevens 2006) was cultivated in parts of South America by 4,000 BP, reaching the Amazonian flood plains as early as 3,000 BP (Roosevelt 1993; Piperno and Pearsall 1998).

The intensification of agriculture and concomitant emergence of different ceramic styles during what is known as the ‘Formative’ period (4,000–2,000 BP) is associated with an increase in population densities, sedentarisation in resource-rich areas and the emergence of socially stratified chiefdoms (Lathrap 1970; Spencer and Redmond 1992; Oliver 2001). This process of population growth and sedentarisation is evidenced, particularly after 1,000 BP, by extensive areas containing rich, dark anthropogenic soils, pottery remains and petroglyphs, often – but not always – along major rivers (Herrera et al. 1992; Eden et al. 1994; Pereira 2001; Petersen et al. 2001; Kern et al. 2003), as well as by earthworks, mounds and causeways in the mouth of the Amazon, the savannahs of Bolivia, Venezuela and Guyana and other upland areas (Roosevelt 1991; Denevan 2001; Pärssinen et al. 2003; Heckenberger 2005; Erickson 2006).

The wide distribution of such language groups as Arawak, Tupí-Guarani and Carib throughout vast areas of lowland Amazonia suggests extensive movements of peoples, language, technology and ideas prior to European conquest. Arawak languages, for example, ranged across much of the subcontinent, forming ‘one of the great diasporas of the ancient world’ (Heckenberger 2002: 99), comparable to the great Bantu or Austronesian diasporas of the Old World. These Amerindian diasporas caught the early attention of linguists and archaeologists, prompting considerable debate and speculation regarding their origin, routes and causes (Lathrap 1970; Meggers 1973; Roosevelt 1994b).

While still contested, most archaeological, and especially linguistic, evidence points to northwestern Amazonia, specifically the riverine area between the upper Amazon (Solimões) in Brazil and the Middle Orinoco in Venezuela, as the centre of origin, or at least early spread, of Arawakan languages (Heckenberger 2002).⁶ According to this model, by about 3,000 BP Arawak groups began to expand across the flood plain areas of the Negro and Orinoco rivers, reaching their broadest distribution by about 1,500 BP (Heckenberger 2002). Evidence from plant domestication, in contrast, points to a southwestern origin for Arawakan expansion (Clement et al., this volume).

Proto-Tupí languages are thought to have originated around the headwaters of the Tapajós and Madeira rivers, in southern Amazonia, and Tupí-Guarani languages a little further north and east from there (Heckenberger and Neves 1998; Jensen 1999; see also Clement et al., this volume). The Tupí-Guarani expansion began 3,000–5,000 BP, eventually reaching and controlling the main river channels and the Atlantic coastal region south of the mouth of the Amazon by about 1,500 BP (Balée 1984; Monteiro 1999). Though not as widely dispersed as Arawak or Tupí-Guarani languages, Carib languages had, by the time of European colonisation, also undergone considerable expansion out of the Guiana uplands, covering much of the area between the north of the Amazon and the Lower Orinoco (Durbin 1977, cited in Santos-Granero 2002: 37) and extending as far west as the Andean foothills (Pärssinen 2003b). In contrast to these widely distributed language stocks, other Amazonian language groups – including many isolates – show a more restricted distribution, suggesting quite different ethnogenetic and historical processes and, in some cases, higher degrees of historical emplacement.⁷ The Gê languages, for example, form a fairly coherent cluster of related languages in the drier sections of the Brazilian Central Plateau, from where they are thought to have originated (Mason 1950, cited in Maybury-Lewis 1971; Urban 1992)

Many of these groups formed multi-ethnic and multilingual regional networks of trade and exchange that connected floodplain and coastal chiefdoms with interfluvial and savannah groups (Arvelo-Jiménez and Biord 1994; de Oliveira 1994; Heinen and García-Castro 2000; Vidal 2000; Renard-Casevitz 2002: 131–32). A similar network of regional exchange linked parts of the high Andes with the Amazon lowlands (Kurella 1998; Taylor 1999; Pärssinen and Siiriäinen 2003; Dudley, this volume), and was itself part of a broader system of exchange, connecting the Pacific coast, the Andes and the Amazon since about 3,000 BP (Renard-Casevitz et al. 1988; Rostain 1999). The existence of large trade networks also meant that some ethnic groups adopted trade as a form of economic specialisation (Lathrap 1973). These regional trade networks also presupposed the ability for individuals to travel long distances. The Cerro de la Sal (‘Salt Mountain’), for example, located just north of the junction of the Chanchamayo and Perené rivers in eastern Peru, was visited by Arawak, Panoan and Andean people from as far away as Ecuador to trade and harvest rock salt (Brown and Fernandez 1992: 180; Renard-Casevitz 2002: 131–32). Likewise, lowland Indians made regular visits to Cuzco, the Inca capital, and Quito (Oberem 1974; Taylor 1999).

While much of this intra- and inter-regional trade involved high-status goods such as precious stones and metals, basic commodities and food surpluses – including dried fish, shellfish and meat, turtle oil and cassava flour – were also exchanged in certain areas (Myers 1981; Whitehead 1994; Zent, this volume). Other items traded among and between groups in Amazonia and Orinoco and between the tropical forest lowlands and the Andes or circum-Caribbean region included arrows and blowpipes, baskets, medicinal plants, arrow poisons, dyes, vanilla and cinnamon, tamed animals and animal products from the lowlands, and cotton cloth, jewellery and metal tools from the highlands (Oberem 1974; Brown and Fernandez 1992; Taylor 1999; see also Zent, this volume). Present-day distribution of crop landraces (Clement et al., this volume) and direct observations among contemporary Amazonians (Kvist and Barfod 1991; Alexiades 1999; Alexiades and Peluso, this volume; Micarelli, this volume) further suggest that such exchanges must have also included plant cuttings, seeds, technology and knowledge (Butt Colson 1985).

Whitehead (1994, 1999) argues that scholars have consistently underestimated the regional scale and supra-ethnic character of pre-colonial Amerindian social organisation (see also Neves 2001 and references therein). Building on contemporary understandings of ethnogenesis and regional trade, Hornborg (2005) proposes that Arawak expansion did not necessarily entail the actual movement of people (sensu Lathrap 1970), but more probably involved linguistic and social exchange and the adoption of a lifestyle by people already established in these areas. Arawak ‘expansion’, Hornborg argues, entailed the diffusion of Arawak as a trade language and a series of social forms – notably river navigation, trade, intensive agriculture, hierarchy and geographically extended identities – all suited to active involvement in an extensive regional exchange system.

By the time the Europeans arrived in South America, then, a diverse range of societies with distinct and varied forms of political organisation and subsistence were established across much of lowland South America, interlinked through a complex and fluid network of trade, war and exchange. The emergence of a ‘differentiated political-economic structure’ (Hornborg 2005: 593) and regional system of exchange in parts of the Amazon by the first millennium BC, of socially ranked societies by ca. AD 400–500, and of complex chiefdoms by around AD 1000 (Spencer and Redmond 1992; Roosevelt et al. 1996; Neves and Petersen 2006), all point to considerable prehistoric social, demographic and political upheavals, including widespread migrations, displacements and shifts in patterns of mobility, all within a highly diversified and human-altered environment (Denevan 2001).

European Conquest and Indigenous Migration in Amazonia

Research over the past decades has shown how post-conquest contact with the expanding state in the ‘tribal zone’ (Ferguson and Whitehead 1992) has often involved a two-phase, recurrent cycle. The first consequences of contact – socio-demographic disruption and spatial dispersion – has followed from the catastrophic impact of ‘virgin soil’ epidemics.⁸ A second stage has involved the pull – often intermittent and uneven but eventually widespread – of indigenous societies into the orbit of the extractive or agro-extractivist economy and/or of colonial or post-colonial states. While often overlapping in time or space, these two stages are distinct, particularly in terms of their influence on ethnoecology, mobility and spatial organisation.

The devastating speed and scale with which infectious diseases repeatedly and systematically decimated Amerindian societies – and their corresponding impact upon their social and institutions and production systems – were not fully grasped until recently. It is almost certain that the Amazonian societies first described by Europeans beginning in the mid-sixteenth century had already undergone substantial demographic and social transformations following the diffusion of early epidemics out of the circum-Caribbean region (Myers 1992). Increased mobility and dispersion have historically offered Amerindians a measure of protection against epidemics, war and colonisation, and have thus enabled strategic mediation of contact with the modern nation state and neighbouring enemy groups (Carneiro 1964; Hugh-Jones 1979; Oostra 1991; Stearman 1995; Fisher 2000; Coimbra et al. 2002; Santos-Granero 2002; Alexiades and Peluso, this volume; Zent, this volume).⁹

In effect, retreat was the only option available to those indigenous societies unable to mount an effective military resistance or who refused shelter and incorporation (for example, by missions), though in effect these three strategies were often adopted simultaneously or sequentially (Taylor 1999). Even active military resistance often entailed the disbanding of large settlements and a strategic reliance on dispersal and mobility (Vidal 2000; Coimbra et al 2002; Athayde et al., this volume; Zent, this volume). As a corollary, the more mobile groups were often those eventually able to resist colonial expansion most effectively and consistently (Wright 1999).

The initial process of depopulation had a complex domino-type effect on indigenous patterns of spatial distribution. Heightened accusations of sorcery and revenge raids, the displacement into areas occupied by neighbouring groups, together with the emergence of a political economy centred around the procurement of steel tools and slave labour, all increased inter- and intra-tribal war, leading to further social atomisation, population loss and heightened mobility (Posey 1987; Ferguson 1995; Alexiades and Peluso 2003; Lyon 2003; Zent, this volume). The profound impact of this process is illustrated by Viveiros de Castro's (1992: 15) characterisation of Araweté history as ‘an incessant movement of fleeing and dispersing’ where ‘a state of war seems to have been the rule and the custom’.

The early colonisation of the Atlantic coast and the main waterways beginning in the sixteenth century led to a collapse of the centralised polities that had established themselves there a few hundred years earlier, and to a retreat from these core areas of colonial economic development. For Tupí-Guarani speakers, the consequent massive exodus west, ‘away from the centres of human extinction’ (Balée 1984: 256), reversed the pattern of eastward migration that had preceded the arrival of Europeans.¹⁰ This pattern of depopulation and dispersal recurred continuously over the following centuries, tracking waves of extractivist activity as these reached new areas or revisited old ones.¹¹ The rubber boom (1850–1916) – Amazonia's largest single extractivist bonanza – set off a wave of depopulation that swept westwards, leading to massive dispersal, spatial reorganisation and widespread retreat towards interfluvial and inaccessible headwater regions.

Colonisation and depopulation through the combined effects of slave raiding and epidemics also brought down the multi-ethnic regional networks that had previously connected much of the Americas. Even though some of these extensive trade networks and inter-ethnic polities initially adapted, thrived, or sometimes even formed, in response to the demand for slave labour and metal tools, most had disappeared by the end of the eighteenth century (Dreyfus 1992; Whitehead 1994: 43–45, 1999; Ferguson 1995; Wright 1999, Zent, this volume). The Andes and the Amazon region also became increasingly disarticulated, contributing to the decline of those groups who had emerged in the context of the intermediation between the two zones (Taylor 1999; Dudley, this volume). In the meantime, however, a number of ethnic groups emerged as important mediators between colonial and indigenous economies, shaping inter-ethnic relationships to this day (Taylor 1999). The Ecuadorean Quijo and Canelo, for example, acted as intermediaries between the colonial highlands and such lowland groups as the Shuar, while the Caribs mediated between the Dutch and inland groups in Guyana (Oberem 1974; Harner 1984; Whitehead 1993).

The initial dispersal and retreat from the agro-extractive frontier often entailed moving out of the most accessible or resource-rich areas. Interfluvial areas became important refuge zones and thus important sites of inter-ethnic contact and assimilation (Zent, this volume). Some of these interfluvial societies were eventually able to expand and occupy those fluvial areas which been emptied of their original indigenous inhabitants (Smole 1976; Henley 1982: 11; Hames 1983; Rival 2002; Alexiades and Peluso 2003; Johnson 2003; Zent, this volume). Coupled with depopulation, the process of spatial dispersion contributed to