THE POLITICS OF SOCIALITY: SOCIAL NETWORKS AND INDIGENOUS MOBILIZATION IN PERUVIAN AMAZONIA By DANNY DAVID PINEDO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2014
Â© 2014 Danny David Pinedo
To my mother
4 ACKNOWLEDGMENTS I am deeply indebted to the Arakmbut people of Puerto Luz who generously shared their lives with me, tolerated my intrusive presence , and answered my insistent questions. In particular, I would like to thank the Tayor i family (Lucho, Maribel, Marcos, Lupe, TomÃ¡s, o , Ana , and Don Pablo ), with whom I lived during my time in P uerto Luz. I learned from them m ost of what I know about the Arakmbut. I would also like to thank Delia Takori , who helped me to conduct the survey. The Irisanehua family gave me shelter and a hearth to cook when I first arrived in Puerto Luz. I am indebted to them too. My g ratitude also goes to Juan de Dios Chimatani, Ezequiel Moqui, Pedro Takae, Roberto Nayori, Erasmo Manya, and Edi Ccotohuanca for their friendship. They were always ready to cha t. In Puerto Maldonado, I must thank FENAMAD for gi ving me permission to conduct this research in some of its constituent communties . J aime Corisepa and Klaus Kicque not only kindly accepted to have an interview with me but also opened their home to me. In FENAMAD I also thank TomÃ¡s Ariqu e, Jorge Payaba, CÃ©sar Auguto Jojaje , and Alicia FernÃ¡ndez for their support. My thanks also to the FE NAMAD personnel for their assistance . I am also grateful to Tom Moore (Director of Centro Eori), Alfredo GarcÃa, and Zoila Arredondo for giving me institutional, logistical and intellectual support, but most importantly for their friendship and companionsh ip during my days in Puerto Maldonado . Tom was kind enough to share with me his inconmensurable knowledge about the Harakmbut. Alfredo , Jenny, Tayana and Nico made me feel part of a family wh en nothing seemed to be working . I also want to thank the good fr iends I made in Puerto: Julio Cusurichi, Edgar Sulca, Ramiro DurÃ¡n, Solveig Firing Lunde, Manolo Ponce de LeÃ³n, Claudia Dura n d , and Carla Merediz .
5 In Gainesville, I would like to thank the members of my supervisory committee, Drs. Marianne Schmink, Anthony Oliver Smith, Brenda Chalfin and Eric Keys. Special thanks are due to my mentor and fr iend Dr. Schmink for her enormous advice, encouragement and friendship. I was very fortunate to have her as a mentor. My long conversatio ns with Tony at the Starbucks in the HUB and i n Coffee Culture helped me to clarify my theoretical doubts. My deep gratitude goes to Nita Bagnal, Karen Jones, Pam Freeman, and Pat King , from the Department of Anthropology , and Patricia Sampaio, from the Tropical Conservation and Developm ent Program . This dissertation would not have been possible without the ir invaluable support. I am also thankful to my friends in Gainesville, without whom my life would have been less the other Rafa (Mendoza), Alfredo, Gina, Xanic, and Francisco. I also thank Alicia PeÃ³n, Joe Feldman, James Cra ndall, Miriam Domingu ez, Jacob Lawson, Zach Cui , Simone Athayde, and Flavia Leite. Fieldwork for this research was financially supported by an Inter American Foundation Grassroots Development Fellowship. A John M. Goggin Research Award, from the Department of Anthropology at the University of Florida, allowed me to return to the field for an additional month . The dissertation writing process received t he financial support of the Bourse Bernard Lelong from the Centre National de la Recherche Scientifique (France). Funding for my doctoral study was generously provided by a Gordon and Betty Moore Graduate Scholarship for Tropical Forest Conservation and th e Department of Anthropology at UF . An Amazon Conservation Leadership Initiative Summer Research Grant provided me with funds for preliminary doctoral research. I must also recognize the love and support of my parents, Gladys and Lucho, who proudly and pat iently waited for this moment to come. I will keep reading, mom! Study ing never
6 ends! Filito gave me the privilege to have a second mom. I not only want to thank her for all she gave us , but also apologize for my long absence. Similarly, I thank my brothers Freddy and Teddy, and my sisters Paola and Sally, for their patience and understanding. I also need to acknowledge the courage of my da ughter MarÃa Jimena during my lo ng absence. You know I love you! Finally, and mos t importantly, I am deeply indebted to Stephanie, for her love and support. She has been not only a great companion, but also an intellectual partner.
7 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 9 LIST OF FIGURES ................................ ................................ ................................ ....................... 10 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 11 ABSTRA CT ................................ ................................ ................................ ................................ ... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 14 Amazonian Sociality ................................ ................................ ................................ ............... 16 Social Networks and Social Movements ................................ ................................ ................ 20 Social Networks and Movement Participation ................................ ................................ 20 Social Networks as Social Capital ................................ ................................ ................... 25 Cultures of Resistance ................................ ................................ ................................ ............ 30 Method s ................................ ................................ ................................ ................................ .. 36 Fieldwork ................................ ................................ ................................ ......................... 36 Interviews ................................ ................................ ................................ ........................ 37 Household Survey ................................ ................................ ................................ ........... 38 Participant Observation ................................ ................................ ................................ ... 38 Arch ival Research ................................ ................................ ................................ ........... 39 Structure of the Dissertation ................................ ................................ ................................ ... 39 2 MADRE DE DIOS: A HISTORY OF INTER ETHNIC RELATIONS ................................ 43 Antisuyo during the Inca Rule ................................ ................................ ................................ 44 Spanish Conquistadors, Coca and Gold ................................ ................................ .................. 50 Cinchona Bark and the Nation State ................................ ................................ ...................... 56 Rubber Boom and the Mission ................................ ................................ ............................... 59 Gold Rush and Oil Boom ................................ ................................ ................................ ....... 67 Conclu sion ................................ ................................ ................................ .............................. 74 3 THE ARAKMBUT: IDENTITY AND LIVELIHOOD ................................ ......................... 80 Place and Population ................................ ................................ ................................ ............... 82 Identity ................................ ................................ ................................ ................................ .... 84 Livelihood ................................ ................................ ................................ ............................... 92 Gold Mining ................................ ................................ ................................ .................... 93 Gardening ................................ ................................ ................................ ...................... 101 Hunting and Fishing ................................ ................................ ................................ ...... 104
8 Conclusion ................................ ................................ ................................ ............................ 109 4 ARAKMBUT SOCIALITY ................................ ................................ ................................ . 112 Kinship and Clan ................................ ................................ ................................ .................. 115 Friends and Compadres ................................ ................................ ................................ ........ 127 Patrons and Clients ................................ ................................ ................................ ............... 13 5 Communit y and Fiestas ................................ ................................ ................................ ........ 140 Leadership ................................ ................................ ................................ ............................. 144 Conclusion ................................ ................................ ................................ ............................ 150 5 SOCIAL NETWORKS AND POLITICAL MOBILIZATION ................................ ........... 154 Indigenous Networks and Struggles for Land ................................ ................................ ...... 158 Networks, Recruitment and Access to Resources ................................ ......................... 159 Organizations as Networks ................................ ................................ ............................ 170 Indigenous Mobilization and Power Relations ................................ ................................ ..... 178 Ethnic and Gender Inequalities ................................ ................................ ..................... 179 Networks of Power ................................ ................................ ................................ ........ 184 Conclusion ................................ ................................ ................................ ............................ 189 6 THE CULTURE OF INDIGENOUS MOBILIZATION ................................ ..................... 192 From Autonomy to Open Resistance ................................ ................................ .................... 194 Raiding and Fleeing ................................ ................................ ................................ ....... 195 The Mission and the School ................................ ................................ .......................... 199 From Open Resistance to Organized Political Mobilization ................................ ................ 205 The Ethnic Federation ................................ ................................ ................................ ... 205 Old So cieties, New Organizations ................................ ................................ ................. 211 Seeing like an NGO ................................ ................................ ................................ ....... 216 Organized Mobilization ................................ ................................ ................................ . 220 Identity Politics ................................ ................................ ................................ .............. 223 Becoming I ndians in Madre de Dios ................................ ................................ ...... 224 Visual signs of indianness ................................ ................................ ...................... 228 Indians and environmentalists ................................ ................................ ................ 231 Conclusion ................................ ................................ ................................ ............................ 234 7 CONCLUSIONS ................................ ................................ ................................ .................. 239 Indigenous Sociality and Networks ................................ ................................ ...................... 239 Networking to Mobilize ................................ ................................ ................................ ........ 243 Political Mobilization as Culture ................................ ................................ .......................... 247 LIST OF REFERENCES ................................ ................................ ................................ ............. 251 BIOGRAPH ICAL SKETCH ................................ ................................ ................................ ....... 268
9 LIST OF TABLES Table page 4 1 Number of people per clan in Puerto Luz, 2012 ................................ .............................. 117 4 2 Ethnic origin of godparents in Puerto Luz, 2008 2012 ................................ .................... 132 5 1 FENAMAD leadership by ethnic group, 1982 2014 ................................ ....................... 180
10 LIST OF FIGURES Figure page 2 1 Madre de Dios Basin and surrounding areas. ................................ ................................ .... 78 2 2 Natural protected areas and territorial reserve for isolated peoples. ................................ .. 79 3 1 Highland miners working with carretilla, Puerto Luz. Photo by Danny Pinedo ............... 96 3 2 Arakmbut native communities. ................................ ................................ ........................ 111 4 1 Residence group, Puerto Luz. Photo by Danny Pinedo ................................ ................... 121 4 2 Puerto Luz village in 2012 ................................ ................................ ............................... 153 6 1 Danny Pinedo ................................ ................................ ................................ ................... 223 6 2 Indi genous demonstration in Puerto Maldonado. Photo by Danny Pinedo ..................... 229
11 LIST OF ABBREVIATIONS AIDESEP CIPA CNA COHAR COHARYIMA COICA COINBAMAD CONAP DANIDA FADEMAD AsociaciÃ³n InterÃ©tnica de Desarrollo de la Selva Peruana Centro de InvestigaciÃ³n y PromociÃ³n de la AmazonÃa ConfederaciÃ³n Nacional Agraria Consejo Harakmbut Consejo Harakmbut, Yine y Machiguenga Coordinadora de Organizaciones IndÃgenas de la Cuenca AmazÃ³nica Consejo IndÃgena del Bajo Madre de Dios ConfederaciÃ³n de Nacionalidades AmazÃ³nicas del PerÃº Danish Intern ational Development Agency FederaciÃ³n Agraria Departamental de Madre de Dios FENAMAD GEF IBIS ILO IPC FederaciÃ³n Nativa del RÃo Madre de Dios y Afluentes Global Environmental Facility Intern ational Benevolence Initiatives International Labor Organization International Petroleum Company IWGIA MISEMA MRTA NGO NORAD REDD RESSOP SIL SINAMOS International Work Group for Indigenous Affairs AsociaciÃ³n de Misioneros Seculares Movimiento Revolucionario TÃºpac Amaru Non governmental Organization Norwegian Age ncy for Development Cooperation Reducing Emissions from Deforestation and Forest Degradation Red Escolar de la Selva del Sur Oriente Peruano Summer Institute of Linguistics Sistema Nacional para la MovilizaciÃ³n Social
12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE POLITICS OF SOCIALITY: SOCIAL NETWORKS AND INDIGENOUS MOBILIZATION IN PERUVIAN AMAZONIA By Danny David Pinedo August 2014 Chair: Marianne Schmink Major: Anthropology This dissertation examines the influence of social networks on political mobilization among indigenous people of Madre de Dios, in the southeastern Amazon lowlands of Peru. Since the early 1 970s, an unprecedented rise in the international price of gold and oil, coupled with neoliberal policies that encouraged foreign investment in mining and oil extraction, spurred the arrival of transnational corporations and thousands of Andean settlers in the region, resulting in massive encroachment on indigenous lands. To defend their lands , indigeno us Amazonians engaged in an ethnopolitical movement, a major expression of which has been the formation of an ethnic federation (FENAMAD). I explore the effec ts that networks based on kinship, friendship and patron client relations have on indigenous people s ability to mobilize resources and supporters, and build organizations. The study also analyzes how power and inequality permeate these networks, and how cultural values inform indigenous forms of political mobilization. To explore these issues, I conducted an ethnographic and historical study of FENAMAD and the Arakmbut community of Puerto Luz. Data were collected through ethnographic and life history inte rviews, participant observation, and archival research. The dissertation argues that
13 Arakmbut sociality, solidarity is not restricted to members of kin groups ; it may be extended to outsiders through friendship, fictive kinship and patronage relations. While kinship networks were more important to facilitate community based mobilization, alliance networks with anthropologists and human rights activists were mor e relevant in bringing autonomous kin groups and communities into larger multiethnic political organizations. Open acts of resistance were replaced by organized yet mostly hidden forms of defiance. The dissertation also concludes that social networks may also engender dilemmas between individual and collective interests, and reproduce existing gender and ethnic inequalities. Godparenthood networks with mining settlers, as well as patron client ties with oil companies, cr eated tensions within Arakmbut communities, hindering their capacity for collective action. Moreover, despite their crucial role in helping communities to get land titles and development projects, ethnic federations have been dominated by more acculturated , male gendered indigenous elites.
14 CHAPTER 1 INTRODUCTION Most studies of Latin American indigenous movements emphasize the role of ethnic identity in catalyzing indigenous grievances an d mobilization (Brown 1993; Conklin and Graham 1995; Jackson 1991 , 199 4 ) . These studies suggest that indigenous peoples, in particular those of the Amazonian region, have adopted Western rhetoric that equates native peoples with a conservation ethic to construct their identities in order to express their political demands an d mobilize public support . In this context, some scholars became interested in the role of social networks in indigenous mobilization. Ethnic identity is seen as facilitating the forging of alliances with powerful non indigenous environmental advocates (Br ysk 1994, 2000 ; Schwartzman and Zimmerman 2005 ). What remains unexplored, however, is the role that more traditional forms of social organization and networks play in indigenous mobilization, and how these articulate with more modern networks with n ational and international acti vists. The Madre de Dios region has been the epicenter of major social changes since it was opened up to the rubber boom in the late nineteenth century. Remnants of indigenous groups demics were concentrated in Dominican missions. In the early 1970s, after a road connected the region with the rest of the country, anthropologists and state agents arrived to promote the organization of indigenous settlements in rporate body based on the collective ownership of land. At the same time, the rise of the international price of gold and the loosening of state control over mining industry spurred a gold rush that has brought to the region mining corporations and thousan ds of impoverished peasant farmers from the nearby Andean highlands. In addition, reforms promoting private investment in the oil industry in the early 1990s have led to a dramatic increase in the number of oil concessions, a process that some have deemed
15 hough indigenous people sought legal protection by having their lands titled, indigenous lands were either invaded or overlapped by mining and oil concessions. To defend indigenous lands, anthropologists and NGOs helped the Arakmbut to form a region wide network of communities ( Federac iÃ³n Nativa del RÃo Madre de Dios y Afluentes , or This dissertation explores the relation s between social networks and political mobilization among Amazonian indigenous people. Through a case study of the Amazonian region of Madre de Dios, southeastern Peru, it ex amines the social ties in which indigenous people engage, the influence of these social networks in their capacity to collectively mobilize with political aims, and the cultural patterns that inform the way this political mobilization is practiced . This is achieved by analyzing the networks based on kinship, friendship and clientelist relations that link indigenous people among each other and with outsiders, and the role that these networks played in mobilizing people and resources for the defense of indigenous rights to land, especially the formation of indigenous organizations. Each research question has been addressed from three different theoretical perspectives. Anthr opological studies of Amazonian sociality have analyzed why some indigenous groups prefer to build intra group social ties, while others are more prone to develop outwar d looking ties. To u nderstand the differential effects of both types of sociality is ke y considering that political mobilization is more likely to be based on network alliances. The impact of social networks in political mobilization have been examined from two theoretical schools. On the one hand, sociologists have studied the role of socia l networks in facilitating individual participation in social movements. On the other hand, students of social capital, particularly from geography and sociology, have analyzed how social networks may help social movement organizations to
16 bring economic an d political benefits to their constituents. Finally, political mobilization can be understood as being informed by a political culture. The forms and expressions of this culture in peasant societies have been studied by scholars of the everyday forms of re sistance. I will review each of these approaches in the next section. Amazonian Socia lity and form social groups, is essential to the understanding of the influence of social networks on indigenous mobilization. Scholars of sociality among Amazonian indigenous groups have focused on the endogenous and exogenous nature of sociality. The kind of social ties in which people engage, and with whom these networks are built ability to mobilize other people. Among contemporary anthropologists, there are two major approaches concerning sociality among Amazonian indigenous peoples. One approach , which Viveiros de Castro the domestic domain, emphasizing the role of consanguinity, endogamy and the solidarity in duced by moral sentiments resulting from intimate relations (Belaunde 2001 ; Gow 1991; M cCallum 2001 ; Overing Kaplan 1975; Santos Granero 1991). According to this perspective, Amazonian societies are mostly concerned about conviviality (amiable and intimate relationships) and cha racterized by the idea of an autonomo us self, rejection of any c oercive rules, and the lack of hierarchical social structures (Overing and Passes 2000). Drawing on her research among the Piaroa of the Venezuelan Orinoco Basin , Over ing (2003: 300) points out that the very creation of the collective is dependent upon such personal autonomy . The ideology of the independent self greatly stresses self control rather than social control, which markedly differs from the Western notion of individualism in which the individual imposes upon society
17 ( Overing 1988 ). Therefore, Amazo sociable life together and avoid dangerous anger, and not upon the building of order and social structures ( Overing and Passes 2000). Overing Kaplan (1975) also argues that, for Amerindians, conviviality has the ability to turn affinal relations into actual kinship relations. Am ong the Piaroa, the process of living together makes those who in the first place are dangerously outsiders (e.g. in laws) be come kin. son for members of a community becoming increasingly similar to one another is that the mutuality of living together creates a certain sort of material , the Piaroa suppress affinal distinctions within the residence group. In fact, for the Piaroa marriage can lead to kinship, since they do not separate affinity from consanguinity (Overing Ka plan 1975: 70). Friendship relations are also compared to consanguineal relations by emphasizing how gift exchange and trade may work as another way of creating close relations that ultimately will turn others into kin (McCallum 2001:96). Among the Piaroa, some kinship terms that carry connotations of friendship can also be applied to trading partners, which is indicative of the fact that friendship and kinship are effectively synonymous (Overing Kaplan 1975:71). Proponents of the second app roach on Amazon ian sociality, which has been deemed as stressing difference (alterity and affinity ), exchange and the metaphor of predation as the key concepts shared by Amazonia n societies (Descola 1994, 1996; Taylor 1996, 2001; Viveiros de Castro 1992) . Drawing on LÃ©vi Straussian structuralism, scholars in this school privilege processes of s ymbolic exchange (such as war and cannibalism, hunting, shamanism and funerary
18 rites) th at play an important role in defining collective identities (Viveiros de Castro 1996: 190). They criticize the conviviality approach for turning descriptions of particular Amazonian groups into a general moral philosophy of Amazonia. In doing so, adherents of this approach are seen as privileging internal relations and their associated moralities of love and conviviality while downplaying and sometimes ignoring relationships with other groups, with their related notions of affinity, predation and vio lence (V iveiros de Castro 1996:189 ). 1 From this perspective, Amazonian sociality is based on exchange, not on the cons ubstantiality that results from sharing everyday life . Affinity, which rather than being understood only as a particular form of kinship relation is thought as a generic value that denotes otherness or alterity, is seen as the key to sociability (Viveiros de Cast ro 1995: 14 , 2001:22 ). In this view, affinity is the natural whereas consanguinity is a cultural construct . Followers of the pred ation approach further assert that while at the local level affinity is encompassed by consanguinity, at the supralocal level affinity encompasses consanguinity and becomes a reference to all kinds of relations between locals and outsiders ( Descola 1996; V iveiros de Castro 1993:181 ). In a recent article, Viveiros de Castro has suggested that as the perspective shifts from the local relations to wider contexts, affinity becomes the overall mode of 24 25). In this context, among Amazonians the emphasis is placed on their relations with ( guests, enemies and trading partners, even animals and spirits ) because in the absence of the affinal other, society cannot reproduce itself. This emphasis on affinity markedly contrasts with 1 Overing and her team have defended themselves against this accusation by arguing that they do not a ctually dismiss issues of supra local relationships, conflict, distrust, and predation. What they actually argue, Overing and Passes assert, is that most of these wider relationships are part of Amazonian narratives of sociality that construed them as asocial, or negative examples of soc iality (Overing and Passes 2000:6) .
19 concern about drawing affines int o consanguineous relations. 2 Some scholars have explored the role of friendship as an alternative mode of relationship in the forging of native Amazonian sociality. This interpretation of friendship distinguishes from the two previous approaches in that he re the importance of friendship is seen lying not in its relation to kinship and affinity, but in its distinction from them (Killick 2008, 2009, 2010; Santos Granero 2007). Santos Granero (2007) argues that while friendship may sometimes be talked about in the idiom of consanguinity and affinity, it is a fundamentally different type of relationship. He goes on to argue that , rather than transforming friends into kin or affines, Amazon ian groups may sometimes turn esteemed kin and affines into formal friends , which renders the tie of friendship more important than the existing kinship ties . In his study of ayompari relations among the AshÃ©ninka of eastern Peru, a trading partnership based on friendship, Killick (2009) contends that friendship ties can be deli berately maintained even if kinship relations can be invoked. Friendship offers an escape from the sometimes oppressive obligations of kinship, thus helping Amazonian people to escape the kinship affinity dichotomy (Killick 2005; Santos Granero 2007). Fri endship networks allow Amerindians to form relations where none exist, that is, with unrelated potentially dangerous others. In this process, by turning fear and predation into friendship, generosity and trust, which are oriented not only to kin but also t o strangers, become the means of avoiding antagonism and controlling unrelated others (Killick 2005). Compared to the predetermined state of kinship relations , friendship emphasize s both free cho ice and 2 Echoing this debate, Descola (2012 ) has suggested the existence of two different forms of exchange among Amazonian societies: equivalent exchange and unilateral taking. Under equivalent exchange marriage is organized b y the exchange of women or goods between different tribes, and even hunting involves a ritual to return the spirit of a killed animal to its own kind. Under unilateral taking, in contrast, interaction is characterized by predation and not reciprocity.
20 consensual nature, a reason why these relations must be constantly maintained by repe ated demonstrations of conviviality and trust . As such, friendship relations are not given and fixed but processual (Santos Granero 2007). Is it possible that, where kinship and friendship coexist, generosity is oriented onl y to relations with close kin? In this context, wha t characterizes friendship with non kin? Is Amazonian sociality restricted to kinship, affinity and friendship? The Arakmbut of southeastern Peru offer a case in which sociality is made up of different coe xisting social ties yet with distinct moralities attached to them. Social Networks and Social Movements Indigenous politics in Latin America has expressed more through social movements than political parties, corporate interests groups, or guerrilla moveme nt. The role of social networks in political mobilization has been the subject of at least two theoretical traditions. First, s ince the mid 1970s, the two main theories in social movements literature , the new social movement theory and the resource mobilization theory , have primarily dealt with the factors that explain why and how social movements form and persist over time, focusing on the factors that encourage individuals to engage in social movements. I n their e xplanations of why people participa te in social movements , these two theories assign different roles to social networks . collective organizations and bring resources to their constituencies. Social N etworks and Movement Participation S ociologists of the resource mobilization theory have long been focused on the crucial role that social networks play in the process of individual participation in social movements. The theory of resource mobilization poi nts out that for a social movement to emerge it is necessary that resources (material, human, cognitive, technical, and organizational) be deployed in order to make the preference structure more attractive for potential participants . T he existence of
21 disco ntent or grievances does not necessarily explain the rise of move ments (McCarthy and Zald 1977). Thus, resource mobilization theorists see s ocial networks as faci litating individual decisions to join a movement (McAdam and Paulsen 1993; Passy and Giugni 20 01; Snow et al. 1980). Social networks are said to provide the mechanisms and dynamics that induce individuals to become involved in social movements. The general consensus is that embedde dness in networks fosters greater levels of movement participation by facilit ating recruitment of activists. Social ties are one of the major channels through which potential activists are given an opportunity to participate in mobilization, and the possibility to convert their political consciousness into action. Thus , individuals with friends and acquaintances already involved in social movements are more inclined to take part in them (della Porta 1988; Gould 1993; McAdam 1986, 1988; Snow et al. 1980). Social networks sha pe the individual preferences or perceptions that form the decision making process regarding participation, and bring potential activists to collective action (Passy 2003). These postulates draw on the rational choice theory, according to which people form or join social movements when they judge it is in their individual interest to do so. In other words, people will become members of a movement when the individual benefits of joining outweigh the co sts of it (Hechter 1987). Th ese postulates have generated a debate wi th proponents of the so called new social movement s theory, which emphasizes the values and emotions that bind individuals as predic tors of movement participation. The new social movement s theory emerged in response to the Marxian class movement paradigm , which upheld class interests as the primary sour ce of conflict and organization . Instead, the new social movements theory argue s capital
22 conflicts gave way to struggles over symbolic resources and identity rights (Touraine 1988). To explain participation in social movements, the new social movements theorists appeal to constructionist approaches that emphasize the role of meaning and bonds of sentiment an d affect in motivating people to participate in collective action. Emotion expressed through ritual, symbol and ideology are considered as having mobilizing power. Nevertheless, social movements based on collective identity and other values have been simpl y dismissed by resource mobilization theory as irrational, since they do not fit in the rational model of cost benefit. In recent years, however, there has been a shift from a narrow application of the rational choice theory towards a more comprehensive m odel that brings together instrumental rationality cultural content . Under this new synthesis, the actor is considered as embedded in a social context of networks and ide ntities (Mueller 1992). S ocial networks build or reinforce individual identities that create potential for participation. The social networks in which actors interact convey meanings (e.g. symbols, rituals, narratives) that build and solidify identities ba sed on certain norms and values . These identities provide individuals with a political consciousness that in turn allows them to get ideologically closer to a given political issue (McAdam and Paulsen 1993; Snow et al. 1986). The loyalt ies and obligations between the individual and the group, which are activated by such collective identities as nationality, race, ethnicity, class, gender, or religion, may potentially lead to collective action (Ferree 1992; Schwartz and Paul 1992). Thus, grievances and calcu lation of costs and benefits are not the result of a single objective reality , but of perceptions or meanings socially constructed within particular social locations. Some new social movements theorists have also tried to merge the rational choice theory w ith collective identity. Melucci (1988), for example, argues that networks facilitate the
23 processes of individual involvement in social movements by reducing the costs of participation. Mobilization never takes place in a void , but within pre existing netwo rks of relationships through which individuals evaluate and recognize what they have in common and if they want to act collectively. I ndividuals interaction (mutual influence, negotiation) within these networks produce the necessary cognitive frame that m otivates individuals for collective action. Thus , motivation to participate and its value are never an exclusively individual variable; they are constructed through interaction in the social networks in which individuals are incorporated. Sociologists emph asize the role of f riendship and acquaintance ties between already involved activists and potential activists in enhancing individual partic ipation in social movements. Nevertheless, what is the role of other social ties such as kinship, f ictive kinship an d clientelist Also, s oc iologists tend to highlight the positive effects social networks may have on prompting engagement in political mobiliza tion. To what extent social networks may have the effect of demobilizing people? Do social networks at different scales (individual and collective ones) differentially affect participation in social movements? Individuals engage in multiple social ties that might have differing effects on social movem ent participation. Some scholars have distinguished different social ties based on their varying effects on collective action. Narayan (1999) contend s that bonding ties (e.g. kinship and ethnic based networks ) lead to overall social fragmentation, while bridging ties (cross cutting networks ) lead to social cohesion they refer to the lo osely knit networks beyond the immediate circle of family and ethnicity.
24 into larger communities. ties to a Mexican squatter movement, Holzner (2004) concludes that residents who avoided clientelist ties (strong ties) with a traditional political party and established extensive networks with other were more able to access information on economic and political resources . Anthropologists studying indigenous mobilization in the Amazon region have found that kinship networks generally play a negative role in mobilizing local groups . Rosengren (2003) , for example, suggest s that the highly atomized nature of their social organization (bonding ties) prevents the Matsigenka of s outhern Peru from forming larger multi community organizations (bridging ties) . T he Matsigenka have had t o rely on the construction of a collective identity as a bound ed ethnic group to mobilize autonomous communities. Veber (1998) , in contrast, describes ho w the AshÃ©ninka of central Peru, despite being organized in as highly autonomous family level groups as the Matsigenka, were able to develop networks among local gro ups based on their attachment to leaders and the establishment of larger settlements around schools . What is the specific role that different social networks play in indigenous mobilization? Are bonding ties always detrimental to indigenous mobili zation? In this dissertation, I explore the varying and sometimes contradictory effects that different social networks have on Arakmbut political mobilization. On the other hand , social networks are not only important for participation in social movements. Social movements require resources to be put in motion and social networks may facilitate access to those resources. What is the role of varying social ties in enabling access to resources for mobilization? What social ties are used to recruit supporters, form organizations and get access to economic resources? These are questions that have been addressed by the
25 second theoretical perspective on the relationship between social n etworks and political mobilization. Social Networks as Social Capital Social capital theory sees social networks as forms of social capital that may be used to accomplish tasks, access goods or services, or gain in status or political power. At least two a pproaches have emerged regarding the nature and functional utility of social capital. From an civic associations as well as norms of trust and reciprocity. Social capital is said to allow individuals and groups to gain access to other forms of capital (economic, human, cultural) and ultimately lead to economic prosperity and democracy. From a political economy view, others posit that as individuals and groups are so cially differentiated, social capital is always unevenly distributed and hence it tends to reproduce power inequalities. The economistic approach to social capital is based on the widely cited works of James Coleman and Robert Putnam . Coleman (1988) define s social capital as networks based on norms or values of reciprocity and trust that reduce transaction costs and foster efficient forms of individual and collective action, thus creating the possibility of transforming social capital into human and financi al capital. For Coleman (1990 ), social capital as a functional value is present in both formal organization s an d informal networks . Of central importance is the concept of reciprocity , ensured by a shared acceptance of effective social sanctions for those who defect , and the idea that social capital facilitate goals that in its absence could not be achieved , or achieved only at a higher cost . Coleman also makes clear that social capital can be built and stored, and its benefits s pread throughout society, including to those who d i d not directly contribute to it ( Coleman 1990: 304).
26 In their now classic work on civic traditions in Italy , Putnam et al. (1993) bring Coleman capital to a broader scale, analyzing t he influence of local level associational practices on political and economic outcomes at the national level . Putnam and his team conclude that differences in social structure account for regional difference s in levels of democracy and economic development . Thus, long term involvement in civic organizations and shared values (like trust and reciprocity) in the northern part of Italy ha ve produced dense , resulting in more democratic governance and better economic performance. In contrast, a social structure based on vertical relations (patronage and clientelist relations) in southern Italy resulted in weak governance and poor economic performance. In explaining these differential economic and political histories, Putnam et al. stress that so cial capital is not necessarily a public good , as political relations based on clientelist ties in southern Italy, as well as crime organizations relying on reciprocity and obligations, are also forms of social capital . Following this economistic approach , some scholars have called attention to the role that access to rights and resources. Within this approach, social capital is seen as facilitating indigenous f the benefit of their member communities. Indigenous federations can assist member communities in a ccessing resources such as official recognition, legal titles for lan d claims, and funds for development projects, thereby enhancing local livelihoods (Perreault 2001). The formation of with state, market, and other civil society a ctors (Bebbington et al. 1993; Bebbington and Perreault 1999; Perreault 2003) , or even through participation in struggles for land and the formation of organizations, from which the relations of trust, interdependence, and obligation
27 necessary for collecti ve action are created (Perreault et al. 1998). The resources, entitlements, and opportunities that these networks bring to local communities can even contribute to reduce poverty and environmental degradatio n (Bebbington 1997). Despite their great influenc e in both academic and international development spheres, economistic approaches to social capital are not exempt from problematic aspects . Fox (1996: 1092), for example, asserts that social capital is more likely to appear under certain political condition s and not under others. The role of horizontal ties and organizations in the formation of social capital is largely conditioned by the political opportunity structure, which determines the possibility of forming alliances to get external support. Since col lective organizations are usually targeted for state repression, support from external alli es is also key to offset the threat of repression. Therefore, historical legacies of horizontal relations are necessary but not sufficient to accumulate social capit al. Historical, social, economic, and political variables need to be considered in any analysis of social capital. Similarly, Putnam do es not provide an explanation of how exactly social capital influences political and economic o utcomes. P erhaps more trou bling , he fails to differentiate between organizations that may provide different amounts of public good ( in his view, choral societies, political organizations an d violent militia groups are all equal expressions of social capital ) . Putnamesque approaches suffer from a voluntaristic point of view , stressing only positive outcomes of social capital. Also, Putnam tends to emphasize formal social relations over more informal ones that may contribute to the formation of social capital as much as civic associations (Fine 1999; Harriss and De Renz io 1997). Particularly, formal organizations such as Amazonian indigenous federations overlay deeply rooted forms of social organization such as kinship networks and traditional forms of leadership associated to them . In these cases, modern
28 organizational structures, perceived to be new forms of civic involvement , may mask traditional social networks and status hierarchies. A more use ful view of social capital needs to be sensitive to the heterogene ity within and among forms of civic associations and, as Fox (1996) emphasizes, to the relations among these organizations, and between them and the state. This approach requires a more nuanced view of social capital and of the actors and organizations wit hin which it inheres, paying attention not only to the forms of social mobilization, but also to the political and cultural contexts within which they are embedded. In this study, I will show how an indigenous federation may result from horizontal ties and principles of democratic governance and , at the same time , reproduce vertical relations such as patron client ties. Horizontal and vertical ties may coexist and are not necessarily mutually exclusive. Putnam and his followers have tended to produce analys is that largely overlook the social differentiation and power relations within which civic organizations and networks are embedded (Fine 1999; Harriss and De Renzio 1997). In the absence of a political economy analysis , enhancing cooperative capacity is only a matter of . T hose actors who have been able to accumulate higher stocks of social capital are more likely to achieve these positive outcomes than those with lower stocks. Under this logic, is interpreted not as the result of fundamental structural inequalities , but as the result of ack of social capital, a deficiency that should be remedied by building social capital endowments rather than with measures to change the social cond itions of inequality (Fine 1999; Harriss and De Renzio 1997). Given that social groups are differentiated in their access to economic capital and political power, what is the role of social networks in
29 reducing or widening these social inequalities? How do indigenous organizations as a form of social capital contribu te to reproduce existing class, gender and ethnic inequalities? From a political economy approach, French sociologist Pierre Bourdieu points to one of the most important theoretical lapses of do minant views of social capital: its lack of attention to the material, symbolic and political dimensions of social networks. Bourdieu (198 6) suggests that even though participati on in historically and culturally embedded networks (social capital) may allow people to accrue economic (material and financial assets) or cultural (symbolic goods, skills, titles) capital, given that endowments of economic and cultural capital are unequally distributed among actors, social capital tends to reproduce existing socia l inequalities or produce new ones. Bourdieu concludes that for this reason social capital accretion does not derive from shaped by the changing constraints of inst itutionalized and culturally embedded social relationships , and their dominant or subordinate position within those social structures. Studies following this line of inquiry suggest how engaging in social networks enables certain individuals and groups to secure more capital than others. For example, the types of networks that some micro finance programs promote do not empower women (Mayoux 2001). The indigenous mobilization for land rights in Madre de Dios is an example of how, while social networks may em power indigenous people in a socially beneficial way, they may also reinforce persistent social inequalities. This abilit y to accumulate or draw upon stock s of social capital is also contingent upon Bourdieu (1984) calls cultural capital : status and relative social power within a given social class. The use of social capital to acquire these benefits is possible as long as it is deployed
30 within a given socio cultural milieu in which that form of social capital becomes meaningful. It is no t enough to have a relationship; both parties must share an understanding of what that relationship means. Thus, from a Bourdieusian perspective, a weakness in the social capital literature is the tendency to portray indigenous organizations as always faci litating access to resources and velihoods . These authors present a quite essentialist view of indigenous movements that associates them only with d emocratic practices and economic development. This literature draws on a popu list rhetoric that overlooks the diversity within and between social movements and their potentially conservative aspects. As I will show through a case study of an indigenous federation in southeastern Peru, ostensibly grassroots social movements may give rise to elite s whose higher status create and accentuate social inequality between them and their constituencies. A more critical and nuanced analysis is needed to take into account not only the social relations underlying indigenous politics, but also ho w indigenous organization s articulate to deeply rooted authoritarian traditions . Central to such questions are the ways in which social networks are embedded in broader political, economic, and cultural processes . Cultures of Resistance Paying attention to the broader political economy structure and the cultural context leads us to another important aspect of indigenous political mobilization: the role of power relations and cultural patterns in shaping forms of mobilization. One of the th eoretical perspect ives that has addressed the issue of the underlying politic al and cultural factors of peasant mobilization is the everyday forms of resistance approach. Political scientist James Scott (1985, 1986, 1990), the major small covert acts of defiance against local elites , such as foot dragging, passive
31 non compliance, deceit, pilfering, slander, sabotage, and arson. By paying attention to long per iods of apparent political calm, larg ely ignored by students of peasant movements, Scott contends that pea precisely during uneventful periods. In this context, is a response to their inability or unwillingness to resort to open political protest. Despite their lack of organized protest, t hese acts are thought to reflect peasant defiance of the legitimacy of ex isting structure s o f domination (Scott 1986:6; 1990: 79 82). Therefore, the way peasantries domination depended upon how permissive the social structure was to resistance. The theory of everyday forms of resistance raises the quest ion of whether a theory formulated for peasant societies can be used to understand the cultural roots of political mobilization among Amazonian indigenous peoples, who have been characterized as tribal rather than peasant societies. In his classic work Pea sants (1966), Eric Wolf argues that what distinguishes peasants from the primitive bands or tribes more often studied by anthropologists is appropriated by a dominant g roup of rulers, the latter exchange their own labor and production. This conceptual distinction, however, can hardly be applied to the Amazonian case in the present conditions. While there still exist bands of hunter gatherers that avoid any contact with t he outside world, the majority of indigenous groups are now well integrated into larger political and economic systems. Brazilian anthropologist Darcy Ribeiro (1970), in analyzing the level of integration of tribal peoples into the Brazilian national socie ty, defined four stages of integration: isolation, intermittent contact, permanent contact, and integration. as the phase in which the natives have lost most of their lands and culture, and have become a
32 labor reserve for urba n centers. In the Peruvian Amazon, while the majority of Amazonian peoples are well integrated into the national society and market economy , they still conserve substantial parts of their culture, language, and land s. Theorists of the everyd ay forms of res istance have focused on the cultural universe that inform practices of resistance. In this respect , there is some variation amo approaches. Scott (1976) sees subsistence , and mil lenarian beliefs as important in the cultural underpinnings of peasant resistance . Kerkvliet (1990) , in contrast, prefers to see everyday resistance as more influenced by mo dern cultural notions such as citizenship and nation state. While he acknowledges that peasants may draw upon traditional cultural values and norms, Kerkvliet argues that peasant struggles may also be informed by the notion of basic rights (rights to human dignity and decent standard s of living) . In a similar vein, Nash (1989) describe s how miner protests in the Bolivian Andes combine primordial beliefs, customs and rituals with modern revolutionary and reactionary politics. The everyday forms of resistance theory has been criticized by its alleged neglect of social change and political organizations . Some scholars, especially Marxist ones, have deemed an on revolutionary struggles that would be conducive to social change (Brass 1991: 176; Gutmann 1993: 87) . Brass (1991: 74; 1996: style neo populist approach that pays too much attention to the idea of the land bound and backward looking middle peasant. It remains unclear, however, whether structural change is only a result of collective political struggles. Also, while it is unlikely that everyday resistance alone will generate structural change, it can certainly do so if combined w ith other factors. T heorists of the
33 everyday resistance do not necessarily see peasant s Scott (1986: 8) argues that hi dden resistance may ultimately contribute to destabilize the policies of dominant groups. Similarly, Guha (1989) points out that longstanding forms of peasant resistance prove to be linked to Indian nat ionalist agitation. The relation ship between hidden defiance and open forms of struggle is also little understood. There is no agreement among ex isting studies as to the causal factors. Adas (1986: 82), for example, suggests that protest through denial (suc h as foot dragging, feigned incompetence, or fleeing to remote areas) dampens social discontent and leads to the fragmentation of the peasants , rendering them more vulnerable to repression by elite groups . Protests of retribution (destruction of crops , a rson, etc.), in contrast, have more potential for open political struggles , although even in this case the organizational skills and ideological sophistication may constitute a serious drawback . P olitical protest needs open resistance in the form of non peasant, elite allies . S ome studies of peasant politics in Latin America have followed this line of analysis, arguing that u rban based political leadership has served as a condition for the transfo rmation of peasant unr est into pol itical movement s (Hobsbawn 1973 ; Landsberger 1974 ). Nonetheless, more recent studies of guerrilla movements in Latin America have described how numerous alliances between left wing political partie s and their would be indigenous supporters have been marked by t ragic misunderstandings (Brown and FernÃ¡ ndez 1991; Rojas Zolezzi 2008). At the same time, other studies have seen the success of peasant political struggle s in the combination of local culture s o f resistance and community leadership and or ganized political mobilization (Smith 1989). Indeed, Kerkvliet (1993: 481) argues that small acts of defiance can pave the way
34 for organized land takeovers, which sometimes are organized by external organizers rather than by local leaders. Theorists of the everyday forms of resistance have also sought to explore what factors other than the presence of urban political activists account for the transformation of hidden into open resistance. For Scott (1986), fear of political repression, along with social frag mentation and th e lack of economic alternatives like migration, is what explains why peasants chose hidden rather than open forms of resistance. Drawing on a case study in the Philippi nes, Kerkvliet (1993: 471) believes that what explained shift to open res istance is a process of national political democratization , which provided the peasants with minimum physical safety and interlocutors in the government. Yashar (1999 , 2005 ), in contrast, argues that neoliberal inspired democratic reforms have catalyzed indigenous movement s in Latin America not because they have provided political openness, but because they have threatened community autonomy. Other scholars assert that people usually engage in both overt and subtle forms of resistance, and that the repert oire of protest on which these groups draw is limited and heavily influenced by social structure and historic traditions ( Eckstein 2001) . Structural and historical conditions explain different forms of rebellion, such as refusing to work, loot ing , demonstr ating in the streets or turn ing to electoral sabotage. Strikes enable factory workers to defy rules of the and dwellings building more effective to protest against high cost s of living. Different for ms of defiance not only coexist, they have changed over time. Sociologist Charles Tilly ( 1978 ) arg ue s that in w estern Europe resistance repertoires have shifted from food riots, resistance to conscription, rebell ion against tax collectors, and organized invasion of fields and forests to demonstrations, protest meetings, strikes, and electoral rallies between the
35 eighteenth and nineteenth centuries. Part of these changes is the emergence of large scale, special pur pose organizations. Tilly (1978) find s the source of these changes in the processes of economic concentration, proletarianization, strengthening of the nation state, and institutionalization of liberal democracy that Europe experienced between the eighteen th and twentieth centuries. In Latin America, while new forms of defiance have appeared on the scene, old forms have persisted. Centuries old types of resistance such as food riots and rural land seizure coexist with more recent forms of protests such as s trikes, demonstrations, and protest meetings. This opportunities provided by dependent development, and the weakness of nation states. In Latin America n c ountries , with few exceptions, proletarianization has been partial and the majority of laborers continue to work in agriculture, are self employed or work for small scale paternalistic enterprises. Additionally, Latin American politics has generally been v ery unstable, oscillating between authoritarian and democratic regimes (Eckstein 2001:10 11 ). In this dissertation, I understand cultural models of indigenous mobilization as shaped by the political economy structures in which they are situated. In this way, the presence or absence of the state, and its repressive forces, shape the forms that indigenous resistance take. The Peruvian Amazon has remained in the margins of the state for centuries , enabling its native dwellers to openly resist integration. Ho w has this political autonomy influenced indigenous resistance in Madre de Dios? What forms of resistance have emerged under these political conditions? Theorists of the everyday forms of resistance tend to assume a priori that democratic openness offers m ore political space for more open forms of political struggle. As I will show in this dissertation, there is no necessary correlation between democratization and less state
36 violence. In this context, even under democratic regimes indigenous people may find it necessary to resort to hidden resistance. Methods To investigate the social networks in which the Arakmbut engage , the influence of these networks on indigenous political mobilization , and the cultural values and practices in which this indigenous acti vism is embedded, I used a diverse methodological approach. This is in accordance to my research questions, which demand an ethnographic and historical approach, and to my eclectic theoretical framework. My attempt to disentangle the cultural meanings boun d up in multiple social relations and practical and symbolic aspects of indigenous mobilization called for an ethnographic approach. At the same time, understanding the origin and evolution of social relations, practices and the symbol ic structures that ma ke them meaningful , ne eded archival analysis. I conducted a case study of the FederaciÃ³n Nativa del RÃo Madre de Dios y Afluentes (FENAMAD ) and one of its constituent communities, the Arakmbut c ommunity of Puerto Luz . I also made brief visi ts to other four Arakmbut communities: San JosÃ© del Karene, Barranco Chico, Boca Inambari and Shint uya. These rapid assessments were meant to be used comparatively to infer more general trends. Fieldwork I conducted the fieldwork for this dissertation from August , 2011 un til October , 2012. Preliminary research for this project was conducted between May and July, 2009 . I returned to Madre de Dios for an additional month of fieldwork in July , 2013. Fieldwork was divided between work in offices are headquartered, and the communities. While in Puerto Maldonado, I conducted archival research in the archives of Centro Eori de InvestigaciÃ³n y PromociÃ³n Regional, FENAMAD, the municipal l ibrary , and
37 personal libraries. Below I detail the resear ch methods of data collection I us ed and relate them to my research questions. Interviews Most of my data were collected through semi structured interviews. While I used interview guides to formulate the questions to my informants, I made open ended questi ons and tried to conduct the interview s as an informal conversation, giving my informants enough freedom for them to ex pand on any topic or refer to topic s not included in the guide. Formal interviews, that is, those that were scheduled with the informant, were recorded with a digital voice recorder, provided a I got the necessary permission. Nevertheless, I also conducted many informal interviews that were not recorded but wrote down on my field notebook. I conducted approximately 80 interviews, both with men and women, although most of my interviews with questioned. I used two types of interviews: ethnographic interviews and oral history interviews. Ethnographic interviews were used to collect data on question about the social networks in which the Arakmbut are involved. Thus, through ethnographic interviews I gathered data on kinship, godparenthood, and patron client relations, the mo ralities that underlie those social ties , the actors that are linked through them, and the economic activities and social contexts to which they are associated . Life history interviews were highly suited for addressing the question of how social networks influenced indigenous mobilization. Thes e interviews documented personal memories of the history of the community and FENAMAD, the building of interpersonal networks , and past participation in indigenous mobilization. In each case, I purposively selected community and federation leaders, individ uals knowledgeable about the history of their communities, and ordinary community members.
38 Household Survey I conducted a structured interview with the members of all households of the community of Puerto Luz. F or the purpose of this survey, I defined hous ehold as those people who eat from the same hearth . Thus , a household could include more than one nuclea r family and house . I surveyed a total of 57 households, covering a range of questions that included demographic aspects, education, and kinship relatio ns. The questions ab out kinship were intended to sup plement ethnographic data by gatherin g information about the clan affiliation of each community member. To carry out this survey, I had the assistance of two people from Puerto Luz, who helped me to gain access to families and community members that I never would have had alone, as well as to translate between Spanish and Harakmbut in the cases where the interviewees were monolingual . These interviews were not digitally recorded; rather, I wrote Participant Observation Interviews were complemented with extensive field notes I took on my observations of, and participation in, daily life activities in FENAMAD and the chosen c ommunities for this study. Participant observation was used to collect data on social networks and political mobilization. While in Puerto Luz, I observed and, as long as I could, participated in such everyday activities as hunting, fishing, gardening , gol d mining, chat gatherings, fiestas, and community assemblies in order to collect information on the social and cultural context of social networ k s . In FENAMAD, attending meetings, hanging out with federation leaders and staff, and helping the federation wi th document reviews , enabled me to collect data on forms of mobilization and the structure and organization practices of the federation. Either in FENAMAD or in the communities, part icipant observation wa s crucial to gaining awareness of gender roles and p ower relations .
39 Archival Research Finally, I networks, and shifts in direction and discourse. This work was intended to address questions concerning the political context in wh ich the federation was formed, the evolution of the alliance networks with researchers, NGOs and other local federations, as well as project orientations . Additionally, documents were useful in reconstructing the ways in which the federation h as embodied bureaucratic forms of managing resources and governing , as well as the ways in which it has discursively constructed notions of identity . The documents reviewed were consulted personal and public libraries such as the municipal l ibrary of Puerto Maldonado, and included assembly minutes, project proposals and reports, state ments, letters, bulletins, newspaper articles, and others . Structure of the Dissertation In Chapter 2, I provide an historical analysis of the inter ethnic relation s in the Madre de Dios region, from Inca times to the present day. The chapter beg ins with a discussion of the fi r s t attempts to colonize the region, which were carried out by the Inca in the fif teenth century in order to establish coca fields and extract gold from the region. Then I describe how the Spaniards , in their search for gold and lands for coca plantation s in the region , were not able to expand the frontier established by the Inca. In th e next section, I examine the change towards an economy based on the extraction of cinchona bark during the colonial period, and the impacts this had on the native population. I also analyze the rubber boom and the arrival of missionaries to the region sin ce the late nineteenth century and the way these processes greatly disrupted the social organization of the native groups . I finally discuss the latest economic changes that have articulated the region to the global demand of gold and oil, and how this has led to an unprecedented migration of mining settlers to the region. I argue that the way the region has been
40 articulated to wider ec onomic and political processes shaped the social relations between natives and outsiders, from trading networks during the Inca and colonial periods, to enslavement and debt peonage during the rubber boom and the missions, and to godparenthood and friendsh ip networks in present days . In Chapter 3, I describe and analyze the identity and livelihood strategies of the Arakmbut co mmunity of Puer to Luz . I start examining both commercial and subsistence production practices, explaining the social organization associated with each activity. I contrast market based gold mining with subsistence based activities such as gardening, hunti ng and fishing, highlighting the way in which they influence social ties within the community. I pay special attention to gold mining and how this economic activity is changing social relationships among community members and between and outsiders . I sugge st that social relations are closely linked to the economic system on which Arakmbut subsistence is based. H ouseholds are economically autonomous and cooperation among them is limited and restricted to members of kin groups. G old mining has introduced some level of monetary relations among kin, but since the means of gold product ion are relatively available to all, households have been able to remain economically autonomous . Chapter 4 examines Arakmbut sociality in the community of Puerto Luz . I begin by discussing kinship and friendship ties among the Arakmbut and how they are associated with the settlement patterns of the community . I also detail friendship, godparenthood and patron client networks the Arakmbut have built with outsiders . Special attentio n is paid to how social ties create and strengthen bonds of solidarity among indigenous people and between them and non indigenous people. I also examine the role of communal fiestas in strengthening solidarity at the level of the community , as well as the cultural underpinnings of Arakmbut leadership. I show
41 that Arakmbut sociality is comprised not only by bonding networks but also by other social ties such as friendship, godparenthood and clientelist ties , which also enable the building of bridging relat ions with outsiders . Conviviality cha racterizes the relations among close kin and friends, while relations beyond the kin group and with outsiders tend to be less convivial and more instrumentalist . In Chapter 5 , I discuss the influence of kinship, friend ship, godparenthood, and patronage networks on mobilization for rights and resources among the Arakmbut people, both at the individual and collective levels . I first examine the role of bonding networks (kinship and friendship ties among the Arakmbut) in facilitating community level mobilization. I then detail how the Arakmbut have relied on bridging networks ( friendship ) with anthropologists , NGOs , international donors, mes tizo organizations and other local and global actors to build multiscale organizations, gain legal rights to land , and get access to development projects . I also analyze the internal conflicts that bonding networks (goodparenthood and clientelism) the Arak mbut have forged with settlers and other non indigenous actors engender within the community. Finally, I present a discussion of the ways FENAMAD constitute s a field of power relations, contributing to reproduce gender and ethnic inequalities among indigen ous peoples. In this chapter , I argue that both bonding and bridging ties have varying effects on indigenous mobilization, the former being more effective at the community level, the latter being more relevant for large scale, organized mobilization. In Ch apter 6, I examine the cultural roots of indigenous political activism. I begin by evaluating how during much of the history of Madre de Dios, indigenous resistance deployed open forms of political struggle . Then I discuss how with the shift to more organi zed forms of mobilization, indigenous political culture came to include new forms of protest, including not
42 only open forms such as demonstrations , but most importantly hidden forms of resistance such as filing suits, appealing to international juridical f rameworks, and constructing collective identities. I contend that the shift from open to hidden forms of indigenous mobilization in Madre de Dios have been closely linked to changes in the broader political context . While the region remained in the frontie rs of the state, where the repressive state apparatuses were weak or non existent, indigenous people were able to resort to open and violent forms of political resistance. With the incorporation of the region in to the Peruvian nation state, indigenous mobi lization became more organized, relying more on hidden forms of resistance. Finally, Chapter 7 summarizes the major arguments of the dissertation and draws conclusions based on the data presented in the previous chapters . I argue that social networks and t heir solidarity morals are crucial in helping communities to build organiz ations, defend land claims, and get access to development projects. I further suggest that individuals also rely on networks to secure economic resources. The use of these networks as a social resource, however, engenders moral dilemmas between individual and collective interests and reproduces ethnic and gender inequalities. I contend that social ties and their underlying moralities may have both pro social and anti social consequences.
43 CHAPTER 2 MADRE DE DIOS: A HISTORY OF INTER ETHNIC RELATIONS Throughout the last five hundred years, the Madre de Dios region has attracted many outsiders seeking to exploit its na tural resource s and native dwellers . This led to the integration of the region in to broader political and economic systems. None theless, the way the region has been articulated to these wider processes has varied in nature, intensity, and geographical reach. What has th e impact of these varying articulation processes been in the way outsiders have interacted with the native populations? How have these forms of interaction evolved through time? In this chapter, I provide a historical overview of the processes of state exp ansion into the Madre de Dios region , and their impact on the relationships between outsiders and the natives. In order to examine current issues of indigenous political mobilization and how they are influenced by social networks, it is essential to consid er the hist orical processes that shaped inter ethnic relations in the region . I suggest that d epending on the forms and intensity of colonization processes, the native population reacted by resisting, avoiding any contact or establishing networks of trade and political alliance with the outsiders. Different broader political and economic contexts h ave shaped the ability of indigenous Amazonians to establish networks among them and with outsiders. This has depended on the degree of political domination and economic exploitation that colonizers have been able to impose on the natives. D uring most of t his history , colonization and resource extraction concentrated on the upper parts of the region , where interaction between outsiders and Indians was marked by horizontal political relations . Despite several attempts of colonization, the lowland rain forest s remained economically and politically independent from the rest of the country until the late nineteenth century, when this area was articulated into global economic processes through rubber exploitation . The way outsiders related to the natives shifted
44 to more intense forms of economic and political domination, shaping the basis for the rise of indigenous politics in the 1970s. This chapter is divided in to four sections that describe what I consider are the most important historical processes through whi ch the region has passed in terms of forms of articulation and associated social relations between colonizers and natives . In th e first section, I analyze how , despite their attempts to subjugate the lowlanders, the Inca ended up establishing only politica l alliances and trading relationships with the natives . In the second se ction, I examine how the Spaniards , also unable to conquer the natives, simply reproduced the spatial and social boundaries established by the Inca . T he third and fourth section s are d evoted to the discussion of the way global demand led to an intensification of resource extraction in the region, especially of cinchona bark and rubber, a nd how this led to more exploitative forms of domination. The fourth section examines the consolidation of state expansion into the region, and the rise in gold exploitation a nd oil development , which threatened indigenous lands and paved the way for indigenous organizing. Antisuyo during the Inca Rule The Inca called the mountainous and lo wland zones north and e ast of Cuzco Antisuyo (Region or District of the A nti) . Anti and C hunchu (or C huncho ) is how the Inca referred to the many tribal groups living in Antisuyo, who wer e distinguished from people of zones in terms of bo th appearance and behavior. 1 Because of its inhospitable environment and indomitable inhabitants, Antisuyo was never conquered by the Inca, remaining as the least 1 T h e Quechua only the Amazonian piedmont northeast of Cuzco . It appear to have referred to mountain ranges, Vilcabamba and Vilcanota, and a river called Antibamba, which was probably part of the Paucartambo R iver. Anti was al so how the AshÃ¡ninka Indians of c entral Peru wer e known (Renard Casevitz et al.1988) . In the colonial period, t he term , who used it interchangeable chunchu Spaniards applied the term to the Amazonian pi edmont of the eastern cordillera and its inhabitants more generally. Anti tended to be used by people in the Cuzco region, while further south the term chunchu or chuncho enjoyed greater popularity.
45 known of the four regions of Tahuantinsuyo, as the Inca empire was called . Spanish chronicler s from the sixteenth and seventeenth centuries, some of them relying on Inca informants, described the Amazonian piedmont and its dwellers as strange , barbarous, and warlike . The Inca themselves, and maybe other Andean highlanders as well, imagined Anti an d Chuncho peoples as savages whose very existence legitimized the Inca role as agents of civilization. In the Inca view, the Chuncho lacked proper dress, agriculture, and architecture. Nevertheless, descriptions of the Chuncho are more a reflection of Inca attitudes than about Amazon cultures. The Inca launched a series of explorations and military campaigns into the Antisuyo region , the most important of which took place in the decades preceding the Spanish invasion (Renard Casevitz 1981: 11 7 8). Th is expan sionist attempts probably responded to the Inca desire to acquire lands ( for raising coca) and valued products such as chonta ( Bactris setosa ciliaba , a dark colored hardwood palm ) , cinnamon, gold, salt, peppers, animal pelts, medicinal herbs and insects , and feathers . Paucartambo , a town located in the easternmost part of the Inca empire, was the gateway to the Antisuyo region . The town controlled not only its adjacent highland valleys but also the northern headwaters of the Alto Madre de Dios River (or Amarumayo , the serpent river , as the Inca called it ) , in the Amazon piedmont, which mark ed the beginnings of Antisuyo ( Renard Casevitz et al. 1988: 81 2). 2 It was to this region that the Inca launched their first expeditions to conquer Antisuyo. The major Inca military incursion into this area was carried out through the CosÃ±ipata and Carabaya valleys by Tupac Inca Yupanqu i between 1473 and 1475 (Lyon 1984: 7). Tupac Inca Yupanqui attempted to annex the headwaters of the Alto Madre de Di os, an area then known as Opatari and now referred to as CosÃ±ipata, where they tempor arily conquered several tribes (Renard Casevitz et al. 1988: 88). 2 The Paucartambo valleys, which coincided with the east ern border of Cusco, are the montaÃ±a area closest to Cusco.
46 Tupac Inca Yupanqui present day Bolivia, but fai led due to both internecine conflicts back in Cusco (a Colla rebellion) and, perhaps, the unrealistic expectations of the endeavor. The dense tropical vegetation and the dispersed yet socially connected native inhabitants foiled the military goals and stra tegies of the invaders. 3 Because of the politically autonomous and fragmentary nature of these tribes, the Inca had to conquer them community by communit y (Renard Casevitz et al. 1988: 88). Political , along with restricted occupation, bore more fruit and were the hallmarks of various circuits of interaction (Saignes 1985). Trade between the Inca and the Antisuyo tribes involved the exchange of highland goods such as metal tools (bronze axes and knives) and decorative articles for lowland items such as feathers, honey, cotton, chonta , palm, copal (tree resin), medicinal plants, among other goods. Nevertheless, while the low landers may have tended to see this exchange as trade, the Inca may have pressed to turn it into tribute . Nevertheless, labor with the provision of gifts, the relationship remained more in terms of trade than of tribute (Lyon 1984: 8 ). Trading laid the base for temporary alliances. Some Inca rulers even married Chuncho women to establish political alliances with the forest tribe s (Renard Casevitz et al. 1988: 86). Everyday contact between highland and lowland peoples appears to have been effective though fluctuating. Trade con sisted of regular yearly exchanges in which groups of lowlanders visited trading posts near the end of the dry season (i.e., from July to September) 3 Battles between the Inca and the Anti persisted in imagery well into the colonial period. Depictions of battles between the two groups can be found on all manner of colonial period wares, especi ally wooden cups called kero and textiles. MontaÃ±a warriors in colonial period works are readily identifiable by facial paint, jaguar skin tunics, and feather headdresses. Also, they use bows and arrows, which are traditional weapons of montaÃ±a inhabitants . In such ascends, the montaÃ±a warriors are signifiers of chaos, the agents of disorder who are properly defeated by the Inca so as to establish order.
47 probably both because it is easier to travel upriver and ford rivers at that time and because that season tends to be a slack period in the work cycle of montaÃ±a dwellers (Lyon 1984 ) . Despite their attempts to conquer the lowland s farther east , the Inca were able to control only the western and southern upper reaches of th e basin, where they established coca fields and extracted gold. E xcept for coca planta tions and a fort built in the zone known as Opatari, the Inca did not establish any permanent settlements in the CosÃ±ipata area (Lyon 1984:7) . As this area was dedicated only to the production of coca and subsis tence crops for the coca cultivators, the population that permanently remained in it probably did not exceed 400, certainly not an intensive settlement (Lyon 1978). Apart from coca plantations, most montaÃ±a area s wi thin the empire produced subsistence crops. Almost every area in the montaÃ±a produced some fruits, wood, coca and a little of gold, along with some luxury or specialty items that were exchanged in the highlands or paid as taxes . The montaÃ±as of the Urubamba Valley, Paucartambo, and Carabaya excelled in the production of coca and gold, and Carabaya apparently was the most important sou rce of Inca gold (Werlich 1968: 62). To establish the c oca plantations , the Inca removed all na tive inhabitants of the area and replaced them by colonists from the highlands (Lyon 1984: 7). Renard Casevitz et al. (1988: 90 1) argue that Matsigenka and Cashinahua were also employed by the Inca as cocacamayos , that is, as coca growers residing permanently close to the coca fie ld s . They also extracted alluvial gold from the headwaters of the Alto Madre de Dios River, and handed it to the Inca as a tribut e (Renard Casevitz et al. 1988: 91). Highland colonists, called mitimaes (or mitmaqkuna ) , were entire populations that were upro oted and moved to new locations by the Inca for economic and political reasons. Th e colonies were politically semi autonomous as they were governed by their own chiefs rather than by provincial authorities. Some of these colonies were in fact military -
48 agri cultural posts established at strategic locations along Antisuyo frontier to protect both highland and other montaÃ±a sett lements from Chuncho marauders (Werlich 1968: 58 60). Additional labor for the montaÃ±a colonies w as provided through the mita or tax service. T ax payers , or mitayos as the Spanish would call them later , were highland laborers who worked only for about three months, after which they returned to their homelands. Mita labor often supplemented that of colonists during peak periods of w ork such as harvest t ime and on public works. There were some montaÃ±a areas, however, that were exploited almost exclusively by mitayos , as was the case of the mines of Carabaya, which was worked primarily by levies of Colla Ind ians from the nearby highlan ds (Werlich 1968: 60 1). This limited settlement of the montaÃ±a and exchange facilitated the more stable and direct access of the Inca to coca , gold and other montaÃ±a crops (Murra 1986: 50). Several factors prevented the Inca from penetrating further into th e lowlands of Antisuyo. The ecological limits of coca growing are perhaps the most important limiting factor for Inca expansion into this region. Coca does not grow w ell below 2,400 meters above sea level, where it is too hot. Also, coca growers from the h ighlands were not able to adapt to the lowland conquest (Renard Casevitz 1981: 1 37 8). For Patricia Lyon (1984: 8), Inca occupation appears to have been limited n ot by altitude, as some writers have suggested, but rather by transportation. Inca occupation ended at a point where ground transportation turned extremely difficult and canoe traffic became a reasonable mode of communication. It was precisely in these fix ed limits where the Inca established frontier posts which could serve both as fortresses or trading posts. Moreover, Inca culture, including patterns of warfare, was based upon land transpo rtation, while water was considered dangerous and beyond control . O f course the Inca could have used canoe -
49 using allies to penetrate into lowland areas, but it is unlikely they would have based their plans of territorial expansion on popu lations they were not able to control. In fact, there was no reason to do so since th e only lowland product in which the Inca were truly interested was coca, which grows best in the upper montaÃ±a, a zone which they did control. Other goods were easily obtainable through t rade with the tropical lowland. Lowland chiefs and communities went up to the interethnic outposts to work for the Inca, attracted by the ir metal goods . T hey stayed in these colonies only for a short time, after which they left and were replaced by other chiefs and communities who came attracted by the same goods. In gener al, as I mentioned above, the Anti were against any form of tribute and servitude, so there were no conquered peoples but peoples who voluntarily and temporarily showed loyalty to the Inc a (Renard Casevitz et al. 1988: 91). When the empire attempted to brea k the status quo by placing colonists beyond the tacitly accepted frontier, the Anti united to expel the intruder s (Renard Casevitz et al. 1988: 94). We do not know with certainty whether the Inca tried to conquer or colonize the Antisuyo, but several Anti groups (AshÃ¡ninka and Matsigenka) response was to unite and prevent the Inca from ruling this regio n (Renard Casevitz et al. 1988: 94). The Cos Ã±ipata region was progressively abandoned when the Inca discovered more secure gateways to the south (Marcapata, S an GabÃ¡n, Inambari River ). After the Tupac Inca Yupanqui military expedition, there were no more Inca incursions into CosÃ±ipata . Taking advantage of the lack of Inca control, the Anti massively and progressively fled the Inca administratio n (Renard Casevit z et al. 1988: 91). In this context, the Harakmbut 4 maintained their political autonomy regarding the Inca and their various attempts of state expansion . Although 4 The Harakmbut speaking peoples are comprised by seven groups: Arakmbut, Wachipaire, Arasaeri, Zapiteri, Kisambaeri, and Toyeri. In this dissertation, I will use the term are common to all the Harakmb ut speaking groups, and the word features that apply specifically to this group.
50 they seemingly lost some land in the upper reaches of their territory, they limited Inca penet ration to trade relationships with highlanders. Spanish Conquistadors, Coca and Gold Sp anish conquistadors came to what is now Peru in search of gold, adventure and glory. Once the y had found the precious metal , these soldiers s tarted fighting among themse lves over the distributi on of the spoils, and searching for more gold. The quest for gold took the Spaniards to distant places , especially into the Amazon Basin. The Spaniards learned from the Inca about the montaÃ±a region and several myths about its citie s and empires ruled by powerful lords who were so rich that washed themselves with gold. It was also believed that the Inca themselves, fleeing from the Spaniards, had carried vast amounts of gold into the forests, where they fo unded flourishing kingdoms . 5 Thus, shortly after the conquest of the Inca, several expeditions (or entradas such as El Dorado and Paititi . These early entradas both economic and political motivations . They served as an outlet for conflicts among the conquistadors over the di vision of wealth. Expeditions to the eastern lowlands were used to reward allies or send enemies away, while losers avoided retaliation by escaping to the Amazon, which decreased pressure over the colonial spoils (Saignes 1981:164; Werlich 1968: 67 ). Between 1561 and 1677, there were a total of 17 entradas Madre de Dios Basin (Wahl 1987: entradas rt artillery captain, headed to the mountains east of Cuzco and reached the Tono River, an affluent of the Alto Mad re de Dios River (Werlich 1968: 69 ). Another 5 This last tale was partially true. Some treasure was carried to secret hiding places, perhaps in the montaÃ±a, before the Spanish sack ed Cuzco, and refugees did establish a neo Inca state in the remote Vilcabamba region, in the upper Amazon north of Cuzco (Werlich 19 68: 66).
51 entrada Ma ldonado, a veteran of thirty years in Peru who received a commission to conquer Carabaya. In 1568, Maldonado penetrated the montaÃ±a of Paucartambo by way of the Tono River, probably the route used earlier by Pedro de Candia, from where he travelled down th e Madre de Dios. He confluence with the Heath River, in the present day border between Peru and Bolivia (Werlich 1968: 127 8). All these expeditions into the lowlands al most invariably failed due to the difficulties of the terrain, the attacks of the Indian s, and most importantly, because the fabled rich lands they had been seeking were never found. Then, in the second half of the sixteenth century, the Spaniards establis hed coca and cane plantations, raised cattle and extracted alluvial gold on the Andean slopes of the basin, in the upper Inambari River ( Carabaya and San GabÃ¡n ), Marcapata, and CosÃ±ipata. 6 When the Spaniards took over this region, there were only a few Ant i and Chuncho, who became reluctant to pay tribute to the Crown, disappearing rapidly. The exchange of goods between the highlanders and the Amazonian nativ es, which had been practiced at the time of the Inca , was interrupted. Rather than continuing these trading relationship, t he Spanish unsuccessfully attempted to subject Amazonian lowlanders to permanent work. W ork ing without any reciprocity Casevitz et al. 1988: 92). The villages and coca fields were then occupied mostly by camayoc and mitimaes from other parts of the empir e (Renard Casevitz et al. 1988: 9 1). 6 During colonial times, and perhaps since the t ime of the Inca, the CosÃ±ipata V Pau the immedia tely adjacent highlands to the we st.
52 S hortly after the conquest , t he Spaniards occupied t he monta Ã±a of Paucartambo , which became an important source of wealth during the first years of Spa nish rule . Around 300 haciendas were established in the valleys of the Tono, CosÃ±ipata and Pilcopata rivers. Ca ptain Gascilaso de la Vega, father of the great chronicle of the Inca , was among the owners of these estates . The Spanish did not found any town in Paucartambo, where there were only Indian villages, and they rarely visited the area because of i ts difficult access. 7 The area valleys produced a great variety of crops maize, manioc, plantains , cacao, fruits, sugar , and cotton and some gold was extracted from local streams . The most important product of the area , however, was coca, which contemporary observers considered to be the most valuable in all the viceroyalty of Peru. 8 S he Spaniards greatly increased t he production of Paucartambo coca leaves , which supplied the market of Cuzco . Actually, most of the 1,500,000 pounds of coca l eaves that Cuzco shipped annually to the Potosi silver mines in the early sev enteenth century were produced in the coca fields of Paucartambo (Werlich 1968: 149 50). The Carabaya region, located at the southern Andean reaches of the Madre de Dios Basin, was rich in gold . Exploited even before the Inca period, Carabaya became the most important gold producing area in the early colonial period (Berthelot 1978 ) . The first Spanish settlement in the area, San Juan del Oro, located in the Tambopata Valley, was fou nded in 1557 or 1558. The mining camps, however, much as today, had to import all their supplies and goods in order to survive. While trade greatly increased in the area, Carabaya remained subject to fighting among miners and the assaults of its surroundin g lowland inhabitants while its costs of 7 During colonial times, residents of most of the Spanish towns in the highlands exploited the warm trans An dean valleys to the east, but very few settlements were established within the monta Ã± a itself (Werlich 1968: 121). 8 In the Amazonian piedmont, from Carabaya northward to the valleys below HuÃ¡nuco, coca was for centuries the important cash crop. It was light in weight, high in value, and there was a ready market for it throughout the Andean area (Werlich 1968: 143).
53 living continued to rise the further one proceed ed in the search of gold. The risk of being attacked by the Chuncho was constant . At one time, San GabÃ¡n , another Spanish town, was completely destroyed by the fierce Sirineyri tribe (Werlich 1 968: 137 140). By the beginning of the seventeenth century, most of the monta Ã±a entered a long period of decline in which the frontier settlements were depopulated and became impoverished. The causes of this decline were many, inc luding high costs of transportation, resource depletion, Chuncho uprisings, and Indian depopulation due to forced labor and devas tating epidemics (Werlich 1968: 132 3). In some areas of the montaÃ±a, the absolute population loss was not as high as in the hig hlands . Early censuses for Paucartambo and Carabaya reported that the montaÃ±a population swelled with the addition of large numbers of Indians who escape d labor service under the Spanish mita in othe r provinces (Werlich 1968: 136). P opulation loss also expl ains, albeit in part, why the Spanish frontier failed to further move eastward during th e first two centuries of colonial rule. With low population pressure in the highlands, and most of the products with high market value coming primarily from the upper r eaches of the Amazon, t here was little economic incentive to push further into t he lower montaÃ±a (Werlich 1968:137). The Paucartambo haciendas were especially vulnerable to labor scarcity and , by the mid seventeen th century, they had also been severely af fected by competition from other provinces. particularly v ulnerable to the labor shortage brought about by the general population decline following the Spanish conquest (Lyon 1978: 21 2). In addition to labor shortage, the harsh conditions under which highlanders had to work accounted for the decline that set in during the eighteenth century. Attacks by lowla nders may have played a role in this decline, as is testified by some several raids and killings committed by Indians between 1576 and 1770 (Lyon 1978: 7). But, overall, most of the native s
54 had been displaced either to lower parts of the basin or to other parts of the empire (Lyon 1978: 21). Moore (1985 a : 75) contends that the rise and fall of b oth gold exploitation and the coca haciendas depended much more upon the prices of goods in the international market than on the attacks of the natives. The Carabaya mining centers began to decline as early as 1570. Yields decreased rapidly while the costs of operation remained high because of the lack of trails or maintenance when ex istent, high costs of fi ghting (Werlich 1968: 142). By 1570, as labor became ever more difficult to obtain, th e mitayos formerly assigned to Carabaya mine operators were diverted to the PotosÃ mines , while the encomenderos and corregidores of Collao began sending their experienced Indian miners and surplus food to Pot osÃ rather than to Carabaya (Werlich 1968: 142). Only a few Indian mita yos were available for work in Carabaya, and only for short period s (Werlich 1968: 143). For the following decades , gold would be extracted by local producers and a few highlanders who penetrated into the area at certain times of the year to extract gold with which to pa y their tribute (Martinez 1969: 79). After the decline of the Carabaya mines, the center of gold extraction moved to Marcapata. Nevertheless, new activities like harvesting of cinchona bark ( cascarilla ) added to the econ omy of the region , although the pr oduction of coca and sugar cane w as still in practice in Paucartambo, Carabaya and Marcapata (CORDEMAD 1986:32) . In the eighteenth century, several naturalists entered the region to explore new sources of cinchona bark. Go ld was extracted in Marcapata well into the eighteenth century, when the area was depopulated after a huge landslide (CORDEMAD 1986:33).
55 Unlike other parts of the Amazon, during t he period prior to the turn of the eighteenth century, missionary work in th e southern frontier was small and isolated, and hence the natives were neither concen trate d in missions nor religiously co nvert ed (Wahl 1987: 99). The montaÃ±a of Paucartambo was the area of eastern Peru most neglected by missionaries, probably because the Spaniards were able to control the coca fields (Werlich 1968: 240 1) . During this time , evangelization depended upon alliances the native lowla nders sought to establish with the Spaniards against enemy tribes. These alliances depended on mutual visits and gift giving. Nevertheless, the lowlanders did not feel obliged to return missionary gifts wi th labor as they would break alliances when they we re no lon ger needed (Saignes 1981: 175). In general, there was little reason to maintain continued interaction with native lowlanders. When existed, relations with the natives took place in relatively equitable terms, to the extent that no means of coercion existed or could be developed (Wahl 1987: 101) . In addition to the production of coca leaves and the extraction of gold, the incursion of individuals or groups fleeing from the Spanish colonial system, as well as missionaries seeking to establish religious posts , into Harakmbut territory had also some impacts . Nonetheless, it is not clear the nature and intensity of these impacts. For example, it is unknown the extent to which native people were affected by the cultural trait s or diseases the outsiders carr ied with them. They may have caused some impacts on indigenous politics by disrupting internal processes of alliance f ormation and feuding, as access to k nowledge of outside diseases or politics, and goods or services, may have become a source of influence for some groups . T his help s explain why some outsiders were welcome and others were not (Wahl 1987 :79 ). What is clear is that raids by lowlanders appeared with the initial disruptions of trade relationships with highlanders and the arrival of the Spanish (Wahl 1987: 72 3).
56 In general, in the Harakmbut , the colonial period did not bring about profound changes in social relations within and among native groups. It can be argued that some Harakmbut groups refused to engage in colonial trading and defensive posts , while some outside groups (mostly highland Indians) entered the ir territory fleeing from the pressures of the colonial systems . But, generally, Spaniards kept the same spatial and social boundaries developed by the Inca for bot h the production of coca l eaves and extraction of gold . The se activities largely involved the same populations and areas that were part of the Inca system, which may explain why this area did not have the same priority that led Spaniards to seek the missio they relied on elsewhere to inc orporate lowlanders (Wahl 1987: 100). Cinchona Bark and the Nation State At the turn of the nineteenth century, coca growing and gold mining had almost disappeared. This situation slightly changed with the ind ependence from Spain in 1821, when new gold miners, colonists and missionaries arrived in the Marcapata region . None theless, gold mining did not reach the importance it had in the early days of Spanish rule. Rather, the colonization process was mainly base d on agriculture, cattle raising, logging , and cinchona bark harvesting (CORDEMAD 1986:34) . In the 1850 s , harvesting of cinchona bark significantly in tensified , as the international demand of this product, especially from industrialized countries like Gre at Britain, increased. During those s debt to Great Britain with gold and cinchona bark from the Amazon . The Inambari and Tambopata river valleys, tributaries of the Madre de Dios to the south , were rich in cinchona trees . W ith the aim to explore new sources of gold and cinchona bark, several scientists, military and other explorers were sent out to the low lands of this region . Between 1860 and 1861, Faustino Maldonado, a cinchona bark trader originally from Loreto, descended the Madre de Dios River and reached the Madeira River, in
57 whose dangerous rapids he dr o w n ed . Some of these explorers established contact with native gr oups of the region, in some cases with violent consequence s (CORDEMAD 1986 ). lowlands also had geopolitical reasons. As conflicts over boundaries arose with Colombia, Ecuador, Bolivia, and, most serious of all, Brazil, the successive governments of the nineteenth century struggled increasingly to develop means of geopolitica l control. While several international treaties had been signed throughout this time, a growing number of complementary mechanisms were developed in the effort to consolidate the boundaries of the newly created nation state. The Amazon region was politically broken up into several departments. Expeditions were sent out to explore the means of communication that might articulate different regions to the rest of the c ountry. Steamships were purchased to offset the ever growing Brazilian control of riverine trade. A number of colonization, immigration, and land grant laws were progressively peoples into the nation state turned into a systematic government concern (Wahl 1987: 118 9). The wars of independence resulted in a political power vacuum, which gave rise to the rebellion of several native groups and their withdrawal from trading relations with outsiders. many parts of the Amazon (Moore 2006). In this context, conflicts that were claimed to exist with native peoples turned out to be a product of opportunistic efforts to master the military support required for the penetration of new areas. T he influx of Bolivian coca and the population decline that followed the w ar s of independence affected even more t he Paucartambo haciend as , which almost disappeared (Lyon 1978: 21 2). It is at this moment that the Wachipaire were blamed
58 for destroying 300 haciendas in Cos Ã±ipata . Nevertheless, t here is no evidence that a single hacienda was des troyed in this way. If the number of haciendas declined, it was largely due to the withdrawal of highland resources and labor from the area. What is certain is that no such a large number of haciendas ever existed in the area, not ev en during peak production years (Lyon 1978) . Although the Wachipaire have, through time, resisted various incursions, it appears that in this case they simpl y occupied those haciendas that had alr eady been abandoned (Wahl 1987: 114 15). In contrast, the Harakmbut in the Carabaya area appear to have attacked some settlements and caused significant uncertainty on the frontier. Although some of these attacks included areas where gold was extracted , gold mining was not the main cause of indigenous incursions or revolts in the area. It appears, rath er, that the growing search in Harakmbut territory for c inchona bark may have caused the uprisings . This extractive industry required that both extensive territories be covered in search of new sources of bark and that laborers who mig ht do this be found ( Wahl 1987: 115). While a number of highland laborers were engaged in cinchona bark gathering through debt peonage relationships , this economy further required not just seasonal work but also the expensive import of foodstuffs . This was so because the territ orial displacements in which cinchona bark harvesting was based precluded the development of settled agricultural production . The cinchona bark of this area must have been an especially intensive and cumbersome enterprise. It is not at all surprising that in response to this , and the search for gold, the previously economically isolated Harakmbut respo nded with hostility (Wahl 1987: 115). The Peruvian state never had the funds required to fully finance on its own missionary endeavors in the lowlands. Yet, al ong with cert ain local sectors of society, the state did provide
59 organizational and infrastructural support that contributed to the development of such efforts. Correspondingly, in those areas where state structures remained unconsolidated, missionaries pl an ideological and material level. Yet, they, like the state, remained limited by the economic developments underway in those areas where they tried to foste r alternate structures (Wahl 1987: 119 20). Rubber Boom and the Mission The late nineteenth century marked a renewed interest of state and private actors in the native people and natural resources of the Madre de Dios region . A bo om in rubber extraction th at had already started in other parts of the Amazon reached the region. The Madre de Dios lowland s were part of a broader Amazonian region rich in rubber trees that attracted thousands of prospectors seeking the so British imperialism, which at the time was at its peak, also penetr ated the region. British investors became interested in the cinchona bark, gold and rubber of the region . Missionaries entered the region and started a systematic process of contact with an d proselytization of the natives . These processes changed dramatically the way the natives had been interacting with outsiders , from trading relations to enslavement and missionization. In Peru, rubber was first extracted in the northern Amazon lowlands. I n the south, r ubber extraction started later and was carried out by Bolivian and Brazilian rubber tappers along the Madre de Dios River and several of its affluents (Wa hl 1987:194). They transported the rubber to Brazil by a hazardous route down the Madeir a and Mamore rivers . As areas of rubber in the northern areas were depleted, rubber traders ventured into southern forests in search of new sources of rubber, first into the Ucayali River basin and ultimately into the Madre de Dios Basin. One of these pion eers was Carlos FermÃn Fitzcarrald , the richest and most powerful rubber baron
60 in the central lowlands of Peru . In 1894, Fitzcarrald used an isthmus that connects the U pper Mishagua, located in the Urubamba River basin, with the upper Manu River, a tributa ry of the Madre de Dios to the north , a new route for rubber transportation to Brazil , which promised to considerably lower the costs of transporting rubber through Iquitos. K o portage his small steamboat, divortium aquarium with the help of a thousand AshÃ¡ninka and Piro slaves (Reyna 1942 :43 5 ), a heroic deed that was brought to the movies by German filmmaker Werner Herzog. Rubber traders not only took over indigenous territories but enslaved their native dwellers. As many other indigenous groups in the area, the Harakmbut were affected since the first years of the rubber boom. Long before the arrival of Fitzcarrald in Madre de Dios, Bolivian and Brazilian rubber tappers decimated several indigenous groups. Large numbers of both higher market value (Wahl 1987:194). A fter crossing the isthmus that bear his name , Fitzcarrald and his men were Indians , probably Piro and Toyeri groups, in the upper Manu River (Gow 2011; Shepard et al. 2010) . To brake the resistance, Fitzcarrald attacked several Mashco vill age s, burning their huts, gardens, and canoes (Reyna 1942 :48 ). The main method used by the rubber traders to recruit indigenous labor was the infamous slave raids ( correrÃas ), which consisted in attacking villages by surprise and kidnapping men who then we re enslaved. Thousands of Indians were recruited through correrÃas and forced into labor for rubber patrons. Groups of Shipibo Conibo and Kichwa Runa, from the central and northern Peruvian Amazon respectively, were also brought into the area as slaves to rubber traders (Rummenhoeller 1984, 1988) .
61 Foreign capital also entered rubber extraction in the Madre de Dios region through the United States owne d Inca Mining Company, which purchased the right to mine gold along the upper Inambari River. By 1896 the c ompany was offered two million acres of land along the Tambopata River for building a trail from Tirapata , in the highlands of Puno , to a navigable point on the Tambopata. At that time, the Peruvian government started granting land concessions to any compa ny that would build roads and river ports, as a way to encourage the building of infrastructure in the remote areas of the Madre de Dios Basin. The Inca Rubber Company, a subsidiary of the Inca Mining, completed the trail by 1908. During the first decade o f the twentieth century, the rubber of Madre de Dios flowed through this trail, up the Tambopata River, towards the coast, from which it was exported (CORDEMAD 1986:50) . Com mercial firms from Arequipa such as Rickets, Braillard, W. Gibson, and Casa Forges, all of them linked to English capital, financed rubber extraction in Madre de Dios during the boom (CORDEMAD 1986:61) . It was in this context that Madre de Dios was incorporated into the state administration in 1901, and Puerto Maldonado was founded in 19 02, in the confluence of the Tambopata and the Madre de Dios rivers (CORDEMAD 1986:51) . Thus, after several centuries of isolation, the discovery of the isthmus of Fitzcarrald opened up the Madre de Dios lowlands to rubber exploitation and articulated it t o the global economy of rubber that developed with the expansion of the automobile industry in the United States and Europe. Fitzcarrald drowned in the rapids of the Upper Ucayali in 1897, leaving Bolivian rubber baron NicolÃ¡s SuÃ¡rez as the only master of the Madre de Dios region (Fifer 1970) . In this context, the Peruvian government sought to en sur e sovereignty over this region by sending scientific expeditions to exp lore navigable routes and issuing concessions for rubber exploitation
62 over vast areas to t he north and south of the Madre de Dios River (CORDEMAD 1986) . Peruvians such as Ernesto Rivero and Carlos LeÃ³n Velarde, and Spanish such as Bernardino Perdiz, Domingo Tronc oso, and MÃ¡ximo Rodriguez , were granted concessions in the Manu, Las Piedras and Tahuamanu rivers , and soon became the emergent elite of the recently incorporated Madre de Dios region (CORDEMAD 1986:56) . Most of them complemented rubber extraction with cattle raising and some intro cachaza distilled spirit made from sugar cane juice. The number of rubber traders increased between 1900 and 1910, expansion that first revolved around caucho rubber ( Castilloa elastica ) and then, as this species was depleted due to destruction of the tree, around shiringa rubber ( Hevea brasiliensis ) and other varieties (CORDEMAD 1986) . The Amazon rubber boom did not last long. It came to its end in the mid 1910s w ith the rise of rubber plantations in the British colonies of Southeast Asia. The Iberia estate in the Tahuamanu River, property of MÃ¡ximo Rodriguez, was one of the few rubber estates that survived the collapse. As other parts of the Amazon Basin, in the s outhern Peruvian Amazon Brazil nut begun to be harvested and traded in order to face the crisis. Madre de Dios had extensive Brazil nut patches and rapidly articulated to the export economy of this product. Other products such as the rubber like balata ( Ma nikara bidentata ) and animal pelts were also exploited, while cattle raising and agriculture continued to be practiced. Hunting was initially practiced to meet the food needs of rubber tappers but in the 1920s it intensified as it was directed to the comme rcialization of skins and pelts, threatening the local fauna (CORDEMAD 1986) . As in other parts of the Amazon, the rubber boom had disastrous consequences for the indigenous populations of Madre de Dios. Thousands died due to murder, enslavement and the
63 ne w diseases brought in by the rubber tappers. The first incursions of Fitzcarrald into Harakmbut territory in the Upper Manu had no indigenous resistance, but when he established permanent camps and looked for indigenous labor, the Harakmbut refused and vio lence ensued . After learning that one of his outposts had been raided by the Mashco, killing more than fifty rubber tapper s , Fitzcarrald ordered his men to massacre an entire Mashco village in retaliation (Reyna 1942:86) . The few people who survived fled t o the forest, abandoning horticulture and taking up gatherers who today are known as avoiding any contact with outsiders, are almost certainly the descendants of these groups (Shepard et al. 2010). As a result, indigenous population decreased dramatically. It was estimated that 60% of the native population in the Manu River died due to disease and maltreatment (Von Hassel 1904:244). The Toyeri and Arasaeri, the two Harakmbut sub groups mos t affected by slave raids, were reduced by 95% between 1894 and 1914 ( Gray 1996:225 ). The headwater tributaries, however, remained as redoubts for some indigenou s groups such as the Arakmbut and the Sapiteri, who this way avoided forced incorporat ion into the rubber production. The rubber boom broke up existing native trade networks and caused inter tribal wars among Harakmbut groups. It has been reported that an independent indigenous rubber trade existed in the Manu area long before Fitzcarrald rediscovered the isthmus in 1893. Traditional exchange networks among native groups were also impacted by indigenous access to Western goods, while differential access to these goods created power imbalances among native groups (Wahl 1987:197 8). At the same time, s into Arakmbut and Sapiter i territories. T hese incursions, and the shortage of resources that followed, unleashed aggressive responses that in turn provoked a series of warfare activities
64 within and among Harakmbut groups (Gray 1996 ) . The loss of trade networks a nd the intensification of inter trib al war among indigenous people exacerbated pre existing ethnic rivalries that would last until the present days. The decline of the rubber economy , however, enabled many indigenous peoples to return to their traditional homelands and way of life. The regio n would keep in relative isolation again through the 1920s, which proved to be beneficial for the indigenous people, who were able to recover from the demographic collapse caused by the rubber holocaust. Upon arriving in Madre de Dios in 1902, Dominican m issionaries established contact with and started proselytization work among several indigenous groups of the region. They organized expeditions on their own but also were guides and interpreters on explorations intended to find and exploit gold and other r esources in the region. The Dominicans operated their missionary work through several missions in which they gradually brought together the indigenous groups they contacted. In order to get financial resources for the mission system, the Dominicans associa ted with rubber barons and companies, although the priests were strongly opposed to the atrocities perpetrated during the rubber boom. The first mission, San Luis del Manu, was opened on the Upper Madre de Dios in 1908, and was sponsored by a rubber baron, Bernardino PerdÃz (Gray 1996:243) . Throughout the 1920s and 1930s, the ja and Arasaeri on the Madre de Dios, Tambopata, and Inambari rivers. Contact with the Arakmbut increased in the mid 1930s and would continue over the next two decades. The m ission of San Miguel de Kaichihu e, which was established on an affluent of the Inambari River in 1943, incorporated many Harakmbut sub groups (Toyeri, Pukirieri, Arasaeri, Sapiteri and Kisambaeri) and served as a base for all the exp editions into the Arakmbut area. The Arakmbut were finally contacted
65 through numerous missionary incursions between 1950 and 1952, and settled in the missions of Pantiacolla and Shintuya, in the Alto Madre de Dios River. The Pantiacolla mission, on the Pal otoa River (a tributary of the Alto Madre de Dios River) was founded in 1954, but after being flooded in 1958 it was moved to its current location in Shintuya, on the mouth of the Shintuya River, also a tributary of the Alto Madre de Dios (Gray 1996) . At t he beginning, t he relationships between the missionaries and the indigenous peoples were based on gift exchange. The Dominicans used trade goods such as knives, machetes and cooking pots to make contact with the indigenous peoples and attract them to the ir mission posts , taking advantage of the indigenous attraction towards those highly desirable goods. Some of these goods were even dropped from airplanes that flew overhead. Some Arakmbut groups moved to get closer to the source of these trade goods, while others travelled as far as the Kaichihue River to attack the mission and raid knives and machetes. Since outside diseases spread out after contact, killing hundreds, the Arakmbut also made their way to the Shintuya mission in search of food and Western med icine (Gray 1996). The Dominicans ran the mission station as a large community farm, with a primary school, medical post and a church. In Shintuya the missionaries concentrated three hundred and fifty Arakmbut and Wachipaeri from different communal house groups or malocas (Torralba 1979: 94). This constituted a major change in the Harakmbut settlement pattern, previously characterized by small groups from fifty to a hundred people living in each maloca . The economy of the mission was based on cattle raisin g, agriculture and logging, and the Harakmbut had to provide labor for these activities, receiving in return clothes, education and medicine from the priests. The priests marketed any surplus produced in order to prevent outside middlemen from exploiting t he Harakmbut, thus acting as intermediaries between the Harakmbut and the
66 outside economic system. The priests also handed out the trade goods or gave support for lumber, cattle or coffee cultivation only to indigenous people who behaved correctly, engende ring resentment among those w ho were not favored (Wahl 1987: 268). Conflicts between the different Arakmbut maloca groups and with the Wachipaire arose due to the highly concentrated settlement pattern and competition for access to resources such as money, goods, women and education, tensions that took the form of sorcery accusations (Wah 1987: 2 72). As a result, Arakmbut groups fled the mission and made their way to their homelands between 1969 and 197 4 (Gray 1986), where they founded the settlements where m ost of them currently live: San JosÃ© del Karene, Puerto Luz, Barranco Chico and Boca del Inambari. The Wachipaire and some Arakmbut remained in the mission, which has persisted to the present days. In the 1940s , the Madre de Dios region experienced a new r ubber boom and an increase in gold mining. The region had remained isolated until World War II, when access to the Asian rubber plantations was blocked by Japan. This caused a new increase in international demand for rubber, which in the Peruvian Amazon wa s met through the CorporaciÃ³n Peruana del Amazonas , a state agency created by an agreement between Peru and the United States which took on the production and commercialization of rubber (CORDEMAD 1986) . The corporation introduced air navigation by building several landing strips, connecting Madre de Dios to the rest of the country. During these years people entered the region in order to mining gold too. In the 1950s, international demand for Brazil nut increased and as a result several trade compan ies of Brazil nut started operating in Puerto Maldonado. These companies introduced mechanical methods of peeling the nut (CORDEMAD 1986) . A subgroup of the Arakmbut, the Kipodneri, were contacted in 1957 by missionary linguists of the Summer Institute of Linguistics (SIL). After staying in Shintuya only for six
67 months in 1958, they returned to their homeland in the upper Karene River, where they founded the commun ity of Puerto Alegre ( Moore 1981 ). SIL missionaries provided lay medical attention and vaccine s to the natives, who in exchange sold gold and pelts to the missionaries. The missionaries then took the items with them to sell in Pucallpa, where their central mission (Yarinacocha) was located. With the money earned, the missionaries bought trade goods (machetes, axes, shotguns, Western clothing, flashlights, etc.) that they sold to the Arakmbut upon returning to Puerto Alegre. This system instituted a monetary economy within the community as the Arakmbut were well aware of the difference between the pr ice of their products and that of the trade goods they obtained in exchange. Cash was also used in their occasional transactions with highland miners who at that time were starting to arrive in the area. Money, however, was used almost exclusively for luxu ry goods, and subsistence continued to be derived from the forest, rivers, and gardens, although traditional tools (bows, arrows and stone axes) were being rep laced by modern ones (Moore 1981 ). Gold Rush and Oil Boom The p ast four decades marked an unprece dented rise in colonization and resource extraction in the Madre de Dios region . While resource prospectors, especially rubber and cinchona bark gatherers, started to colonize the region since the turn of the twentieth century, it was not until the 1970s t hat, due to a rice in the international price of gold and the implementation of liberal reforms in the extractive industries, Madre de Dios has been massively colonized by mining settlers. This period also witnessed the arrival of both mining and oil trans national corporations. This period experienced gold extraction at a scale never seen before , and the beginning of a new extractive industry in the region, oil development. The consequences for the indigenous peoples have been disastrous, as many communitie s saw their territories invaded, resulting in several conflicts with mining settlers and corporations. These conflicts
68 eventually led the indigenous people to organize themselves for the pursuit of legal titles for their lands. In the 1930s and 1940s, the region of Madre de Dios experienced an increase in gold mining. Gold miners settled in this region in succes sive migratory waves. They penetrated into the region first through the Inambari River in the 1930s , and continued to the Kaychihu e, Huepetuhe, and Pukiri rivers by 1943, reaching the Karene River, the heart of Arakmbut territory, by 1955 (Pacuri and Moore 1992). The number of miners, however, remained low until 1965 , when the building of a r oad connecting Cuz co to Puerto Maldonado was complet ed, impr the international market through the ports on the Pacific coast. The new road allowed an increased migration from th e highlands, especially from Cuz co, Puno , and ApurÃmac. Brazil nut harvesting and logging also experienced an increase. But it was not until the boom in gold prices in the early 1970s that landless Andean peasants massively migrated to Madre de Dios, attracted by the gold bearing beaches mostly l ocated in Arakmbut territories. The natives themselves also started to mine gold within their own territories. B etween 1972 and 1974, the price of gold in Madre de Dios tripled due to an extraordinary rise in the international price of gold, and an increas e in state gold productivity in the region. This increase in gold mining profitability not only encouraged the Arakmbut to mine gold within their own lands, but also increased the migration of Andean settlers to the region, especia lly to the Madre de Dios River and several of its tributar i es . The government promoted gold mining through the Banco Minero (Mining Bank), which became the only legal purchaser of gold and provider of tools, food supplies , and other provisions at subsidized prices (Moore 1979) . De spite the state promotion of g old mining through subsidies and prospection of new
69 mining areas, the number of mining settlers remained relatively low. The main mining areas remained under state control and only those who had been previously registered by t he Banco Minero were able to mine gold in those areas . As the Banco Minero provided tools and supplies at subsidized prices with which independent traders were unable to compete, an increase in the numb er of traders was put in check. Th e Arakmbut sold gold to and obtain ed trade goods from the Banco Minero . Also, in 1973 the government banned the commercialization of pelts of animals in danger of extinction, which had previously been an important source of income for the Arakmbut (Moore 1979). Since gold min ing provided the Arakmbut with more goods and more rapidly, it became an important part of their livelihood, to the point that now the Arakmbut people define themselves as miners. Also, the relationship between the mining settlers and the indigenous dwelle rs was one of peaceful coexistence. The relative harmony was such that the Andean miners taught the natives some mining techniques such as the use of mercury to separate the gold from the sand . This situation changed in 1978 when the military government of Francisco Morales BermÃºdez , of liberal orientation, passed a new mining law (Law 22178), which reestablished granting of mining concessions and suspended the Banco Minero an effort to serve the interests of large national and transnational mining companies, the new law gave small miners (highland and indigenous ones) only 30 days to apply for a mining concession with a preference right. Since the small miners were not able to fulfill this requirement, they lost access to the mi ning areas they had been previously exploiting. As a result, almost all indigenous lands were petitioned as mining concessions by medium sized miners and large mining companies, which started to arrive in the area. Additionally, due to a new rise in gold p rice in the early 1980s highland mining colonists started to arrive in the Arakmbut territory by the
70 thousands. The number of these small miners increased from 1,000 in 1972 to 20,000 in the early 1980s. They, however, were pushed to informality or to work for the mining entrepreneurs (Moore 1983, 2003; Pacuri and Moore 1992). One of the m ain consequences of the incorporation of gold mining into Arakmbut livelihood was the monetization and commodification of economic relationships. The Arakmbut began to get involved in monetary relationships, both with other indigenous and with non indigenous people (Gray 1997b). Some Arakmbut became patrons, employing highland workers under wage labor relationships in gold mining activities. T hose who had more outside worke rs and made larger profits than others sold capital materials such as extra tubes for pumping water in the gold placers. Working for money also became common. For example, s ome people began making money by offering services such as making wooden furniture for others or hiring out their chain saws to cut wood for houses. To some extent, payment for services or goods introduced some aspects of individualism in the communities (Gray 1997b). Another effect of the introduction of gold mining was an increasing re liance on commodities. Several commodities such as salt, sugar, and cooking oil, which in the past were along with meat, manioc, and plantains. Gaining money has thus become an essential motive in daily labor (Gray 1997b). The effect of this increased use of money (or gold ) as a med ium of exchange was to place fixed values on certain commodities. Small stores started to sell drink and groceries for money or gold to other Arakmbut and foreign miners appeared in some communities. These stores institutionalized the exchange of commodities at fixed rates (Gray 1997b).
71 The earliest attempt to drill for oil in Madre de Dios goes back to 1920, when International Petroleum Company (IPC), a subsidiary of Standard Oil of New Jersey (now Exxon), started conducting geological explorations in the region (Moore 1996). After several decades of inactivity in Madre de Dios, IPC resumed oil exploration in this region in 1967, in an ar ea on which the Manu National Park is now located (Moore 1980). Between 1970s and 1980s, several transnational oil companies ( Company , and Royal Dutch Shell Petroleum Company) arrived in the area to co nduct seismic exploration (Moore 1980, 1984 a, b ) . This intensification of oil exploration resulted in several conflicts with the indigenous peoples inhabiting the area. In 1967, while in a reconnaissance mission, a group of geologists escorted by Per uvian military had a violent encounter with a small group of isolated indigenous people , probably Yaminahua . 9 Twenty Indians were shot to death (Moore 1980, 1996) . Ande s Petroleum Company displaced several Toy eri families from an area the company had chosen as a base of operation in the Madre de Dios River. Cities Service built several facilities within the Do minican m ission of Shint uya in the Upper Madre de Dios , and the A rakmbut community of Puerto Alegre, in the Karene River , where the SIL missiona ries had built a landing strip . There was fri ction between oil workers, who numbered more than a thousand people, and nati ves due to theft , aggression to native women, disagreement in barter and commercial transactions, alcohol consumption, and presence of prostitutes (Moore 1980). In 1984, during the preparations for a visit of President Fe rnando Belaunde Terry to the Shell 9 Indigenous peoples in voluntary isolation are groups that chose to remain isolated from the outside world. They descend from people who survived enslavement during the Amazon rubber boom and retreated to the most remote parts of the forest, where they have remained to this day. The majority live in the southeast of Peru and maintain a nomadic or semi nomadic lifestyle, moving across large areas to hunt, fish and gather.
72 exploration zone in the u Mashco Piro years before repelled Fitzcarrald, attac ked with arrows and bows a crew of heliport construction workers. Members of the Peruvian Navy who accompanied the workers responded to the attack with gunfire from helicopters, killing two Indians (Moore 1984a, b ). on gave a new impulse to oil development in the Amazon by liberalizing even further the energy and mining sector. In 1993 government passed the Law of Organic Hydrocarbons (Law 26221), which eliminated state controls on and encouraged foreign i nvestment in oil development. These reform attracted t he arrival in Madre de Dios of a consortium of three oil corporations (Mobil, Exxon and Elf), which in 1996 signed with the government a seven year contract to explore hydrocarbo ns in blocks 77 and 78 ( Kraft et al. 1997). Block 78, located in the Tambopata and Karene river regions, covered 1.5 million hectares, affecting the proposed area for the Amarakaeri Communal Reserve and the Tambopata Candamo Reserved Zone, both located over ancestral territories of the Harakmbut isolated indigenous peoples. The presence of this consortium and its investment on oil exploration in the area halted for a while the establishment of the Amarakaire reserve. The Mobil led consortium withdrew from Madre de Dios in 1999 , after drilling an exploratory well. A year later, a temporary reserved zone for the Amarakaeri reserve was established. The new governments of Alberto Paniagua and Alejandro Toledo, more open to indigenous demands, offered a favorable context for the creation of the communal reserve, which was finally established in April 2002 , after 10 years of petitioning and several mass demonstrations in Puerto Maldonado. With 40 3,814.28 hectares, the reserve was created for the use of the surrounding Harakmbut, Yine, and Matsigenka communities. That same year , the
73 Territorial Reserve for Peoples in Voluntary Isolation of Madre de Dios was also created. Encompassing 829,941 hectar es, this reserve was intended to protect the isolated Yura and Kugapakori Indians of the northeast of the Manu National Park. The political instability that followed the disagreements between Fuj Shell al gas deposits, in the lowlands of Cuz co, discouraged major foreign investments in the Amazon. This situation changed in the 2000s when presidents Alejandro Toledo and Alan Garcia openly supported the return of oil corporations. In 2005 the government gra nted two transnational corporations the rights to explore oil and natural gas over three blocks in Madre de Dios. Texas based Hunt Oil Company was granted a concession over hectare Block 76, while SAPET, a Peruvian venture of China National P etroleum, These medium size corporations had together concession rights over about half of the territory of the Madre de Dios department (53.6%). As in the case of Mobil in the 1990s, these new concessions greatly overlaid indigenous terr itories of the region, which caused conflicts with native communities and FENAMAD. Block 76, for much of the Amarakaeri Communal Reserve , while blocks 113 an d 111 cover ed part of the Reserve for Peoples in Voluntary Isolation of Madre de Dios, territory of three native communities , and even part of the city of Puerto Maldonado. After pressure from FENAMAD, however, in 2006 SAPET a greed to withdraw from Block 1 13, while the government removed Block 113 from those offered to oil companies . Hunt Oil, however, continued oil exploration and by the time I conducted fieldwork in Puerto Luz, it was preparing to open three exploratory well within the Amarak a eri reserve.
74 Conclusion In this chapter , I have presented the historical context against which we may understand current processes of indigenous politics in Madre de Dios. I hav e attempted to demonstrate not only that the current structures of power and domination in which indigenous people are embedded have deep historical roots, but also that the forms of indigenous mobilization and the role of social networks in them have depended on broader political contexts and the changes they have experienced through time. Whil e the economic activities that outsiders have undertaken in Madre de Dios have articulated the region to broader economic processes, their i mpact on the native peoples has varied. Apart from the Inca, whose main interest in the region was to gain lands for growing coca, the economies that outsiders carried out in the region were based on an extractive logic, with little or no interest in developing the region or benefitting their local populations. Unlike the coastal and highland region s , which w ere complet ely subjected to colonial domination, the Madre de Dios region , and in general all the Peruvian Amazon, remained politically and economically autonomous through the late nineteenth century. Despite their several attempts to subjugate the Madre de Dios lowl ands in the fifteenth and sixteenth centuries, the Inca and Spanish were able to annex only the upper parts of the b asin, in the montaÃ±a areas of CosÃ±ipata, Carabaya and Marcapata. Even in the areas that were under the control of the Inca and Spanish, the native population resisted attempts of domination and remained largely independent from external forces for much of its history. During this period, the natives were not even affected by the epidemics of Western diseases that decimated aboriginal populatio ns in other parts of the country. The economic systems that the Inca and Spanish established on Madre de Dios had no significant effects on the native populations. The Inca established coca fields and extracted gold
75 from the area , but instead of utilizing native labor, they brought Indian laborers from the nearby highlands. Dur ing the c olonial period, the Spanish largely reproduced this system. T he coca fields of CosÃ±ipata were the primary supplier s of coca leaves for the Cuzco ma rket , while the Carabaya mines produced gold to pay tribute to the Spanish Crown. These activities did not use native labor, but mobilized highland Indian laborers. Under these economic conditions, indigenous social organization , as well as trading and political networks among tribes, were not disrupted . In this context, political alliances enabled indigenous people to organize collective resistance and ensure political autonomy . This political autonomy in turn allowed indigenous people to interact with the o utsiders through more horizontal relations such as trading. I argue resistance to external domination. This situation dramatically changed in the ninetee nth century , when the region was articulated to processes associated with the increasing global demand of newly industrialized countries such as Great Britain and the United States for natural resources. Cinchona bark gatherers, rubber traders, and gold pr ospectors arrived in the area, interacting with the natives in more vertical ways . Most of the Madre de Dios indigenous groups were not contacted until the late nineteenth century, and some of them, like the Arakmbut , were contacted as late as the 1950s. R ubber traders, in particular, enslaved, displaced and murdered large numbers of native people. At the same time, Dominican missionaries arrived in the region and, after contacting previously autonomous indigenous groups, concentrated them in mission posts. These were not direct representatives of the state, but in many ways they allied with the state to pave the way for future state expansion into the area. Resource extractors and missionaries had larger impacts on
76 the native people in terms of a drastic re duction in population, due to enslavement and epidemics, threats to local livelihoods , and violent conflicts. In the Amazon region, the major effects of epidemics took place when indigenous peoples were concentrated in missions, which in the case of Madre de Dios were not established until the early twentieth century. The impact of these changes on indigenous social organization was dramatic. Some groups were literally wiped out due to enslavement, maltreatment, and diseases. Resistance became more violent , which in some cases led to inter tribal alliance . But , in general, indigenous alliance and trading networks broke down and conflicts among indigenous groups accentuated. Several groups, fleeing the rubber holocaust, or searching for the Western goods brought by rubber traders or missionaries, penetrated into the territories of the ir traditional enemies, which increased competition for natural resources and tribal tensions . Spatial c oncentration in mission posts also caused serious disruption of indigen ous social organization. The Arakmbut abandoned the use of malocas and their initiation rituals , while they started to relate wit h the missionaries through debt peonage . Traditional ethnic tensions accentuated due to concentration of several maloca groups . As we will see in C hapters 5 and 6, the break down of indigenous social networks, as well as engagement in new ones based on clientelist ties, would have a large impact on the indigenous ability to mobilize. Unlike the previous years, the second half of th e twentieth century witne ssed an unprecedented rise in state expansion and migration of colonists into Madre de Dios. The state increased its presence by building highways that connected the region to the rest of the country, and promoting resource extract ion, especially gold and oil. The rise in the international price of gold not only triggered the invasion of indigenous lands by thousands of mining settlers, but also represented the opportunity for the Arakmbut to engage t hemselves in the market and avoi d debt
77 peonage relations with outsiders. In C hap t er 3 , I will examine the impact of gold mining on the Arakmbut traditional economy and social organization, and explore the potential effects this may have on indigenous politics.
78 Figure 2 1. Ma dre de Dios Basin and surrounding areas .
79 Figure 2 2. Natural protected areas and territorial reserve for isolated peoples .
80 CHAPTER 3 THE ARAKMBUT: IDENTITY AND LIVELIHOOD When in March 2012, FENAMAD allied with the FederaciÃ³n Minera de Madre de Dios (FEDEMIN) to launch a paro minero mining strike ), many people criticized the decision. The protestors were against recently passed legislation that banned gold mining outside a legal mini ng corridor . The law was intended to stop decades of environmental destruction, for which the m iner s represented by FEDEMIN , most of whom are illegal, are blamed . Attacks a gainst FENAMAD came not only from environmentalists, but from other indigenous organ ization s like the ConfederaciÃ³n de Nacionalidades AmazÃ³nicas del PerÃº ( CONAP ), one of the two umbrella networks of indigenous federations in the country . FENAMAD, however, defended indigenous Arakmbut e conomy has relied on gold mining since the 1970s, when the government set up a branch of the Banco Minero in Boca Colorado , at the mouth of the Karene River , in order to promote gold production in the area . Due to the high price of gold, gold mining has remained as a very profitable activity, becoming the most important source of cash. Money offered t he Arakmbut the means to acquire the goods brought by the outsiders, such as machetes, shotguns, and Western c lothing, and upon which they had become dependent. Over the years, the Arakmbut have become depend ent on money not only for buy ing prestige goods, but for thei r subsistence. The Arakmbut need money to buy food, clothing, and pay the ir education , and gold mining is for now the only relatively reliable source of cash. In the present conditions, the Arakmbut are certain that if gold prices dropped and gold mining declined, they would not be able to survive. How has the incorporation of gold mining i mpacted on the Arakmbut social organization? Have the Arakmbut s t opped being Arakmbut because of these changes?
81 Some students of the Arakmbut argue that although articulation to the national economy has introduced some changes, their essential social struc tures remain intact (Gray 1997a; Fuentes 1982; Moore 2003). Gray (1997a) contends that while increased involvement in monetary relations has introduced some levels of individualism and social inequality in the communities , social relations among the Arakmb ut have not been fundamentally affected because the goods and services that are bought with money come from the outside, while the internal distribution of goods remains untouched. Similarly, Moore (2003) points out that while Arakmbut individuals involved in gold mining employ highland workers under monetary relationships, labor relationships with other Arakmbut follow kinship and clan relationships. Also, even though the Arakmbut use their profits from gold trade to purchase motor pumps and other equipmen t and consumption goods, generosity remains as the main pattern of resource distribution , preventing capitalist accumulation from occurring . Fuentes (1982) asserts that access to means of production (land) and reproduction (women) still depends on belongin g to a kin group. The penetration of gold and money, however deep and violent, seems not to have disrupted Arakmbut social structures. There are, however, other authors who suggest that market integration may actually have the effect of forging reciprocal relationships instead of simply disrupting them. Questioning the classic dic hotomy between personal and non personal relationships a s the main division between non market and market economy, Plattner (1985) asserts that market relationships ca n also depend on personal, long term, open ended and generalized relationships, with a strong affective, non instrumental component. These relationships are economically preferable to the i mpersonal transactions because markets are usually characterized by imperfect information about goods, tran sactions and traders. Long tern trading pa rtnerships, where economic
82 self interest and general ized reciprocity intermingle, are intended to cope with the uncertainties of the market. In this chapter, I focus my attention on the identity and livelihood strategies of the Arakmbut. The chapter is divided in to two sections. I n the first section , I exami ne how the identity of the Arakmbut has changed over ti me. I argue that while the Arakmbut have partly constructed their i dentity based on primordial cultural traits such as kinship ties or language, for the most part they ha ve constructed their identity through their contact with outsiders and larger economic and political processes . In the second section , I describe and analyze the subsistence and market oriented economic practices of the Arakmbut , and the social relations associated with them . I pay special attention to the regional gold rush in which the Arakmbut are currently involved , and how this is affecting their social rela tions and identity within and outside the community . I suggest that while gold mining has affected some aspects of the Arakmbut social organization, identity has not d isappeared and old social relations have not been totally disrupted , while new one s have been created . This is no t because of an alleged indigeno us resistance to change, but because of the highly a tomized nature of Arakmbut social organization and the cons tr ucted nature of their identity. Place and Population The tropical forests of the Madre de Dios River basin is located in the southeastern Peruvian Amazon. The region, which co vers an extensive area of 85,183 km 2 (8. 5 million hectares), is dominated by th e eastern slopes of the Vilcanota mountain range to the southwest , Oriental, and, farther east and north, the lowland rain fores ts. The altitude ranges from 3,967 meters above sea level in the Andean mountain range to 160 meters in the Amazon lowlands. According to the last national census conducted in 2007, the region has a population of
83 approximately 112,814 people, out of which about 60,000 people live in Puerto Maldonado, the capital city (INEI 2008 a ) . Approximately 4 ,000 people (3.6 % of the region ) identify themselves as indigenous people (INEI 2008 b ) . The Madr e de Dios Basin is home to nine ethnic groups belonging to five linguistic families: Har akmbut (Harakmbut or Hate), Matsigenka and Piro or Yine (Arawak), Amahuaca, Yaminahua and Shipibo Conibo (Pano) ja and IÃ±apari (Takana), and Kichwa Runa (Quechua) (Moore 2003). The Shipibo Conibo and Kichwa Runa were brought into the area during the rubber boom at the turn the twentieth century . This population is currently distributed in 31 native communities that occupy 5.5% (430,750 hectares) legal rec ognition and titles granted by the state. More recently, two other groups have appeared sporadically: the Yora (a Panoan group also called Nahua) and the Mashc o Piro (probably an Arawakan speaking people). The indigenous group known as the Harakmbut is th e largest of the seven indigenous groups of Madre de Dios. The term Harakmbut refers t o the language spoken by about 2 ,000 people (INEI 2008 b ) . According to Patricia Lyon (1975), the Harakmbut speak a language called HatÃ©, which she considers to be indepen dent from the other languages of the region and consisting of several dialects. Anthropologists have divided the Harakmbut speaking people into six groups: Arakmbut, Kisambaeri , Wachipaire, Arasaer i, Sapiteri and Toyeri (Califano 1982). At the time these g roups were contacted by rubber traders and missionaries, they occupied an area of 600,000 hectares, stretching between the Alto Madre de Dios and Inambari rivers (Gray 1997a) .
84 The Arakmbut are the most numerous group, perhaps because they were contacted by Dominican missionaries as late as the 1940s and, because of that, were not affected by the rubber boom enslavement and epidemics as severely as the other groups. The Wachipair e are the second largest group, although their numbers were decimated by a smallpox epidemic in 1948, during the construction of the road to Cuzco (Lyon 1975: 225). The Arasaeri, Kisambaeri, Sapiteri and T oyeri are the smallest groups since only a few famil ies managed to survive the ravages of the rubber and gold booms as well as intensive missionary activity between 1900 and 1940. All these groups are now distributed in 12 native communities (Figure 2 1) . The Puerto Luz community has a population of 357 peo ple, according to a census I conducted between 2011 and 2012. Since 1980, t he settlement is an officially recognized native community . Puerto Luz is the largest native community in Madre de Dios, with a demarcated territory of 56,873 hectares, out of which 38,784 hectares of cultivable lands are titled . 1 The community is located in the District of Madre de Dios, within the Province of Manu. In 2012, the community had an elementary school, a kindergarten and a health post. Puerto Luz is inhabited by indigeno us peopl e who identify themselves and are identified by others as Arakmbut people; but who are the Arakmbut? How are they different from other indigenous Amazonians? Identity Norwegian anth ropologist F r edrik Barth ( 1969) argues that ethnic identi t ies are defined according to changing circumstances. This means that ethnic boundaries are not fixed, but change depending on factors such as personal attraction, tactic needs, common interests or moral 1 In 1978, the 1974 Law of Native Communities was replaced by a new law that granted native communities legal titles only for cultivable lands. Since the enactment of this law, community titles have not included forest lan ds, which can be used but owned.
85 obligations. Barth does not reject the existence of or primordia l ethnic traits, but he believes that these characteristics can be used or discarded to construct ethnic identities, depending on their importance for meeting tactic needs. Barth also suggests that ethnic identities depend not only on h ow a group defines itself, but also on how other s define them. Drawing on these ideas, I suggest that how the Arakmbut define their ethnic identity depends not so much on their primordial cultural and social features, but on their interaction with outsider s . According to Moore (1975), the terms used to refer to the different Harakmbut speaking groups are not terms they used to refer to themselves. The Harakmbut, in fact, do not have terms to refer to themselves. Instead, they use the term nuestros paisanos countrymen). The names of the six Harakmbut speaking groups are geographical terms of reference used by each group to refer to These names did not have historical continuity since the Harakmbut groups change d their location very often and intermarried. For example, the homeland of the Arakmbut was the Zapite mountain, from which they migrated into the headwaters of the Ishiriwe and Karene rivers ( Califano (1982) . At the time they were contacted, the Arakmbut were divided into two groups: the Wandakweri, who lived in the Wandakwe River in seven communal houses or malocas , and the Kipodneri, who lived in five malocas on the Kipodnue River, a parallel affluent o f the Karene River (Gray 1997a: 18). The Wandakweri l ive now in the communities of Shintuya, Boca Inambari, Barranco Chico , and San JosÃ© del Karene, while the Kipodneri live in the community of Puerto Luz. 2 Furthermore, t he name of a group changed with the place of birth of an individual and it was not relat ed to historical descent. For example, the Arakmbut of Puerto Luz refer to those of 2 Thomas Moore (1975) contends that the Kipodnue River groups were in fact two groups: the Kareneri and the Kipodneri. The former were the first group to move from the Kipodnue to the Karene River in 1952 and 1955. The latt er moved to the Karene in 1956 and 1965. In Puerto Alegre, each group occupied a different section of the village, the Kareneri living in the center of the village, and the Kipodneri living in the lower part of the village.
86 San JosÃ© del Karene as Shintuyeri or Shintuyanos because the younger generation was born or grew up in Shintuya and not in the headwaters of the Wandakwe. Other names used did not refer necessarily to a specific place, such as Wandakweri for those who lived on the Wandakwe River, but were applied to several groups in accordance to their location with respect to the other groups. which m does not refer to any group in particular; only to those who live downstream from wherever one is . Nevertheless, Toyeri became the term used to refer to the group that once lived on the Madre de Dios River. Arakmbut indigenous groups, although the term can also refer to native people who are not considered Harakmbut either because they are very different or because they are hostile t o the Arakmbut. Arakmbut people were considered enemies of the Arakmbut and were often at war with them. In pre contact times, Taka used to attack malocas in dawn ra ids to kill the men and capture or steal the women (Gray 1996). The term seems to have appeared during the rubber boom of the turn of the nineteenth century, when groups of Toyeri, Arasaeri, and Sapiteri, fleeing from the rubber gatherers or other Harakmbu t groups, penetrated into Arakmbut territory seeking refuge, leading to an unprecedented period of fighting between groups as a result of the shortage of resources. Although the term has fell into disuse, some Arakmbut may still use the term to refer to in dividuals who are identified with groups with which the Arakmbut used to be at war. In Puerto Luz, for example, a man who came from the Chilive River was called Taka. Throughout history, Harakmbut speaking groups have been known by multiple names. Many o f these names were put by the outsiders who first contacted them, such as the Inca ,
87 Spanish c onquistadors, rubber traders, missionaries , and anthropologists . Thus, the Harakmbut speaking peoples have been successively named Chuncho, Opatari, Mashco and mor e recently Harakmbut. Chuncho (originally Chunchu) is how the Inca used to call the tribal groups living i n the forested areas east of Cuz co. The ter m had a derogative connotation as it literally means and it seems to have been used to specifical ly designate the native groups living in the headwaters of the Madre de Dios River, particularly the Wachipaeri of the CosÃ±ipata Valley . Over time the term generalized to include all native groups of the Amazon and now is part of the folk vocabulary. In ac cordance with its original meaning, Chuncho is also used to refer to any shy person or with no education or manners, regardless of their ethnic origin. I still remember adult person. The term Chuncho and its associated culture also became part of the Andean folklore in several parts of Cusco, including the famous V irgen del Carmen festival in Paucartambo , which includes some dance companies that represent the Chuncho. Pau cartambo was the easternmost Inca town, and had a close contact with the native peoples living in the adjacent Alto Madre de Dios River area. During the colonial period, the Wachipaire were called Opatari by the first Spanish conquistadors who ventured in t o this area. There seems to have been some large settlement known as Opatari, which took its name from the people of the region. Nevertheless, whether this was an indigenous settlement or an Inca outpost used by the conquistadors is unclear. In the late co lonial period, another term, Mashco, was added to the list of names used to refer to Harakmbut speaking peoples. The term, apparently of Conibo origin, has been mainly used to refer to Arawakan speaking Piro. Lyon (1975) suggests that Mashco was the name g iven to an Arawakan speaking Piro group that in the late nineteenth century was living in the Isthmus
88 of Fitzcarrald. It appears that the word was first used by Franciscan priest Manuel Biedma in 1686, probably in reference to a group of Piro found in the Mishagua River, an eastern tributary of the Ucayali River (Biedma 1981). The term, however, came to be used to refer to any isolated or warlike group in Madre de Dios. Then , Dominican missionaries working in Madre de Dios started to use Mashco as an ethnic denomination for Harakmbut speaking Arasaeri and Arakmbut, groups who are completely unrelated to the Piro (Califano 1982) . Shepard et al. (2010) suggest that Harakmbut speaking Toyeri were among the Mashco decimated by rubber baron Carlos Fermin Fitzcarr ald in the upper Manu River in 1894. The term Mashco was used throughout the nineteenth century and until quite recently it was used instead of Opatari to refer to the Harakmbut. The use of local terms by outsiders to name native groups at the moment of co ntact has being s encountered, outsiders asked the natives who they were, and the nat ives, confused by such a question, which did not make much sense for them, merely answered by giving them the terms names, who often provided pejorative terms. In the late 1960s, a process of indigenous organizing started all around the Peruvian Amazon, which was accompanied by a process of identity construction. Indigenous groups appropriated and, in many cases, questioned and reformulated the ethnic names they wer e given in colonial times , a process in which anthropologists and linguists studying indigenous Amazonians played an important role. In Madre de Dios, scholars clarified that the Mashco were in fact several groups (Amarakaeri, Wachipaire, Sapiteri, Arasaer i, Toyeri, and Kisambaeri) with
89 a distinctive language (Califano 1982; Lyon 1975; Moore 1975). In many cases, indigenous groups changed their names when they thought them to imply a derogative connotation, as happened with the term Mashco, which the Harakm but considered to be an insult. The Harakmbut of the Kipodnue and Wandakwe rivers preferred to be called Amarakaeri because this term better described their geographical location before contact with outsiders. But when a version that equated Amarakaeri wit h a term of abuse used by the Wachipaire and the Matsigenka gained strength , it was changed for Arakmbut. English anthropologist Andrew Gray, who conducted field research in San JosÃ© del Karene, narrated that in 1991 92 the elders asked him to refer to the would reflect in his published ethnographies about the group, in which he used Arakmbut instead of Amarakaeri (Gray 1997a). 3 (Lyon 1 967 ), its meaning may vary according to which people are being contrasted. In its narrowest meaning, the term is used to distinguish the Arakmbut from the other Harakmbut speaking peoples. When I asked my informants in Puerto Luz what distinguish ed them fr om the other Harakmbut they mentioned primarily variations in the language. In its broadest meaning, the term Arakmbut can be applied Amazon (Gray 1997a: 121). Thus, a person is still considered an Arakmbut even if one of her parents identifies with other indigenous ethnicity. This is the case of some people in Puerto Luz, who were born to Matsigenka and Arakmbut parents. 3 Cases of native Amazonians cha nging their names are many . One very well known c ase is the one of the Campa of c entral Peru, who changed their name for AshÃ¡ninka. These cases also include those in which the spelling of the names was changed to more accurately reflect the pronunciation i n the native language, as in the name es demand non indigenous people discard their past names, which had a colonial origin, and adopt the correct ones.
90 In contrast, all the H arakmbut a miko to refer to all individuals who are not Harakmbut, in particular, and not native Amazonians, in general. Accordin g to Califano (1990), the term a miko is of relatively recent origin, dating back to the first ye ars of the Republican period (second half of the nineteenth century), although Harakmbut mythical tradition situat ed its origin at the time of the contacts with the Inca. The term seems to have derive d from the Spanish word amigo (friend) , which is what ou tsiders shouted to the Harakmbut in order to establish a peaceful contact. Currently, the Harakmbut distinguish between two main categories of a miko. One includes white Peruvians coming mainly from the coast, and highlander Peruvians coming from the Andes, the latter distinctively known as w ahaipi to a gente b lanca Harakmbut indigenous people, regardle ss of their ethnicity. Thus, even Andean highlanders are commonly referred to as gente blanca. I suggest that this term also originated during the first years of the Republican period, when most of the explorers, military, rubber traders, and missionaries who first contacted Amazonian groups were white people. These more primordial identities do not preclude the addition of others, more situational identities that do not necessarily erase the pre existing ones. In 1995, Andrew Gray returned to San JosÃ© del Karene for a brief visit and found the community involved in a conflict with illegal miners who had established themselves in the confluence of the Pukiri River with the Karene, coming to invade their community territory and plunder their resources. Seeing themselves as different
91 from Peruvians seems to have been a common feature of indigenous identity. 4 Yet, during the same period, people in San JosÃ© were alarmed about the possi ould be invaded by Ecuadorian armed forces. That year, an armed conflict between Peru and Ecuador over national borders had broken up, and some Arakmbut young men who had been in the military as conscripts were saying that the ir country was in danger and would join the military forces to fight for Peru. In the first context, the Arakmbut clearly distinguished themselves from the Peruvians; in the second, they identified as Peruvians. differentiation from Peruvia ns does not seem to be the case anymore. Along with their more primordial indigenous identities, the Arakmbut clearly identify themselves as Peruvians, as I was able to confirm in Puerto Luz. One Sunday afternoon, all adult males in the residence group whe re I lived gathered to watch on TV a soccer game between Peru and Colombia. The match was part of the qualification competition for the World Cup, so there was great expectation. I myself, as a long life soccer fan, joined the group. While watching the gam e, the Arakmbut were cheering so enthusiastically for the Peruvian team that it occurred to me to ask them if they considered themselves to be Peruvians. The expression of confusion in their faces was a reflection of the meaningless nature of my question. On another occasion, I asked a young Arakmbut whether he thought of himself as a Peruvian. He Peruvians too 5 This may be a more recent characteristic of indigenous identity, one that results from their demands to have the same political rights as any other citizen of Peru. 4 Smith (1996 ), for example , d escribes how the YÃ¡nesha of c entral Peru considered Peruvian s as different from them. The YÃ¡ nesha were even dubious about the human nature of Peruvians because of their aggressive and abusive behavior . 5 Peru and Chile were at war between 1879 and 1885, wa r in which Peru was defeated. Since then, a sense of revanchism towards Chile is inculcated in schools and the armed forces.
92 of social change to destroy Arakmbut culture, but because the Arakmbut constantly recreate their identity. Actually, since their incorporation into modernity, many cultural traits of the Arakmbut have been lost: the use of malocas, body painting, puberty r ites, songs, myths, etc. As we will see in the next section, the main economic activity among the Arakmbut is now gold mining, an activity that was unknown to them until the 1960s. The Arakmbut of today bear little resemblance to the Arakmbut of the time o f contact. Nevertheless, despite all these changes, the Arakmbut have chosen to keep identifying themselves as Arakmbut. This process of identity construction will play an important role in political mobilization. Livelihood Like other Amazonian indigenous groups, the Arakmbut of Puerto Luz have a mixed gardening) with a relatively more recent activity: gold mining. As they became increasingly dependent on money and other commodities, the Arakmbut started to mine gold within their communal lands, thus incorporating gold mining as a key component of their livelihood. Both kinds of activities depend on the territory ( wandari ) and are equally important in securing a livelihoo d for the Arakmbut. Traditional activities provide families with food, building materials, and medicinal plants, which are directly obtained from the forests, rivers and gardens. Gold mining is the main source for cash income, although families can get som e money by selling cash crops such as manioc and plantain s . With the money earned through the sale of gold ( wakupe ), the Arakmbut buy products that are not produced locally, such as manufactured food, capi tal goods, and beer . Very few work timber since the majority lacks the economic capital necessary to buy the equipment and pay laborers that this activity requires.
93 Within the household, most daily activities are performed both by men and women , although some have marked gender divisions. Hunting is almost a strictly male activity, but women may hunt some small game species. C ooking, basketry and weaving are carried out only by women. In other activities, there are certain tasks that belong to the domain of women and others that have to be done by men. The male head of household, his married sons and sons in law provide the household with game meat. It is the female head of the household, often with the help of her daughters and daughters in law, who is in charge of distributing the meat to the household and residence group members. Both men and women cooperate in fishing with the poison plant called barbasco, but only men fish with arrows and nets. Horticulture is considered a female activity, but men generally carry out the heavy work in the gardens. Gold m ining is basically a male activity but women can help their husbands washing the gravel or cooking while in the mining camps. Gold Mining Gold has been mined in Madre de Dios since pre Columbian times. There were gold diggers on the trail to Tambopata in 1 907 and in the 1930s the Arasaeri faced the consequences of a boom in prices. Nevertheless, since the 1970s gold has been extracted in Madre de Dios on an unprecedented scale. When the different Arakmbut groups escaped from the mission of Shintuya and esta blished in their current locations, they did a small amount of gold washing. Gold miners migrated from the Andes and established in the Karene River in the 1950s and 1960s, although at a small scale. The Arakmbut learned to wash gold from them, with whom t hey maintained relatively peaceful relations. But it was not until 1978, when a new boom in the international price of gold took place, that all the Arakmbut regularly began to mine within their community lands.
94 The gold is annually washed down from the An des Mountains in the rainy season to be deposited on banks of the rivers , where it mixes with sand and pebbles . It can also be found inland in the dried up beds of old streams . Inland deposits yield more gold but their exploitation requires better equipmen t because it is necessary to dig large quantities of earth, while beach placers can be worked with less sophisticated equipment but produce less gold. The Arakmbut work gold more intensively in the rainy season (November to April), which is when the river s rise and flood the beaches, depositing gold that becomes available once the water recedes. After a floodin g, families sail up the Huasorok o River in a competition for the beaches where the river has deposited gold. Finding a good gold deposit before some one else does is key because once an Arakmbut has appropriated a beach, he will not share it with other Arakmbut. In the dry season, beaches are worked less. Those who have appropriate machinery can work gold in inland placers throughout the year. To find a gold placer, either in recently flooded beaches or in inland areas, the Arakmbut test the ground by sieving sand in the water or prodding with a rod to find old stream beds. The method of finding gold deposits in upland areas is si milar to hunting. Small groups go into the forest and search for places to test. In Puerto Luz, the richest gold placers are loca ted on the banks of the Huasorok o River. The majority of people work gold in this area, either o n the beaches or in inland placers . Only a few Arakmbut work gold on the Karene River, which is consid ered to contain less gold . Since getting to the Huasorok o River takes a two hour upriver sail, the Arakmbut build camps in this area where they stay whil e working gold . Some camps i n upland placers have been used by some families since the Arakmbut moved from Puerto Alegre to Puerto Luz in the mid 1970s, exerting a quasi private property over them. T he se camps have also been passed on from parents to their
95 children. In some cases, ki nsmen (mostly siblings) have built their camps close to each other, somehow reproducing the settlement pattern of residence groups in the village. In these cam ps, families build huts and clear gardens to provide themselves with food while extracting gold. They also hunt and f ish in the surrounding forests and water bodies, which contain more game and fish because are more distant from the village. Each time they stay in these camps for one or two weeks, after which they return to the village on their way t o Delta Uno, where they sell the gold and buy supplies. After spending a few days in the village, the Arakmbut come back to their camps to continue working gold. During school vacation, which coincides with the rainy season, most of the children help their parents to wash gold in the mining camps. For this reason, during this season entire families spend most of their time working gold in the camps, a time during which the village mostly remains empty and quiet. In some cases, the owner of the equipment may hire some Arakmbut or Andean workers to do the job. T he method most practiced on the beaches is the one that the locals call carretillear Once the natives have found good amounts of gold on a recently flooded beach, they dig out with pick and shovel, l oad wheelbarrows with gravel, and convey it up to a metal sieve attached to a wooden chute , where a man stands with a hose attached to a motor pump. By using the water sucked out of the river by the motor pump, this man washes the gravel and separates the sand from the pebbles. A piece of rug placed on the sie ve captures the black sand that contains the gold, which is then preserved in buckets and later mixed with mercury, which draws together the gold dust. The mixture is then panned by the riverside and t he mercury is burnt off, leaving the gold.
96 Figure 3 1. Highland miners working with carretilla, Puerto Luz . Photo by Danny Pinedo Usually, this method requires th e work of four or five people . Three or four people load wheelbarrows with gravel and take them to the sieve, while one stands at the sieve and was hes the gravel. B ecause the equipment used in this method is relatively cheap, most people in Puerto Luz extract gold with it . According to Gray (1996:263), i n the early 1980s, when motor pumps were scarce, this method employed large groups of men linked by clan ties who would work toge ther around a pump . 6 As people began to afford pumps, the groups became smaller, nowadays consisting primarily of household membe rs, although they can also include other kin and friends. For example, a carretilla group might be formed by a m an, his wife and their 6 I find it very unlikely that these groups were formed only by people linked by clan ties. Given that residence groups (or clan cl usters) include not only clan related people but also in laws and friends, I do not think that in laws and friends were excluded from these groups only because they belonged to a different clan.
97 children, a man and his sons, adult siblings, in laws, as well as several combinations of them. When women participate in gold mining, they do so helping their husbands wash the gravel or cooking while they stay in the mining camp. In the past decade, the Arakmbut have adopted more sophisticated mining technology, such as the chupadera and caranchera methods. The chupadera method is used only in inland are as, near water sources such as stream s, swamp s or lake s . After the area selected is cleared, the soil is blasted away with water pulled from the nearby water source by a 20 horse power motor. With another motor, a 180 horse power motor, the eroded soil is sucked into another hose that spits the sandy mix onto a sieve that sifts it into a chute covered by rugs. The heavy deposits, which contain the gold, get stuck onto the rugs. Every four or five ho urs, the rugs are emptied into buckets and the muddy clumps are later mixed with mercury . The mix is then burnt in mini stoves to remove the mercury and extract the gold. This method can be used throughout the year because it does not depend on river flooding that only occurs in th e rainy season. This method usually requires the work of two people, one to operate the high pressure hose , called marakero and other to operate the hose that conveys the gravel up to the sieve. Unlike chupadera , the caranchera method is used only on the riverbeds. It consists in vacuuming up auriferous material from below the surface of a river using a suction hose. To do this, the operator of the hose has to dive into the water. A motor then takes the sucked material up to a hopper l ocated on land, which includes a sieve and a chute with a rug where the gold sand is deposited. The caranchera can be operated even by a single person. While work for chupadera teams may be provided by the household, most peo ple prefer to hire workers. The greater productivity and profitability of this method make it possible to hire about three or four laborers. Young people are generally preferred as workers, who may include
98 people of Andean origin or other Arakmbut from Puerto Luz. Andean workers come mo stly from the nearby highland and lowland provinces of Cusco, such as Sicuani or Quillabamba. Arakmbut clusters of the village, but in any case are always from a lo wer socio economic status. Workers are usually hired for periods of 15 days, after which they may start a new work period with the same patron, work for another patron or leave the community to work elsewhere. The employer or patrÃ³n provides equipment, tra nsportation, fuel, food, and lodging, and wages are based on several arrangements. One arrangement is that 80% of the gold mined during a 24 hour period goes to the employer, while workers earn the remaining 20%. Each period of 24 hours (one day and one ni ght) is worked by turns. Additionally, some employers may permit their workers to labor at their own cost on days of rest the benefits going directly and solely to the workers. This system is known as the chichiqueo . The workers make an agreement among the mselves to undertake this labor and cover the costs of the operation. Between them, they pay for the petrol required by the pump, the mercury for the amalgam and other items. The gold mined by the end of the day is distributed amongst those who participate d in the chichiqueo . Because the equipment they require is expensive, and their consumption of petrol is high, chupadera and caranchera methods involve a significant capital investment that not all the families can afford. Therefore, in Puerto Luz only a h andful of families practice these types of gold mining. Most owners of chupaderas have purchased their equipment by accumulating some capital through the rent they receive from mining settlers extracting gold from areas controlled by Puerto Luz. These mine rs have mined gold within these areas since the early 2000s. By this invitado lly
99 known as regalÃa RegalÃas are not paid only to the individual owners of gold deposits but also to the community as a whole. This money goes to a communal fund that is used to cover the travel expenses of community offic ials or to support community members in need. RegalÃ as consist of a fixed amount of money and do not depend on the production of gold during a certain week or month. In 2012, invitados weekly paid 100 soles ( US $35 . 7) to individuals , and 100 soles to community per month . Invitados mine gold mainly in two zones under the control of Puerto Luz: La Cumbre and the Huasorok o River. While La Cumbre is an area located outside Puerto Luz titled lands, the Arakmbut consider it to be part of their ancestral terr itory, used as a hunting and gathering ground since before Puerto Luz lands were demarcated and titled. Gold in this area was discovered by people from Puerto Luz, but it was exploited sporadically and on a small scale. In the early 2000s, this area was in vaded by hundreds of Andean settlers, which brought about several violent confrontat ions with the Puerto Luz people, who several times tried to expel them. The conflict was solved when the miners proposed to pay the Arakmbut rent in exchange for the right to mine gold in La Cumbre. Unlike La Cumbre, the areas in the Huasoroco River lie within the titled lands of Puerto Luz and have only recently been mined by invitados . During the peak years of gold production in La Cumbre, the number of invitados reached several hundreds, and each Puerto Luz family came to have between three and five invitados . The people of Puerto Luz still remember the golden years in which large amounts of gold were extracted from this area and regalÃas became the most important source of cash income. The boom, however, lasted only for a few years and once gold deposits were exhausted the number of invitados decreased, along with the regalÃas . In 2012, there were approximately 5 0 invitados in
100 La C umbre and 10 on the Huasorok o River, but these numbers fluctuate because the miners move out to other camps and then come back. The relationship between the invitados and the Arakmbut is often marked by tension and distrust. The Arakmbut often complain that their invitados lie about the amount of gold they have extracted. Tensions increased especially after the producti on of gold diminished, while the amount of the regalÃas remained fixed. For this reason, in order to foster trust and strengthen the relations among them, mining settlers and Arakmb ut engaged in friendship and godparenthood relationships. These social ties enabled the settlers to gain legitimate rights to extract gold from Arakmbut lands, while they assured the Arakmbut not only a relatively stable source of cash income but also addi tional economic support in times of emerge ncy. As we will see in Chapter 5 , these social ties would have a negative impact on the Arakmbut land rights when the community, fearing to lose their gold deposits at the hands of the invitados , decided to expel t hem from the community. The use of more efficient technology in gold mining, and the associated possibility of profiting more, has brought about incipient capitalist relations and some level of social differentiation among the Arakmbut of Puerto Luz. Those who own chupaderas and can extract more gold not only can afford to hire laborers but also show several signs of wealth such as bigger houses and the possession of capital and prestige goods such as motorbikes, boats with long tail outboard engines, power generators, chainsaws, satellite TVs, among others. Nevertheless, this rise of capitalist relations is still partial because the youth who are hired do not depend exclusively on chupadera owners to survive. They can still work gold with their parents or o ther mining teams, and over time it is relatively easy for everybody to ge t a motor pump of their own, or associate with others to get it, and avoid wage labor. Further, since only five people
101 posses chupaderas in Puerto Luz, social differentiation remains low and Puerto Luz can be characterized as a relatively egalitarian community. Gold used to be sold at Boca Colorado, in the confluence of the Karene and Madre de Dios rivers, where the state opened a branch of the Banco Minero in the 1970s. But in the ea rly 2000 s , when a trail that connected Puerto Luz with Delta Uno was opened, Boca Colorado was replaced by the closer mining town of Delta Uno. In Delta Uno, Puerto Luz men not only sell their gold but also purchase petrol, hoses and other items for gold mining, as well as manufactured food, clothing, school supplies, etc. Gardening Gardening provides an important part of the Arakmbut diet. Each household cultivates its own gardens, which can be opened wherever free, suitable land is found within the commu nity territory. Gardens are located in two differen t ecological zones, upland and flooding zones, and crops vary depending on the zone. Crops such as pineapple, barbasco, manioc, and rice are grown on upland areas near the village, while plantains, papaya, manioc and other crops are planted on the more distant, flooding areas. The Arakmbut also open gardens in their mining camps, which are located along the banks of the Huasorok o River and some of its tributaries. Since the households spend much time in the ir mining camps, and because the mining camps are relatively far away from the village, they use these gardens to supply themselves with crops while working gold. Most of the households also have little gardens close to their huts in the village, where the y grow pepper, tobacco, cocona, herbs, and fruits. The preparation of a garden takes place in the dry season , and includes clearing, burning and sowing . At the beginnin g of the dry season, around May or June, the men of the household pick an area outside t he village and clear it. A fter vegetation has been allowed t o dry for a few months, the burn it . The first sowing is carried out at the end of the dry season, by September.
102 Sowing, weeding, and harvesting are responsibilities of women, although single men may perform these tasks too. Except for clearing, in which sometimes a man and his wife may work together, men do not spend m uch time in the gardens. M en may help out whenever their wives need extra help, but generally all the routine maintenance of garden s is left to the women. Since t he supply of crops for the kitchen is responsibility of women, they go to the garden most of the days, carrying the produce back to the house in a string bag . The wet season is when less work in gardening is needed . At the en d of the wet season and at the beginning of the dry, which elapses for about a month, men an d women cooperate again to remove the weeds in the old gardens and choose the area where they will open new ones . Other than that, men and women work at their respe ctive tasks and are more frequently apart than together. Thus , cooperation between the men and women of a household takes place more in the transition between the seasons. Each household has about five chacras ( gardens ) (Gray 1996: 45). The number of gardens increases with the size of the household. A newly married couple will share the gardens of the er if the marriage is virilocal or the wife has com e from out side the community. A s the family size increases its members will clear gardens their own garden, depending less on their in laws gardens . Couples with a large family have well established g ardens of their own and are fully independent from thei r in laws (Gray 1996: 74). Diversification of crops, however, is equally important in determining the number of gardens. A new couple starts growing mani oc, plantains, and papaya, obtaining other crops from their in laws in return for help they offer them w ith the horticultural work. As it becomes more independent, the family diversifies its range of crops to include barbasco, pineapple, potatoes, maize, peach palm, and sugar cane (Gray 1996: 74) . The more independent the household, the more autonomous and di ve rse are the gardens .
103 There is not much cooperation between households in gardening and, when it exists, it is confined to the most laborious tasks. A household is able to provide the labor force necessary to perform all the gardening tasks (from clearing to harvesting) . Nevertheless, help may be provided by close kin or neighbors with heavy activities , especially with c learing gardens , which is considered to be hardest activity. These offers of help are appreciated and c onsidered generous . The routine gar den work of planti ng, weeding, and harvesting gives women with an opportunity to be generous in their own right. Older women, in particular, share cuttings and seeds with their younger sisters, d aughters, and daughters in law (Gray 1997a: 143). In the pre m ission days, each household in a communal house or hak used to have between seven and eight gardens and it could get help from other households that were linked by clan ties, especially with sowing and planti ng of manioc stems (Moore 2003: 64 5). Most of w hat is produced in the gardens is redistributed and consumed by the members of the household. C rops are not distributed between h ouseholds . Among households, the norm is rather balanced reciprocity . If a woman runs out of manioc or plantains , she may ask for a loan from a close relative , in which case the loan must be repaid as soon as possible at the risk of getting the reputation as a scrounger (Gray 1996). The surplus produce can be sold in local markets, which makes gardening the second source of cash income after gold mining. Crops that are sold include, but are not limited to, manioc and plantains. The families sell these crops to mining settlers working in La Cumbre and to traders in Delta Uno. In the last decade, the intensification of gol d mining may have negatively impacted the participation of men in gardening. The intensity of these impacts, however, varies depending on the technology used in gold mining. Thus, involvement in gardening has decreased among men who work gold in upland are as, which does not depend on flooding and hence can be practiced
104 throughout the year. Men who work with chupaderas in upland zones spend most of the time in their mining camps, and their cooperation with women in gardening has decreased. In Puerto Luz, how ever, only a few men work with chupaderas , which sharply contrasts with San JosÃ© del Karene, where the majority of men work gold with this technology. In San JosÃ©, my informants told me that abandonment of gardening due to increasing investment of time in gold mining is such that most of the daily diet is provided not by crops grown locally but by industrial food among people who extract gold with carretilla becaus e gold mining with this technology has its peak during the rainy season, which is when horticulture requires less work. Clearing and burning, which are male activities, take place at the beginning of the dry season, which is when gold mining activities dec rease. My informants coincide in pointing out that cooperation in gardening has diminished with the rise of gold mining. It seems that with an increasing time i nvested in gold mining activities, it became harder for the Arakmbut to devote time to help others in their gardens. A consequence of this may be the fact that currently some Arakmbut hire others as laborers during peak season. Hunting and Fishing In Puerto Luz, hunting and fishing provide one of the most important sources of prot ein. Hunting is carried out primarily by adult males, whether individually or collectively. 7 Collective hunting used to involve both small and large hunting teams, but over time it has become a family based activity. In the past, all members of the community gathered to hunt the white lipped peccary, which is found in large herds that may number 200 individuals. Today 7 Occasionally, women may hunt small animals when going to th eir gardens or while fishing. Sometimes, hunting is carried out by widows in need of feeding their children.
105 community based hunting is very rare, having been replaced by smaller groups formed by men from the same residence group. When a youn husband helps his father in law with hunting. The main tool used in hunting is the shotgun, which replaced the bow and arrow, whose use is now restricted to fishing. Hunting is more frequent during the wet season. During this time, the fruits that animals eat come out, and the leaves on the hunting trails are softer and make less noise, which is crucial to not scare the prey. This seasonality does not represent any conflict with gardening since this acti vity reaches its peak in the dry season, and because men participate less in gardening. This is not the case for gold mining, whose seasonality very much overlaps that of hunting. Like hunting, the peak season for gold mining is the rainy season. One strat egy to cope with this conflict between both activities is to hunt while working gold in the more distant mining camps. When men go to their camps they carry with them their shotguns to hunt while in their spare time. Hunting while working gold in these cam ps is also very convenient because the forests that surround these camps are far away from the village and hence contain more game. Since Puerto Luz has well conserved forests close to the village, hunters do not need to go too far to find game animals. Fo r this reason, hunters keep their expeditions to within a day. They use hunting trails that are connected to salt licks, which are areas that attract most of the animals, and other areas where animals abound because of the presence of certain edible fruits that attract some species. I was told that each family has its own hunting trails, which can even be passed on from one generation to the other, but it seems that at least some of them are shared by several families of the community. These trails, which a re located in inland areas, are also used to explore gold deposits in old river beds.
106 Game meat is primarily for the consumption of the members of the household. No game meat is sold in the market. When large pieces of wild game are hunted, such as a tapi r or a friends and neighbors can also be recipients of the meat. Me at sharing is certainly the primary expression of generosity among close relatives and friends and it contributes to strengthen solidarity bonds among members of the same clan and residence group. A household with a dependent son in law may increase its po tential to produce and distribute meat, which enhances its prestige. There are two types of fishing: fishing with bow and arrow or net, which is practiced individually, and fishing with barbasco, which is a collective activity. Barbasco fishing consists in pounding barbasco roots in order to produce a milky juice which, when mixed with water from a stream or a lake, deprives the fish of oxygen and forces them to the surface for air . This is why barbasco fishing is better in the d ry season, when the rivers are low. During this season, the fish are more concentrated together, and the barbasco ( kumo ) works well in the shallow water. Before the decision to go fishing with barbasco is made, some members of the residence group are sent to a nearby pond to see if there are enough fish. If so, the barbasco is dug up from the gardens and taken to a stream or lake. Barbasco is a root that is grown on high land away from the river. Barbasco fishing groups can be either exclusively female or c onsist of people of both sexes. When groups are mixed, the men pound the root with stones and then place the crushed
107 roots into string bags. 8 Usually, young men or women are responsible for mixing the barbasco with the water. The older men and those boys w ho have bows and arrows stand scattered along the bank waiting for the large fish to appear on the surface, while the women armed with machetes follow the barbasco dye and dispatch anything that comes their way. Barbasco fishing is carried out mainly by th e members of a residence group. Sometimes all the residents of the community gather to fish with barbasco. Community fishing takes place during the celebrations of the founding of the community, in which fish are used to prepare food that will be sold duri ng the fiesta to raise funds for the community. Nevertheless, community fishing is becoming rare, as even during the anniversary fiestas residence groups, and not the community, are the units that organize these fishing events, mostly with the purpose of s elling food and making some money. The Arakmbut also organize longer fishing expeditions to further areas in the Upper Karene. During the dry season, several households from the same residence group travel upriver and make temporary camps where they stay f or some weeks fishing, hunting and gathering. Fishing with barbasco is an important activity during these expeditions. I n the last decades , h unting and fishing have experienced some changes, especially due to the incorporation of gold mining into the loca l economy. As the Arakmbut gained more access to money, manufactured food that can be purchased in the market of Delta Uno has become increasingly important in the local diet, displacing fish and game meat to a secondary place. The impact of gold mining on hunting and fishing, however, varies in each community. In San JosÃ© del Karene and Barranco Chico, for example, where gold extraction takes place at a larger scale 8 In the past, these bags ( wempu ), which are manufactured entirely by the women, were made from twine, which comes from a species of setico palm. Nowadays the setico is scarce and women have to walk long distances to get the twines, so they prefer to use fibers that they take from industrial bags ( costales ) that they buy in the market.
108 and yields more profits, the Arakmbut hunt or fish less since they prefer to buy food rathe r than produce it. An additional factor that discourages men from going hunting is the destruction that gold mining has caused to nearby forests and the game animals they contain, which decreases hunting yields since men have to walk further in order to fi nd prey. The gold miners have penetrated deep into the forest looking for new gold deposits , an d the animals have frightened away due to the constant noise produced by the motor pumps. The increasing investment of time in gold mining has certainly had a ne gative impact on hunting too . Additionally, until recently, hunting and marriage were intimately connected as hunting ability was an important criterion to select a husband. Hunting was more important as it provided a source of prestige. A man would becom e attractive to a woman for his capacity to bring meat to the household. When a young couple decided to get married , the girl accepted and cooked the meat that her prospective husband had hunted and offered to her. This practice was called chindoi . In the last decades, the source of prestige has shifted from hunting to gold mining, as it is the ability to find good gold placers and produce gold that gives prestige to a man. Despite these transformations, hunting and fishing continue to be important in Puert o Luz , both as a source of protein and as a social context for the practice of generosity and cooperation. The impacts of gold mining on these activities, however, are greater in communities like Barranco Chico and San JosÃ© del Karene, where industrial food have replaced game and fish, and hunting and fishing as activities have disappeared either because almost all time is invested in gold mining or because nearby forests have been damaged. In Puerto Luz, e ven though gold mining have caused some impact s on these subsistence activities, such as the increased consumption of manufactured food in detrimental of game meat and fish, hunting and fish still play an important role in the community. Hunting and fishing are an important source of
109 solidarity and cont ribute to strengthen the networks among kin and friends . In Chapter 5, we will see how this solidarity is important for political mobilization. Conclusion Amo ng the Arakmbut , articulation to the market economy has taken place through the incorporation of gold mining to their livelihood strategies. Gardening, hunting, fishing, and gathering have remained as subsistence activities. S cholars have argued that although the A rakmbut increased involve ment in market economy has introduced some levels of monetary relationships , indi vidualism, social inequality, and commodity exchang e at fixed rates , it has not disrupted Arakmbut social structure because social relations among them are based on k inship and clan ties and not o n monetary relations, and access to reso urces is still secured through inte rnal distribution and generosity (Gray 1997a; Fuentes 1982; Moore 2003) . Drawing on my data on Puerto Luz, I argue that these assertions are only partially true. Two characteristics of the Arakmbut during my fieldwork con trast with the above explanations . First, the sharing of game meat and cooperation in barbasco fishing take place only among members of the residence gr oup . As we will see in more detail in Chapter 4, distribution and cooperation beyond the residence group is ve ry limited. Furthermore, the food that is provided through meat distribution and fish is not the most important part of the diet. Game meat and fish have became a complement for industrial food in the daily diet, which for the most part is supplied b y food purchased in the market. Due to their small quantities, shared meat and fish do not meet all the dietary needs of a household. Thus, meat distribution and barbasco fishing basically constitute social mechanisms to strengthen social solidarity among members of discrete kin groups. Second, today the Arakmbut hire other Arakmbut as workers in gold mining, so monetary relationships among them are now very common. This does not mean that all relationships
110 among the Arakmbut are now expressed in monetary t erms. As I already mentioned, part of the food and labor needed are distributed through sharing and generosity. Some Arakmbut express their generosity towards other Arakmbut by helping them in heavy gardening tasks such as clearing. This practice, however, is not very common and takes place only among people closely linked by kinship ties. The Arakmbut do not need to work for others in order to subsist. The basic means of subsistence, such as land and labor, as well as access to a motor pump due to the high prices of gold are available to the majority in the community Drawing on Plattner (1985) , who asserts that market integration may have the effect of forging social relations and not only disrupting them, I suggest that rather than assuming a prior i that participation in the market economy necessarily diminishes reciprocity relations, or that kinship based relationship s resist market relationships, it is necessary to examine what social relations are being affected by what forms of participation in the market. In this chapter, I have shown how gold mining has prompted the development of godparenthood ties between the Arakmbut and mining settlers. While this social tie, based on a reciprocity morality, has strengthened solidarity betwee n Arakmbut and their compadres , it has divided loyalties within the community and hindered unified action at the village level. Understanding the effects of these reciprocity ties with outsiders is key to better understand ing Arakmbut sociality and its effects on politic al mobilization.
111 Figure 3 2. Arakmbut native communities .
112 CHAPTER 4 ARAKMBUT SOCIALITY On my first arrival in Puerto Luz, the mototaxista (motrobike taxi driver) dropped me in a compound of houses that I believed to be the whole village. Federico, the head of the community, accommodate d me in his house, which was lo cated close to the soccer field , o n the other side of th e village . It took me a few days to realize t hat the village was divide d into a center located around the soccer field, and several clusters of houses scattered throughout the village, me eting, it surprised m e to see attendees occupying specific portions of the communal house, forming s maller groupings, a pattern I would observe again during commun al fiestas . Later I would realize that all these groups were formed by people closely related by kinship and frien dship ties. Nevertheless, unlike other Amazonian native groups that show a disp ersed settlement pattern, the A rakmbut of Puerto Luz live in a fairly nucleated village. Dur ing my first days in the village , I would also notice another important and seemingl y contradictory characteristic of the Arakmbut social life . Despite their strong kinship oriented organization, the Arakmbut seemed not to avoid forming relations with outsiders. Motorbikes transport ed Arakmbut people back and forth between Puerto Luz and the mining town of Delta Uno. Andean miners pass ed the Huasoroko River on a daily basis. Many outsiders were friends with the Arakmbut, and some of them had become compadres or married Arak mbut women and moved to the community. How do these distinct characteristics of the Arakmbut society their desire for seclusion in discrete kinship based house compounds and their proclivity to develop intimate relations with outsiders form part of the same pattern of sociality?
113 A nthropologists have addressed these questions from two main theoretical perspectives . Scholars in the first approach argue that among indigenous Amazonian s sociality is constructed in the local domain, where consanguinity ties and the moral sentiments derived from them prevail (Belaunde 2001 ; Gow 1991; McCallum 2001 ; Overing Kaplan 1975; Santos Granero 1991). They argue that Amazon ian societies value personal autonomy and reject coercive social structures (Overing and Passes 2000) . This ideology of an independent self, expressed in the emphasis of self control, is what allows individuals to live a convivial communal life (Overing 19 88; Overing and Passes 2000) . Therefore, Amazonian peoples see k to live a sociable life, avoiding anger and oppressive rules. In their search for conviviality, Amerindians seek to turn affinal and even friendship relations into consanguineal relations, whi ch result s from living together (McCallum 2001; Overing 2003; Overing Kaplan 1975). Adherents of the second approach posit that Amazonian indigenous sociality is rather constructed in the supra local sphere, where the most important social relations are th ose that stress affinity and alterity ( Descola 1994 , 1996 ; Taylor 1996, 2001; Viveiros de Castro 1992) . At the inter local level, social relations are based on exchange, which sometimes turns into competition and predation. In this context, collective iden tity is defined through contrast with and antagonism towards . From this perspective, the importance of affinity lies not in its function as a particular form of kinship but in its meaning of otherness . S ociety needs the affinal other in order t o reproduce itself (Descola 1996 ; Viveiros de Castro 1993, 1995, 1996 , 2001) . More recently, a third approach has explored the role of friendship as an alternative mode of relationship in the forging of nat ive Amazonian sociality. S cholars in this perspect ive see the importance of friendship not in its relation to kinship and affinity, but in its distinction from them (Killick 2008, 2009, 2010; Santos Granero 2007). Santos Granero (2007) argues that rather than
114 transforming friends into kin or affines, frie ndship may sometimes be more important than the existing kinship ties. Friendship offers an alternative to the often oppressive kinship obligations, while allowing the formation of relations with dangerous outsiders, avoiding antagonism (Killick 2005; Sant os Granero 2007). Two major changes in Arakmbut social life is their spatial concentration in larger settlements and their increasing contact with outsiders. What has the impact of these social changes been o n their sociality? When not all members of a settlement group are linked by consanguineal ties, what characterizes sociality? Is it possible that we can find different social ities within a single settlement group ? I f friendship offers a n alternative to the consanguinity affinity dichotomy, what distinguishes friendship among kinsmen and among non kin? Is Amazonian sociality restricted to kinship, affinity and friendship? What about patron client relations? The Arakmbut of Puerto Luz offer a case in which sociality is made up of different coexisti ng social ties yet each with distinct moralities attached to them. In this chapter, I discuss the main characteristics of Arakmbut sociality, the moralities associated with it , and its underlying economic, political and cultural determinants. First , I desc ribe kinship and affinity ties, and their role in organizing social relations and settlement patterns within the community. Then the chapter documents the building up of friendship and godparenthood relations among the Arakmbut and between them and outside rs, and how these social ties are mainly used to gain access to resources. In the next section, I examine how, unlike friendship and godparenthood, relations based on clientel ism are more vertical yet their exploitative character depends with whom the rela tion is established . For each type of social tie, I discuss how they create and strengthen bonds of solidarity betwe en the parties involved . In the final section, the chapter focuses on the different spaces of socializing, especially fiestas, both at the l ocal and supra local levels. I rely on a perspective that without discarding the role of
115 consangu ineal and affinal relations, attempts to provide a more nuanced explanat ion of the specific conditions o n which one or another prevails. I argue that Arakmbut sociality goes beyond the consanguineal affinal dichotomy to incorporate not only friendship but other social ties such as godparenthood and patronage. The moralities attached to each social tie, however, vary with social distance. Kinship and Clan In Puer to Luz, k inship ties and the social relationships to which they give rise operate mostly at two levels of social organization: the household and the res idence group. The smallest social and spatial unit, is formed by all those who eat at the same hearth. These are usually the members of a nuclear family (father, mother and children), but sometimes they may also include the spouses and children of married sons or daughters . This is because some children may have formed their own families yet they neither have built a house of their own nor have they prepared a garden yet. In this case, there is only one hearth in the house and all its dwellers share the food. The members of the household, however, may dwell in separate houses, so household and house are not necessarily the same. While male and female members of the household cooperate in most daily activities, there is still a sexual division of labor. Hunting is primarily a male activity, whereas cooking, basketry and weaving are carried out only by women. Even though horticulture is considered a female activity, men generally help their wives by carrying out the heavy work in the gardens. The male head of the household, his married sons and son in laws provide the household with game meat, but it is the female head of the household, often with the help of her daughters and daughters in law, who is in charge of distributing the meat to the household members. Gold mining is ba sically a male activity, but women can help their husbands by washing the gravel or cooking.
116 Marriage is regulated by the clan system, which certainly is the most important element in Arakmbut social organization. 1 The Arakmbut have seven clans, but only s ix of them are r epresented in Puerto Luz (Table 4 names and totems refer to their mythological origin, and all members of a clan claim to be descended fro m a common ancestor ( Moore 1975). In a household, the children belong to the her siblings always belong to a different clan . Clan exogamy is at the base of the Harakmbut system of marriage. The norm is that members of a clan take their women from other clans. A survey I conducted in Puerto Luz indicates that this norm is still followed by the villagers. The Arakmbut believe that sexual relations between members of the same clan could lead to the deformity of newborns, ill ness, or even death (Gray 1996: 83). One of my informants told me that the punishment for the the illness and death of a close relative. In pre contact times, the pr eferential form of marriage was the exchange of sisters (cross cousins) between clans (Helb erg 1993), and the norm in post marital residence was uxorilocal ity . After marriage, the husband had to provide the bride service, which consisted in working for his father in law for two to four years. After the son in law had carried out the bride service, the new couple moved to a new house nearby. Today post marita l residence is mostly virilocal and , with t he exception of Puerto Luz, the exchange of sisters between clans has become rather rare because of the lack of women in most Arakmbut communities (Gray 1997a). 1 Not all the Harakmbut speakin g peoples have clans. Whil e clans are present among the Arakmbut , Sapiteri and Wachipaeri, they are absent among the Arasaeri and Toyeri (Gray 1996: 79).
117 Table 4 1. Number of people per clan in Puerto Luz , 2012 Clan name Number of people Huandignpana Sahueron Singperi Yaromba Total 2 32 93 10 123 21 281 * Source: Survey conducted by the author *This number does not include yernos and other outsiders who reside in the community. I n communities experiencing an imbalance in the numbers of men and women c ompetition for women has become a common source of internal tensions. This is so because finding an appropriate spouse is the most difficult activity in the life cycle of the Arakmbut. Conflicts arise between the families of suitors competing for the same women and between households with marriageable women and those without. Relationships among young men sharing the same marriageable women become competitive, a situation that may involve either siblings or members of different households. These tensions us ually expand when kinfolk beyond the household of the rival suitors get involved in the dispute. Whether because suitors were rejected or because the postmarital arrangements were not satisfactory, the way these tensions are solved is with one party leavin g for ano other Arakmbut community (Gray 1997a). For instance, two brothers from San JosÃ© del Karene, where a shortage of women has been a rec urrent problem, got married to Puerto Luz women, after which they opted for an uxorilocal postmarital residence, me in Puerto Luz. Even in a community like Puerto Luz, where the ratio of men to women is more balanced, suitors from other communities may also subtract from the potential wives. This ha s been a main source of conflict between Puerto Luz and San JosÃ© for several years.
118 In the last decade, an additional factor contributing to these tensions has been the increasing number of Arakmbut women who are married to outsiders. Marriage between Arak mbut women and outsiders offsets the increase of potential wives for Arakmbut men. Mining settlers have used marriage as a strategy to obtain access to and work gold within the community lands. Indeed, some miners who became invitados ended up getting marr ied with the daughters of their Arakmbut partners. Rights to land are granted only to members of the community , and only Arakmbut people, even if they have come from other communities, can yernos in law) are not allo wed to become members of the community, which means they neither enjoy full rights of access to the community resources, nor can they attend community assemblies and participate in decision making. Nevertheless, in practice yernos , who numbered eight in 20 12, are entitled to mine gold in the mining camps of their fathers in law. I was told that in the past Arakmbut women were not allowed to marry outsiders, a norm that was used as a strategy to keep outsiders out of the community lands. Outsiders who courte d Arakmbut women used to be harassed until they gave up their undertakings. Over time this local norm weakened, mostly because of the increasing interest of Arakmbut women in getting married to outsiders, whose greater attractiveness stems from their equal ly greater access to economic resources and their purportedly hard work ethic, compared to the Arakmbut, who would be taken to drink and would not project into the future. In order to legitimize their rights to use community lands, yernos are required by t he Arakmbut to provide services to the community. For example, during the preparations for a fiesta to celebrate the community anniversary, yernos were asked to collect fire wood to cook food that would be sold in the fiesta. On another occasion, community members gathered in an assembly complained that yernos who were working gold with their own chupaderas were not contributing to the community.
119 Households are clustered in several kin based groupings spread throughout the villag e (Figure 4 1). Following Ro sengren (1987 ), I call these clusters residence groups, which are comprised by siblings and their respective families living in close proximity, although they can also include elderly parents and some affinal and other close relatives, thus forming extende d families. These relatives, married to sisters or daughters, provide meat and give political support. The siblings in the same residential group are referred to by the name of their father, which was turned by the Dominican priests into a surname after ba ptism. While in Puerto Luz, I used to live Javier) and one sister (Guadalupe) and their respective households, and three additional households of relatives. The elderly father of the brothers and sister lived in his own hut but ate husband. The residence group also included two men who spend their t ime between Pue rto Luz and Puerto Maldonado: Lucho, a fourth Tayori brother, and because, as clans are patrilineal, all siblings living in the same block belon g to t he same clan. Clans , however, are not territorial as their members are dispersed among the various clusters of the community. 2 In Puerto Luz , residence groups formed as a result of physical separation, which is how the Arakmbut solve their internal conflicts , common to many A mazonian indigenous peoples. After fleeing the mission of Shintuya in 1969, the Kupodneri Arakmbut established in the Upper Karene, where they founded the village of Puerto Alegre. Puerto Alegre had a nucleated settlement patter n, with most of the houses arranged around a soccer field. In 1975, the entire 2 Gray found in San JosÃ© del Karene an additional kin group called wambet , which is a smaller, dispersed group whose members are greatly linked by affinity relationships in which women are more influential (Gray 1996, 1997a, b). In Puerto Luz, I could not identify wambet groups.
120 settlement moved to Puerto Luz, its current location, located midway between Puerto Alegre and Boca Colorado, at the mouth of the Karene River, where a post of the Banco Minero was set up in 1973. In their new location, the Arakmbut would have greater access to the Banco Minero and better gold placers. At the beginning, the Arakmbut built their houses around a soccer field, reproducing the settlement pattern they had in Puerto Al egre. But as the population grew, conflicts between families became very common. My informants mentioned gossip and theft as the primary causes of these con flicts. To avoid these tensions most of the households detached themselves from the group and moved to separate parts of the community. Over time the children of these households got married and opted mainly for a virilocal postmarital residence, building th e preferent ial norm for marriage was the exchange of sisters . Sister exchange enabled men to forfeit their bride service obligatio n, increasing the possibility that brothers living together form residential groups. As I described in Chapter 3, e ach househo ld has also an encampment where they wash gold. These camps are located far from the village, along the banks of the Huasoroko River. There, a man, his close kin and sometimes some highland workers spend most of the year working gold, returning to the vill age on their way to Delta Uno every one or two weeks to sell the gold extracted, buy goods , and catch up on news. These mining camps reproduce the cluster division in the village, since households related by kin ties tend to build their camps close to each other. In Arakmbut culture, c lans provide a fundamental source of collective identity and solidarity to their members. The norm is that, since clan members regard themselves as descended from a common ancestor, they are expected to help each other (Gray 1 996; Moore 1975). For example, when a woman distributes meat that her husband has brought into the house,
121 the meat. Also, when there is a dispute between members o f different clans, people are expected to support those members of their own clan, regardless of who is right or wrong in the dispute. Thus, clans with more members often dominate the smaller ones, which sometimes detach from the community to join members of the same clan who live in other communities (Moore 1975). Even the skills and qualities of a person are claimed to be shared by his or her clan mates, while any criticism to a member of a clan may be perceived as an insult to the whole group, generating its vigorous defense by its membe rs (Gray 1996: 80 1). Figure 4 1. Residence group, Puerto Luz . Photo by Danny Pinedo Nevertheless, in Puerto Luz I did not see clans playing a role other than prescribing the rules of a marriage partner. It seems that due to some factors among which we can include
122 increasing participation in monetary economy through gold mining, clan solid arity declined and became restricted to the members of residential groups. I was able to observe several quarrels that involve members of the same clan. This does not mean that in the past clan solidarity always prevailed, and tensions between members of t he same clan did not exist. It seems that clan solidarity was stronger when young men were not seeking spouses, and weaker when they c ompeted over women (Gray 1997a: 227). Moreover, clan solidarity may have varied according to the prevalence of virilocal or uxorilocal marriage arrangements. Uxorilocal marriage emphasizes solidarity between a man and his in laws because he establishes residence at his important becau se men remain at their residential groups, forming clan block s . Nevertheless, while I found virilocal marriage being prevalent in Puerto Luz, clan solidarity was not so important. Everyday interaction is primarily with other households in the residence groups. While the different residence groups in the community are clearly separated from each other, they are located within walking distance. During my time as a dweller of one of these residence groups, however, I hardly saw people from other residence g roups coming to visit. Interaction among households in a residence group takes place in the form of generosity and cooperation based on generalized reciprocity, meaning that someone who provides goods and services to others does not necessarily expect a re turn in the short term (Sahlins 1965) . Generosity is expressed through meat sharing, while cooperation among households includes collective fishing. Since each household is to a large extent self sufficient in providing the means for its subsistence, coope ration among households in a residence group is basically intended to strengthen solidarity. Redistribution of food among domestic units in indigenous Amazonia has been described as a form of informal gift exchange (Smith 2002). Some scholars, however, con tend that among
123 some n ative Amazonians meat sharing is not a form of reciprocity. Hvalkof and Veber (2005), for example, point out that for the AshÃ©ninka of Gran Pajonal food sharing is a moral imperative verybody share everything with offers something in exchange for a future retribution. Among the AshÃ©ninka all members of a community have the right to demand and re ceive from others who in turn have the obligation to demand and receive, however, is not necessarily accompanied by an obligation to reciprocate. 3 Gray (1997a: 140) suggests for the Arakmbut that while meat sharing is mainly an act of generosity that does not require an immediate response, it is still a form of gift exchange since reating an exchange. For the Arakmbut, prestige and respect are gained by being generous with others, which is expressed by satisfying th e desire of others (Gray 1997a: 138). Generosity allows people to counteract accusations of selfishness and meanness whi ch arise from acts that give rise to jealousy in others, especially holding back material goods or knowledge. The spheres of activity in which generosity qualities can be recognized va ry between genders (Gray 1997a: 140). Men show generosity mainly through the distribution of meat, while women do so through the sharing of products that not all wo men grow such as peanuts, sweet potatoes, papa del monte , or barbasco among close relat ives and neighbors (Gray 1997a: 139). In Puerto Luz, w hen a man brings in game meat, his wife or other female members of his household chop it into several pieces, keep the best pieces and generous ly distribute the rest among residents of the same cluster. Meat sharing will depend on the availability of meat from 3 This contrasts with other descriptions of the AshÃ©ninka, which depict this people as valuing generosity more than unilateral taking (see Killick 2005).
124 large game animals s uch as a tapir or a peccary, since these species provide enough meat to feed households are not expected to reciprocate sometime in the future, a s the hun ter di stributes the meat as an expression of generosity . In practice, some households are more generous than others and hence meat may flow more from specific households to the others. This happened in the residence group in which I resided, where a young Arakmb ut was a better hunter than the adult men of the cluster and hence hunted more. I observed that this household distributed more meat than the others. S ince each family is able to provide the necessary labor to cultivate its gardens , from clearing the forest to harvesting , gardening does not involve reciprocal relations among housholds . Neverthele ss, households that lack enough labor force to meet their production needs may receive generous aid from relatives and neighbors. This happens when people are helped with so me heavy tasks in their gardens . For example, an old woman would be helped by her sons and sons in law to plant barbasco in her field, a service for which she will not provide a retribution. A man can can get help from his brothers to plant b arbasco , without the expectation of return in the near future. One of my informants pointed to the fact that generosity in gardening may take place when a woman needs help in her gardens because her husband is taken to drink and does not help her with heav y tasks like clearing. Sometimes, however, generosity may involve some degree of reciprocity. For instance, l observed how a woman was helped to sow and harvest peanuts by female relatives from a neighboring residence group who in return took part of the h arvest with them, which indicates that reciprocity does not imply an exchange of the same goods and services. 4 4 Reciprocity among the Arakmbut may encompasse more than the exchange of goods and services. It also incorporates mutual visits (in the case of the fiesta for commu nity founding anniversary ) and the exchang e of
125 Generally, generosity is practiced by those who live in the same residence group. M eat is rarely shared with people outside the residence group, even if they belong to the same kin group. In other words, generosity is restricted not only to other Arakmbut, but also to those who are close kin and friends. The Arakmbut are seldom generous with non Arakmbut people, as I myself was able to experience, which may have been related to the fact that the Arakmbut do not normally turn outsiders into kin by calling them by kinship names. During my stay in Puerto Luz, only one man used to call me buy brother, and even in this case the relationship was not driven by generosity. 5 At first I thought this lack of kin reference was because of my own personality, but then I realized that it was indicative of a general avoidance by the Arakmbut of using kin terms to refer to unrelated a nd specifically non Arakmbut in comers. Also, despite my intention to keep my relations with my hosts free of monetary transactions, they made it clear from the very beginning that their relationship with me would not be based on generosity a nd that I would have to pay for the food and lodging they provided me. Some aspects of generosity and reciprocity may have broken down due to individualistic behavior that the Arakmbut adopted in the missions or with t he increasing monetization of the economy. For example, some people prefer not to help others in their gardens because that would take time away from their work in gold mining . In the past, when a man found a gold deposit, whether on a recently flooded beach or on a dried stream bed, he sh ared the information with close relatives, to whom he invited to mine together. This was common when money was scarce and only a few individuals owned motor pumps. Gold mining was practiced by a few groups women among clans (Helberg 1996) . Nevertheless, since the number of Arakmbut women marrying mestizos has increased in the last years, which is provoking a gender imbalance within the community, the reciprocal exchange of women between clans has become very difficult to accomplish. 5
126 formed around motor pumps. But when the majority o f people became able to afford motor pumps, large groups of kinsmen stopped cooperating in gold mining, which turned into a household activity . In the rainy season, when the river floods beaches families compete for gold deposits rather than sharing them. One morning, after a rainstorm that lasted all night long, I was woke n up by the constant noise of boats sailing up the Karene River. When I inquired Guadalupe for the reason of the unusual high activity in the river, she told me that the river had overflo w ed and people were in a race for gaining the best gold deposits that were formed. Nevertheless, the notion of the Arakmbut as generous people persist even after these changes. One of my informants s aid that, unlike the Arakmbut, a mikos are selfish and nev er share a gold placer. Fishing with the poison plant called barbasco is another activity carried out at the level of the residence group. During the dry season, all households of a residence group, including elders and children, gather to fish with barbas co . This type of fishing requires the cooperation of several households in blocking a section of a stream, pounding the plant roots, introducing the juice to the water, and catching the fish that surface down the stream. While fishing with barbasco is a co llective endeavor, each family catches and consumes its own fish. The residence group is also where everyday socializing takes place. Socializing occurs in the form of gatherings for chatting and beer drinking. In the evening, after having finished a work day people go to the river to have a bath. Upon returning from the river and before dinner, adults and children cluster in small groups outside their huts to chat. Before the community got a power generator and most of the families purchased televisions, c hat gatherings went on for hours. Now they do not take too long as people return to their huts as soon as the power generator is turned on and electricity is available to power televisions. On the other hand, beer drinking is perhaps the most important mec hanism of socialization. S haring drinks involves mainly people who live in the same cluster and hence happen to be members of the same kin
127 group or friends. When doing fieldwork in Puerto Luz , I was struck by the high level of alcohol consumption among mal e adults and even youngsters. When men were not working in their mining camps and stayed in the village for some days, it seemed to me that they were drinking all the time. For this reason, at the beginning I had a bad attitude towards beer drinking, not o nly because running into drunk people at every corner of the village was annoying , but also because that was preventing me from finding sober informants. But as my fieldwork developed I came to realize the importance of this activity for Arakmbut socializa tion. By talking, getting drunk together and, in general, having a good time, social bonds are created and strengthened. During beer drinking, people also have the opportunity to show generosity by offering beer to their partners. Friends and Compadres Ne arly as important as kinship ties, friendship and godparenthood describe many relationships among some A rakmbut people and between them and outsiders. These social ties are based on moralities of generosity, reciprocity and loyalty, but their intensity depen d s on the social distance of the people engaged in the relationships. These moralities stre ngthen pre existing relationships or create new one s when no previous relationships existed. In this sense, these social ties are essential social resources tha t assure individuals and groups access to otherwise unavailable material resources . Long interested in kinship ties, anthropology has paid relatively little attention to friendship and the culturally diverse set of relationships that it describe s (Bell an d Coleman 1999). The culturally rooted ideologies of friendship, as well as the diversity of spaces and historical contexts in which they have been produced, make it difficult for anthropologists to define and analyze friendship relations (Killick and Desa i 2010) . Nevertheless, while friendship should definitively be understood as culturally specific, there is still room for some universal
128 features. In this sense, in most cultures friendship is a relationship that stands in contrast to kinship , and is forme d , in some cases voluntarily , between relatively autonomous individuals who consi der themselves of equal status, and are engaged in reciprocal, affectionate relations (Killick and Desai 2010). Unlike friendship in Western culture, where indebtedness is not considered to be part of the relationship between friends, among Amazonian groups such as the AshÃ¡ ninka friendship is not based on affectionate sentiments, but on instrumental relations led by the desire of getting highly valued goods , as well as on gift exchange and the log term relations to which mutual indebtedness gives rise (Killick 2008, 2010). In his definition of friendship in the Amazon context, however, Santos Granero (2007:3) prefers to emphasize the mutual generosity that link the individuals involved. I suggest that the prevalence of one morality or the other depends on the social distance between the people engaged in friendship. Within th e community, the Arakmbut develop f riendsh ip ties mainly with those with whom they interact the most. As we have seen, the y interact more with people living in their extent, friendship ties may involve dwe llers of other residence groups. Friends may include relatives and unrelated people. Friendship and kinship are not mutually exclusive, as one person may befriend his consanguineal or affinal kin. Sometimes, friendship may take precedence over relations wi th kin, as it is precisely its voluntary nature and the lack of the oppressive and hierarchical relations associated with kinship that make friendship more attractive and sometimes more important than kinship (Santos Granero 2007). Friendship is based on m orals of mutuality, trust and loyalty. Among Arakmbut people, friendship is primarily based on generalized reciprocity and deferred exchange, although some more balanced reciprocal exchange may also be present . Mutual obligations vary depending on the soci al distance among those related by friendship. A man or a woman would be more
129 generous with friends that are close relatives or neighbors. In this case, when Arakmbut friends exchange gifts, there is no expectation of immediate and equal return. Thus, food sharing not only include kin people but also friends who reside in the same cluster of households. The more distant the social relation between the friends, the more balanced the exchange. Friendship among kin is what Eric Wolf (1966 a :10 a type of relation that satisfies some emotional need in each party involved, very common in solidary groups like lineages and communities. With the articulation of the region in broader economic and social processes, the A rakmbut not only entered in conflictive relations with outsiders, but developed social ties with them. It has been suggested that economic transformations associated with the spread of capitalism, particularly in Western societies, prompted a rise in the i mportance of friendship (Adams and All an 1998:9 10; Giddens 1992, 2000 ). While friendship may have been equally important before the Madre de Dios region experienced rapid capitalist transformatio ns, it did increase contact with a larger spectrum of outsid e actors. T he articulation of the region to global capitalist processes through gold mining brought to the are a thousands of mining settlers, traders, and state officials . The greater incorporation into the state since the 1950s also meant the arrival of a wide array of other outside actors, such as missionaries, school teachers, anthropologists, health workers, NGO staff, and corporation employees. To a greater or lesser extent, the Arakmbut have befriend ed all of these o utsiders . As we have seen in Chapt er 3 , the Harakmbut call outsiders a mikos , a term that seems to have derived from the Spanish word amigo (friend), which is how rubber traders and Dominican missionaries called the Harakmbut to gain their trust when they first contacted them. While the ter m a miko resembles its origin in the Spanish word amigo , it does not mean that the Arakmbut consider every outsider a friend. It took me some quite time to gain the confidence of some
130 Arakmbut, and after more than a year and a half of field work among them, I can say that I became friends with only a few of them. I suggest, howev er, that despite what the term a miko may suggest, the fact is that the Arakmbut have never avoided contact with outsiders. Fri endship with outsiders is driven by a more instrumentalis t logic (interest in getting some benefits from the relation), and hence it is b ased on balanced reciprocity . With this kind of friendship, t rust is gained through long term reciprocal exchange . Friendship between the Arakmbut and outsiders is best described by what Wolf (1966 a :12) has called , in which gaining access to resources is vital, although this does not preclude the exis tence of some level of affect . Some relations among Arakmbut people and between them and outsider s are also governed by compadrazgo ( godparenthood ) ties . In most Catholic countries of Latin America and the Mediterranean region, g odparenthood results from several rituals of sponsorship, among which baptism is the most important. When godparenthood is b uilt upon the ritual of baptism, parents invite another person (usually of higher status) to be their chi godmother . In this case, g odparenthood consists of three sets of relationships: parent child, godparent godchild, and parent godparent ( Gudeman 197 1 ; Mintz and Wolf 1950 ) . Moore (pers onal com munication ) suggests that among the Harakmbut, the institution of compadrazgo was introduced by Andean mining settlers as a strategy to get access to indigenous lands. I suggest, however, that compadrazgo was not unknown to the Harakmbut. They may have become familiar with this Catholic institution when they were concentrated in missions by Dominican priest s in the 1950s and 1960s. T he missionaries baptized indigenous chi ldren as part of their evangelization project , for which parents wer e required to provide godparents . In Puerto Luz, godparenthood ties result only from baptism since the Arakmbut do not perform other ritual sponsorships required by the Catholic Church, su ch as first communion, confirmation or marriage. Every year, during the celebrations of the anniversary of the community foundation,
131 the priest of the Boca Colorado parish church visits Puerto Luz to baptize children. During the central day of the fiesta, several children are baptized. Godparenthood i s also based on a morality of reciprocity that furthers social solidarity . Parents and g odparents become linked by a quasi religious bond tha t i mplies obligations such as gift giving and a strong moral prohibition on coming into conflict with each other. Because of the solidarity bonds they create, godparenthood ties are used as a social resource to gain access to land, capital , and other scarc e resources. This explains why the Arakmbut prefer to choose outsiders as compadres since they have a greater socio economic status. While godparenthood can link together people of either differing or equal status (Mintz and Wolf 1950; Foster 1969; Paul 1942), the first type is more common among the Arakmbut and other Amazonian groups such as the AshÃ©ninka (Killick 2005, 2008, 2010) . Table 4 2 shows that between 2008 and 2012 the majority of baptismal godparents in Puerto Luz were outsiders. Fifty eight p ercent of godparents are Andean or mestizo people, while only 27% are Arakmbut. Outsiders usually have more access to resources and networks with the outside world. Godparenthood relationships with gente b lanca enable the Arakmbut to get access to economic capital, which is scarce for the Arakmbut but not for outsiders. For settlers and mestizos, godparenthood allows people to get access to co mmunity lands or indigenous labo r. Even when other Arakmbut are chosen for compadrazgo relations, the godparents are of a higher status than the parents. Arakmbut people who own chupaderas are most desirable as god parents. I also learned of an Arakmbut young man who was the godfather of several children in the community. This Arakmbut was a prosperous merchant and owner of a cloth es store in Lima.
132 Table 4 2. Ethnic origin of godparents in Puerto Luz, 2008 2012 Ethnicity Number Percent Andean/mestizo Arakmbut Mixed Total 41 27 3 71 58 38 4 100 Source: Elaborated with data from baptismal records of the Boca Colorado parish. Many godparenhood relations with minin g settlers were built since the early 2000s, after a group o f miners established themselves in an area called La Cumbre, which despite being located outside the titled lands of the community it was used as a hunting and gathering ground. This invasion gave rise to violent confrontations with the community , but the conflict was finally solved when the Arakmbut allowed the settlers to extra ct gold in the area in exchange for rent or regalÃas, as it is known locally. 6 The invitados , began to pay weekly rent to community members on an individual basis. 7 During the years of greater production of gold in La Cumb re , each Arakmbut had between two and three invitados , although there were some who came to have five or six. The producti vity of gold in the area declined and most of the invitados left, but the Arakmbut still recall the good old days in which they made l arge amounts of money through the regalÃas. Over time, the Arakmbut and their invitados engaged in relationships of friendship and compadrazgo in order to strengthen the bonds between them and assure a stable access to capital and land. The interest in for ming godparenthood relation s was mutual. While the settlers used the relationship to legitimize their access to indigenous lands, the Arakmbut used them to gain a regular source of income and aid in times of need. Indeed, the Arakmbut started to resort to their 6 This event and its further consequences will be described in detail in Chapter 5. 7 Invitados also have to pay rent to the community, money that goes to a communal fund that is mainly used to cover the costs of the constant trips that community leaders make to Puerto Maldonado in order to attend meetings called by FENAMAD.
133 invitados in the event of an emergency. An Arakmbut from Puerto Luz described to me how he getting worse, what are we going to do? The money we have is not enough; we have to go see gave us t Even though invitados should pay rent to have access to community land s, the relationship between them and the Arakmbut is still framed in terms of friendship and godparenthood. This implies that the relationship is seen in terms of reciprocity , which involves not only the exchange of rent for the right to extract gold within community lands , but also the provision of gifts (goods and services) on the part of the invitados. Wolf (1966 a :16) asserts that when reciprocity between instrumental friends becomes too imbalanced, friendship may turn into a patron client relation. Similarly, s ome scholars argue t hat godparenthood relations, especially those formed between indigenous and mestizo p eople , may become more vertical and evolve into clientelism (Osborn 1968 ; van den Berghe and van d en Berghe 1966 ). This is more appare nt when the person who has the lower status , usually the indigenous one , tries to withdraw from the relation when it has become oppressive, and the higher status person seeks to enforce clientelism by involving his partner in deb t . The difference in status , which characterizes godparenthood r elations between indigenous and mestizo persons, reinforces the vertical nature of the relation, turning it into more clientelist . This is not the case for the Arakmbut, as their friendship and godparenthood relationships with the invitados lack a componen t of coercion or exploitation , . The Arakmbut have the freedom to terminate the relation whenever they feel it involves a highly disadvantage exchange.
134 Godparenthood r elations w ith settlers not only became a relatively reliable sour ce of financial capital; they also created economic dependence and prompted social differentiation within the community. Until the early 2000s, the Arakmbut practiced an artisanal gold mining, and productivity and profits remained small. For most Arakmbut , regalÃas supplemented the income they were already obtaining through their own work in mining, but for some they represented the opportunity to improve their mining technology and substantially in crease their profits. RegalÃas offered the Arakmbut the opportunity to rapidly access economic capital with which they would be able to buy the equipment necessary to mine gold at a more productive scale, especially chupaderas . Over time, only a few Arakmb ut managed to get enough capital, buy chupaderas , and break the economic dependence upon their invitados , although they continued demanding regalÃ as from them. For the majority, however, regalÃ as continued to be the only means to obtain cash income. School teach ers are also good prospects for godparenthood relations. Most of the teachers who run the elementary school and kindergarten in Puerto Luz are outsiders, primarily of Andean origin. They are very valued by the Arakmbut for their knowledge on and netw orks with the outside world. For this reason, they are usually asked to provide several services to the community, such as to write the minutes during community meetings, write letters to state officials and the oil company, or drive the sick to the health center in Delta Uno. School teachers are very influential actors in the community. They can even call work parties to repair or paint the school, or carry the milk provided by state programs, work parties that work better than the ones called by the commu nity officials, since school teachers have better mechanisms to punish defection (e.g. preventing children from attending class). Due to their great influence, school teachers are asked to become compadres by many Arakmbut. For example, in a conflict betwe en
135 the school principal and a n Arakmbut extended family, the principal was supported by a network of Arakmbut families among which many had compadrazgo ties with him. Godparenthood relations, however, are not as stable as one would expect them to be. During the first ye ars of the relationship the tie is stronger , but over time it may get weaker or vanish, esp ecially if the compadre leaves t he village. Many informants told me they used to have compadres , but they no longer had them. Most settlers come t o the area seeking to mine gold, and when the when the gold is gone they move to another place, which makes compadrazgo relations more difficult to sustain. Some of my info rmants told me they had not seen their compadres since they left the community sever al years ago. When gold in La Cumbre became scarce and productivity went down, most of the invitados left . Another source of instability in the compadrazgo with invitados is the tensions tha t this relationship often entails. My Arakmbut informants told me that an important reason for these tensions is that invitados often cheat on them about the amount of gold they extract weekly, and therefore pay the Arakmbut less regalÃas than they should . The Arakmbut do not trust a mikos because they are believed to lie very often. These tensions accentuated when gold production in La Cum bre decreased and the Arakmbut did not lower the amount of the regalÃas accordingly. It is precisely to countervail the lack of trust and make the relation more reliable that the Arakmbu t sought to engage in godparenthood relations with the invitados. Patrons and Clients An additional strategy that outsiders use to gain access to the community lands and res ources is debt peonage . Debt pe onage is a form of clientelist or patr on client rela tion based on indebtedness. Clientelism has been defined as the vertical relation between actors with unequal power who exchange different types of economic resources for political support, based on a sense of solidarity that creates the obligation to be l oyal to the other party (Eisenstadt and
136 Roniger 1984). Patron client ties specifically link A rakmbut with outsiders, such as timber patrons and state and corporation actors. There are no patronage relations among Arakmbut people. Locally called habilitaciÃ³ n , debt peonage is very common to many parts of the Amazon region. In Puerto Luz, it is primarily established by mestizo patrons interested in exploiting the , and other goods that are need ed to extract the timber, and the A rakmbut has to sell the extracted timber to the patron at a price set by the latter. Unlike compadrazgo relations with mining settlers, very few Arakmbut engage in debt peonage relationships with timber patrons. There are two reasons for this. First, the Arakmbut prefer gold mining to logging because the former provides faster returns. The shortest time one has to wait until enough gold has been extracted and is ready to be sold is one week, while in logging one has to wait several months to see any profit. Second, the Arakmbut would rather have friendship and compadrazgo relationships with outsiders than patron client relationships. This is because the latter involve more vertical relationshi ps and the A rakmbut people tend to avoid social relationships that place them in an openly subordinate position. They prefer more horizontal relationships such as friendship and godparenthood because these relationships offer them more freedom of withdrawi ng the engagement if they feel it is turning into a exploitative relationship , without experiencing any social sanction . Even though patron client ties imply a more vertical relation than friendship and compadrazgo , they are not the exploitative relations they used to be in other parts of the Amazon. Indeed, the literature describes how the system of habilitaciÃ³n (also referred to as aviamento in the Brazilian Amazon) economically exploited Amazonian indigenous people by creating debts that they were not ab le to pay (Brown and FernÃ¡ ndez 1991 ; Santos Granero and Barclay 2000; Varese 2002 ) . In Upper Madre de Dios, after the decline of the rubber boom and the missions,
137 patrons never had unlimited power over the indigenous population. When the Arakmbut were unde r the influence of missions, the missionaries controlled access to trading goods and the in direct contact with patrons and merchants. When the uence declined , gold mining and the setting up of a post of the Banco Minero in Boca Colorado, where they could sell to control access to goods and establish de bt peonage relations (Moore 1981) . A s gold mining represents the main source of cash income, the Arakmbut do not depend on debt peonage with a timber patron to get access to money and other desired goods. Therefore, if debt peonage terms turn too disadvantageous for the Arakmbut, they may terminate the relation without experiencing any social sanction . As a result , the Arakmbut are in a good position to negotiate more advantageous terms in the ir relationship with timber patrons . One morning, I was alarme d by the screams of a woman whom I would later that day learn was the wife of a mestizo timber patron. The woman was chastising her husband for having received though initially Marcos had been delivering some timber to the patron, he was not able to either extract more timber or give him the money back because his tractor bro ke down. In order to help Marcos complete the timber extract ion, the patron gave Marcos additional money, which irritated his wife even more and caused her rage to burst. Marcos had told them he would pay them back with the sale of two chainsaws he owned, but the sale could not be accomplished. Marcos had spent t he money advanced by the patron pa ying some debts to the boarding school his daughter was attending in Cuzco. Guadalupe told me that if they did not pay the debt their daughter would no longer be al lowed to stay in the school. The patron and his wife spent some days in the
138 community, during which they begged and even cried to have their money back, but Marcos ey were not able to pay the debt. Another patron who advanced them money on credit for timber chainsaw as part of the payment, but she was told the chainsaw had been pawned. This example shows that while Marcos was economically indebted to the mestizo patron, he was not politically obligated to him, as he had the chance to choose when he would give the money back to the patron. Although their greater access t o capita l makes patrons economically more powerful, they lack the political power that may assure them complete dominance over the Arakmbut. This explains why iinvitados have to resort to social ties such as godparenthood to strengthen the ir relation with the Arakmbut and make them more durable. A very different situation is the clientelist relation Hunt Oil has establish ed with the people of Puerto Luz. Compared to timber patrons, the oil company is tremendously more powerful, not only economically but pol itically, and may use that power to incline the balance of exchange almost absolutely to their side. According to the Peruvian legislation, titleholders of oil and mining concessions are granted the right to explore and exploit the mineral and hydrocarbon resources lying in the subsoil, but not any rights with respect to the surface land. When a concession overlaps surface land owned by third parties, the titleholder of the concession may either buy the land or obtain an easement, both of which re quire dire ct agreement with the landowner. If such agreement is not reached within 30 days, the titleholder of th e concession may request that state authorities impose an easement. If approved, the indemnification is appraised and the easement is granted. Hunt Oil w as granted Block 76 in 2005, which greatly overlapped the Amarakaeri Communal R eserve and som e of its bordering native communities . When they began seismic exploration, several seismic lines went through Puerto Luz
139 community territory. In compensation for the damages produced by these seismic lines, the Puerto Luz demanded the provision of several goods and services. Between 2011 and 2012, this wer generator, supplies for the school, and support to the sick through health campaigns and emergency evacua tions to Puerto Maldonado or Cuz co. Since not all in Puerto Luz favored the presence of the oil company within the reserve and the communities , the oil company used the compensation and gift giving to gain the oppo nents the community, such as presents for the children and fruitcake during Christmas time or other material rewards such as transport. The need for economic resources in some cases led the community members to accept the compensation and the gifts, even though by doing so they were implicitly legitimizing the presense of the company . According to Fox (1 994), the survival needs of the poorest members of society make them especially vul nerable to clientelist incentives. But in the case of Puerto Luz, it was apparent to me that for the majority of people the allure of gifts such as gasoline for the power ge nerator played a crucial role. Some of these clientelist relations take place at the individual level, since one of the strategies used by the oil company has been coopting indigenous leaders through gift giving. Community officials were particularly chosen with whom to develop patron client relations because of their capacity to convince other members of their communities. While these relations were kept secret, my informants mentioned to me some signs that they believed to be indicative of their exis tence, such as leaders being seen wheel drive trucks , or having some drinks with the company staff in local bars.
140 Community and Fiestas All the residence groups together form the settlement group. Although the Arakmbut have a relativ ely nucleated settlement pattern, interaction between households at the settlement group level is limited. Activities that involve all residents of the village are confined to communal meetings and fiestas. The communal meeting or assembly is not an indige nous in 1980. Since then, all family heads are required to attend these assemblies, which are held whenever making a decision concerning co mmunity affairs is needed. E ven during these meetings , however, family groups do not mix together; they occupy particular areas of the communal house where the assembly is held. The fiesta to celebrate the anniversary of the community founding attracts almost all the reside nts of the village. Cumbia music played by a hired band and the conspicuous consumption of beer are attractions that nobody wants to miss. This fiesta, which is held every December 8, is the only occasion in which the whole community gathers to dance and d rink beer. Nevertheless, as with the communal assemblies, I observed that during this event smaller groups based on kin and friendship ties formed to share drinks. It is common that when men get drunk, fights occur between members of different clans. I sug gest that t hese work as an informal mechanism to reduce existing tensions between kin groups . These fiestas are also the social space for i nteraction with the Arakmbut of the neighboring community of San JosÃ© del Karene. Members of Puerto Luz and San JosÃ© visit each other to participate in the anniversary fiestas the two communities organize each year . Since indigenous communities are widely spread out in the region, interaction with more distant
141 communities is less frequent. 8 Some families take advantage o f these events to visit some relatives they may have in the host community. Besides beer drinking, these fiestas include dancing and a soccer tournament with teams from Puerto Luz, San JosÃ© and Delta Uno. Soccer games and fights between drunk people also e nable members of Puerto Luz and San JosÃ© del Karene to reduce tensions caused by tribal rivalries, strengthening solidarity bonds between them. According to Moore (2003: 87), soccer games and drinking replaced the traditional rites that encouraged contact b etween malocas in pre contact times. Anniversary fiestas also provide an opportunity to strengthen relations with Andean miners and mototaxi drivers. As we have seen, tensi ons with invitados stem from disagreement about the amount of the regalÃas they give to the Arakmbut, who consider it does not represent the actual amount of gold they extract. Relations with the mototaxi mototaxi are always tense since the Arakmbut believ e the cost of the fare is too high the service bad. Miners and mototaxi drivers are usually asked to s upport the organization of the fiestas by donating beer, transporting it from Delta Uno or buying food and drinks that are sold during the fiesta to raise funds for the community. According to my informants, this support is asked in reciprocity for the authorization to provide transportation service between Delta Uno and Puerto Luz. Also, miners and mototaxi drivers are invited to participate in the soccer tournament organized during the fiestas. These soccer games also serve to alleviate tensions with the invitados and mototaxi drivers. While fiestas are an important space for socialization with outsiders, t he Arakmbut build most of their social relations w ith outsiders in the mining town of Delta Uno. Delta Uno was founded by mining settlers in the early 2000s, and as a result of the last gold rush in the region it 8 Since the establishment of a federation of communities, federation sponsored meeting s have allowed for interaction between communities of all ethn ic groups of Madre de Dios. I wi ll talk about this in C hapter 5 .
142 grew rapidly, be coming an important trading post in the area. Delta Uno is like any other fro ntier mining town in the Amazon: dusty (but muddy in the rainy season), chaotic, dangerous and with scarce state presence. Despite that, the role it plays in the area cannot be more important. Delta Uno provides the Arakmbut with a market where they can se ll their gold and purchase trading goods in their several gold shops and stores. The town also provides multiple services through restaurants, hotels, call centers, cybercafes, and brothels. It also constitutes a hub where indigenous people from different communities, miners and traders meet and catch up on news. Motorbikes provide daily transportation from Puerto Luz to Delta Uno for just 20 soles (US$ 7.12), using a winding trail that is covered in just half an hour. The Arakmbut travel daily to Delta Uno in order to sell their gold and purchase food, clothing, gasoline, mining equipment, and other goods. In Delta Uno they also come across their friends and relatives from San JosÃ© del Karene. They often take advantage of these encounters to have some drink s in the numerous bars existing in town. Families may also go there to spend the weekend. In Delta Uno the Arakmbut build networks with traders and miners, and reinforce their ties with San JosÃ© del Karene. While both men and women travel to Delta Uno, men do so more often. This is because men control the selling of gold, which is one of the main reasons why men visit Delta Uno. As men travel more, they have more opportunities to forge these networks. Cooperation at the settlement group level is minimum and in some cases, such as with barbasco fishing, it has significantly diminished. One form of cooperation at the village level is communal work parties. Communal labor is called by the community officials in order to repair comm unity infrastructure ( e.g. tra ils, bridges or the communal hall) or cut the grass in common areas like the soccer field or around the communal house and school. As with communal meetings, male heads of each nuclear family are obliged to provide labor to the work party or pay a fine. Ne vertheless, the level of attendance to labor parties is low. On one occasion, I
143 observed how the community officials had to hire a man who owned a lawn mower to cut the grass in the soccer field because very few people showed up in the labor party called t o do it. Communal fishing used to be a common practice in the past. All households of the community gathered to fish with barbasco in the Karen e River, which due to its large size required the participation of several households to close the river and catc h all the fish. This type of fishing was commonly organized to provide fish for consumption during communal fiestas . Today, however, it is rather rare . Population growth and increasing participation in gold mining seem to have made it more difficult to undertake collective action at the level of the settlement group. A census conducted in 1978 showed that there were 111 people living in Puerto Luz (Torralba 1979). In 2012, the population of Puerto Luz was 357 people, which repre se nts an increase of 221%. I n thirty four years Puerto Luz tripled its population. In this context, generosity and barbasco fishing have become activities that involve only the dwellers of residence groups. In the 1980s and 1990s, when anthropologist Andre w Gray conducted fieldwork among the Arakmbut of San JosÃ© del Karene, a man could be generous not only with members of his kin group but also with the community at large. In this community, s ome individuals used to offer fiestas and distribute beer to the whole community. Those who were known to have profited from gold mining were socially pressed to redistribute their profits by offering a party and inviting beer to everyone. By the time of my fieldwork, people preferred to keep their money for themselves instead of offering parties to the entire community. The community was too big for a single individual to offer everyone beer. One of my informants recalled that in the 1970s if someone hunted a tapir, he would share the meat with the entire community. No w game meat from a large animal was Also , with families spending most of the ir time in the mining camps it was more difficult for them to organize communal
144 fishing, which had become less common t han fishing parties involving only the households of residence groups. Work in the mining camps is also the main cause for the low tur out for community work parties. I argue that the the main function of the settlement group level is to provide the Arakmb ut with an instance to strengthen ties that are weak because of the lack of interaction among households beyond the residence groups. With so few ties between residence groups, the Arakmbut use fiestas to build ties between these groups and ease kin based tensions. Native communities are artificial assemblages of indigenous families who get together for specific purposes , su ch as gain access to a school or health post. In some parts of the Peruvian Amazon, the creation of native communities did not even r esult in the spatial concentration of families, who keep living in a dispersed manner. In fact, the Law of Native Communities doees not require indigenous families to live close to each other. Among the Arakmbut, and many other indigenous groups of the Ama of the Arakmbut social organization what explains the lack or social ties, or their weakness, at the village level, and not the break down of an alledgelly collectivist culture . Leadership The day after my first arrival in Puerto Luz, Fede rico spent almost all day in Del ta Uno. Upon his return, he referred to me he had been talking to the managers of Rosario Flores, a regionally famous folk singer who those days had been performing in Delta Uno. The celebratio ns of the anniversary of Puerto Luz find a good folk singer or cumbia band to entertain the fiesta. In the last years, the Arakmbut have been celebrating the community anniversary by holdin g a communal fiesta in which the main attraction is the performing of a well known singer or orchestra. It is the duty of the head of the community to hire these musicians, which has become an important source of prestige. The
145 more famous the artist, the m ore pre stige he gets. Rosario Flores was a pretty famous Andean folk singer in southern Peru, and hiring her would take not only a significant amount of money but also ability to convince her managers to perform in Puerto Luz. If Federico managed to hire R osario Flores, he would be able to demonstrate his generosity towards the community, which would yield him prestige. As in the past, leadership is still based on proof of skill and generosity towards the community, and the prestige they gra nt to the lead er. Nevertheless, the ways to express these abilities and generosity have changed. Indigenous Amazonians are well known for their appreciation for personal autonomy and a deeply held dislike of any form of subordination and authority. In part, this form of political organization has to do with the autonomy that households show in the provision of the necessary means of subsistence. There is no need for an overall coordinated action , except when it comes to defense against a common external threat. Leaders w ith coercive power over their fellow people emerge only in response to an outside threat or to lead a revo lt against an oppressive external regime. This explains why indigenous leaders are primarily war leaders whose authority is temporary, for it lasts as long as the external threat persists (LÃ©vi Strauss 1967; Clastres 1989; Pr ice 1981 g part ly upon their ability to speak well in public (Hvalkof and Veber 2005 ). The authority of the se leaders is not based on orderi ng others what to do , but rather on their ability to put forward an idea about what should be done and persuade others about it . Leaders do not concentrate powers of domination, but work as a focus point for others and, as such, their most important ability is to attract followers and keep them living near them. Traditionally, the Arakmbut h ad three categories of leaders: shamanic leaders, political leaders, and general leaders (Gray 1997a) . The shamanic leaders included those who were recognized to have a capacity to cure ( ) , and those who were recognized for their
146 ability to dream ( wayorokeri ). Both gain ed respect and influence for their knowledge and generosity in helping the community with health issues. Political leaders were called wantupa , a term that denoted respect and influence. The wantupa defended the community politically and sometimes even physically from threats from other indigenous c ommunities, both Arakmbut and non Arakmbut . 9 It seems that in the past, the wantupa was a warrior or military leader who brought together one or more malocas to fight an enemy . The wantupa was respected for his physical strength and ability to respond to h is group was very influential within his group, his power and position could weaken if he attempted to exert it in a coercive manner. General leaders are called wairi , a term used to refer to a respected ma n. The source of t his respect was the wairi community. A wairi was a successful person in hunting and go ld mining who widely distributed the fruits of his success. Although he d id not need to be a good speaker, a wairi exerted great influence in collective decision making. This influence, however, would continue as long as he was able to demonstrate generosity and concern about the community. To some extent, the wairi could also involve in the political and phys ical defense of the community. The authority of all these leaders expressed itself in the ability to tell others what to do, or decide or determine activities through propositions, an influence that was baaed on a notion of power that rested on social rela tions, respect and prestige. Arakmbut politics , however, is not only about personal qualities, but also about access to resources (Gray 1997a:156). People desire material goods, knowledge, and a pleasant life. Having the means to obtain such benefits is pr oof that a person have skills. These skills in turn show that the person has knowledge of the spirit world and outside sources of cash, goods, and 9 A form of warrior chief called ohpu appears to have existed prior to the rise of the wantupa , but the term is no t used anymore (Gray 1997a: 150).
147 services, all of which is widely recognized by the community. Access to resources does not depend on politica l control , but on skill and knowledge, which in most cases is available to all. As I already expalined, p restige and respect are gained through generosity to other people. The advantages of prestige are that prestigious men are liked, and the Arakmbut list en to what they say, which ensures that they have influence in the community. For men, the activities that constitute sources of generosity is the distribution of meat and the fruits of gold mining, and prestige increases as generosity is shown to broader social units, such as the clan or the community. Successful men demonstrated his generosity by distributing their gold profits through fiestas. 10 Generosity is the primary attribute of a respected and influential person, and is the only way of counteracting accusations of selfishness. When in 1980 Puerto Luz was officially recognized as a native community, a new structure of political offices was introduced. In accordance with the 1974 Law of Native Comm unities, this structure included a president, a secret ary, a treasurer, and a spokesperson. The first Arakmbut who were elected for these offices were young men who had some command of Spanish, had some formal education, had served mandatory military conscription, and because of all that were familiar with and inclined to deal with outside affairs. At the beginning, older men referred to these officials as wantupa , but in general holdi ng these offices did not grant much prestige. This was because the holders of these offices were yout h and had little experience, and hence had nothing to distribute. Because of their lack of personal resources, these young officials relied on donations from the community members to perform their duties , which often exposed them to accusations of misappro priating money. This was in contradiction 10 Another source of prestige among the Arakmbut is the ability to accumulate knowledge . This knowledge not only includes abilities in production activities such as hunting, house building, gathering or gardening, but also proficiency in performing stories, myths, songs and jokes. Oratory does not have the same value as in other Amazonian soci eties. The ability to speak well is in fact another form of entertainment, but what is important is not giving a long speech but a brief presentation containing jokes (Gray 1997a:145 7).
148 with the traditional notion of a generous leader, and the Arakmbut often refused to subsidize the few Arakmbut offer ed themselves for election, and those who did it were motivated by the desire of making contacts with non indigenous Peruvians. Also, t he responsibilities that men had for the welfare of their household s were so important that most frequently they refused to take on the new positions . In the early 1990s, elders still had a great influence in the politics of the community. Th is i nfluence remained important even though a new form of leadership, one based on the command of Spanish and familiarity with the nati onal culture, was already in course at that time. Community officials used to work together with and listen respectfully to the elders. They were consulted every time an important decision had to be made. These two forms of leadership worked simultaneously , and each one filled a political niche . Traditional leaders mediated internal political relationships, while officials of the community were purely boundary leaders, mediating connections with the outside world (Gray 1986:21) . The president and other comm unity officials constituted a form of foreign affairs ministry that dealt with external relations without taking important decisions (Thomas Moore, personal communication) . Although the president of the community is not an Arakmbut political authority, he became responsible for community affairs. While these community officials cannot decide for themselves, they may use the information they control to influence decisions. Because of this, community officials may end up making personal deals with outsiders a t the expense of wider interests. This is why outsiders like Hunt Oil take community officials for a drink in order to entice them with deals and to try to influence the community. When I visited Puerto Luz in 2011, the elders who were active in the 1980s and 1990s were much older and most of them had already passed away. Th ose who were still alive were too
149 old and their influence was much lesser than in the past. Their knowledge of the new problems faced by the Arakmbut and of the even more complex forms o f solvin g them was limited, and their role in decision making was almost nil. Those who were young leaders three decades ago are now playing the role of elders, but in this case their political influence comes out not of their knowledge of the traditional Arakmbut culture , but of their experience with the outside world . To a certain extent, there has been a fusion of traditional and modern notions of leadership. Wairi are now those who have previously been elected as presidents of the community , or have he ld a position in FENAMAD. These men are also the heads of their own residence groups and, as will be seen in Chapter 5, play an important role in mobilizing their kinsmen when facing an external threat. Their influence and prestige are still valued in term s of the generosity proved to the community, but today generosity does not necessarily come s from distribution of meat or drinks in a fiesta. These leaders express their generosity by using their knowledge of the outside world to b ring benefits to the comm unity. For this reason , a leader is subject to criticism if he does not show efficacy in accomplishing his duties and his prestige may decrease as soon as he is unable to prove skill and generosity. This is why Federico was so concerned about hiring folk s inger Rosario Flores, as that would please the community and increase his prestige . Sometimes, community officials, in their desire of satisfying the During my days in Puer to Luz, I was called to a meeting with some leaders and school teachers of the community. The purpose of the meeting was to ask me to advice Federico about his role as a leader. They were concer ned about the fact that he had been giving in to the pressure by those families who were in favor of the presence of Hunt Oil in the community.
150 Conclusion In this chapter, I have illustrated the main components of Arakmbut sociality and the ir economic , political , and cultural underpinnings . I have shown that, among the Arakmbut, sociality consists of a complex combination of social ties (kinship, friendship, and godparenthood), which give rise to both altruistic and instrumentalist relations . The prevalence of one type of sociality or the o ther depends on social distance. Kinship and friendship ties are characterized by convivial interaction and generosity only among co residents of clan clusters. C onviviality among close kin and friends is associated with a settlement pattern characterized b y the dispersion of the population in smaller, discrete compounds of households based on kin and clan affiliations. Beyond these residence groups, people interact less and hence social relations are less convivial and signaled by tensions stemming from env y, competition over women . In the language of Robert Putnam and his followers, there is little social capital at the settelement group level. The settlement group level, especially through fiestas, works to strengthen ties and release tensions among reside nce groups. Among the Arakmbut, there is also a strong inclination to develop ties with outsiders, especially with Andean mining settlers. Relations with outsiders are built primarily through friendship, godparenthood and patron client ties. Unlike kinship relations, ties with outsiders tend to be based on balanced reciprocity and therefore are more instrumentalist. In some cases, however, friendship ties with outsiders may be stronger than those with kin or other Arakmbut residing in a different cluster of households. One of the main motivations to establish relations with outsiders is to gain access to resources that the Arakmbut lack, especially financial resources, since outsiders generally have more access to economic resources and political influence. As relations with outsiders are driven by balanced reciprocity, any provision of goods or services made by the Arakmbut to outsiders is accompanied by the expectation of a
151 retribution. There is no generosity with outsiders. At the same time, any taking fro m strangers is followed by a retribution in the near future. There is no unilateral taking. In contrast with the predatory theory of Amazonian sociality, and some descriptions of Amazonian peoples as overly generous with strangers, the Arakmbut show neithe r generosity nor predatory forms of engaging with outsiders. Related to this feature of Arakmbut sociality is their avoidance of relations that put them in a subordinate position to outsiders . Nevertheless, some Arakmbut engage in debt peonage with timber patrons , while the entire community is in a client position regarding Hunt Oil. Debt peonage with timbermen is explained by t he fact that relationships with patrons are no longer as exploitati ve as they have been during the rubber boom. Since the decline o f the rubber boom in the region, timber and other patrons have not regained complete control of access to trading goods, which prevented them from creating bonds of dependence among indigenous people . Gold mining provides the Arakmbut not only with faster economic returns but also direct access to the market. Clientelist relations with Hunt Oil , in contrast, are more vertical and exploitative as the company exerts more power than timber patrons , and the Arakmbut have no control of the terms o n whi ch the relationship takes place . Unlike theories of Amazonian sociality that emphasize either conviviality or predation as the main concern of Amazonian peoples, the Arakmbut case shows a more sophisti cated combination o f both features. In the current situation, i t is very unlikely that Amazonian indigenous groups exhibit either a convivial or predatory form of relating among thems elves and to others. A n overemphasis on either form of sociality can only stand on social contexts that no longer exist for the majority of indigenous peoples of the Amazon: the complete isolation from outsiders or utter hostility towards strangers. On the one hand, not all kin or community relations are marked by conviviality an d sociability; on the other , even though social ties with outsiders
152 seem to be less altruistic, there is no predatory taking. With the exception of indigenous groups in voluntary isolation, whose con tact with the outside world would bring their decimatio n , most Amazonian societies live now in a world where the boundaries with the national society are inevitably shrinking, and where interactions with outsiders have in some cases become a way of surviving.
153 Figure 4 2. Puerto Luz village in 2012
154 CHAPTER 5 SOCIAL NETWORKS AND POLITICAL MOBILIZATION On the morning of June 5, 2009, Peruvians awoke to the shocking news of a massacre that had claimed the lives of 34 people near the northern city of Bagua. The violence erupted two months after thousands of AwajÃºn and WampÃs Indians, supported by mestizo f armers, had begun blockading a highway, demanding the repeal of decrees issued in 2008 by President Alan GarcÃa to facilitate corporate access to indigenous lands and resources in the Peruvian Amazon. The unrest had spread. In Madre de Dios, FENAMAD organi zed several strikes in support of these demands. While the area has a history of indigenous insurgency, the uprising culminating in the tragedy of Bagua seemed to mark the emergence of Amazonian peoples as a pivotal force ique 2009). The infamous events in Bagua aroused my interest in the role of social relation ships and the resulting networks in mobilizing thousands of indigenous Amazonians. These networks were based primarily on new Amazonian alliances with Andean indige nous peoples and a range of nonindigenous partners. Before 2009, the last successful Andean Amazonian alliance in Peru dated back to the mid 1700s, when Juan Santos Atahualpa, an Inca descendant, led the AshÃ¡ninka people of the central Peruvian Amazon in a rebellion against the Spanish. Since then, attempts at collaboration had been undermined by cultural perceptions stemming from the 2009 Amazonian uprising that social networks took on greater importance for indigenous mobilization. S ociologists of the resource mobilization theory have long been interested in th e influence of social networks o n social movement s . They have focused on the role that social networks play in the process of individual participation in social movements. According to these
155 scholars, social networks facilitate individual decisions to join a movement (McAdam and Paulsen 1993; Passy and Giugni 2001; Snow et al. 1980). N etworks based on frie ndship with individuals already e ngaged in a movement increase incentives for participation among potential activists (della Porta 1988; Gould 1993; McAdam 1986, 1988; Snow et al. 1980). Social networks also shape the individual preferences or perceptions in the decision making process regarding participation in collective action (Passy 2003). Resource mobilization theori sts contend that the meanings inhering in social networks build or reinforce individual identities, loyalties, and obligations , creating potential for movement participation ( Ferree 1992 ; McAdam and Paulsen 1993; Mueller 1992; Schwartz and Paul 1992 ; Snow et al. 1986 ). S ocial networks , however, are not only important for participation in social movements. They may facilitate acces s to th e resources required to put s ocial movements in motion . This issue has been explored by s ocial capital theorists , who see social networks as forms of social capital that may be used to access resources and gain political power. Among these scholars , two approaches are clearly distinguishable. From an economistic approach, some theorists view that result s from engagement in civic associations as well as values and norms of trust and reciprocity. As public good, s ocia l capital is said to allow individuals and groups to gain access to other forms of capi tal (economic, human, and cultural), and ultimately lead to economic prosperity and democracy ( Coleman 1988, 1990; Putnam et al. 1993 ) . Following this approach, some sch olars have c alled attention to the role social capital rights and resources. Within this perspective , social capital is seen as facilitating indigenous ability to access resources such as official recognition, legal titles for land claims, and funds for development projects, thereby enhancing local livelihoods ( Bebbington 1997;
156 Bebbington and Perreault 1999 ; Bebbington et al. 1993 ; Perreault 2001 , 2003 ; P erreault et al. 1998 ). How are social networks positive in increasing resource access? What is the role of varying social ties in enabling access to resources for mobilization? Both schol ars of social capital and resource mobilization theories tend to focus on formal networks, or to emphasize only some social ties like those based on friendship and acquaintance ship . The economistic approach of social capital , in particular, fails to distinguish between different forms of social capital, and to recognize the importance of informal social capital, such as kinship networks (Fine 1999; Harriss and De Renz io 1997). Individuals engage in multiple social ties that might have differing effects on social movement participation. Some s cholars have disting based strengthen solidarity within small groups, the latter bring social cohesion to the broader society (Granovetter 1973; Holzner 2004; Narayan 1999). What is the role of kinship, godparenthood, and clientelism Do different social networks differentially affect participation in social m ovements? H ow do new network alliances interface with traditional kin based networks among indigenous groups suc h as the Arakmbut ? From a political economy approach , French sociologist Pierre Bourdieu (1986) posit s that as individuals and groups are social ly differentiated, social capital is always unevenly distributed and hence it tends to reproduce power inequalities. In this sense, so cial capital accretion derive neither from rational choice nor haped by the ir place in social structure . Studies in this line of analysis suggest that social networks may enable certain individuals and groups to secure more capital than others (Mayoux 2001;
157 Wilshusen 2009). P olitical economists have critisized the ec o nomistic approach of social capital for its failure of consideration of the political conditions under which social capital is more likely to appear. Social capital tends to appear where external allies that provide support against state repression are ava ilable ( Fox 1996:1092) . Both resource mobilization and social capital theorists tend to use a voluntaristic approach, highlight ing only the positive effects social networks may have on prompting engagement in , and facilitating access to resources for, political mobiliza tion. The economistic approach to social capital, in particular, has been criticized for overlooking social differentiation and power relations within networks . Scholars in this vein have interpreted underdevelopment as a result of low s tocks of social capital, rather than as a result of social inequality (Fine 1999; Harriss and De Renzio 1997) . In what ways do power and inequality permeate networks? Given that social groups are differentiated in their access to economic capital and polit ical power, what is the role of social networ ks in reducing or widening social inequalities? How are indigenous organizations affected by existing ethnic and gender inequalities? The indigenous mobilization for land rights in Madre de Dios gives us the opp ortunity to analyze not only the varying effects of social networks on political mobilization, but also how power relations influence networks. In this chapter, I examine the influence of kinship, friendship, godparenthood , and clientelist networks on mobi lization for land rights and res ources among the Arakmbut , both at the individual and collective level. First, I discuss the traditional role of kinship networks (bonding ties) in mobilizing residence groups at the community leve l. Then I examine how the b uilding up of friendship networks with outsiders (bridging ties) enabled the Arakmbut to create inter ethnic organizations, get access to economic resources and technical assistance, and gain leg al rights to land. The chapter also analyze s the down side of social networks , focusing on how
158 networks the Arakmbut have forged with non indigenous actors at the individual level engendered conflicts withi n the community. Finally, I examine the way s in which FENAMAD represents a contentious arena and how it contributes to repr oduce existing power inequalities. I show t hat while bridging networks were more relevant in forming alliances and organizations, they also reproduced existing ethnic and gender cleavages . Indigenous Networks and Struggles for Land In 19 75, the reformist military government of Juan Velasco Alvarado was overthrown by another military regimen led by General Francisco Morales BermÃºdez, who passed legislation that liberalized extractive industries (mining, oil, and lumber) and relaxed constra ints on land ownership (Dean 2002:211). I n 1978 , the new military government enacted a new mining law that eliminated state control of the mining activity , and promoted private investment by reestablishin g the concession grating system . Several mining conc essions , most of which overlapped Arakmbut communities, were granted to large mining companies and medium sized miners. In addition to this, a new rise in the international price of gold in the early 1080s, led to the arrival of thousands of empoverished settlers from the nearby highlands. Unable to secure concessions, these small miners were put to work for large holders of mining concessions (Pacuri and Moore 1992). Many Arakmbut communities sa w their lands encroached by corporations and small miners. A lthough by that time the Law of Native Communities, which officially recognized Amazonian indigenous settlements and granted titles to their lands, had already been passed, no Arakmbut settlement had their lands titled. The presence of m ining companies an d settlers caused serious impacts on Arakmbut livelihoods, such as soil erosion in the more fertile agricultural areas and forests, and the loss of important mining areas, which threatened the Arakmbut ability to making a living through gold mining (Moor e 1983) . The Arakmbu t responded by collectively de fend ing their lands , which led
159 t o numerous conflicts and violent confrontations with the miners . In this process, social networks have played both positive and negative roles in facilitating indigenous mobi lization. Networks, Recruitment and Access to Resources When the Arakmbut faced a common threat, especially the encroachment of their lands by outsiders, networks based on kinship ties played an important role in facilitating collective action . While in th e everyday life solidarity bonds are restricted to members of kin groups , a common threat may prompt solidarity bonds t hat overrides the divisive effects of kinship and bind all members of the community together . This community level solidarity is based on the fact that land is not owned by kin groups but co owned by all members of the community. Therefore, any external threat to this common property is perceived as affecting the interests of the majority or all members of the community. Once this community solidarity is set into motion, kin solidarity enhances mobilization at the settl ement group level . Kin ship solidarity enables heads of residence groups to mobilize their own kin smen and friends, which is facilitated by the concentrated settlement pattern t hat characterizes these clusters. Unlike other Amazonian indigenous groups that have a dispersed settlement pattern, the Arakmbut live in concentrated settlements. After fleeing the mis sion of Shintuya, the different maloca groups returned to their homelan d s and founded new settlem ent s in which they did not reproduce their precontact social organization in separate malocas . Instead, the new settlements were formed by members of several malocas , and the organization in malocas was abandoned and replaced by individual houses . In Puerto Luz, t hese houses were initially arranged around a soccer field but over the years they dispersed and formed smaller re sidence groups . Although the Puerto Luz settlemen t is divided in to several residence groups, the settlement pattern is still rather concentrated, as the distance between residence group s is short. In the event of an external threat, spatial proximity between households within and
160 between residence groups facilitates communication and a rapid reaction. Even work in the more distant mining camps is not an obstacle for mobilization, since people constantly move back and forth between the village and the camps, which improves the spreading of information. Soc ial and cultural changes that led to concentrated settlement patterns brought about wider networks of alliance and communication among the Arakmbut. The shift from dispersed malocas to concentrated villages was encouraged by the Arakmbut residence in the mission of Shintuya. In Shintuya, most of the maloca groups split up in different settlements that established themselves close to each other within the radius of the mission (Fuentes 1982 : 167) . In Gran Pajonal, the AshÃ©ninka formed larger settlements for different reasons, but the effects on networks and communication were the same (Veber 1998) . Traditional small and widely dispersed local groups gathered around headmen who took a leading role in securing schools . Over time, the school building process led to the emergence of solidarities that tied many scattered settlements and facilitated the formation of networks of communication and cooperation towards common goals. This in turn paved the way for efforts to orga nize and legally constitute native communities. This involvement in the process of community organizing and land titling explains the successful AshÃ©ninka resistance to colonization and the forced recruitment attempts by the Maoist inspired Sendero Luminos o (Shi nning Path) guerrilla (Hvalkof 1994 ) . These new solidarities enabled the Arakmbut of Puerto Luz to collectively expelled the small miners and mining companies that have sought to enter the community lands to extract gold since the late 1970s (Moor e 1 983). One of the most recent examples of this occurred in 2003, when hundreds of Andea n miners settled in La Cumbre, an area the Arakmbut of Puerto Luz used as a hunting and gathering ground. Although this area was part of the Amarakaire
161 located outside the titled lands of the community, the Arakmbut considered it to be part of their ancestral territory. In response to this threat, nearly all men, women, elders and even children of the community, holding bow s and arrows, rallied to the area in order to throw the intruders off the community lands. The head of Puerto Luz at that time described the rally as follows: One comunero went hunting into the forest and found mining colonists who were exploring the area, holding crowbars and machetes. After that, he found them making a trail. Then he found that the colonists had cleared the forest to establish themselves in that area. It was like two hectares. They had already cleared all the forest there. Then the comunero came back to see me and said that colonists from comuneros and suggested that we go there. Everybody agr and everybody, women, men, youth, children went there. When we got there we realized that it was true that they had cleared like two hectares of forest, it was in an upland area. We found them clearing more forest with their machetes. Then we And then we stopped them. Kinship networks are also important in en abling the formation of alliances between Puerto Luz and the neighboring community of San JosÃ© del Karene. Some families from both communities are linked by kinship and marriage ties . Due to the lack of marriageable women in San JosÃ© , some men from this co mmunity married Puerto Luz women and moved to this community. These ties create d solidari ty bonds between members of both communities, facilitating political alliances when confronting common threats. In 1992, for instance, about seventy five Arakmbut men from both communities sailed down the Karene River to the trading building material that municipal officials had offered for the two communities to construct new school s. The Arakmbut had learned the municipal official s were selling off the buil ding material (Gray 1997a).
162 Nevertheless, this solidarity is activated as long as a perception of a common threat exists, becoming dormant after that perception fades away. When not in the face of a common enemy, everyday life within the community is characterized by household autonomy and tensions between households, even among members of the same clan. Alliance between Puerto Luz and San JosÃ© may turn into hostile rivalry, as occur red when members of San JosÃ© entered the Puerto Luz lands to harvest caÃ±a brava, a woody stalk used to make arrows. Fearing there would not be enough of this resource for their own use, the Puerto Luz people refused to allow San JosÃ© to take the canes (Gra y 1997b). Some networks have the effect of inhibiting rather than prompting community solidarity. The rent ( regalÃas ) that invitados pay for the right to mine gold within the community has been a stable source of capital for many Arakmbut families. When over the years the Arakmbut developed friendship and compadrazgo ties with their invitados , community level solidarity was difficult to achieve. The personal loyalties between the Arakmbut and their invitados that these ties engendered collided with the i nterests of the community as a whole. Thus, w hen some Arakmbut of Puerto Luz, fearing to lose their communal lands at the hands of the invitados , decide d to expel them from the community , those with friendship and god parenthood ties to their invitados oppo s ed the decision, which l ed to the failure of these attempts. Thus , the se social ties have been beneficial at the family level but harmful at the community level. Also, the rent invitados pay to the community may have also played a role in discouraging the se initiatives. This rent is the most important source of funds for the community coffers, which is used primarily to cover the travel expenses of the headmen to Puerto Maldonado or to support community members in need.
163 The Arakmbut of Puerto Luz have also been unable to forge community level solidarity against Hunt Oil. During my stay in Puerto Luz, most of the families favor ed the presence of this corporation within the community lands , while only a few were in opposition to it. My interviews suggest that in this case the difficulty to reach unity among all members of the community mainly stems from the patron client r elationship that the company established with the community . Although it is obliged by law to compensate Puer to Luz for damages to its land s and environment, the compa ny support . Thus, Arakmbut families supported Hunt Oil as long as the company provided the community with petrol for its power generator, medical attention, supplies for the school , and gifts for the children . Another factor that accounted for the lack of community solidarity against Hunt Oil was the lack of a shared perception of the oil company as a threat to the collective interests of the community members. In a context where ties among households of different residence groups are weak, unity is difficult to achieve if there is not a common perception of external threat. This lack of perception of Hunt Oil as a common threat was a resul present itself as bringing benefits to the community , minimiz ing the potential environmental damages that oil development may bring to the Amarakaeri reserve and community lands . This not only caused internal tension in Puerto Luz but also worked as a divide and conquer strategy aimed at F ENAMAD . FENAMAD considers the presence of this corporation as a threat to one of the most pristine rain forests in the region and the indigenous livelihoods derived from it. tactics are very common among multinational oil companies willing to drill for oil in territories of native Amazonians. region, ARCO employed not only the distribution of jobs and gifts, but also the provisi on of
164 development projects to generate compliance and weaken resistance to its operations among indigenous communities of the Pastaza River (Sawyer 2004). Even though in its campaign against Hunt Oil, FENAMAD informed the communities about the threats that oil development represented to the natural environm ent within the reserve, there were still some community members who believe d that benefits outweigh ed the threats and therefore chose to support the company. These differences in opinion cut across k in groups since members of the s ame residence group or clans showed different positions regarding the oil corporation. Not all networks with outsiders , however, have negatively affected the collective interests of the Arakmbut. When the liberal reforms of Morales BermÃºdez brought increasing numbers of mining companies and settlers to Arakmbut lands , the need for securing legal rights to land and forging more permanent, supra community alliances became more apparent. Nevertheless, geographical distance betwe en Arakmbut settlements, ethnic differences, historical conflicts between ethnic groups, and the lack of knowledge and resources were major obstacles for organization and mobilization. In this context, non kin ties proved to be more relevant in facilitatin g the building up of networks not only between different communities and ethnic groups but also with non indig enous actors. Networks with non indigenous people were instrumental in helping the Arakmbut to set up organizations that would help them secure land titles and funds to carry out development projects. In the early 1980s, s everal local ethnic federations and the AsociaciÃ³n InterÃ©tnica de Desarrollo de la Selva Peruana (AIDESEP), an umbrella organization that encompasses local federations, already exist ed and offered an organizational model for native Amazonians . The federation a loose association of autonomous settlement s of a particular ethnic group (Smith 1985, 1996) , was becoming the representative organization common to the majority of
165 Amazonian indigenous peoples . T he FederaciÃ³n Shuar in southeastern Ecuador and the Cong reso de Comunidades Amuesha in central Peru, the first ethnic federations to be formed in the Amazon, had been founded in the 1960s . The Arakmbut , however, had neither p revious experience at forming organization s beyond the settlement g roup nor funds to undertake it. During the 1970s and 1980s, several Peruvian and foreign anthropologists , such as Thomas Moore and Andrew Gray, conducted research among Harakmbut communitie s, building up close and enduring relations of friendship with them. Some anthropologists have maintained these ties for decades, and even though they left Madre de Dios after completing research , they keep returning very often to visit and help the native s . Anthropologists were familiar with the model of ethnic federation and its potential for supporting indigenous people to legally defend their lands and promote development. Thus, the Arakmbut relied on friendship networks that linked them to anthropologi sts and human rights activists, including local and international non governmental organizations (NGOs) and donor agencies , to obtain the technical and financial resources needed to form FENAMAD, which was established in January 1982. Anthropologists had no t only the knowledge needed to create a federation but also networks with dono r agencies and already existing federations. For example, an NGO helped the Arakmbut make contact with AIDESEP, which sent one of its leaders to support the formation of FENAMAD (Moore 1985 b ). The forging of networks with outsiders to create ethnic federations was a common feature in the Amazon. In most cases, the formation of these federations wa s encouraged by urban based non indigenous individuals and organizations such as anthropologists, missionaries, United States Peace Corps workers, and political parties. In Ecuador, Salesian missionaries promoted the creation of the Federa ciÃ³n Shuar to counteract the growing advance of colonization into Shuar territory, which threaten ed to deprive the mission of converts and labor (Salazar 1981). In Peru,
166 this role was played primarily by anthr opologists such as Richard Smith, who played a key role in the formation of the Congreso Amuesha , the first ethnic federation to be formed in Pe ru (Smith 1996). In Madre de Dios, Thomas Moore, an American anthropologist who arrived in the region in 1973 to conduct doctoral research among the Arakmbut, encouraged and provided technical advice for the establishment of FENAMAD. Centro Eori de Investi gaciÃ³n y PromociÃ³n Regional (Centro Eori ) , an NGO founded by Moore, supported FENAMAD with financial aid and technical assistance for many years. During its initial years, the federation operated within unding was made available through the support of Oxfam UK and Oxfam America, the federation moved to its own building. 1 Several reasons explain why the Arakmbut chose to forge networks with human rights activists, and not with state officials , political p arties , or missionaries, to garner resources for the formation of an ethnic federation. First, the state , missions and political parties have historically been dominated by white or mestizo people who have often shown racism and authoritarianism towards in digenous people. In the Amazon, the way they often approached indigenous people has been marked by paternalism, clie n telist relations , and attempts to co opt indigenous organizations . Most left wing parties saw the form ation of indigenous federations in the Amaz on with suspicion and hostility, while others tried to absorb and control them (Smith 1996) . Sometimes the natives, led by messianic beliefs in external salvation from colonial explo i tation , allied with Marxist guerrillas only t o be brutally repressed by the Peruvian armed forces (Brown and FernÃ¡ ndez 1991) . More lethal revolutionary parties, such as Shining Path and the Movimiento Revolucionario TÃºpac Amaru (MRTA), forcefully recruited, enslave d, and massacred AshÃ¡ninka Indians (Rojas Zolezzi 2008). Al l these negative experiences resonated 1 Currently , the federation owns a modern building equipped with tel ephone, internet, secretary and an auditorium for meetings.
167 throughout the Amazon, creating distrust towards leftist political parties. In Madre de Dios, n etworks with activi sts and NGOs , especially international ones , allowed natives t o bypass vertical relations with politica l parties , allowing resources and information to flow more freely. The newly created federation, however, faced obstacles stemming from tra ditional relations associated with the old mission system. One of these obstables was the great control the Spanish priest of Shintuya, Javier Ignacio Irairoz , had over the Arakmbut and Wachipaeri who lived in the mission . 2 Father Ignacio had developed a patronage relation with the natives of S hintuya, where he ran a sawmill, raised cattle, and had a monopoly on gas oli ne distribution in the area (Moore 1984 b ) . rojects required the natives to work everyday, but the natives were not used to this style of work and ran away after working for a few days. To secure native labor, Father Ignacio endebted the natives b y selling them goods like chainsaw s on credit and all owing them to pay them off with labor (Fuentes 1982). When FENAMAD was formed, Father Ignacio became one of its major opponen t s, for the meetings the federation called for in Puerto Maldonado often distracted the native s from their duties with the sawmill. To prevent the native delegates from attending the se meetings, the priest scheduled labor i n the sawmill the same day s the federation meeting s were s et up , or simply he did not sell gasoline to the delegates. According to Moore (personal communication), the Shintuya natives saw in FENAMAD an opportunity to free themselves from the priest. To a certain extent, the federation contributed to break up the patronage relations between priests and natives . Once FENAM AD was formed, it served as a resource to defend indigenous rights, especially rights to land, and bring development to their member communities. Since its inception 2 The Arakmbut and Wachipaire that remained in Shintuya eventually formed a native community that was officially registered in 1974.
168 land and natural resources (Moore 1987; Wahl 1985) . To accomplish this, the federation resorted to the Law of Native Co mmunities , which had been passed in 1974 by the reformist government of Juan Velasco Alvarado as part of a series of structural reforms inten ded to diffuse social discontent in rural areas. The law was the first state attempt to confer legal recognition for settlements and collective land rights to the native inhabitants of the Amazon region (Dean 2002). The Arakmbut realized that their recogni their lands would enable them to better protect their territories from outsiders. The role of FENAMAD in implementing the titling process was to sign agreements with state agencies in charge of titling indige nous lands and to obtain financial and technical support from NGOs. Land titling was implemented through projects run by the federation or NGOs, which supported indigenous settlements with the long and complex process of registration, land demarcation and territory extension. Several Arakmbut settlements were officially registered and received land titles through these projects, including the communities of Puerto Luz, San JosÃ© del Karene and Barranco Chico, which had their lands titled in 1986. The federat ion also helped the Arakmbut to obtain concessions for gold deposits located within their communities, and to get the support of NGOs and Congressmen to draw up and pass bills of mining legislation. Applying for mining concessions within community lands wa s used as a strategy to halt the advance of mining settlers into indigenous lands. FENAMAD also managed a number of development projects intended to make communities more economically autonomous. These projects covered a wide array of issues, including edu cation, agricultural production, health, environmental conservation, ecotourism and the protection of indigenous peoples in voluntary isolation. To obtain financial and technical assistance to implement these projects, federation leaders forged networks wi th international
169 donor agencies, mainly with Scandinavian organizations such as Norwegian Age ncy for Development Cooperation ( NORAD ), Danish International Development Agency ( DANIDA ) , and Rainforest Foundation Norway, but also with others such as Bread for the World, W. Alton Jones Foundation, Global Environmental Facility ( GEF ) , and International Benevolence Initiatives (IBIS ). Funding received through these projects helped the federa tion to cover its administrative costs. Emphasis was placed on educational projects that provided indigenous youth with scholarships for high school and college education, waiver of application tests to universities, and training in bilingual education. Th the Territorial and Sustainable Development in Madre de Dios Program (also called Plan Karene), initiated in the early 1990s . The project was funded by DANIDA and implemented by FENAMAD with the support of the Internatio nal Work Group for Indigen ous Affairs (IWGIA). British anthropologist Andrew Gray, who had carried out doctoral dissertation in San JosÃ© del Karene between the 1980s and the 1990s, had conatcts with IWGIA and played a key role in securing the funds. The pr oject had two main foci: territorial defense and econom ic sustainability based on self sufficiency. The territorial component incor porated a strategy to defend Arakmbut lands by strengthening its legally recognized boundaries, removing settlers through lega l action , and providing training in indigenous rights and Peruvian law . The economic componen t included revolving funds for breeding small animals , which was intended to improve local diet with the consumption of protein (IWGIA 1999) . While the land defense component succeded in t itling three communities and evicting numerous settlers, t he impact of the project in promoting self sufficiency was limited, since the Arakmbut eventually became increasingly dependent on gold mining to secure their livelihoo d (IWGIA 1999) .
170 These projects , I contend, were less important in bringing material benefits to the communities than in building capacity, both at the individual and the organizational level. Several individuals who received higher education through FENAMA projects eventually became federation off icials or were hired as staff for the projects implemented by the federation . Participation in the management of these projects also became a source of training, money and social status, which in turn enabl ed so me individuals to forge and expand their personal social networks , and eventually get positions within the local and regional governments. Through the management of these projects the federation also enhanced its prestige as a competent administrator, whic h in turn increased its ability to attract more international funding. The indigenous individuals who led the formation of FENAMAD and became its officers belonged to a political elite whose authority no longer derives from traditional sources of knowledge and influence, but from their mastery of intercultural relations. Indeed, they received formal education at schools in the Dominican missions of Shintuya and El Pilar , in Madre de Dios, and Quillabamba and Sepahua , in Cuz co. Additionally, after finishing school, many of them served mandatory conscription in the army. These experiences provided them the necessary skills to successfully navigate in the national society: literacy, command of the Spanish language , and exposure to the national culture in its re gional variant. Thus, upon returning to their communities, these individuals led the process for the official recognition of their settlements and, due to their knowledge of the state bureaucratic structure, were appointed as their first headmen. They also cultivated friendship ties with the anthropologists who would eventually help to create FENAMAD. Organizations as Networks While FENAMAD was primarily an initiative of Arakmbut communities, it became and worked as a multi ethnic federation. Originally, Thomas Moore and Andrew Gray promoted the
171 idea of forming an Arakmbut et hnic federation , but the Arakmbut rapidly realized the importance of establishing alliance s with the other ethnic groups of the region to consolidate the nascent organizati on, and opted for a multi ethnic federation . Thus, when the fede ration was created in 1982, the Arakmbut were supported by the Yine of Diamante and the Matsigenka of Shirigayoc (FENAMAD 2007). Also, as the federation proved to be efficient in helping their member communities to get official recognition and land titles, communities from other ethnic groups constituent base grew from seven to 32 communities from all over the M adre de Dios region. This multi ethnic character contrasted with originally thought as associations of settlements belonging to a single ethni ethnic character was reflected in the name it adopted, which re fers to a geographical area (Madre de Dios) rather than to a particular ethn icity. Nevertheless, except for including delegates from all ethnic groups Even though F shared a history of land invasion by outsiders , their ethnic differences have worked against the building of inter community solidarity. One of the main reasons for this is the rivalry, distrust and conflict that have long characte riz ed relations among ethnic groups in Madre de Dios . In pre Columbian times, most of these groups considered themselves enemies and remained in a per manent state of war, even among groups of the same ling uistic family such as the Harakmbut. These conflict s were exacerbated by the incursion of rubber traders and missionaries into indigenous territories . For example, Zapiteri and Aras aeri groups, fleeing the rubber gatherers, penetrated into Arakmbut territory, causing conflicts over natural resources. Many of these groups fought over the trading goods that missionaries distributed among some groups to attract them to their mission s . These inter ethnic
172 conflicts persisted even after the Harakmbut were concentrated in Dominican missions in the late 1950s. The c oncentration of groups in settlements larger than the traditional Harakmbut settlements also caused conflicts over the nearby natural resources, while competition over the trading goods and women controlled by the priests brought about accusation s of sorcery and even murders (Gray 1997b). Today people from the Arakmbut communities of Puerto Luz and San JosÃ© del Karene treat each other with suspicion and sometimes with some disdain. A man of San JosÃ© once told me that he considered the people of Pue In Puerto Luz, people call those of Shintuya sorcerers, clearly alluding to the Wachipaire who live there, who are widely known for their reputation as sorcerers. Interestingly, when I was in Shintuya, the people had the s ame opinion about Puerto Luz. In this context, FENAMAD has contributed to create bonds of solidarity between its member communities. By constantly bringing together the heads of its constituent communities, who otherwise would not have had much contact, t he federation has improved communication between distant indigenous settlements. During meetings called by the federation , community delegates not only exchange information and make collective decisions , but also, more importantly, construct a common ident ity and strengthen their solidarity bonds . 3 Moreover, the federation c alls, funds and leads protests in the form of strikes, rallies and demonstr ations in Puerto Maldonado, in which community members participate . Engagement in these protests also enables i ndigenous people to put aside their ethnic rivalries and build identity and solidarity . Rick Fantasia (1988) argues that solidarity does not result automatically from collective interests but from relationships that emerge in action, and especially in stru ggles. Thus, through the 3 As people from these different ethnic groups are bilingual, speaking both their own native languages and the national language of Spanish, linguistic diversity was not an issue. Spanish is used as a lingua franca that facilitates communication among member s of different ethnic groups.
173 creation of networks among its member communities, the federation has contributed to lessen the divisive effects of ethnic based parochialism, making po ssible the existence of a multi ethnic federation. FENAMAD has also been key in lessening the costs of mobilization of widely dispersed communities in a large region. Due to long distances between most communities and Puerto Maldonado, mobilization involves high costs of transportation, food and lodging for the protestors. The federat ion not only provides protestors with shelter by allowing them to stay in its own building while in Puerto Maldonado , but also, when available, allocates funds to co ver part of these costs . The costs of mobilization significantly decreased with the growth in communication infrastructure that the Madre de Dios region has experienced in the last decade. Today almost all native communities in the region have a public satellite telephone or access to cell phone and Internet service in nearby towns such as Delta Uno or Boca Colorado. Through telephone or e mail, information between the federation and its member communities has improved and it is now fluid and faster . In additi on, the completion of the Inter oceanic highway in 2011, an asphalted road that connects the Brazilian state of Acre with the Pacif ic coast of Peru, has substantiall y reduced geographical distance between Puerto Maldonado and many communities located in the middle and lower part of the Madre de Dios Basin. For example, in the 1980s it took a t wo to three day boat ride to get from Puerto Luz to Puerto Maldonado. Now the same distance can be covered in four hours, and leaders of Puerto Luz constantly travel back and forth from their community to Puerto Maldonado, where they attend meetings or par ticipate in protests. These improvements in communication have considerably reduced time and costs of coordination for mobilization and transportation of protestors.
174 On the other hand, the mobilization of also depended on e xisting kinship and ethnic ties linking federation leaders and members of communities . Kinship networks are used by federation leaders to forge political allian ces and recruit supporters for mobilization. Federation leaders mobilize their own kinsmen and t hose groups with which they establish alliances. When federation leaders call for a mobilization, it is very likely that their kinfolk will join, while other kin groups have to be persuaded to participate. This is accomplished by convincing the leaders of these kin groups, who in turn will mobilize their respective kinsmen. Once federation leaders manage to form an alliance with specific kin groups, these groups become their followers. For example, in 2000 a federation leader from the Shipibo group mobilized six communities to throw a Korean mining company off the lands of the community of Aras aire. The six communities that supported the mobilization were primarily Shipibo communities . When the federation leadership is dominated by a particular ethn ic group, federation led mobilization is often supported by communities of the same ethnic group. But FENAMAD has not only relied on existing networks, but also has built new alliance networks with other sectors of the popular movement in Madre de Dios . Th ese networks have been crucial in strengthening the federation started forging these networks in the early 1990s and over the years the range of these alliances has been expanded to encompass almost all rur al and urban popular organizations in the region. While its main ally has been the FederaciÃ³n Agraria Departamental de Madre de Dios (FADEMAD), an organization representing peasant farmers who have come from the nearby Andean highlands, FENAMAD has also es tablished alliances with loggers, miners, Brazil nut harvesters, school teachers, the settlements on the outskirts of Puerto Maldonado , and even mototaxi drivers. In 2000, the federation joined with all these organizations to create a
175 coordinating body cal led Asamblea Regional de Madre de Dios (GarcÃa 2000), which later changed its name to Alianza de Federaciones de Madre de Dios. In 20 08, the Alianza joined the the Frente para la Defensa de los Intereses de Madre de Dios, the ConfederaciÃ³n General de Traba jadores del PerÃº, and other social and political organizations to form an even higher level of coordination called ComitÃ© de Lucha de Madre de Dios . Mobilizing as part of the larger popular movement in the region has proved to be extremely advantageous for this. In 2002, after 10 years of petitioning and several mass demonstrations organized by FENAMAD with the support of the Alianza de Federaciones , the Amarakaeri Communal Reserve, a natural p rotected area intended to benefit the surrounding Harakmbut, Yine and Matsigenka peoples, was finally established. In 2008, the Alianza de Federaciones launched a regional strike to make local demands and support an Amazon wide uprising led by AIDESEP agai nst a package of legislative decrees passed to ease the implementation of the Free Trade Agreement with the United States. Puerto Maldonado was paralyzed during three days in which the regional government offices were occupied and ultimately burned down. T hese and other protests throughout the Amazon obliged the Congress to repeal two of the most controversial de crees. A year after , however, little had been done to revoke the remaining legislative decrees. AIDESEP then organized new demonstrations in differ ent parts of the Amazon. In June, shortly after the tragic events of Bagua, a new regional strike in protest for the massacre and demanding the definitive revocation of the decrees was organized by the ComitÃ© de Lucha de Madre de Dios , ending up in a publi symbolically burned down signs representing the legislative decrees.
176 Two other events illustra te the r ole of different social networks in indigenous mobilization . The first one took place in October 2009, when 200 native s from seven communities rallied to SalvaciÃ³n, where they would occupy Hunt O the company to halt its operation s within the Amarakaeri Communal Reserve. Hunt Oil had installed in 2007 its base camp in SalvaciÃ³n, a small town in the Madre de Dios province of Manu, in preparation for s eismic exploration activities . Indigenous leaders sailed upriver for three days , collecting members of eight communities in their way up to Shintuya, from which they would continue by road. The mobilization , led by FENAMAD, also involved communities that do not border the reserve and hence have no t been designated as beneficiaries, such as Boca Inambari, which nonetheless showed solidarity with the communities directly affec ted by the company . Manuel Kameno, a n old Arakmbut leader of Boca Inambari, took a leading role in mobilizing not only members of his kin group but natives of several other communities. The mobilization was carried out not only by Arakmbut communities, but also by Yine and Matsigenka Upon arriving at Shintuya, the prote sters were received with insults by the Shintuya dwellers, who had previously signed an agreement with Hunt Oil to accept US $30,000 in compensation for allowing the co mpany explore for oil within their title d lands. In SalvaciÃ³n, small pea sant settlers of the area, group ed in the FederaciÃ³n de Campes inos de KosÃ±ipata, asserted activities, expressing their solidarity with the F ifty policemen were sent to SalvaciÃ³n, who dissuaded the protesters from installations. Instead, the natives d emanded a meeting with high executives of Hunt Oil and high rank ing government officials, who never
177 arrived at SalvaciÃ³n. Hunt Oil would continue its operations in the rese rve, and by 20 12 it was preparing the opening of eight exploratory wells. Two years later, in March 2012 , FENAMAD allied with FEDEMIN , a body representing mestizo miners, to launch a mobilization against some decrees that banned mining activities taking pl ace outside a designated 500,000 government enacted these decrees in response to the environmental devastation caused by decades of illegal mining in Madre de Dios. Both miners and native communities that make a living through gold mining feared for their livelihoods. FENAMAD then demanded the government approve a bill allowing indigenous people to legally work gold within their communities. As the government did not meet these claims, indigenous people from ten native communities joined thousands of miners in a two protesters killed. The indigenous leaders who promoted the strike mobilized their own kin groups and allied communities. As the ten communities that parti cipated in the strike fought over the right to make a living through gold mining, this economic activity worked as a source of identity. Communities that are not involved in gold mining neither expressed solidarity with the mining communities nor support ed the strike. FENAMAD broke up its alliance with FEDEMIN when the miners abruptly suspended a negotiation process with state officials. Although its alliance with the miners, most of whom practice mining illegally, was highly criticized by conservationists and other sectors of civil society, including other indigenous organizations, FENAMAD succeeded in forcing the government to negotiat e the bill . Most of these networks, however, are temporary since they are based on changing and fragi le al liances. This has to do with at least three factors. First, a mong Amazonian indigenous
178 people, politics consist of ever shifting alliances . Indigenous people belong to different social units: household, residence group, community, ethnic group, federation, and nation state, each with its own sense of identity and solidarity. Indigenous people may emphasize one set of identities and solidarities depending on the circumstances. When conflicts internal to the community arise, household or residence groups ntity prevail and networks with outsiders may take precedence over internal ties. In the face of conflicts with outsiders, indigenous peoples may invoke their community, ethnic, federation or even national identities. Second, alliances with some social act ors become imperative under certain political junctures and, because of this, they have specific goals. When these circumstances change or the goals are accomplished, those alliances become less advantageo us and come to their end. And third , alliances brea k easily when one of the allies does not fulfill the accords on which the alliance has been based. There are some networks, however, that are more durable, such as the ones b uilt on alliances with anthropologists and FADEMAD. But networks are no t only unst able and dynamic; they are also the field of power relations, as we will see in the next section. Indigenous Mobilization and Power Relations While FENAMAD has been successful in securing land titles and development projects for indigenous peoples , it has also contributed to reproduce some existing social inequalities. Since its founding, the federation has experienced internal tensions stemming from the struggle among ethnic groups over its political and economic control. Power differences shape differenti s resources and government . As a result , a male gendered indigenous elite from the communities more integrated into the national society and market economy dominate the federation .
179 Ethnic and Gender Inequalities During s first years of existence , geographical proximity to Puerto Maldonado, where the federation was headquartered, determined differential access to its political and economic control. Puerto Maldonado, located in the lower part of the Madre de pital city of the Madre de Dios Region , it has access to communication facilities and is where state agencies and NGOs , with which the federation has to deal , are located . Arakmbut leader s, who live in the Upper Madre de Dios , could not afford their movement to Puerto Maldonado and visit the member communities, widely dispersed throughout the region. This discouraged the Ha rakmbut from running for election as FENAMAD officials. In this con text, leaders of ethnic groups of the ja, Shipibo , and Amahuaca) took advantage of their proximity to Puerto Maldonado to gain acces s to key leadership positions (president, vice president and s ecretary) and rapidly took over the control o f the federation. Representatives from communities closer to Puerto Maldonado would dominate FENAMAD from 1985 to 1995, as is evidenced by the elected presidents during this period, who all belonged to the ja people (Table 5 1) . During this period, wi th the support of Centro Eori the federation obtained financial resources and started to implement its first development projects. C ommunities from the L ower Madre de Dios benefitted more from these projects, either because leaders prioritized communities from their own ethnic groups as beneficiaries , or because they were unable to reach the more distant communities. The Ha rakmbut resented that, despite being the initiators of the federation, they were not able to control project resources . Because of this, they began complaining of being excluded from their benefits. Matsigenka communities also felt their demands were not being served and, instigated by an NGO working with Matsigenka communities in the neighboring Urubamba watershed, expressed a desire to
180 s plit with FENAMAD and adhere to another federation (GarcÃa 2003) . In response to these tensions, and as an attempt to withdraw from FENAMAD and form their own organization , the Ha rakmbut creat ed the C onsejo Harakmbut ( COHAR) in 1993. The explicit use of th e term Harakmbut in the name of this new organization was indicative of the ethnic character of the tensions that originated the split. Ethnic identity would remain important for COHAR even when it later incorporated the Matsigenka and Yine groups and shif ted its name to Consejo Harakmbut, Yine y Machigueng a (COHARYIMA). Table 5 1. FENAMAD leadership by ethnic g roup, 1982 201 4 Period ame Ethnic g roup 1982 1984 1985 1986 1987 1988 1988 1989 1989 1991 1991 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2006 2007 2009 2010 2012 2013 2014 Pedro Quique Roberto MasÃas FÃ©lix Kuakuibehue Miguel Pesha Miguel Pesha Miguel Pesha Antonio Iviche Antonio Iviche Antonio Iviche Victor Pesha Jorge Payaba Antonio Iviche Jaime Corisepa Klaus Kicque A rakmbut Arakmbut Arakmbut Arakmbut Shipibo Arakmbut Arakmbut Arakmbut Source: FENAMAD (2007) . In 1995 , the Harakmbut regained control of FENAMAD. At that time, increasing profits from gold mining and availability of pr oject funds made it easier for leaders from the distant Ha rakmbut communities to establish residence in Puerto Maldonado . With the Harakmbut control of the federation, its resources started to be allocated primarily to communities of the upper region. As with the Harakmbut ten years ago, thi s caused the communities of the lower part to complain of their lack of attention by the federation. ja leader of FENAMAD recalls how a rice husking machine purchased by Plan Karene was ha nded out to an
181 Arakmbut community, where rice was not produced. Tensions reached their peak in 2006 when, during a congress, a list dominated by candidates of the Upper Madre de Dios won the election. Leaders of the lower part responded by creating the Con sejo IndÃgena del Bajo Madre de Dios (COINBAMAD). A Shipibo leader and founder of COINBAMAD described the formation of this organization in these terms: There was no organization that represented the lower communities, because it is true that the work FEN AMAD was doing no longer reached all the communities. organization working up there and they have th eir reserves, they have their repeat, we got together and formed COINBAMAD with seven communitie s. Even though over time economic and communication improvements have reduced , there are still some communities that because of their greater isolation participate less in the , for instance, with the Matsigenka communities of Yomybato and Tayakome, which are located on the Manu River, Delegates of these communities are never elected for key positions within the federation , as most of them lack Western education, command of Spanish and speech, and familiarity with the national culture, which are highly valued as qualifications for bei ng a leader. Leaders of more acculturated communities take advantage of these limitations to politicall y influence delegates from isolated they are less integrated into t he market economy and therefore rarely can afford the high costs of transportation to Pu erto Maldonado, members of more distant communities do not have the chance to express their demands and influence the federation policies . Since the federation is usually short on funds, its
182 leaders do not visit these communities very often, nor can they pay for their for traveling to Puerto Maldonado. Only communities that can raise communal funds, such as those that re ceive rents from mining settlers, are able to cover the travel costs of their delegates. Another expression of ethnic based political struggle s took place when Antonio Iviche, four times president of FENAMAD, ra n for the presidency of the Madre de Dios Reg ional Government in the 2010 elections . This was the first time in the history of the region that indigenous people directly participate d in an electoral process. Iviche was proposed as a candidate by the delegates who attended a FENAMAD congress held in B oca Inambari in 2009. didacy , the delegates formed Pueblo Unido (United People), a political all iance with no affiliation to an existing political party. Indigenous communities wanted to break the dependence upon mestizo poli tical parties that did not represent indigenous The delegates then sought to get the support o f the A lianza de Federaciones for to gain votes , but the different social groups represented by the A lianza could not agree on a single candidate . Also, some Shipibo leaders like Julio Cusurichi and Juana Payaba, who rejected representing the indigenous people , decided to ru n for election with other political parties or coalitions. This divided the FENAMAD also reinforces social inequalities by reproducing gender differences already existing in its member commu nities. Women have become increasingly involved in the political affairs of their communities, their participation in FENAMAD is less than that of men. N ow it is be coming fairly common to see women expressing the ir opinions in communal assemblies or being 4 Many factors 4 Both men and women attend community assemblies and participate in them, but they differ in the way they participate. When men participate, they stand up, step forward and express their opinio n about a topic being
183 have contributed to the empowerment of indigenous women in the region, among which we can mention their increased access to schooling and their participation in state pro grams Popular Canteens, all of which require female participation. Nevertheless, women have still less access to leading positions. Like men from the more distant c ommunities, they generally have less schooling and are less fluent in Spanish than men, and their level of socialization with members of the regional society is limited. Further, housekeeping and childcare, two activities that belong almost exclusively to female participation in politics. Even if a woman is willing to engage in politics, her husband would not let her do that, which explains why those women who have succeeded in politics are mostly single . These gender inequalities at the community level are reproduced at the level of the federation. Within FENAMAD, w omen are usu ally chosen for minor positions and responsibilities within the federat pokesperson, f iscal or the evelopment area. Despite their increasing leadership skills, wome n have never been appointed as presidents of the federation , nor even as vice p residents . When a new leadership is to be elected, the lists of candidates a re made by men, who usually look for other male delegates. If female delegates are invited to be candidates, it is only for secondary positions. Even during protests and manifestations, female participants are relegated to subordinate roles such as cooking or bringing water to the protestors. discussed. N ot all men speak out during assemblies. Most of them just attend these meetings and remain quiet , although they may make discrete comments . Women stand close to their husbands and while they never step forward to speak out, whe never they express an opinio n they do it to protest, almost always by yelling.
184 Differences in status between men and women also affect the kind of networks they are able to build up. Unable to access leading positions within FENAMAD, women cannot develop the networks with key acto rs that only th ose positions offer. Spaces and opportunities that permit leaders to build t hese networks, such as meetings and workshops, are available only to those in This does not mean that female leaders are not able to build up their own networks, but these are not the kind of networks that bring economic and political advantage. Also, women have been less involved in projects implemented by FENAMAD in its base communi tie s, which means they had less access to the training provided by those projects such as learning h ow to use a computer or manage a budget. By excluding them from these projects , women were prevented from acquiring the qualifications needed for leadership positions, as well as from a ccessing networks facilitated by those projects. Networks of Power The three indigenous federations currently existing in Madre de Dios differ in the networks they build up and rely on to get access to resources. The conflict be tween FENAMAD and COHAR was solved when the Ha rakmbut resumed control of the federation and agreed to recognize COHAR both as part of the federation and as an autonomous organization (GarcÃa 2003). But COHAR, and later COHARYIMA, have actually operated mor e under the umbrella of FENAMAD, depending on its resources and networks to operate. Thus, when the federation found itself short of funds and was no longer able to support it, COHARYIMA stopped working. COINBAMAD, in contrast, remained as an autonomous fe deration and receive d no financial support from FENAMAD . Therefore, COINBAMAD had to build its own network with influential individuals and organizations in order to get financial and technical resources. During my fieldwork, were apply ing for funds and expanding their network of
185 support in order to car ry out development projects to benefit communities of the Lower M adre de Dios . Amazonian indigenous people have been historically excluded from positions within the mun icipal and regional governments, which has prevented them from influencing public policies and diverting state resources towards native communities. In this context, indigenous federations have sometimes resorted to clientelist networks with state officials in order to gain a ccess to resources. This is partly because governments regularly pour resour ces into their own clientelist networks for the purposes of securing votes (Molineux 2002). The building of these networks, however, may depend on the sympathies that indigenous leaders may have towards some individuals in a certain moment . For example, in 2012 COINBAMAD leaders were willing to cultivate networks with and receive support from the p resident of the Regi onal Government, while some leaders within FENAMAD overtly opposed him and rejected any relation with him . In the past decade, however, indigenous people have sought to participate more in the local government . They have seen p articipation in government a s a strategy to influence decision making and serve indigenous interests. This greater political participation , however, does not necessarily reflect a substantial democratization of the political system. Indigenous people are offered positions of little i mportance and , i n most cases , the ir inc l usion in electoral lists responds to the necessity of complying with indigenous quotas that are required by law (Espinos a 2012 ). A s we have seen , no indigenous leader ha s so far been successful in being elected as mayor or regional president in Madre de Dios . Despite their minor importance, these positions facilitate the acquisition of some state resources. For this reason, FENAMAD and COINBAMAD compete with each other for the same state resources by promoting thei r own representatives in the local government . As a result,
186 b oth FENAMAD and COINBAMAD have managed to place their candidates in positions within the regional government, the municipality , and the state mining agency. In the process , however, mutual distru st and resentment have emerged among both federations . FENAMAD leaders, for despite their ability to influence the state and NGOs in favor of indigenous interests. Th us, while putting indigenous people in the local government has helped indigenous organizations to gain access to state resources, competition for resources has ero ded solidarity ties between these two federations and the communities they represent. While the clientelist networks that some n ative communi ties maintain with Hunt Oil have enhance eroded communit FENAMAD, which has had negative impacts on the indigenous mobilization for land rights. For example, I was able to observe how networks between some communities and Hunt Oil weakened s campaign against oil drilling within the Amarakaeri Communal Reserve . As we have seen, t consent for its oil project, including giving gifts to communities, hiring indigenous people as employees, and even bribing indigenous leaders. Some individuals engaged in these networks in ord interests, resulting in tensions within communities and between communities and the federation, sometimes exacerbating existing kinship and ethnic based rivalries. As a re sult, the federation has not been able to keep communities together against the oil consortium. Some community members have fram ed their decision to support Hunt Oil within the NAMAD
187 implement a project to provide the community with electricity or gasoline for our power ommunity of Shintuya, who was interests to be a betrayal, but the Wachi paire justified it by arguing a lack of reciprocity on the part of the indigenous organization. In a conversation with his sister, who is a fierce opponent of oil development within the reserve FENAMAD? Does Patron client relations are not only exclusive to relations between indigenous organizations and state and corporate actors. To a certain extent, implementing land titling and development projects may h ave at times led to sem i clientelist relations with its member communities. Upon returning to Puerto Maldonado, after some weeks of fieldwork in Puer to Luz, I went to see the president of FENAMAD. When I was telling him how my findings could help improve th e relationship between the federation and the with the federa tion as one based on patron client relations , in which allegiance to the federation is exchanged for support for territorial defense and development projects. In a certain way, FENAMAD replaced the role played before by the missionaries in providing communities with Western goods and services such as education, health care , and work in lumbering and cattle ranching projects federation, and ultimately of solidarity between communities, depend upon how much the
188 Alt hough not based on clientelist sometimes t aken place in a vertical way . Over time, the federation has become fully dependent on project funds to cover its administrative and operational costs, and n ow it functions primarily as an NGO. In the last decade, donor agencies have funded the federation almost exclusively to implement projects aimed at promoting t he creation and management of reserves for the several indigenous groups in voluntary isolation existing in the region. As oil development and logging have increasingly threatened the rights and lives of these indigenous groups, their protection not only a ttracted the interest of donor agencies but also became a primary goal for the federation. Now many communities complain about the fact that issues such as territorial defense, health care or education no longer receive the attention the federation used to give them in previous years current emphasis on the protection of the isolated indig enous peoples. Thus, donor agencies have had an ambiguous impact: on the one hand, with their funding they have maintained FENAMAD alive as a network of communities but, on the other hand, they have generated tensions by prioritizing some topics over other s. Patron client relations with rubber traders and missionaries and now they also permeate their interact ion not only with the state and corporations , but also with FENAMAD , a perspective stemming from the way indigenous people first related to outsiders. The strategy that rubber traders and missionaries used to attract and secure native labor was the delivery of free gifts, which then turned into enslavement or debt peonage. This form of articulation to the wid er society established a patronage style of relating to
189 Conclusion In this chapter, I have presented a case study of the FederaciÃ³n Nativa del RÃo Madre de Dios y Afluentes and one of its constit uent communities, Puerto Luz, in order to show how social networks influenc e the Arakmbut mobilization for their land rights. I have attempted to shed light on the mechanisms through which social relations facilit ate recruitment of activists , the creation of organizations and networks, and access to financial and technical assistance. I have also tried to illustrate how organizations created for the def ense of land rights constitute contentious field s where prevalent ethnic and gender inequalities are repro duced . Sociologists focus on the role of social ti es in recruiting activists for social movement s , especially ties with friends and acquaintances. My research advances this literature by specifying the different ial effects of social ties on political mobilization. This chapter showed that recruiting participants in social movements can also be a function of kinship ties. Even though households are highly autonomous, and cooperati on between them is limited, external threat may activate solida rity bonds and bringing different residence groups together . In this process, kinship ties (bonding networks) among members of residence groups enhance communication and coordination for mobilization . Kinship ties are also important to forge solidarity and encourage political alliance with the neighboring community of San JosÃ© del Karene. L eaders mobilize their kinfolk by resorting to kinship solidarity. I ncreasing encroachment of miners on Arakmbut lands in the late 1970s and early 1980s rendered forging alliances among several communit ies and ethnic groups more necessary . It became clear that to defend indigenous lands kinship networks and the number of individuals they could mobilize were not enough. W hile kinship ties are central to mobilize people at t he community level, at a much broader scale they have the effect of dividing rather than uniting commun ities. In this context, b ridging ties have been more crucial to bringing kin groups and
190 communities into larger multiethnic political organizations. Frie ndship networks with anthropologists were crucial to secure the resources ( know ledge and networks with state officials , donor agencies , and other federations) necessary to create organizations and navigate the bureaucracy of the land titling process. In this sense, social networks not only enhance individual participation in mobili zation. They play a key role in facilitating access to resources for the formation of organizations and the implementation of development projects . N etworks with anthropologi sts offered the Arakmbut more horizontal relations and enabled them to circunvent disadvantageous patronage relations with poltical parties and the state. Therefore, d ifferent social networks play different ro les at different scales. On the other hand, soc ial networks are not only social resources that help individuals and groups to make rational decisions and reduce transaction costs, allowing society as a whole to achieve economic development and democratic governance, as economistic approaches to social capital argue. From a Bourdieusian perspective, my study enriches social capital theory by presenting a case of the down side of social networks . Indigenous federations like FENAMAD empower indigenous people by enhancing their capacity to mobilize constitu ent communities, get land titles and financial resources, implement development projects, and defend indigenous off groups to maintain their privileges by controllin g the organization and benefi t ting more of its resources . The position of individuals and groups within the social structure influence the ability to construct and accumulate more social capital. While ethnic federations became powerful tools to advance indigenous interests, they have also worked as political arenas for inter nal and external domination. Once FENAMAD was created and started to operate as an inter community coalition, tensions based on ethnic
1 91 differences and rivalries arose. Leaders from different ethnic groups struggled over the resources that became available through n ew networ ks built with national and international NGOs , leading to attempts by the subordinate groups to create their own organizations. Furthermore, ethnic groups more integrated in the national political and economic system dominated communities from the more isolated and less acculturated ethnic groups. As a result, members of the latter groups have had less access to the f of its projects . Similarl y, despite their increasing political experience , women are still less successful in accessing key positions within the federati ons . Consequently, ethnic and gender inequalities that existed in native communities before the creation of FENAMAD are reproduc ed within this organization. Finally, networks not only create solidarity and bring social benefits. They may also create tensions and have antisocial effects. N etworks between federations and st ate and corporation actors tend to be built upon patr on client relations. State officials and corporation staff use the distribution o f goods and services to gain political allegiance or compliance with their development projects . To a certain extent, even FENAMAD uses development project loyalty . While these networks may bring economic benefits to federations and communities, dilemmas have arisen to pit individuals against the group, and communities against FENAMAD, threatening to weaken indige nous mobilization . In t he same way, w hile fr iendship and godparenthood netw o r ks wtith invitados may enhance the ability to gain access to resources at an individual level, they may also erode community solida rity and create internal divis ion, mobilize collectivelly .
192 CHAPTER 6 THE CULTURE OF INDIGENOUS MOBILIZATION Indigenous mobilization is not simply the result of material forces or the selfish calculus of economic interests . It also repres ents modes of political culture . The way the Arakmbut have built networks with a nthropologists and NGOs to form organization s is part of an indigenous tactic of making political alliances against common enemies, deeply rooted in their culture. Today, in order to protest the social and political change that ha s been imposed on the Arakmbut for decades , they combine c ulturally specific practices such a s raids and body paints with more activist and union oriented politics such as marches, speech acts, and the framing of demands within global discourses. Neverthel ess, political culture has not been widely explored to explain how social movements come into being and develop. In this sense, a thorough comprehension of issues of culture may contribute to our understanding of not only why rural people resist but also w hy they resist the way they do. The culture of resistance has been widely studied using the theory of everyday forms of peasant resistance. Political scientist James Scott, perhaps the most salient representative of this approach, asserts that peasants become involved in everyday forms of resistance (small covert acts of defiance against local elites) because of their unwillingness to resort to open political protest. He further contends that even in the absence of organized protest movements, peasants are able to refuse to accept the legitimacy of the existing structure of domination (Scott 1986: 6; 1990: 79 82). Scott (1976) emphas izes the importance of cultural elements in peasant resistance (such as norms of reciprocity, the right to subsistence and millenarian beliefs). Kerkvliet (1990), however, is more inclined to see everyday resistance as a product of modern cul tural influences, associated with notions of citizenship and nation state. While acknowledging
193 that peasants may draw upon cultural values and norms rooted in the past, Kerkvliet argues that peasant struggles may be informed by the notion of basic rights ( rights to human dignity and decent standard s economy. While we have started to understand hidden forms of resistance, their relations with open forms of struggle remain almost completely unclear. Existing studies seem to point in di fferent directions. Adas (1986: 82) , for example, suggests that protests ( such as foot dragging, feigned incompetence, fleeing to remote areas) may work to dissipat e social discontent and crea te division among the peasants , leaving them more vulnerable to repression. In his opinion, may be more effective in engender ing open political struggles. H e also argues that even in the presence of open resistance, the lack of formal organization and sophisticated ideologies may represent serious obstacles . In this context, political protest depends on external p olitical leadership . Some studies of peasant politics in Latin America have come to the same conclusion and have suggested that urban based political leadership is a condition for the transformation of peasant unrest into a political movement (Hobsbawn 1973). Nevertheless, more recent literature on g uerrilla movements in Latin America shows that political alliances between urban left wing political le aders and their would be indigenous supporters may end up in sometimes tragic misunderstandings (Brown and FernÃ¡ ndez 1991; Rojas Zolezzi 2008). At the same time, other studies show that local cultures of resistance and community based leadership, which were intrinsically linked to peasant political mobilization, may be at the base of successful political
194 struggle (S mith 1989). Indee d, Kerkvliet (1993: 481) argues that small acts of defiance led by local organizers may in fact prepare the ground for more open political struggles . Scholars have also pointed to broader political contexts as the crucial factor behind the transformation of hidden into open res istance. Kerkvliet (1993: 471) describes how, in the Philippines, it was not until a process of national political democratization started, which dissipated the risk of repression, that peasant leaders found interlocuto rs in the local and national government. In Latin America, neoliberal democratization prompted the politization of indigenous people because it threatened community autonomy and not because it brought political freedom ( Yashar 1999 , 2005 ). W hat are the rel ationships between broader socio political processes, changes in social organization, and the tactics of Madre de Dios indigenous peoples in defense of their rights? In this chapter, I examine Arakmbut ways of doing politics and their historical and cultur al roots . Specifically, I discuss old indigenous forms of resistance , and how these have articulated to new principles of political participation that took form in historically particular modes of mobilizing against predatory state policies. I argue that, in consistence with what theorists of the everyday forms of peasant resistance suggest, the case of the Arakmbut political mobilization shows the presence or absence of a repressive state as the determining facto r in the evolution of indigenous forms of re sistance. I n the following sections , I first discuss how while the Arakmbut remained in the frontiers of the state, they openly resisted attempts to subjugate them. Then I examine how , with their incorporation to the nation state , the Arakmbut resorted to more hidden forms of resi stance . From Autonomy to Open Resistance For much of its history, the region that is currently known as Madre de Dios remained politically and economically autonomous from wider processes. The different s tates that were
195 formed in what is now Peru Inca, colonial and republican state s attempted successively to Republican period, that this region was significantly inco rporated into the national state. Unlike the Andean indigenous peoples, who first were subject to the centralized Inca state, and then became integrated to the colonial regime, Amazonian native peoples maintained their independence from Incas and Spaniards , and hence were not articulated to a central system of domination (Remy 1994) . Thus, since the presence of state institutions has been marginal until very recently, Madre de Dios remained as a frontier region. The weakness of state repression, and in some cases the lack of it, enabled the native population to develop open forms of resistance. Raiding and Fleeing One of the most common traditi onal forms of resistance among n ative Amazonians are raids. When outsiders attempted to invade their territories and subjugate them, the Madre de Dios indigenous peoples responded by defending themselves or attacking the invaders. Several accounts describe how indigenous groups violently repelled Inca armies, Spanish conquistadors, rubber gatherers, missionaries, and go ld prospectors. In 1400 A.D., Tupac Inca Yupanqui and 10,000 soldiers reached the headwaters of the Madre de Dios River, in Wachipaire territory, where they stayed for two years in preparation for an expedition to the Moxos region, in present day Bolivia. There, the bellicose Chuncho mercilessly exterminated t he Inca troops (de la V ega 2006 ). Unable to subjugate the natives, the Inca resorted to trade and marriage alliances to assure access to the lowland natural resources they demanded. After the Spanish c onquest of the Inca, several expeditions to the Madre de Dios region in search of the legendary treasures of Paititi were attacked by hostile Indians. Indian attacks, along with harsh climate, natural barriers and unknown diseases, contributed to the failu re of
196 these endeavors (Califano 1982). The attacks on both Inca and Spanish expeditions contributed to forge the fame of the Mashco as tireless warriors who defended the impenetrability of their territory with great violence. For this reason, the Spanish d id not use native labor in plantations and mines, but brought laborers from the nearby highlands. Unable to defeat the natives and colonize the lowlands, the Spaniards only controlled th e upper parts of the basin , where they established coca plantations an d extracted gold. While these posts were raided by hostile Indians, the major resistance strateg y used by the natives was fleeing to the periphery of these areas ty, thus forming an elastic frontier with the highland laborers (Moore 2006: 19). With the decline in the production of coca and gold in the late sixteenth century, the Madre de Dios region entered a long period of relative calm that lasted for the next t wo centuries. During this time, the presence of outsiders in the region was minimal and hence aggression to the natives decreased. This changed in the nineteenth century when the recently independent Peru promoted the exploration and incorporation of the M adre de Dios Basin. This new policy responded to the increasing interest in exploiting natural resources (especially c inchona bark and rubber) and the necessity to protect the national borders against the increasing threat of neighboring countries. This ne w period marked an increase in confrontations between outsiders and native populations, and in the violence of these encounters. In 1896, rubber baron Carlos FermÃn Fitzcarrald and his flotilla of native guides, who had penetrated into the upper Manu River , were attacked by fiercely resistant Mashcos, losing 50 men . In retaliation, Fitzcarrald ordered his men to massacre an entire village (Reyna 1942). In some cases, several ethnic groups would get together to face the aggression of outsiders, as occurred i n 1873 when
197 ce the abusive expedition of Cuz co prefe ct Colonel Baltazar La Torre (GÃ¶ hring 1877). In this period, i ndigenous resistance involved not only attacks and eviction, but also avoiding confrontation. One of these forms of resistance has been isolation from external aggression. After their violent encounter with rubber gatherers at the turn of the twentieth century, the Arakmbut retreated into the headwaters of several tributaries of the Madre de Dios River and were not contacted again until the 1950s. The groups that are currently known as in voluntary isolation fled the rubber traders who between 188 0 and 1920 enslaved many indigenous groups in Madre de Dios (Huertas 2002). These groups took refuge in remote areas of the forest. Despite some exceptions, the first contacts between Dominican missionaries and Harakmbut groups were characterized by violent confrontations, resulting in the killing of priests and native people (Alvarez 1940 ) . Nevertheless, this period did not last long and the natives accepted to be relocated in mission posts, mostly attracted by Western trading goods and medicine. At that time, hundreds of Harakmbut were dying due to the new diseases brought by the outsiders. Fleeing continued to be a resistance tact ic even wh en the Harakmbut were concentrated in mission posts. When accusati ons of sorcery and murder aggravated the tensions between several ethnic and maloca groups in the mission of Shintuya, they fled the mission and returned to their homelands, where they found ed new settlements. This strategy was used also when the Arakmbut faced new threats in their new settlements. One of the reasons why the Arakmbut of Puerto Alegre moved to Puerto Luz was the installation in the former of a supply depot by the Japanese oil company Peru Cities Service, which turned this remote village into a
198 center of air traffic for oil exploration in the area. This caused frequent conflicts between the akmbut to move out (Moore 1981: 140 141). 1 The invasion of indigenous lands by mining corporations and settlers in the 1970s and 1980s brought about a return to open and violent forms of resistance. To repel the invaders, the natives resorted to physical expulsion of outsiders , which is a practi ce widely used by indigenous people to defend their lands. In mid 1978, two large mining companies occupied the lands of the community of Boca Inambari. One of them brought 300 workers to wash gold within the communit y lands, while the other operate d two t ractors to extract gold. Violent confrontations between the Indians and miners ensued (Moore 1985 b : 173). This conflict led in 1978 to the first large indigenous mobilizat ion towards Puerto Maldonado. Almost all members of Boca Inambari, supported by severa l indigenous people from other communities, rallied to Puerto Maldonado to demand the government their land rights be respected (Moore 1985 b : 173). In 1981, after a mining company left the community due to economic failure, Puerto Luz managed to throw all mining settler s out of its lands (Moore 1985 b : 174). Attacks also included raids into mining camps. Extreme cases occurred when the Arakmbut, driven by anger, pillaged the mining camps or burnt the machinery down . The Arakmbut resorted to these forms of res istance especially when the authorities did not respond to their complaints. The miners often responded by filing criminal suits against the Arakmbut, for which the former always had the supp ort of the police (GarcÃa 2003: 282). These violent forms of resis tance were coupled by o ther non violent forms of land defense , which included depriving the miners of their equipment and the opening of gardens in 1 Village fissioning is also a mechanism used by many indigenous groups to solve internal conflicts. When a conflict between households or kin groups arises, one of the groups will avoid confrontation by leaving the village and establishing elsewhere, sometimes in other part of the com munity. Many new settlements have been formed this way. The settlement pattern of Puerto Luz is the result of the dispersal of households due to conflicts that arose from the concentration of people.
199 strategic places. In order to force mining settlers to leave their communities, the Arakmbut took their equip ment away and in some cases hid it in the forest or sank it in t he river. On the othe r hand, the Arakmbut have opened asion. By opening gardens the A rakmbut attempted to establish their possession over land a nd discourage the miners from encroaching on it (Moore 1985 b : 174). In some cases, the Arakmbut have received the support of outsiders. In Puerto Luz, for example, the Arakmbut were supported by father Adolfo Torralba, a Dominican priest who resided in this community from the mid 1970s to 1986. Torralba chose to carry out his proselytization work by residing in Puerto Luz because the strategy of bringing together many groups to Shintuya and other missions failed. In Puerto Luz, Torralba not only promoted the creation of a school in the community and delivered to the Indians some trading goods such as shotguns, he also advised the Arakmbut about strategies to defend their territory (R umme nhoeller et al. 1991: 224). The Mission and the S chool Two of the mo st imp ortant sources of social change among the Arakmbut have been the mission and the school. Missionaries and schoolteachers privileged patronage ways to relate to the natives, which prevented the natives from developing and strengthening ties among them. This caused cultural and institutional changes that did not encourage solidarity and cooperation among the natives, but competition and individualism. This in turn accentuated tensions within maloca groups and among them, negatively affecting the possibility o f mobilization beyond the settlement groups. In pre mission times, all residents of a settlement group used to live in one communal house or maloca . Malocas were not isolated from each other as fiestas that brought several malocas together would frequently take place, encouraging contact and the exchange of
200 information, songs and s tories between them (Gray 1996: 227). This settlement pattern changed after the Harakmbu t were concentrated in missions . In the Shintuya mission , about th ree hundred and fifty Harakmbut were brought together, which was a major change in their lifestyle since the maloca groups used to comprise between fifty a nd a hundred people. Because this caused tensions , most of the maloca groups split up and made their own settlements within the mission, with those who lived nearer the houses of the missionaries being more integrated into the spher e of the mission (Fuentes 1982: 167). Furthermore, the missionaries encouraged the norm of living in nuclear family huts. Aft er fleeing the mission and returning to their homelands, the Arakmbut no longer lived in malocas but in individual huts . When the Arakmbut founded P uerto Alegre , they preserved to some extent the effect of living in a maloca by arranging their new houses a round a soccer pitch that took the plac e of the open space in the cent e r of the maloca . But further tensions among families led them to spread out and form smaller residence groups. They no longer performed their rites of initiation either, which were aban doned while in Shintuya as they were ridiculed by the priests and the more acculturated indigenous groups. Arakmbut ceremonies were replaced by Christian rituals that took place ever y Sunday in the church. These rituals made sense primarily in the context of living in a maloca , and once that was gone all aspects of the culture associate d to it disappeared (Gray 1996: 256). practice they promoted competition among i ndividua ls instead of mutual aid . This was apparent through the patronage way in which missionaries related to the natives, a nd the competition that this generated over access to resources and even women that the missionaries controlled . During the period of conta ct, the Dominicans attracted the Harakmbut to the missions with gifts of
201 highly desirable goods, but once they were brought to the missions those goods became commodities that were available only to those who behaved according to Christian precepts (Wahl 1 987). The priests not only reward ed appropriate behavior with the promise of an afterlife; those who conformed to the structures of the mission were considered worthy of benefitting from the advantages held by the power of the priests, such as Western good s or support in the lumber mill, cattle ranch, or coffee cultivation. To ensure this, the missionaries kept the interaction between the natives and dealers, traders and other representatives of the national capitalist system to a minimum. This not only pre vented the natives from being exploited by outsiders but assured the missionaries the advantageous role of intermediaries between the natives and the outside world of Peruvian society (Gray 1996). The natives had to compete for access to resources such as money, goods, jobs, education, and grace from God. They also had to compete to gain access to women, as Dominican priests controlled marriage and decided whom men could marry. The power held by the missionaries ultimately bred resentment among those who lo st out (Wahl 1987: 268). Conflict thus constantly arose between the different Arakmbut maloca groups and with the Wachipaire in competition for access to resources and women. While initially the Arakmbut got on well with the Wachipaire , and even joined in t heir masato fiestas, disputes between the two groups eventually arose due to competition for resources. 2 The problems over women arose from the shortage of eligible female marriage partners, and the priests, unaware of Harakmbut marriage practices, arranged inappropriate marriages. These tensions expressed themselves through accusations of sorcery rather tha n aggression against the priests, and at least two women were murdered by the Arakmbut. Nevertheless, the accusations were directed to those who were 2 Masato is a drink made of fermented manioc common to many Amazonian indigenous groups. The Harakmbut did not know masato, but the Huachipari adopted it from the Matsigenka, who were their neighbors.
202 closer to the missionaries (Fuentes 1982:186; Gray 1996: 252). In this context, it was easy for the priests to exploit these conflicts in order to obtain compliance with their plans for the mission. With the Protestant SIL missionary linguists the situation was quite similar. In their dealings with the Arakmbut of Puerto Alegre, SIL members also emph asized indi vidualism. Wage labo r and trade were always dyadic relationships between one native and one SIL person. Missionaries requested the Arakmbut for labor on an individual basis and without involving corporate groups such as the residence groups or clans. The A rakmbut thus learned to respond to this, and competition developed among members of the same kin groups, in a manner previous ly unknown to them (Moore 1979: 117 18). Furthermore, like the Dominicans, SIL missionaries had access to a seemingly endless supply of exciting new goods and services, and to some extent they were able to satisfy the new needs created among the natives. Thus, the SIL missionaries took on a patronage role as they had the power to facilitate or impede access to desired trade goods accor ding to the degree of cooperation the natives gave them in their projects. Despite his efforts to avoid favoritism, Robert Tripp, the SIL missionary who spent most time in Puerto Alegre, was not always able to prevent some manipulation and paternalism from occurring (Moore 1979). Both Dominican and SIL missionaries ran schools in Arakmbut communities, with very similar impacts on the ability of natives to cooperate with each other. Both organizations started to compete for the right to provide schooling to indigenous children right after the Arakmbut were contacted. The Dominicans not only installed boarding schools in their missions, but also founded in 1953 the AsociaciÃ³n de Misioneros Seculares (MISEMA), a mobile team of educators, while SIL established i ts government approved bilingual schools throughout much of the central and northern Peruvian Amazon. When the SIL missionaries started to expand their
203 school network, the Dominicans turned MISEMA into Red Escolar de la Selva del Sur Oriente Peruano (RESSO P), which built schools in the newly formed Arakmbut communities that would be ran by lay mi ssionary teachers (Aikman 1999: 42 3). Puerto Alegre received the attention of SIL, which opened a bilingual school in 1973 , ran by a native teacher from the communi ty who had previously attended primary school in the mission of Shintuya and spoke some Spanish (Moore 1981). The SIL school was closed when the agreement between SIL and the government came to its end and was not renewed. Then, after moving to Puerto Luz the Arakmbut petitioned for and were sent a RESSOP lay m issionary teacher (Aikman 1999: 44). The Puerto Luz primary school has since then been run by RESSOP, but teachers, who are mostly of Andean origin, are now provided by the Ministry of Education. Mesti zo school teachers identify themselves with the national culture and not with the Arakmbut . They consider themselves to be ethnically and culturally superior to the Arakmbut, who are thought to be in need of development . According to the teachers , moderniz ation means the adoption of the national language, values and customs , a way of life of which they consider themselves to be a role model (Aikman 1999: 57). In this way, t he content of class lessons oft en ignore Arakmbut history and knowledge, promoting ins tead Peruvian history and a sense of defending the country. Since the children are taught to identify with the national culture rather than with their own culture, the school system deprives t he Arakmbut of the cultural elements with which they may build and rebuild an ethnic identity that could generate solidarity ties among families. To carry out projects to modernize the community, teachers call meetings to plan the project s and their implem lack of resources, not only teaching materials and adequate infrastructure, but also labor. This is
204 e school or the community, and parents are required to provide free labor for these projects. Most of the times, either the turn out is low or the parents simply reject the projects for finding them in conflict with their own culture. This makes access to labor one of the main bottlenecks for the compare the Arakmbut schoolchildren and parents with those in mestizo settlements, viewing the differences in terms of degrees of willingness to cooperate and work towards common goals. While the Arakmbut are seen as being lazy and backward, the mestizos are considered to be cooperative and hardworking . The differences between Arakmbut and mestizo social organization a nd cosmology are not given any consideration (Aikman 1999: 62), nor even the In such conditions, teachers have develop ed strategies for coping with labor shortage. In good relations with some members of the community. One approach is to establish formal relations with select ed individuals through compadrazgo ties. In Puerto Luz, teachers use this institution to form alliances with certain individuals and families. Over the years these teachers have nurtured a ritual kin network that could be mobilized for support for school i nitiatives. support to their projects, a relation that is reinforced through other forms of gift exchange. As with the priests in the times of the mission, th lientelist ties to recruit labor and support brings about resentment among those who are left outside those networks, resentment that in the end undermines the ability of the Arakmbut to build solidarity bonds.
205 From Open Resistance to O rganized Political Mobilization Until the 1950s, development of the Amazon region was of secondary interest for the Peruvian elites. This changed dramatically in the 1960s , when several public policies aimed at incorporating and developing this portion of the country led to massive investments in roads and communications infrastructure, colonization schemes, extension programs in agriculture and cattle raising, and projects of petroleum and other resource extraction (Smith 1982, 2005; AramburÃº 1984; Chirif 1985). One of the main consequences of these changes was the encroachment of indigenous lands at unprecedented levels. Thousands of Andean colonists, cattle ranchers, and mining and oil corporations invaded the lands of indigenous settlements. Under these new circumstances, the traditional ways in which native people had been protecting their lands, based on physical expulsion of invaders and fleeing, became insufficient. Supported by outsider allies, especially anthropologists and missionaries, in the late 1960s indigenous Amazonians formed organizations that brought together several indigenous settlements of the same ethnic groups. The roots of these new organizations were clearly indigenous, as they worked as councils of leaders from each indigenous settl ement. In the early 1970s, as part of state policies that sought to promote popular participation in politics, a government led by progressive military promoted the formation of native communities and federations among indigenous Amazonians. These state pr omoted organizations impinged on already existing indigenous councils, making them legitimate interlocutors of the state. A s a result, indigenous resistance became less open and violent, and more organized and hidden. Raids and attacks were replaced by law suits, demonstrations, and appropriation of global discourses. The Ethnic Federation Organizing has become the main expression of indigenous political mobilization in the Amazon . ethnic
206 federations, a loose association of autonomous settlements of a particular ethnic group, represented by their delegates and an elected leadership (Smith 1985, 1996). The idea of creating ethnic federations in the Amazon stretches back to mid centur y and was encouraged i n most cases by urban based non indigenous individuals and organizations such as anthropologists, missionaries, United States Peace Corps workers, state officials, NGOs, and political parties. The formation of the FederaciÃ³n Shuar in E cuador in 1964 is recognized as the first major landmark in the trend toward organized indigenous struggle. Salesian missionaries planned and carried out the establishment of the FederaciÃ³n Shuar to counteract the growing advance of colonization into Shuar territory, which threatened to deprive missionaries of converts and labor (Salazar 1981). The next significant milestone in the r egion came in 1969 when YÃ¡ nesha communities in the central Peruvian Amazon formed the Congreso Amuesha. Richard Smith, an Amer ican anthropologist who arrived in the country in 1966 as a volunteer with the United States Peace Corps, was key in the formation of the Congreso Amuesha. Since the YÃ¡nesha had lost most of their lands due to massive migration of Andean colonists into the ir territory in the late 1950s, Smith helped r epresentatives of twenty YÃ¡nesha communities to form the Amuesha Congress in 1969 (Smith 1996). In other cases, the presence of S IL educational and civic programs among Am azonians, operative since the mid 1950s, inspired ethnic organizational activity. This increase in militant mobilization among indigenous Amazonians was part of a larger process that included highland peasant farmers. This upsurge in indigenous mobilization brought about grater demands for agrarian reforms and increased migration of rur al landless to the urban centers. The response of the elites was to redirect highland migration to the Amazon forests as an alternative to land tenure reforms . To expedite the colonization of the Amazon, the state invested
207 in the building of penetration ro ads into the forest, thus placing this re mote area under political control and integrating it into the national society and market economy. In many cases, missionaries were commissioned by the state to carry out this endeavor (Smith 1985:16). Nearly all th e first ethnic federations began as meetings of headmen or representatives of different settlements of a particular ethnic group who were looking for common strategies to defend their land base and ethnic identities (Chirif 1974; Salazar 1977). While these initial meetings evolved in different ways, depending on the level of acculturation, participation in the market economy, and ethnic particularities, they have shared several features. They have basically worked as alliances of landed corporate groups wit h a tradition of cooperation in political, economic and social issues (Smith 1985:17). By the 1970s , emer gent native leaders from all major ethnic groups in the Peruvian Amazon formed their own local ethnic federations, whose organizational st ructure share d common elements with the Congreso Amuesha: (1) a voluntary alliance of autonomous settlements; (2) an elected leadership representative of and accountable to the constituencies; (3) a combination of political representation and lobbying for providing goo ds and services; (4) a unity based on ethnic identity; and (5) autonomy regard ing outside actors (Smith 1996: 94 5). Acoording to Smith (1985:17), t he reason why Amazonian native people opted for the model of ethnic federation has to do with the fact that t he areas where these organizations proliferated were peripheral to or outside of Inca and Spanish states, and for that reason indigenous peoples were able to preserve a strong sense of their e thnic identities . Nevertheless , indigenous organizing not only r esponded to the interests of indi genous peoples and non indigenous supporters; it was also a result of state intervention. This to ok place in the 1970s, under Velasco milita ry administration . Velasco was part of a reformist faction of the military who saw i ts task as breaking the stranglehold of the traditional oligarchy and
208 instituting a redistributive form of economic development. As political parties and civilian leaders were seen as incapable of implementing these urgently needed reforms, Velasco install ed style system of interest representation that was intended to move beyond clientelism while controlling social mobilization (Stokes 1995:33). Thus, in order to replace the traditional role political parties played as intermediaries between the state and grassroots organizations , Velasco government encouraged popular participation in state institutions and decision making processes (Fran co 1979). To accomplish its aim of creating a new structure of local poli tical activism and attracting popular loyalty, Velasco government created in 1971 the Sistema Nacional para la MovilizaciÃ³n Social (SINAMOS) , a centralized mobilization agency . SINAMOS hired Peruvian anthropologist Stefano Varese to found a Division of Na tive Communities of the Jungle, a special section intended to deal with the specific problems of Amazonian indigenous communities. Believing that the solution for most of indigenous people s problems would come with their organization, SINAMOS personnel de fined as their main task to help indigenous people organize themselves (Smith 1974 :31 ). Thus, SINAMOS fostered the creation of two forms or modalities of political organization: (1) the class based peasa nt like grated into a national level umbrella organization called ConfederaciÃ³n Nacional Agraria (CNA), also created with the help of SINAMOS (Guerra GarcÃa 1983); and (2) ethnic At that time, the issue o f whether indigenous organizing should be based on class or ethnic terms was intensively debated (Espinosa 2004). The administration of General Francisco Morales BermÃºdez closed down SINAMOS in 1978, and prioritized colonization and road building rather than titling of indigenous lands in the
209 Amazon . The same year, a group of anthropologists who formerly worked for SINAMOS founded the Centro de InvestigaciÃ³n y PromociÃ³n de la AmazonÃa (CIPA), an NGO that would play an important role in the formation of in digenous organizations in many parts of the Peruvian Amazon, including Madre de Dios. The return to a democratically elected government i n 1980 , after twelve years of dictatorship, did not interrupt the neoliberal economic policies initiated by Morales Ber mÃºdez. Nonetheless, t he political rhetoric of democracy that accompanied these policies, which , gave a new impulse to the organiza tion of native Amazonians . NGOs such as CIPA shared this rhetoric an d promoted the participation of social organizations in th e public space, by helping in the institutionalization of indigenous organizations. In Madre de Dios, the formation of an ethnic federation was the result of the intervention of at least three exte r nal actors: the state, anthropologists , and NGOs . SINAMOS came to Madre de Dios in 1975, but it created neither agrarian leagues nor ethnic federations, since it focused its work on the registration and titling of several indigenous settlements in accorda nce with the 1974 Law of Native Communities (Moore 1985 b : 175; Rummenhoeller et al. 1991). In 1979, CIPA signed an agreement with the Ministry of Agriculture to demarcate the territory of four indigenous settlements in Madre de Dios, most of them Harakmbut. By forming native communities, SINAMOS and CIPA promoted the adoption of the bureaucratic and hierarchical model of organization ( governing boards and communal assemblies ) that would later influence the model adopted by FENAMAD. In this context, anthropologist Thomas Moore played an important part in indigenous organizing in Madre de Dios. During the 1970s and 1980s, it was considered that so called ance tactics were insufficient to protect indigenous lands, and that a mo re
210 organized political action was requi red (Rummenhoeller et al. 1991: 390). Then a doctoral student at the New School of Social Science in New York, Moore ar rived in Madre de Dios in 1971 to conduct d issertation research among the Arakmbut of Puerto Alegre . At that time, Peru Cities Service had installed a supply depot in Puerto Alegre, leading to conflicts between the oil workers and the Arakmbut. The n the Arakmbut asked Moore for support against the abuses of the oil company. In coordination with Stefano Varese, Moore took an Arakmbut delegate to Lima to get the support of SINAMOS against the oil company . This experience gave the Arakmbut a clear idea of the effectiven ess of networks with outsiders, and the need to form an organization began to be discusse d. By the lat e 19 70s, several ethnic federations, among them the Consejo Aguaruna Hu ambisa, the Congreso Amuesha , and AIDESEP had already been formed , and served as a model for the creation of other regional and local indigenous organizations all around th e Pe ruvian Amazon. Thus, between 1980 and 1981, with the financial support of Oxfam UK, some Arakmbut delegates were taken to Lima to hold some meetings with s (GarcÃa 2003:279; Gray 1986: 82). Also, in 1981 AIDESEP sent one of its leaders, a Yanesha Indian , to support the organization of a federatio n in Madre de Dios (Moore 1985 b : 175). Moore, AIDESEP and Oxfam UK promoted the model that represented only the interests of the Arakmbut, a model that contrasted with the union like model of other indigenous federations created at the time by the CNA . The Arakmbut, however, opted for an inter ethnic alliance because they considered they had their own mechanisms to affirm their identity internally, while the federation would be devoted to deal with external issues. The result of the meetin gs with AIDESEP and other federations, as well as various other local inter community meetings , was the first FENAMAD congress hel d at Boca del Karene in January 1982 (Gr ay 1986: 82).
211 While initially ethnic federations reflected indigenous forms of social organization, eventually they had to adjust to formal models imposed by the state. James Scott (1998) argues that statecraft is a project of legibility. W hether through the work of forest scientists or urban planners, state making agents simplify and standardize societies that for them look chaotic and complex. The Peruvian state demanded Amazonian indigenous people to fulfill formal their organizations . Thus, ethnic federations adopted juridical formulae from national legislati on (e.g., civil associations) that made indigenous peoples legally visible and v alid interlocutors for the state. For example, in 1986 FENAMAD was legally At the same time, the demand for political participation prompted indigenous people to use these juridical resources to i nstitutionalize themselves , fulfill ment of t he legal formalities. The emergence of organized indigenous politics in Madre de Dios was a direct result of indigenous political system. In the early 1970s, a military dictatorship not only allowed indigenous peoples to organize themselves, but also was an active advocate of indigenous organizing. With the return to a democratic regime in 1980s, the new government became uninterested in indigenous organizing, but there was enough political freedom f or anthropologists like Tho mas Moore and NGOs like Oxfam UK to support the Arakmbut to find organizational models in accordance with their own culture and needs. This proves that we should take the role of political regime with caution. In this case, a mil itary regime did not necessarily mean violent repression and, as will be seen later in this chapter, a democrat ic regime not always is exempt from repressive forms of political control. Old Societies, New Organizations no few difficulties for its adoption by the
212 objectives. Although there are several historical examples of the different Harakmbut groups making alliances to fight non indigenous aggressors, such as Baltazar La Torre in the 1870s, and the inter ethnic mobilization towards Puerto Maldonado in support of Boca Inambari in 1978, the idea of an organization that brings together members from different territorial and ethnic g roups who historically have not been friends was not well received by the elders. The social structure of many indigenous peoples of lowland South America is characterized by independence, individualism, egalitarism, and even hostil ity among groups (Salaz ar 1981: 589). As we have seen in Chapter 4, within Arakmbut communities sociality is characterized by an stress on personal autonomy, an aversion to hierarchies, and a rejection of coercion (Overing 1988 ) . Prior to the formation of FENAMAD, the Arakmbut ha d nearly no experience with a level of organization beyond the community level. In the pre contact period, the different malocas within a territorial group would band together for celebrations that could last several days, but beyond the maloca the only fo rm of organization seems to have been the war leader. This person, however, led his group only as long as the conflict lasted. This lack of larger organizational structures, as well as the extremely antagonistic nature of relations between different groups , would prevent any kind of spontaneous collaboration and the creation of indigenous organizations. This explains why there have been few examples of unified organizations beyond the level of the community or territorial group. In fact, in the literature t here are no examples of independently and spontaneously created indigenous organizations that have arisen from the level of communities. When Amazonian indigenous people finally adopted the model of the ethnic federation, they understood it in their own c ultural terms. Indigenous peoples conceived of missionaries and anthropologists as leaders whose abilities they wanted to harness. They not only accepted these
213 agents of change but also encouraged them to assume a leadership role, at least during the first years of the organization. The Arakmbut use the word wairi to refer not only to a respected Arakmbut men but also to non (Gray 19 86:117). Indeed , Gray (1986: 108) mentions that some Arakmbut beli eved that the roles of political leaders need not be played by indigenous peoples but by non indigenous support groups and advisors to traditional leaders. Still, non indigenous leaders are expected to behave according to indigenous models of leadership. U nlike Western conception s of leadership as hierarchical in nature and predicated on relations of subordination and domination, indigenous people conceive of leaders as providing ideas and taking initiative on behalf of their groups. It is important to no te, however, that it is the indigenous people who decide to take advantage of the skills and financial resources of non indigenous actors and not the other way around. The leadership role that these actors play is limited in scope and they could in no way coerce or impose their will on anybody, and their position is not permanent but dependent on their continuing success which, in turn, is indicative of their limited power. They are followed as long as it is worthwhile, but if they fail or are no longer per ceived as sufficiently powerful, they could be abandoned. This happened to Thomas Moore in the FENAMAD congress held in Santa Rosa de Huacaria in 1995. After the final session, the newly elected president of the federation, Antonio Iviche, approached Moore and told him that they were thankful for all Centro Eori had done for FENAMAD but they would not need it anymore. This also explains why th e role of external advisors has mostly been played by anthropologists and other human rights activists, d worldview. Indigenous Amazoni ans have been more reluctant to establish alliances with urban based, left -
214 wing p olitical parties, which have seen indigenous peoples with a paternalistic and sometimes racist attitude. Many indigenous peoples do not share the assumption that political leadership should be necessarily exercised by people of their same ethnic backgroun d. This is probably derived, in part, from a concept of the other that rather than emphasize the inferiority of the other, tolerate it and st rongly desire it (Overing 1996: 53). Therefore, wh en Gray (1986:117) asserts that outside agents only became wairi a enough to take on the power which thre missing the point. It is possible that the main interest of the Arakmbut was in the efficacy of that person as a leader rather than in ethnicity. Indeed, it is clear that ethnic federations became more popular with indigenous people once their efficac y was demonstrated. Gray (1986:115) reports that the d its utility in a conflict over land invasions. Furthermore, when FENAMAD was created and the Arakmbut had to choose between two organizational models, they opted for the one that reported more benefits. One of these models organized FENAMAD on a centra lized basis with a permanent office in Puerto Maldonado that could be managed by indigenous representatives, probably students who had finished their courses at secondary school there. The office would operate to put pressure on the government to respect i ndigenous rights; would act as a commercial center, helping communities sell their produce in th e city such as wood, crops or handicrafts; and would be a cultural center and hostel for those staying in town or in need of medical treatment. The other model viewed FENAMAD as decentralized and each community as a satellite of the federation. The officers would make regular visits to the communities and discuss their needs and requests during
215 frequently held general assemblies. Decisions would be taken in more participative fashion and there would be no central base. At the end, the first option was chosen (Gray 1986) , probably because that would provide personal benefits to community members . The correlation of legitimacy with ethnicity, however, is manifested in the attitudes of both the friends and enemies of indigenous organizations. Attacks on indigenous organizations and political movements often focus on the white people behind the curtains who are supposed to be in control. Enemies of the indigenous move ment often attempt to portray outside agents as manipulators of the indigenous people. In Peru, for example, whenever indigenous people protest, government officials state that they are being manipulated by leftist NGOs for political purposes. It is not s urprising that indigenous organizations are not financed by their members, a situation that is also normal among the Arakmbut (Gray 1986: 111). But this cannot be explained only by arguing that indigenous communities are too poor to redistribute surpluses t o their representative organizations. In fact, there are examples in which it has been perfectly feasible that indigenous people finance their own organizations, such as the Piaroa anizaciÃ³n IndÃgena Piaroa Unidos del Sipapo ( OIPUS ) that was largely f inanced by its member communities, or the Enxet of Paraguay who, despite their great poverty, are capable of generating central funds, at least within communities (Kidd 1997). A leader should help his group, and this may be a common phenomenon in lowland S outh America given the frequently observed requirement that indigenous leaders be generous. Given the existence of well funded non indigenous leaders it would not make sense if communities themselves were to provide mon ey for their own organizations.
216 The creation of organizatio ns was also a result of a pursuit for independence from anthropologists and NGOs. In fact, as I explained in Chapter 5, the formation of the Consejo Harakmbut entro FENAMAD, but a process common to Peruvian Amazonia. Evaristo Nugkuag, one of the initial founders of AIDESEP and its several times president, considers the birth of AIDESEP as a strategy to set up a native agenda without the influence of CIPA (Greene 2006: 341). The incorporation of legal formulas that allowed indigenous federations to have a state recognized political status is not in contradiction with thei r main objective: the defense of rights and culture. But arguing that the formalization is at odds with the aims of indigenous organizations is ignoring the dramatic changes indigenous peoples have experienced through a long history of engag ement with the national society . On the o ther hand, this view does not consider the possibility that formal organizations used by the Indians to uphold their demands be understood and practiced from their own cultural and political perspective. Seeing like an NGO The organizational culture of ethnic federations is not confined to the building of networks o f ethnic groups and communities; it also incorporates centralized and hierarchical governing and decision making structures that are taken from burea ucratic styles of civic life. As Amazonian Indians were increasingly exposed to the national society, they adopted bureaucratic structures that represent the ne plus ultra of Latin A merican civic life (Brown 1993: 312). With the formation of indigenous fe derations the number of ministry like positions rapidly proliferated (Chaumeil 1990: 109 ), and this is a characteristic not only of ethnic federations but also of every pular
217 Canteens, and Parents Associations) , in which each official has a formal bureaucratic title, ranging from president and secretary to treasurer. An important structural element shared by ethnic federations is an elected leadership representative of an d accountable to th eir constituencies (Smith 1996: 94 5). FENAMAD is managed by a governing board ( junta directiva ) c omposed of a president, a vice p resident, a secretary, a treasurer, a spokesperson , and a f iscal. Each member is assigned one or two areas o f work, which currently are seven: organization, territory and natural resources, economy and production, education and culture, health, indigenous women, and indigenous people in voluntary isolation . lage members meets at least once a year to elect a governing board that will act on its behalf and to make decisions about the programmatic goals of the organization (Smith 1985). FENAMAD holds an assembly or congress of community delegates every two years . In this congress, a new leadership is elected and the main goals, plans and policies of the organization are set by delegates of all the constituent communities. The main mechanism of decision making is universal voting, and each delegate has the right t o vote in any decision or election. This coincides with the notion that most people have of democratic governance, but some deficiencies can be identified. Initially, Moore worked with Arakmbut leaders to find a formula that would enable the federation to keep a broad ethnic representation i n the governing board. In those years, candidates were chosen from and by all delegates of the assembly, and the goal was to include members of all ethnic groups, emphasizing the qualification of the candidates as the main criterion for their election. But T his leader promoted the use of planchas electorales (electoral lists) in
218 Pl anchas electorales consist of forming up to three lists of candidates who are chosen from among the delegates attending the congress, who will then run for the offices et negotiations in which political alliances, rat her than ethnic representativeness , play a key role. These alliances are based on ethnic and friendship ties and, a ccording to a federation leader I interviewed , sometimes on the exchange of votes for moneta ry rewards . 3 The federation has much of the protocol of the state bureaucracy. The calling and holding of assemblies, the election of the governing board, and the duties of the elected officials are Peruvian civil code. The secretary records all the decisions and agreements made in each assembly in a minutes book, which has been previously legalized by a notary. The style in which the minutes are written follows the byzantine format of all official do cuments. Once FENAMAD got external support to design and fund development projects, a bureaucratic apparatus was needed to successfully undertake those projects. Thus, the management and implementation of development projects involved the incorporation of personnel in a wide range of hierarchical positions, such as project coordinators, specialists, managers, accountants, secretaries, etc. Development projects also required a wide array of literacy practices, such as agenda and report writing, planning meet ings and project documents, accounting and administration, media dissemination of information, and myriad other activities associated with projects a nd accountability (Aikman 2001: 114). Indigenous leaders had to learn these procedures and in fact some of t hem occupied leading positions within the projects, 3 Procedures to elect leaders have also been used as a political tool to pave the way for the election of certa in candidates. For example, an A rakmbut man who had been proposed as a candidate for the federation presidency was disquali ( FENAMAD 1994) .
219 especially those who had college degrees, although the most specialized and technical tasks were left for non indigenous professionals that were hired by the projects. Over time, project bureaucracies bec ame an integral part of the very functioning of FENAMAD. Until 1997, the year in which FENAMAD began to implement its largest accountability system. In fact, some tensions wi th Centro Eori regarding funds management were among the reasons why FENAMAD decided to terminate its relations hip with this NGO (GarcÃa 2013: which was key to helping the fe deration get funding for new projects. Now the federation operates as an hybrid of an ethnic federation, with a governing board comprised by leaders democratically elected, and an NGO with a salaried staff of lawyers, anthropologists, foresters, agronomist s, communicators, Global Information System (GIS) specialists, and extensionists. This NGO style of functioning has not been exempt from some confusion among the cies do not co ntr ibute funds for its operational expenses, FENAMAD found itself forced to rely on an NGO style structure and depend on project funds to survive as an organizatio n and fulfill its objectives. Therefore, besides its function as a representative organizatio n, defending the territorial rights of the communities, FENAMAD has played a role in providing development projects and other goods and services to their constituencies. Although much of the projects and funds have been devoted to the defense of communiti deman ds from their constituencies that tend to view the federation as a provider of health care , education and production services. One of the main consequences of this has been the
220 development of patronage relations between FENAMAD and the communities, as the latter have conditioned their allegiance to the federation on the provision of goods and services. The confusion and tension are wel l expressed in the answer an ex leader of FENAMAD gave to one of my questions about the relations between the federation and the communities. When I asked his opinion about the fact that some communities were asking FENAMAD to provide them with electricity or gasoline for their power generator in exchange for their support against Hunt Oil, Organized Mobilization While spontaneous community based mobilization still takes place, now FENAMAD centralizes most of the political m obilization of its constituent communities. The federation began to advise community members about their legal rights an d tactics of land defense. Th us, th e Arakmbut received training on how to watch their territories by patr olling the rivers in outboard canoes along the community borders. Patrolling allowed the Arakmbut to expel many outsiders who were attempting to seize community lands (Moore 1985 b : 174). In a sense, organized mobilization led by the federation has taken p reeminen ce over the less organized community mobilization. When I asked an Arakmbut man from Puerto Luz why the community had not yet taken any action against some miners found inside the community, he explained: Organized mobilization has also included many forms of street protest. While there were rallies to Puerto Mal donado before FENAMAD was created , street demonstrations in this city became a common form of resistance only after the formation of the federati on in 1982 . Until then, each community spontaneously and individually mobilized every time its members experienced a threat to their land rights. Now the federation calls and leads strikes and demonstrations in Puerto Maldonado . While in Puerto Maldonado, community members may
221 community and food are covered by the federation. protest was not fast. The main obstacle for the mobilization of indigenous people to Puerto Maldonado was their lack of identity cards ( Libreta Electoral ). Every time they travelled from their communities to Puerto Maldonado for a mobilization or other purposes, those without documents were arrested by the police at checkpoints . Most of the times, the policemen subjected them to humiliating abuse, such as to set them clean the bathrooms or clear the garden of the police station . Therefore, one of the main tasks of FENAMAD and its allied anthropologist s and NGOs was to help indigenous people to get identity cards and to file suits against abusive policemen . Once this problem was overco me, indigenous people experienced more freedom to protest in the streets of Puerto Maldonado. Street protest is an organized form of mobilization that involves the joint action of several communities. Its purpose is basically to have indigenous demands heard by regional and national authorities that are headquartered in the capit al city of the region. Street protest in cludes many forms of expressing disagreement with state policies and decisions affecting indigenous rights. These forms include marches, road blockades, chants, tire burning, banner holding, building oc cupations, and public speeches. The different forms of street protest are not of indigenous origin. Many of them are actua lly practiced by protestors all around the world and have been around for many decades. In Peru, this style of protesting became popular in the 1970s, when class based unions organized sev eral strikes in many cities under the influence of leftist political parties (Sulmont 1977) . These modalities of protest were reproduced by Madre de Dios unions that were active in the 1970s and 1980s . In 1980, several unions representing different labor g roups had been organized
222 or reorganized in Madre de Dios (Moore 1985 b ). FADEMAD, a n organization based on a union . FENAMAD learned modes of street protests from all these actors. Although street protest has been widely used by FENAMAD, it is not the main form of indigenous resistance. M arches and demonstrations may sometimes occur in a peaceful manner, but most of the times they result in v iolent confrontations with anti riot police men, sometimes with tragic consequences like the shooting and killing of some protesters. Because of the violent repression they may provoke, open forms of protest are always the last resort, negotiation and dialogue being the first options. Strikes, with their many forms of street protest, are generally called when several attempts to have the government hear indigenous demands have been exhausted. The shift to an organized form of indigenous mobilization in the Amazon has generally involved a reduction in the level of violence. Many founding leaders of the Aguaruna people of northern Peru who organized the Consejo Aguaruna Huambisa acknowledge the shift war like politics (Greene 2006 : 339 40). Rather than street protest or community border patrol , and despite the rebellious culture of indigenous people, the main forms of organized resistance are more subtle in nature, such as framing indigenous demands in global discourses, making all iances with powerful outsiders, and appealing to international legal frameworks to hold state authorities to their own normative commitments (e .g. I nternational L abor O rganization consistent with Scott (1986) s theory in that repress ive political regimes prevent the rise of open resistance. Nevertheless, open re sistance did not necessarily ari se with the advent of democratic openness , as Kerkvliet (1993) describes for the Philippines . FENAMAD was created under the
223 democratically elected government of Fernan do Belaunde Terry, a government more repressive than the previous military regimes. In fact, un like the case of Otavalo analyz ed by Korovkin (2000), in which hidden forms of resistance preceded organized political action, the case of FENAMAD reveals a reverse process in which organizing rather meant the shift from open to hidden resistance. Figure 6 1. . Photo by Danny Pinedo Identity Pol itics influenced from the start. This influence came from linkages with international actors and the
224 forged alliances with two global actors, h uman rights and environmental activists, turning the Amazonian indigenous movement into an ecological ethnic polit ical international movement. This global articulation was not based only on alliances with international activists, but also on the appropriation of transnational discourses of environmentalism that connect indigenous peoples and tropical rainforests (Conk lin and Graham 1995). Becoming Indians in Madre de Dios One of the key factors that enabled the various indigenous groups in Madre de Dios to collectively mobilize is the construction of an ethnic identity. When mining corporations and n indigenous lands dramatically increased in the late 1970s and early 1980s, the defense of indigenous lands through the mobilization of single communities became insufficient. The union of several communities and ethnic groups in a federation to face the threat of outsiders became an urgent necessity . Although the different Madre de Dios indigenous peoples were able to organize around a common political agenda through FENAMAD, many problems remained unsolved , such as maintaining unity among peoples who had previously been enemies. Rivalry and tensions between kin groups, settlements, and ethnic groups were a major obstacle f or an inter community and inter ethnic indigenous coalition. In order to overcome the ership promoted the construction of an identity that rather than resort ing ethnic groups in Madre de Dios. To forge their own identity discourse, F ENAMAD leaders relied on previously existing narratives on the unity of Madre de Dios indigenous peoples that are actually culturally diverse. This has been more apparent in the case of Harakmbut speaking peoples. Since the first accounts on the Harakmbut, this group has been described as divided into six sub groups with their own dialect and other discrete cultural features. Thus, it is commonly said that the Harakmbut are
225 comprised by the Arakmbut, Wachipaeri, Sapiteri, Toyeri, Arasaeri and Kisambaeri, na mes that etymologically referred to the places these groups originally inhabited and wer e, and still are, used for self denomination. As we have seen in Chapter 3, a ll these denominations were not in fact ethnic names but a way to denote a group by its geog raphic position. The Mashco denomination, as well as that of the six Harakmbut sub groups , were created through the contact with outsiders, from the rubber gatherers and missionaries to anthropologists. Even the notion of fierceness and warlike nature of the Mashco Indians was invented in the midst of violent contact with rubber gatherers, explorers and missionaries. These groups present marked cultural differences, among which language is the most important, as each group speaks their own dialect, but differences also include particular body paints, orna ments, clothing, and cosmology (FENAMAD 1992). Despite these cultural differentiation , the Harakmbut have been depicted as pertaining to a single ethnic group or linguistic family. The Dominican missiona ries spread the idea that the diverse Harakmbut groups made up a single socio cultural conglomerate. Anthropologists also significantly contributed to create th is idea of ethnic unity . The early ethnography on the Harakmbut gr oups identified them as a unit (Lyon 1967 ) . Subsequent studies found enough evidence of linguistic similarities among the Harakmbut dialects that the y were classified as a single linguistic family called Harakmbut Hate (Lyon 1975). Thus, the Harakmbut came to be recognized as a socio c ultural entity encompassing various subgroups linked by similar kinship systems, languages, myths, and customs (Califano 1982 ; Gray 1996, 1997a; Helberg 1996 ; Moore 1975, 2003; Wahl 1987) . Thus, the previously called Mashco became the new self denominated Harakmbut. These narratives of the Harakmbut unity became a key element in the social construction of an identity comprising not only all Harakmbut groups, but also all ethnic groups of Madre de Dios.
226 Ethnic identity as a culturally constructed category ra ther than as a given cultural feature has been emphasized by several scholars (Jackson 1991). An event that vividly illustrates the constructed nature of ethnic identity was when the Arakmbut people of San JosÃ© del Karene and Puerto Luz, accompanied by FEN AMAD representatives, Centro Eori staff , and British anthropologist Andrew Gray, held a meeting in the latter community to discuss c ommon problems. One of the issues addressed during the meeting was whether the Arakmbut should be called Arakmbut instead of Amarakaeiri, as they had been called since the first contacts. After much deliberation they opted for Arakmbut since Amarakaeri is a derogative term that means murderer (Gray 1997b). The interesting thing here is that the word Arakmbut as an identificatio n marker is used only before an outside audience, since in everyday conversation among the Arakmbut the term is used only to deno During public speeches, statements, reports and other written documents FENAMAD any specific ethnicity. When they need to refer to a particular community, they use the name of the community and very rarely the ethnic group with which the members of that community identify t hemselves . In Peru, indÃgena (indigenous person) has historically referred primarily to Andean indigenous peoples. Its use was largely absent in Ama zonia where the Velasco government adopted the tern nativo (native) into official usage in order to replace the colonial Spanish and Quechua vulgarisms used to refer to the lowlanders ( salvajes , chunchos , jÃvaros , and so on). 4 The law that in 1974 conferred Amazonian indigenous settlements legal recognition and land ti tles incorporated the term comunidades nativas (native communities) for newly registered indigenous settlements. In line with this shift, in 1982 the Harakmbut and their supporters 4 Velasco did the same with the use of to denote the indigenous people of Andean communities. His government replaced the then der ogative term for the more neutral word (peasant). Thus, the name of Andean peasant farmers shifted from to
227 FederaciÃ³n de Comunidades Nativas del RÃo Madre de Dios y Aflue ntes its Tributaries), avoiding any reference to particular ethnicities. Today urban based white and mestizo sectors of Peruvian society call indigenous themselves. In 1984, AIDESEP hosted a meeting in Lima which was attended by newly created indigenous organizations from four other Amazonian countries. The result of this meeting was the creatio n of the Coordinadora de Organizaciones IndÃgenas de la Cuenca AmazÃ³nica (COICA), the transnational organization that now represents Amazonian peoples from nine South American countries. This meeting served as an initial point of dissemination of the now i international agreements (the ILO dissemination of this international legal terminology is now evident in every day forms. While Nevertheless, this identity is not a monolithic discourse impervious to ethnic based factionalism. Discrete ethnic identities emerge especially in the context of tensions within Harakmbut boast of still having their own l anguage, while ridiculing the Shipibo for having lost their native language and now nearly no Shipibo speaks it. We have seen in Chapter 5 that w hen leaders from the lower ja and Shipibo, took control of FENAMAD, the Harakmbu t formed COHAR, an organization that, unlike FENAMAD, emphasized ethnic affiliation. When Julio Cusurichi, a Shipibo leader, won the Goldman Price
228 for his contribution to the defense of indigenous rights, especially for the creation of a reserve for indige nous people living in voluntary isolation, many Arakmbut leaders accused him of individually appropriating the authorship of collective indigenous struggles. The Arakmbut particularly resented the fact that achievements made mostly by Arakmbut people were acknowledged in a leader from a different ethnic group. Visual signs of i ndianness When I asked Mankehue, an elder of Puerto Luz, what he recalled the most from the time of the Dominican missions, he responded that the priests taught the Harakmbut to wear clothes. The most acculturated Harakmbut groups in the mission considered wearing Western clothing as a sign of being civilized and often ridiculed those who still wore traditional costumes. Paradoxically, forty years after they fled the mission, the Harak mbut take off their Western outfits and put on their feathered costumes to go out to the streets and express their political demands. Headdresses, body paint, feathered ornaments, and bows and arrows once played a role in traditional rituals and warfare. T oday many traditional rituals are no longer performed, while intertribal war is not practiced anymore. Nevertheless, indigenous peoples still use these artifacts, although in a very different context. Objects that embody the qualities associated with being Indian are essential cultural elements of indigenous activism. As markers of identity, this paraphernalia is used to mobilize not only non indigenous supporters but also to forge solidarity among indigenous groups. When indigenous people rally on the stre ets of Puerto Maldonado, organize meetings and office , they take their mass produced Western clothing off, paint their faces and bodies, and put on colorful costume speak of their condition as Indians, such as wearing feathered headdresses and cotton tunics,
229 wielding spears, bows and arrows, and showing off body paints and nakedness. The purpos e of this exhibition of visual signs of indianess is to legitimize indigenous people demands before a non indig enous public. Indigenous people seek to make sure that their look conforms to the image that non Indian, urban based sectors of Peruvian societ y have formed of native Amazonians, who are imagined through stereotypes that represent indigenous Amazonians as semi naked people wearing feathered ornaments and wielding bows and arrows. Fi gure 6 2. Indigenous demonstration in Puerto Maldonado . Photo by Danny Pinedo In a context where the issue of authenticity is at stake, indigenous people wear visual signs of indigeneity to affirm their quality of Indians. In Peru, the issue of who is an Indian is a very sensitive one since indigenous people demand from the state certain rights that are granted
230 and the titling of collective lands are granted only to settlements whose inha bitants fulfill the requirement o f being Indians. The need to affirm indigenous identity through images of indigeneity stems from the fact that the condition of being Indian is related to a traditional way of life that is connected to a conservationist ethics. Indigenous people will be co nsidered as authentic Indians as long as they show external signs of indianness, which means that they still have a lifeway that is traditional and hence ecologically harmonious. This is evident when Madre de Dios indigenous peoples are represented in para des held during festive days such as Independence Day or the anniversary of the founding of Puerto Maldonado. Primary school children use stereotyped images to represent Indians with semi naked and painted bodies and wearing feathered ornaments and grass s kirts. There is an obsessive tendency of representing Indians through these stereotypes even though it is so evident that th e majority of indigenous people no longer look like that. The fact that indigenous people do not use visual signs of indianness on a daily basis has been used by some representatives of the regional elite to deny the authenticity of indigenous leaders and, hence, to question the legitimacy of their claims for land rights. The manager of one of the largest logging companies in Madre de Dios talked to me about Antonio Iviche, an claim to the state he wears his feathers. But during a normal day you can see him wearing jeans, calling from his cell pho people sees visual signs of indianness as incompatible with most of the material expressions of modernity. By providing a source of identity, distinguishing Indians from non Indians, bo dy image creates a sense of solidarity between the different indigenous groups of Madre de Dios. I suggest
231 that by sharing an identity that is constructed through a body image that is distinctive of an indigenous culture, indigenous groups experience a sen se of communion that is based on a feeling of belonging and goes beyond the sharing of interests. Nevertheless, body image may also serve to emphasize difference among indigenous groups. Since members of each ethnic group use their distinctive body image, public manifes tations of protest may provide the context identify the Shipibo people by their cotton tunics called cushma or the Arakmbut by their body pain ts and feather ornaments. An emphasis on discrete ethnicities rather than on an all encompassing indigeneity as the basis for identity becomes more relevant when internal tensions among different groups accentuate. Indians and environmentalists Nevertheles s, FENAMAD's experience with the international environmental movement represents an exception to the way in which this articulation took place in the majority of Amazonian countries, including other lowland regions of Peru. Despite its efforts to legitimiz e image of an organization concerned environmental activists and NGOs have been very difficult to forge. FENAMA about the conservation of biological diversity, complicating any synergy between the two actors. As a result, d uring the first two decades of funded by a conservation funding agency. In the 1980s, the relation of FENAMAD with environmental NGOs working in Madre de Dios was marked by vertical interaction, prejudice and mistrust, even though at that ti me the environmental movement increasingly acknowledged indigenous people as allies rather than
232 enemies of conservation. At that time, the work of conservation NGOs in Madre de Dios was focused on the creation and management of natural protected areas, esp ecially national parks and the federation and these NGOs arose when the former began to demand its participation in decision making concerning the Manu National Park, especially where the park threatened the territorial rights of bordering native communities or groups in voluntary isolation localized within or near it. Often federation leaders were not included in the planning proces ses for this park (GarcÃa 2003: 299 30 0). In the early 1990s, as a result of its increasing demands for inclusion, FENAMAD was finally allowed to participate in the planning processes for natural protected areas. After the creation in 1990 of the Tambopata Candamo Reserved Zone, a democratic p lanning process was opened in which FENAMAD along with other local unions took active participation in consultation and decision making activities. Two forums were held in 1992 and 1993 in Puerto Maldonado to discuss issues related to the planning of the r eserved zone, including registration, tit ling and enlargement of lands for three native communities adjacent to the reserved zone. FENAMAD not only took active part in these debates but also co organized the second forum (GarcÃa 2003: 300 1) cipation in local conservation planning was not only a result of its struggles for inclusion but also of shared interests with environmentalists. The conditions for the building up of a quasi alliance with environmental NGOs did not emerge until the second half of the 1990s, when the federation organized a campaign against the plans of the consortium led by Mobil Oil Corporation to explore for hydrocarbons within two biologically diverse areas of Madre de Dios. Since they were also against oil development i n these areas, environmentalists
233 became sympathetic to the federation il . In 1996 the consortium had been granted two oil blocks for seismic exploration , one overlapped the Tampopata Candamo Reserved Zone and the proposed area for t he Am arakaeri Communal Reserve (La Torre 1999:120), and the other an area that served as a refuge for two indigenous groups in voluntary isolation . from mining set tlers and defend the rights of isolated gro ups against Mobil received the support of several national and international organizations . This gar ner ed international attention and acknowledgement for the federation , which in 1997 was awarded the BartolomÃ© de las Casas prize. enhanced interna rights and the conservation of for ests within ethnic territories served to attract funds from at least one major international environmental funding agency. In 2002, the Global Environmental Facility (GEF) agreed to fund FENAMAD with US$989,010 to implement a project for conserving biodiversity within the Amarakaeri Communal Reserve, which had been created that same ye ar . The project, known as Conservation and Sustainable Use of Biodiversity in the Am arakaeri Reserve and Neighboring Indigenous Lands, was aimed at carrying out sustainable co management of the reserve by loca l communities and the state. This has been the only FENAMAD project support ed by an international environmental agency to implement conservation activities within indigenous areas. environmental movement did not last long. O nce the GEF funded project ended in 2007, environmental NGOs resumed their vertical style of interaction wit h indigenous people. At that time, the environmentalist agenda shifted from an emphasis on resource management and biodiversity conservation to a concern about reducing the impacts of climate change. Thus, several NGOs sought to carry out REDD
234 (Reducing Em issions from Deforestation and Forest Degradation) projects within the forests of some native communities, but without requesting permission from FENAMAD . The federation protested against this new attempt to exclude indigenous people from conservation plan ning, and after launching its own REDD initiative called Indigenous REDD, it managed to force NGOs to partner with the federation to implement these projects within indigenous lands. Neither drawing on the lang uage of environmentalism nor building allian ces with ability to mobilize foreign support. This was more a result of the international acknowledgment ve communities and isolated indigenous groups, as well as of its struggles to be included in the regional planning processes of natural protected areas that affected indigenous rights. This way, FENAMAD did not change its project agenda to reflect internat ional concern for environmental conservation, but it remained adhered to Conclusion In the above discussion, the cultural underpinnings of indigenous mobilization in Madre de Dios ha ve been described and analyzed from a historical perspective . The proponents of the theory of everyday forms of resistance argue that the way the peasantry has historically resisted elite domination has been shaped by broader political regimes ( Scott 1985, 1986, 1990 ; Kerkvliet 1993 ). My data show that the cultural patterns of indigenous mobilization in Madre de Dios have evolved from open, spontaneous forms of resistance to hidden, organized political struggle. The most important factor in explaining this particula r evolvement is the nature of state formation in this particular portion of the Peruvian Amazon. While th e region remained outside of neighboring predatory state formations, the natives were able to re pel any attempt of domination by deploy ing o vert ly violent forms of resistance. When th e region was effectively incorpor ated
235 in to the Peruvian nation state in the twentieth century, state repression increased and the native groups resorted to more hidden forms of mobilization. While politically aut onomous, indigenous groups of the region put in to practice forms of resistance that were spontaneous and confrontational . Two common resistance tactics at this time were raids and isolation. While indigenous Amazon ians were at times able to form int er triba l alliance s to face external aggression, resistance was mo stly based on localized self defense. In this context, the lack of large political assemblages worked in favor of the natives, since th is forced outsider s to conquer each group individually. Local fo rm s of resistance have survived until the present day , as ma ny communities still practice them . In the 1960s, the state increased its presence in the southeastern Amazon lowlands. In order to facilitate access to the natural resources of the area, especial ly gold, a road that connected Puerto Maldonado to Cuzco was completed in 1965. Since t he late 1970s, state reforms that liberalized mining and oil industries prompted the arrival of corporations and thousan ds of settlers in Arakmbut lands. In this context, traditional ways of indigenous mobilization became insufficient to hold back massive invasions of lands . While the need for more permanent and larger organizations became urgent, t he autonomous nature of indigenous settlem ents, ethnic rivalry , and geographic distance worked again st the formation of networks of settlements and ethnic groups. Similarly, t he patronage way in which missionaries and school teachers interact ed with indigenous people , and the ideologies of individ ualism to which patronage gave rise, made it even more difficult to forge solidarity within and between local indigenous settlements. In this context, ethnic federations emerged as an alternative model that met the requirements for an organization beyond l ocal group s . Larger organizations made it possible to improve communication among local settlements and forge solidarity among them.
236 Organized mobilization brought a new way of doing politics . This new political culture was based on bureaucratic and hiera r chical structures of governance . M ini s t erial like po sitions proliferated wherever a new federation was created, whi le universal voting was adopted for the election of governing boards. In this context, FENAMAD evolved into a n NGO like organization, bec omin g dependent on externa l funds for development projects and hire qualified personnel to implement them. To a certain extent, this organizational change responded to external pressures towards making indigenous organizations more legible and hence legitimate interlocutors of state and other outside actors. While traditional forms of resistance are still in practice at the community level, FENAMAD centralized and rationalized political mobilization, introducing globalized forms of protest such as rallies, roa d blockades, speeches, chants and banner holding . These forms of political regimes have proved to be extremely repressive against native Amazonians. In this context, fighting in the legal arena and framing indigenous demands into globalized discourses became the main forms of resistance. Legal recognition of indigenous settlements, titling of community lands, and creation of communal reserves are all more effective str ategies to defend land rights than strikes and demonstrations. Global narratives have been appropriated and reformulated not only to influe nce public opinion and gain non indigenous support , but also to forge solidarity bonds among autonom ous kin groups and communities. In this process, FENAMAD has been more successful in capitalizing on identity and human rights discourses rather t han on environmental narratives . Although ethnic federations eventually became more bureaucratic, they were understood and , to a certain extent, adapted to more traditional forms of leadership . Indigenous leaders are
237 now chosen for their command of the Spanish language and familiarity with the national culture, and some of them may even reproduce mo re Western, authoritarian styles of government, but they are still pressured by their followers to show generosity in the distribution of goods and services. The role of anthropologists and other activists is still seen through this cultural lens. The moral obligation to be gene rous, alon g with a clientelist culture inherited from missionaries and school teachers, has led FENAMAD to use the distribution of development projects to gain P atron client relations are deeply rooted in the political culture of the country, permeating all aspect s and levels of political life, including ethnic federations. The theorists of the everyday form of resistance point to political repression as the main factor explaining the form peasants choose to resist (Scott 1985, 1986, 1 990; Kerkvliet 1986). In this sense, they argue that democratic regimes offer better political conditions for the emergence of overt forms of resistance. This chapter has shown that political regime should be taken with cau tion. In Peru, not only did the m ilitary regime of Juan Velasco Alvarado not repress grassroots organization s , but rather it promoted them as a strategy to gain popular support and control popular classes. Many Amazonian ethnic fe derations were formed under this military dictatorship. The return to democracy in 1980 provided more political openness for the creation of popular organizations and the participation of activists and NGOs in popular organizing. A t the same time , however, democratic governments showed repressive styles of politic al control that are usually associated with authoritarian regimes. Some scholars of the everyday forms of resistance theory emphasize the role of traditional cultural elements in shaping peasant resistance, such as reciprocity, the right to subsistence, an d hidden forms of resistance (Scott 1976, 1977). Others consider resistance as a product of both cultural values rooted in the past and modern cultural influences , such as notions
238 of citizenship and nation state (Kerkvliet 1990). This study is consistent w ith the latter approach in that it presents a case in which indigenous mobilization combines modern with traditional forms of resistance. The Arakmbut adopted formal and bureaucratic organizational models, borrowed notions from the human rights discourse ( self determination, territorial rights), appeal to international legal frameworks (I 69), and employ open tactics of protest such as strikes and demonstrations . At the same time, they follow political practices deeply rooted in local and na tional culture, such as the use of interethnic political alliances to get support for mobilization , and of generosity and patronage to gain access to desired resources.
239 CHAPTER 7 CONCLUSIONS This dissertation explore d indigenous sociality and its implications for political mobilization among indigenous people of the Peruvian Amazon. In particular, the study sought to understand how the different forms of social relations in which indigenous people engage, and the symbo lic meanings they embody, influence their capacity to mobilize people and resources with the aim to defend their land rights. In this respect, the dissertation has focused on social networks built on kinship (including fictive kinship), friendship, and pat ron client relations and their role in facilitating and/or inhibiting the formation of organizations, recruitment of supporters , and access to resources for mobilization. The study has also sought to explore the cultural traditions that inform indigenous m obilization and the political conjunctures that made it possible for them to appear and evolve. The general theoretical literature on this subject and specifically in the context of the Amazon is inconclusive on several vital questions. The dissertation so ught to answer three of these questions: (1) what are the m ain characteristics of sociality among the Arakmbut? (2) H ow does this sociality influence their capac ity to mobilize for land rights? And (3) what cultural traditions and political conditions gave rise to the ways they mobilize? Indigenous Sociality and Networks The dissertation shows that, among the Arakmbut, sociality is characterized by a combination of altruistic and instrumentalist relations. The prevalence of one type of sociality or the othe r depends on the degree of closeness, both social and spatial, among individuals. Thus, relations among individuals linked by close ties of kinship and friendship are characterized by conviviality and generosity . Since close kin and friends tend to live in close proximity, forming discrete compounds of huts, conviviality and generosity are confined to households living in the
240 same residence group. Thus, Arakmbut sociality is also associated with a settlement pattern characterized by the division of the vill age into smaller kin based clusters of households. Among the Arakmbut of Puerto Luz , there are six patrilineal and exogamic clans whose members are dispersed among different residence groups. Since clans are exogamic and residence after marriage is mainly virilocal, residence groups are comprised by male adult siblings and their respective households. Thus, in a residence group men tend to be of the same clan, while women belong to different clans. Residents of the same cluster show generosity to each other through food sharing and help in gardening, with no expectation of return. The residence group is also the space where everyday socializing and cooperation in the form of barbasco fishing take place. All these activities strengthen the bonds of solidarity among households of residence groups. Beyond residence groups, relations among kin and friends become less convivial and more instrumentalist, in the sense that the provision of help is accompanied by an expectation of reciprocation. At the level of the s ettlement, interaction is limited and confined to assemblies, fiestas, work parties and barbasco fishing, although the frequency and importance of each of them vary. Community assemblies and fiestas seem to be more important, the former being a regular for m of discussing and making decisions about collective issues, and the latter an annu al mechanism to socialize and relax tensions among all kin groups in the community. Community work parties and fishing with barbasco have decreased in importance in the pas t decade and nowadays are carried out less fre quently. In general, despite their diminished importance, these forms of interaction contribute to smooth things over and strengthen solidarity bonds at the level of the settlement group. The Arakmbut also show a strong tendency to establish social ties with outsiders. Outsiders with whom the Arakmbut develop ties include mining settlers, school teachers,
241 loggers, traders, and anthropologists. Relations with these actors are built mainly through friendship, godp arenthood and patron client ties. These social ties are based on morals of reciprocity, which means that any provision of goods or services to the other party entails an expectation of retribution. In general, these social ties are governed by an instrumen talist motivation since they are seen as a social mechanism to gain access to recourses. Miners have capital but lack lands to mine gold, while the Arakmbut own lands but lack capital. This explains why the Arakmbut look for outsiders, especially settlers and school teachers, as godparents. This reciprocal exchange may create solidarity bonds that in some cases are stronger and more important than those among kin. Although this reciprocity is accompanied by the ideology of an equitable exchange, in practic e the quality and amount of what is returned may vary according to the social tie and parties involved. Friendship and godparenthood are generally horizontal relations characterized by balanced reciprocity. Patron client ties give rise to more vertical rel ations and, hence, reciprocal exchange is less balanced. Nevertheless, a distinction should be made between patronage relations with loggers, more in the form of debt peonage, and clientelist relations with Hunt Oil . The Arakmbut tend to a void relations that place them i n a subordinate position , which is reflected in the fact t hat deb peonage with loggers is not as exploitative as in other parts of the Amazon. Even though timber patrons have comparatively more access to financial capital, they do not cont rol access to trading goods, which can be acquired by the Arakmbut through participation in the market. This is not the case with clientelist relations with Hunt Oil, which relies on its disproportionately higher economic power to impose the terms of the e xchange. Sociality among indigenous Amazonians has been analyzed from two major approaches: one that argues that Amazonian sociality is based on individual autonomy and the search for a
242 convivial life in community by turning af fines into kinfolk ( Belaunde 2001; Gow 1991; McCallum 2001; Overing Kaplan 1975; Santos Granero 1991 ), and an other that views affinity (non kin) and predatory relations as more important for Amazonian sociality (Descola 1994; Taylor 1996; Viveiros de Castro 1992 ). This study is only p artially consistent with both approaches. On the one hand, my findings show that the Arakmbut sociality is characterized by personal autonomy and aversion to any form of structural constraint, coupled with an emphasis on conviviality, mainly expressed thro ugh generosity, among kinfolk . Unlike the conviviality approach, however, the Arakmbut show a strong tendency to establish ties with affines and other outsiders. Since pre Columbian times, the Harakmbut have engaged in trading, marriage and war relationshi ps with Andean highlanders. Nevertheless, they do not seem to practice turning outsiders into kin by using kinship teknonyns. On the other hand, unlike the predatory approach, Arakmbut not based on unilateral taking, but on b alanced reciprocity. This is consistent with their aversion to engage in lopsided relations. This morality of balanced reciprocity with outsiders is attached to friendship, god parenthood and patron client relations, which in some cases take preeminence ov er relations to other Arakmbut in the community. Nonetheless, a distinction should be made between debt peonage relations with timber patrons and clientelist relations with corporation or state actors. In the latter case, the relation is marked by an extre me inequality in political and economic power, which results in a highly unbalanced reciprocal exchange that is more detrimental to the Arakmbut. The Arakmbut do not freely engage in these relations, which are usually imposed by the state, and lack the pow er to negotiate more equitable terms of the
243 with outside actors other than the traditional outsiders (settlers, timber patrons), such as state, corporate, and NGO actors, interact with local understandings of sociality. Networking to Mobilize Over the past four decades, the Arakmbut lands have attracted highland settlers and corporations seeking to exploit their rich gold and hydrocarbon resources. The presence on A rakmbut lands of mining companies and settlers, and more recently of oil corporations, has represented a threat to the ir livelihoods and a violati on of their land rights. How have hi ghly atomized, kin based groups , with a strong tendency to establish ties with outsiders, organized themselves to defend from this threat? In this dissertation , I have demonstrat ed that social ties have played a crucial role in mobilizing indigenous communities. Nevertheless, social networks have had differential influence on po litical mobilization depending on the type of social tie and scale. When the Arakmbut of Puerto Luz experienced encroachment on their lands by mining companies and settlers, they have been able to overcome their kinship based divisions, and collectively ex pelled invaders from their community lands. At least two factors have contributed to this. First, the Arakmbut experience shared incentive s to collectively defend the ir land base . These incentives come from two factors: (1) all community members depend on land for their livelihoods (gold mining, horticulture , and hunting fishing); a nd (2) community members are co owners of those lands. Therefore, any invasion of those lands is perceived as a threat to the interests of all community members. Second, once a co mmon threat is perceived and a decision in favor of self defense has been made, kinship solidarity facilitates (1) recruitment of members of residence groups; and (2) alliances with the neighboring community of San JosÃ© del Karene. This collect ive action, h owever, is temporary and lasts as long as the common threat is
244 perceived. Once this common threat is gone, social organization returns to its atomized state and community solidarity remains latent . Bridging ties with outsiders have had differential effects on indigenous mobilization, depending on the type of social relations and the actors involved . These social relations are based on friendship, godparenthood and patron client ties, and have linked Arakmbut people with missionaries, anthropologists , mining settlers, school teachers , and corporation employees. At the individual level, engagement in relations with outsiders has been beneficial for the Arakmbut as it has enabled them to gain access to scarce resources such as money, trading goods, jobs , formal education, health care, gifts, etc. At the community level, however, these relationships have exacerbated existing rivalries among kin groups and created conflicts between individual and collective interests, weakening community level solidarity. godparents and patrons has overridden solidarity among kin groups , i nhibiting their capacity to mobilize against external threats. My research has shown that social ties with anthropologists and other indigenous rights activi sts have had more positive effects on indigenous mobilization. Networks with these actors, primarily based on friendship, were crucial when an escalation in land invasions rendered the creation of more permanent coalitions of kin groups imperative. The Ara kmbut relied on networks with activists to gain access to the technical and financial assistance necessary to form an ethnic federation that would help them to defend their land rights more effectively. Once it was created in 1982, FENAMAD provided communi ties with an ad hoc institutional platform to forge networks with international donor agencies, and channel technical and financial resources to fund and implement land titling projects and other benefits such as bilingual education, health care , agricultu ral projects, and legal defense. The small population that characterizes ethnic
245 groups in Madre de Dios required that FENAMAD play a role as a coordinating body of a multiethnic network of communities. In this context, the federation not only improved comm unication between communities, but was central in leading and funding indigenous mobilizations. Nevertheless, while FENAMAD has operated as a powerful tool to advance indigenous interests in Madre de Dios, it has also reproduced existing social inequalitie s among indigenous groups and between them and outsiders. Ethnic groups have tended to compete for the political control of the federation and its economic resources, leading to the creation of competing federations by secessionist groups. In this context, I argue that the creation of organizations is not the result of rational planning but of political struggle. Ethnic tensions also played a part in the . As the manageme nt of an indigenous federation requires specialized knowledge on how to deal with the outside world, FENAMAD has come to be dominated by indigenous leaders from ethnic groups more integrated in the national society and hence more educated and acculturated. Thus, those who have received formal education in mission schools, have been enrolled in mandatory military conscription, have f orged personal networks with outside s , and have learned how to navigate the intricate state and NGO bureaucracies, became part of an indigenous elite that has dominated FENAMAD since its inception. Moreover, their experience as federation officials and project staff provided these leaders with further opportunities to acquire more specialized knowledge and expand their personal ne tworks, enabling some of the m to occupy positions in larger indigenous federations or the regional government. Subordinated groups such as women and members of less integrated and acculturated communities have been less successful in gaining access to FENA
246 participation in local com munities and the federation has undoubtedly increased in the past decade, women are still less represented in leading positions. Members of the more isolated communities have even l ess chances to gain experience and be elected as leaders. Sociologists have emphasi zed the role of social networks in recruiting activists for social movement s , especially ties with friends and acquaintances (della Porta 1988; Gould 1993; McAdam 1986, 1988 ; Snow et al. 1980). My research contributes to this literature by specifying the different kinds of social ties affecting political mobilization and their differential effects among an Amazonian indigenous group. Chapter 5 showed that the mobilization of participants in a social movement can also be achieved through kinship ties. Among the Arakmbut, leaders mobilize their kinfolk by appealing to clan solidarity. Moreover, social networks not only enhance individual participation in mobilization. They may a lso play a key role in facilitating the formation of social movement organizations and access t o the resources necessary for them . This role has been played primarily through bridging networks with anthropologists and other human rights acti vists. Friendsh ip ties with non indigenous actors were more critical to create organiz ations capable of forging multicommunity and inter ethnic coalitions, as well as gaining access to resources for mobilization. In contrast with relationships with missionaries and politic al parties, which have traditionally been built on patronage and an authoritarian culture, friendship networks with anthropologists offered more horizontal reciprocal relations. My research also advances sociological theory on the role of social networks in social movements by defining the differential effect s social networks have at varying scales. W hile kinship ties are central to mobilize people at the community level, their ro le becomes less effective at a much broader scale. At an inter community level, long term ethnic tensions prevail over kinship ties. By creating personal loyalties, godparenthood and cliente list networks may
247 enhance the ability to gain access to resources a t an individual level, but they may also undermine political mobilization at a collective level. On the other hand, an important school of social capital theory sees social networks only as enhancing collective action and access to resources (human, capita l , and cultural capital) (Coleman 1988 , 1990 ; Putnam et al. 1993). From this perspective, social networks are seen as social capital that can be accumulated and help individuals and groups to reduce transaction costs and make rational decisions, allowing s ociety as a whole to achieve economic development and democratic governance. Applying this approach to indigenous federations in Latin America, some scholars have argued that organiza tions as social capital enhance livelihoods and access to land rights (Bebbington and Perreault 1999; Bebbington et al. 1993; Perreault 2001, 2003) . From a Bourdieusian perspective, my study enriches social capital theory by presenting a cas e of the down side of social networks . While indigenous federations l ike FENAMAD represent a powerful institutional tool to mobilize individuals and resources for the defen se of indigenous land rights, they also enable better off indigenous groups to maintain their privileges by controlling the organization, appropriating i ts resources, and building personal networks. The differential position of individuals and communities within the prevailing social structure influence the ability of some to construct and accumulate more social capital. More isolated and less acculturated communities, as well as women, occupy the lower strata and hence have fewer opportunities to participate in and benefit from the organization. Political Mobilization as Culture The ways the A rakmbut mobilize have their roots in a combination of traditional and modern cultural models of resistance and protest. These cultural patterns have been shaped by the broader political and economic processes to which the region has been articulated throughout hi story. In this sense, it can be argued that indigenous mobilization in Madre de Dios, and
248 perhaps all around the Peruvian Amazon, has evolved from open to hidden forms of res istance. For centuries, the Madre de Dios region remained politically and economic ally autonomous, a condition that allowed indigenous dwellers to resist external domination and develop a culture of resistance that is still prevailing. In this context, indig enous resistance was based on practices associated with intertribal war such rai ds and selfisolation, albeit some forms of inter tribal alliances were also forged to face external aggression. Circumstances changed with the increasing articulation of Madre de Dios to the national implemented to improve communic a t ion to the area and encourage private investment in resource extraction. Poverty in nearby highland areas, as well as th e rise in the international price of gold, led to massive migration into the area, affecting indigenous land rights. Under these new circunstances, t raditional ways of mobilization became unable to ensure the defense of indigenous land rights and cultural integrity. Land rights defense increasingly required more permanent and larger alliances among different communities and ethnic groups. Initially , deeply rooted cultural patterns worked against the adoption of more or ganized forms of resistance. T he need for formal organizations faced much resistance stemming from the nature of indigenous social organization and new cultural models of behavior. The autonomous n ature of indigenous settlements and ethnic rivalry made it difficult to create networks beyond the settlement group. The culture of patronage introduced by missionaries, patrons and school teachers exacerbated individualist attitudes and weakened solidarity among loca l settlements and ethnic groups . Articulat ion to the nation state increased contact not only with outsiders but also with new cultural models of resistance. These new models were primarily based on organizing.
249 Organizing was also promoted by Velasco authoritarian regime in its attempt to gain support from and coopt lower classes. F ormalization r endered indigenous organizations more legible to the state, making indigenous pe ople legitimate interlocutors. In this context, the ethnic federation provided the Arakmbut with an alternative model that met the requirements for an organization beyond the residence group. Surprisingly, this Velasco military dicatorship provided enough associational freedom that enabled indigenous people to make contact and build networks with anthropologists and NGOs. Networks wit h anthropologists and human rights activists, largely based on the ancient indigenou s culture of establishing inter tribal and interethnic alliances, finally made it possible to create FENAMAD. Indigenous networks consolidated with the return to a democratically elected government in the 1980s , which also provided the political space for anthropologists and human rights acti vists to foster the creation of organizat ions in ways that responded to indigenous interests. Organizing, especially through t he fomation of ethnic federations, enabled indigenous peoples to strengthen communication among local settlements and ethnic groups. Autonomous communities and rival ethnic groups started to forge solidarity bonds through participation in mobilization fost ered by FENAMAD . Nevertheless, this solidarity, albeit temporary and goal specific, still coexists with individualism and tribal rivalries and tensions. The adoption of formal organizations inaugurated a new form of indigenous politics. L eadership and dec ision making be came hierarchical and bureaucratic, although these new structures were combined with more traditional notions of leadership as the generous distribution of goods. FENAMAD c entralized and rationalized political mobilization , introducing globa l forms of protest such as strikes, rallies, road blockades, speeches, and banner holding , although traditional forms of resistance are still p racticed at the community level. Nevertheless,
250 en in nature . I mportant hidden tactics of indigenous activism include invoking national and i nternational legal frameworks on indigenous rights, titling indigenous lands, and framing indigenous demands into global narratives of identity or human rights , all of them used to legitim ize indigenous political claims. In this sense, this dissertation advances the theory of everyday forms of resistanc e by providing a more nuanced examination of the pol itcal and cultural factors underlying the different forms of indigenous resistan c e in an Amazonian region. Theorists of the everydaty foms of resi stance tend to emphasize dichotomic interpretations of these factors: forms of resistance are whether hidden or open, traditional or modern; politcal regimes are either democratic or autoritharian, repressive or tolerant, leaving little or no space for complex and sometimes contradictory combina tions (Scott 1985, 1986, 1990; Kerkvliet 1986). This study shows that politcal organizing can result either from democrat ic openness or authoritatian , as actors may use local organizations to coopt lower class movements. It also shows that even under democratic regimes repression ma y characterize state relations with subjects , rendering hidden resistance more relevant.
251 LIST OF REFERENCES Adams, Rebecca G., and Graham Allan 1998 Contextualizing Friendship. In Placing Friend s h ip in Context. Rebecca G. Adams and Graham Allan, eds. Pp. 1 17. Cambridge: Cambridge University Press. Adas, Michael 1986 From Footd ragging to Flight: The Evasive History of Peasant Avoidance Protest in South and Southeast Asia. The Journal of Peasant Studies 13(2):64 86. Aikman, Sheila 1999 Intercultural Education and Literacy: An Ethnographic Study of Indigenous Knowledge and Learn ing in t he Peruvian Amazon. Amsterdam : John Be n jamin s . 2001 Literacies, Languages and Developments in Peruvian Amazonia. In Literacy and Development: Ethnographic Perspectives. Brian V. Street, ed. Pp. 103 120. London: Routledge. AramburÃº , Carlos E. 1984 Expansion of the Agrarian and Demographic Frontier in the Peruvian Selva. In Frontier Expansion in Amazonia. Marianne Schmink and Charles H. Wood, eds. Pp. 153 179. Gainesville: University Press of Florida. Barth, Fredrik 1969 Ethnic Groups and Boundaries: The Social Organization of Culture Difference. Propect Height s : Waveland Press. Bebbington, Anthony 1997 Social Capital and Rural Intensification: Local Organizations and Islands of Sustainability in the Rural Andes. The Geographical Journal 163(2):189 197. Bebbington, Anthony J., Herna n Carr asco, Lourdes Peralbo, Galo RamÃ³ n, Jorge Trujillo, and Vi ctor Torres 1993 Fragile Lands, Fragile Organizations: Indian Organizations and the Politics of Sustainability in Ecuador. Transactions of the Ins titute of British Geographers 18(2):179 196. Bebbington, Anthony, and Thomas Perreault 1999 Social Capital, Development, and Access to Resources in Highland Ecuador. Economic Geography 75(4):395 418. Belaunde, Luisa Elvira 2001 Viviendo Bien: GÃ©nero y Fertilidad entre los Airo Pai de la AmazonÃa Peruana. Lima: Centro AmazÃ³nico de AntropologÃa y AplicaciÃ³n PrÃ¡ctica.
252 Bell, Sandra, and Simon Coleman 1999 The Anthropology of Friendship: Enduring Themes and Future Possibilities. In The Anthropology of Friendship. Sandra Bell and Simon Coleman, eds. Pp. 1 19. Oxford: Berg. Berthelot, Jean 1978 L'Explo i tation de s MÃ©taux PrÃ©cieux au Temps des Incas. Annales. Ã‰conomies, SociÃ©tÃ©s, Civilisations 33(5 6):948 966. Biedma, Manuel 1981 La Conquista Franciscana del Alto Ucayali. Lima: Editorial Milla Batres. Bourdieu, Pierre 1984 Distinction: A Social Critique of Judgement and Taste. Cambridge: Har vard University Press. 1986 The Forms of Capital. In Handbook of Theory and Research for the Sociology of Education. J. Richardson, ed. Pp. 241 258. Westport: Greenwood. Brass, Tom 1991 Moral Economists, Subalterns, New Social Movements, and the (Re )Em ergence of (Post )Modernised (Mi ddle) Peasant. The Journal of Peasant Studies 18(2):173 205. Brown, Michael 1993 Facing the State, Facing the World: Amazonia's Native Leaders and the New Politics of Identity. L'Homme 33(2 4):307 326. Brown, Michael F., and Eduardo FernÃ¡ndez 1991 War of Shadows: The Struggle for Utopia in the Peruvian Amazon. Berkeley: University of California Press. Brysk, Alison 1994 Acting Globally: Indian Rights and International Politics in Latin America. In Indigenous Peoples and Democracy in Latin America. Donna Lee Van Cott, ed. Pp. 29 51. New York: St. Martin's Press. 2000 From Tribal Village to Global Village: Indian Rights and International Relations in Latin America. Stanford: Stanford University Press. Califano, Mario 19 82 EtnografÃa de los Mashco de la AmazonÃa Sud Occidental del PerÃº. Buenos Aires: FundaciÃ³n para la EducaciÃ³n, la Ciencia y la Cultura. 1990 El Concepto de Acto Ã‰tico JurÃdico y la PercepciÃ³n del Blanco entre los Mashco (PerÃº). In El Aborigen y el Derecho en el Pasado y el Presente. Abelardo Levaggi, ed. Pp. 43 103. Buenos Aires: Universidad del Museo Social Argentino.
253 Chaumeil, Jean Pierre 1990 "Les Nouveaux Chefs...": Pratiques Po litiques et Organisations IndigÃ¨ nes en Ama zoni e PÃ©ruvienne. Problemes d' AmÃ©rique Latine 96:95 113. Chirif, Alberto 1974 Los Congresos de las Comunidades Nativ as: El Inicio de la ParticipaciÃ³ n. ParticipaciÃ³n 3(5):46 53. 1985 25 AÃ±os de PolÃtica de Desarrollo Rural en la AmazonÃa Peruana y sus Repercusiones en las Sociedades IndÃgenas de la RegiÃ³n. In Balances AmazÃ³nicos: Enfoques AntropolÃ³gicos. Jurg Gasche and JosÃ© M. Arroyo, eds. Pp. 171 223. Iquitos: CIAAP and UNAP. Clastr es, Pierre 1989 Society against the State: Essays in Political Anthropology. New York: Zone Books. Coleman, James S. 1988 Social Capital in the Creation of Human Capital. The American Journal of Sociology 94(S95 S120). 1990 Foundations of Social Theory. Cambridge: Belknap Press. Conklin, Beth A., and Laura R. Graham 1995 The Shifting Middle Ground: Amazonian Indians and Eco Politics. American Anthropologist 97(4):695 710. CORDEMAD 1986 Madre de Dios: El PerÃº Desconocido. Lima: CorporaciÃ³n Departamental de Desarrollo de Madre de Dios. Dean, Bartholomew 2002 State Power and Indigenous Peoples in the Peruvian Amazon: A Lost Decade, 1990 2000. In The Politics of Ethnicity: Indigenous Peoples in Latin American States. David Maybury Lewis, ed. Pp. 199 237. Cambridge: Harvard University Press. de la Vega, Garcilaso 2006 Royal Commentaries of the Incas and the General History of Peru. Indianapolis: Hackett. della Porta, Donatella 1988 Recruitment Processes in Cladestine Political Organizations : Italian Left Wing Terrorism. In From S tructure to Action : Comparing Social Movement Research across Cultures . Bert Klandermans, Hanspeter Kriesi, and S idney Tarrow, eds. International Social Movement Research: A Research Annual, Vol. 1. Pp. 155 169 . Greewinch: JAI Press.
254 Descola, Philippe 1994 In the Society of Nature: A Native Ecology in Amazonia. Cambridge: Cambridge University Press. 1996 The Spears of Twilight: Life and Death in t he Amazon Jungle. New York: New P ress. 2012 Beyond Nature and Cu lture: Forms of Attachment. HAU: Journal of Ethnographic Theory 2(1):447 471. Eckstein, Susan 2001 Power and Popular Protest in Latin America. In Power and Popular Protest: Latin American Social Movements. Susan Eckstein, ed. Pp. 1 60. Berkeley: Univer s i ty of California Press. Eisenstadt, S. N., and L. Roniger 1984 Patrons, Clients and Friends: Interpersonal Relations and the Structure of Trust in Society. Cambridge: Cambridge University Press. Espinosa, "scar 2004 Indigenous Politics in the Peruvian Amazon: An Anthropological and Historical Approach to Shipibo Politica l Organizations. PhD Dissertation, New School University. 2012 El Sistema de Cuota Electoral IndÃgena en la AmazonÃa: Problemas y Limitaciones. In ParticipaciÃ³n Electoral IndÃgena y Cuo ta Nativa en el PerÃº: Aportes para el Debate. Alicia del Ãguila and Milagros Suito, eds. Pp. 37 46. Lima: Insttituto Internacional para la Democracia y la Asistencia Electoral. Fantasia, Rick 1988 Cultures of Solidarity: Consciousness, Action, and Contem porary American Workers. Berkeley: University of California Press. FENAMAD 1992 Informe TÃ©cnico sobre la Propuesta para el Establecimiento de la Reserva Comunal Amarakaeri. Puerto Maldonado: DirecciÃ³n Subregional Agricultura Ma dre de Dios. 1994 Â¿QuÃ© pasÃ³ en Nuestro VIII Congreso? AclaraciÃ³n y Deslinde. Avance IndÃgena 4(5):6 8. 2007 25 AÃ±os de Lucha y ReivindicaciÃ³n IndÃgena: Trayectoria de FENAMAD. Avance IndÃgena 15:6 15. Ferree, Myra Marx 1992 The Political Context of Rationality: Rational Choic e Theory and Resource Mobilization. In Frontiers in Social Movement Theory. Aldon D. Morris and Carol McClurg Mueller, eds. Pp. 29 52. New Haven: Yale University Press. Fifer, J. Valerie 1970 The Empire Builders: A History of the Bolivian Rubber Boom and the Rise of the House of Suarez. Journal of Latin American Studies 2(2):113 146.
255 Fine, Ben 1999 T he Developmental State is Dead Long Live Social Capital? Development and Change 30(1 19). Foster, George M. 1969 Godparents and Social Networks in Tzintzun tzan. Southwestern Journal of Anthropology 25(3):261 278. Fox, Jonathan 1994 The Difficult Transition from Clientelism to Citizenship: Lessons from Mexico. World Politics 46(2):151 184. 1996 How Does Civil Society Thicken? The Political Construction of Social Capital in Mexico. World Development 24(6):1089 1103. Franco, Carlos 1979 PerÃº: ParticipaciÃ³n Popular. Lima: Centro de Estudios para el Desarrollo y la ParticipaciÃ³n. Fuentes, Aldo 1982 Parentesco y Relaciones de ProducciÃ³n en una Comunidad Hara kmbut en el Sur Oriente Peruano. Unpublished MS. GarcÃa, Alfredo 2000 Madre de Dios: Cuando el Mosquito Pica. Asuntos IndÃgenas 3:46 51. 2003 FENAMAD 20 AÃ±os DespuÃ©s: Apuntes s obre el Movimiento IndÃgena AmazÃ³nico en Madre de Dios. In Los Pueblos IndÃgen as de Madre de Dios: Historia, EtnografÃa y Coyuntura. Beatriz Huertas and Alfredo GarcÃa, eds. Pp. 274 309. Lima: International Work Group for Indigenous Affairs. Giddens, Anthony 1992 The Transformation of Intimacy: Sexuality, Love, and Erotism in Mode r n Societies. Cambridge: Polity Press. 2000 Runaway World: How Globalization is Reshaping our Lives. New York: Routledge. GÃ¶ hring, Herman 1877 Informe al Supremo Gobierno del PerÃº sobre la ExpediciÃ³n a los Va lles de Paucartambo en 1873 al M ando del Coron el D. Baltazar La Torre. Lima: Imprenta del Estado. Gould, Roger V. 1993 Collective Action and Network Structure. American Sociological Review 58(2):182 196.
256 Gow, Peter 1991 Of Mixed Blood: Kinship and History in Peruvian Amazonia. Oxford: Clarendon Press. 2011 "Me deixa em Paz!" Um Relato EtnogrÃ¡fico Preliminar sobre o Isolamento VoluntÃ¡rio dos Mashco. Revista de Antropologia 54(11 46). Granovetter, Mark S. 1973 The Strength of Weak Ties. American Journal of Sociology 78(6):1360 1380. Gray, Andrew 1986 Development among the Amarakaeri of Southeastern Peru. Copenhagen: International Work Group for Indigenous Affairs. 1996 The Arakmbut: Mythology, Spirituality, and History in an Amazonian Commun ity. Providence: Berghahn Books. 1997a The Last Shaman: Change in an Amazonian Community. Providence: Berghahn Books. 1997b Indigenous Rights and Development: Self Determination in an Amaz onian Community. Providence: Ber ghahn Books. Greene, Shane 2006 G etting over the Andes: The Geo Eco Politics of Indigenous Movements in Peru's Twenty First Century Empire. Journal of Latin American Studies 38:327 354. Guerra GarcÃa, Francisco 1983 SINAMOS y la PromociÃ³n de la ParticipaciÃ³n. In El PerÃº de Velasco. Carl os Franco, ed. Pp. 681 708. Lima: Centro de Estudios para el Desarrollo y la ParticipaciÃ³n. Gudeman, Stephen 1971 The Compadrazgo as a Reflection of the Natural and Spiritual Person. Proceedings of the Royal Anthropological Institute of Great Britain and Ireland:45 71. Guha, Ramachandra 1989 Saboteurs in the Forest: Colonialism and Peasant Resistance in the Indian Himalaya. In Everyday Forms of Peasant Resistance. Forrest D. Colburn, ed. Pp. 64 92. Armonk: M. E. Sharpe. Gutmann, Matthew C. 1993 Rituals of Resistance: A Critique of the Theory of Everyday Forms of Resistance. Latin American Perspectives 20(2):74 92. Harriss, John, and Paolo De Renzio 1997 'Missing Link' or Analytically Missing?: The Concept of Social Capital. Journal of Internati onal Development 9(7):919 937.
257 Hechter, Michael 1987 Principles of Group Solidarity. Berkeley: University of California Press. Helberg, Heinrich 1993 TerminologÃa de Parentesco Harakmbut. AmazonÃa Peruana 12(23):107 140. 1996 Mbaisik en la Pe numbra d el Atardecer: Literatura Oral del Pueblo Harakmbut. Lima: Centro AmazÃ³nico de AntropologÃa y AplicaciÃ³n PrÃ¡ctica. Hobsbawn, Eric J. 1973 Peasants and Politics. Journal of Peasant Studies 1(1):3 22. Holzner, Claudio A. 2004 The End of Clientelism? Strong Ties and Weak Networks in a Mexican Squatter Movement. Mobilization: An International Quarterly 9(3):223 240. Huertas, Beatriz 2004 Indigenous Peoples in Isolation in the Peruvian Amazon: Their Struggle for Survival and Freedom. Copenhagen: I ntern ational W ork Group for I ndigenous A ffairs . Hvalkof, SÃ¸ ren 1994 The A shaninka Disaster and Struggle The Forgotten War in the Peruvian Amazon. Indigenous Affairs 2:20 32. Hvalkof, SÃ¸ ren, and Hanne Veber 2005 Los AshÃ©ninka del Gran Pajonal. In GuÃa EtnogrÃ¡fica de la Alta AmazonÃa. Fernando Santos and Frederica Barclay, eds. Pp. 75 279. PanamÃ¡: Smithsonian Tropical Research Institute. INEI 2008a Censos Nacionales 2007 XI de PoblaciÃ³n y VI de Vivienda. Resultados Definitivos: Depart amento de Mad re de Dios. Vol. 1. Lima: Instituto Nacional de EstadÃstica e InformÃ¡tica. 2008 b II Censo de Comunidades IndÃgenas de la AmazonÃa Peruana 2007: Resultados D efinitivos. Vol. I. Lima: I nstituto N acional de E stadÃstica e I nformÃ¡tica . IWGIA 1999 Informe de EvaluaciÃ³n Proyecto Plan Karene, Madre de Dios. Unpublished MS. Jackson, Jean E. 1991 Being and Becoming an Indian in the VaupÃ©s. In Nation States and Indians in Latin America. Greg Urban and Joel Sherzer, eds. Pp. 131 155. Austin: University of Texas Pr ess. 1994 Becoming Indians: The Politics of Tukanoan Ethnicity. In Amazonian Indians: From Prehistory to the Present. Anna Roosevelt, ed. Pp. 383 398 . Tucson : The University of Arizona Press.
258 Joseph, Gilbert M. 1990 On the Trail of Latin American Bandit s: A Reexamination of Peasant Resistance. Latin American Research Review 25(3):7 53. Kerkvliet, Benedict J. 1990 Everyday Politics in the Philippines: Class and Status Relations in a Central Luzon Village. Berkeley: University of California Press. 1993 Claiming the Land: Everyday Politics in the Philippines with Comparisons to Indonesia, Peru, Portugal, and Rusia. The Journal of Peasant Studies 20(3):459 491. Kidd, Stephen W. 1997 Indigenous Organizations and Invisible Whitemen. Paper presented at the Congreso Internacional de Americanistas, Quito, Ecuador. Killick, Evan 2005 Living Apart: Separation and Soci a l i ty amongst the AshÃ©ninka of Peruvian Amazonia. PhD Dissertation, The London School of Economics and Political Science. 2008 Godparents and Tra ding Partners: Social and Economic Relations in Peruvian Amazonia. Journal of Latin American Studies 40:303 328. 2009 AshÃ©ninka Amity: A Study of Social Relations in an Amazonian Society. Journal of the Royal Anthropological Institute 15(4):701 718. 2010 A yompari , Compadre , Amigo : Forms of Fellowship in Peruvian Amazonia. In The Ways of Friendship: Anthropological Perspectives. Amit Desai and Evan Killick, eds. Pp. 46 68. New York: Berghahn Books. Killick, Evan, and Amit Desai 2010 Introduction: Valuing F riendship. In The Ways of Friendship: Anthropological Perspectives. Amit Desai and Evan Killick, eds. Pp. 1 19. New York: Berghahn Books. Korovkin, Tanya 2000 Weak Weapons, Strong Weapons? Hidden Resistance and Political Protest in Rural Ecuador. Journal of Peasant Studies 27(3):1 29. Kraft, Karen, Richard Piland, and Avecita ChicchÃ³n 1997 The Politics of Oil, Conservation and Human Rights in the Southeastern Peruvian Amazon: The Case of Block 78. Unpublished MS, Conservation International Peru Program. Landsberger , Henry A. 1974 Peasant Unrest: Themes and Variations. In Rural Protest: Peasant Movements and Social Change. Henry A. Landsberger, ed. Pp. 1 64. London: Macmillan . La Torre, Lily 1999 All We Want is to Live in Peace: Lessons Learned from O il Operation s in Indigenous Territories of the Peruvian Amazon. Lim a: International Union for Conservation of Nature .
259 LÃ©vi Strauss, Claude 1967 The Social and Psychological Aspects of Chiefta i nship in a Primitive T ribe: The Nambik uara of Northwestern Mato Grosso. In Comparative Political Systems. Ronald Cohen and John Middleton, eds. Pp. 45 62. New York: The Natural History Press. Lyon, Patricia J. 1967 Singing as Social Interaction among the Wachipaire of Eastern Peru. PhD Dissertation, University of Cal ifornia, Berkeley. 1975 DislocaciÃ³n Tribal y Clasificaciones LinguÃsticas en la Zona del RÃo Madre de Dios. In Actas y Memorias, XXXIX Congreso Internacional de Americanistas, Lima, 2 9 de Agosto, 1970. Pp. 185 207. Lima: Instituto de Estudios Peruanos. 1978 The Atackers or the Atacked? The Invention of " Hostile Savages" in the Valleys of Paucartambo, Cuzco, Peru. Paper presented at the International Colloquium on the Defense of the Latin Ameircan Indian Cultures and its Present Projections, Notre Dame, Indiana. 1984 An Imaginary Frontier; Prehistoric Highland Interchange in the Southern Peruvian Andes. Proceedings of the 12th Annual Conference of the Archaeological Association of the University of Calgary, Calgary, 1984, pp. 3 18. The University of Calg ary. McAdam, Doug 1986 Recruitment to High Risk Activism: The Case of Freedom Summer. American Journal of Sociology 92:64 90. 1988 Micromobilization Contexts and R ecruitment to Activism. In From S tructure to Action: Comparing Social Movement Research ac ross Cultures. Bert Klandermans, Hanspeter Kriesi, and S idney Tarrow, eds. Pp. 125 154 . International Social Movement Research: A Research Annual, Vol. 1. Greewinch: JAI Press. McAdam, Doug, and Ronnelle Paulsen 1993 Specifying the Relationship Between S ocial Ties and Activism. The American Journal of Sociology 99(3):640 667. McCallum, Cecilia 2001 Gender and Sociality in Amazonia: How Real People are Made. Oxford: Berg. McCarthy, John D., and Mayer N. Zald 1977 Resource Mobilization and Social Movements: A Partial Theory. The American Journal of Sociology 82(6):1212 1241. MartÃnez, HÃ©ctor 1969 Las Migraciones AltiplÃ¡nicas y la ColonizaciÃ³n del Tambopata. Lima: Centro de Estudios de PoblaciÃ³n y Desarrollo.
260 Melucci, Alberto 1988 Getting Inv olved: Identity and Mobilization in Social Movements. In From Structure to Action: Comparing Social Movement Research across Cultures. Bert Klandermans, Hanspeter Kriesi, and Sidney Tarrow, eds. Pp. 329 348. International Social Movement Research: A Resear ch Annual, Vol. 1. Greenwich: JAI Press . Mintz, Sidney W., and Eric R. Wolf 1950 An Analysis of Ritual Co Parenthood (Compadrazgo). Southwestern Journal of Anthropology 6(4):341 368. Molyneaux, Maxime 2002 Gender and the Silences of Social Capital: Les sons from Latin America. Development and Change 33(2):167 188. Moore, Thomas 1975 Resumen de la OrganizaciÃ³n Social y ReligiÃ³n Harakmbut. Unpublished MS. 1979 SIL and a "New found Tribe": The Amarakaeri Experience. Dialectical Anthropology 4(2):113 125. 1980 Transnacionales en Madre de Dios: Implicancias para las Comunidades Nativas. Shupihui 16:451 463. 1981 SIL and a "New f ound Tribe": The Amarakaeri Experience. In Is God an American? An Anthropological Perspective on the Missionary Work of the Sum mer I nstitute of Linguistics. SÃ¸ ren Hvalkof and Peter Aaby, eds. Pp. 133 143. Copenhagen: International Work Group for Indigenous Affairs. 1983 SituaciÃ³n de los Nativos frente a la MinerÃa AurÃfera en Madre de Dios. Shupihui 28:413 426. 1984a Peru: Manu Nation al Park and Indigenous People Endangered. IWGIA Newsletter 37:126 135. 1 984b Peru: Parks, People, and P e t roleum. Cultural Survival Quarterly 8(1):82 83. 1985 a Informe Preliminar sobre el Impacto de la MinerÃa AurÃfera de las Poblaciones IndÃgenas de Madr e de Dios PerÃº. Unpublished MS. 1985 b Movimientos Populares en Madre de Dios y RegionalizaciÃ³n. In PromociÃ³n Campesina, RegionalizaciÃ³ n y Movimientos Sociales. Maria Isabel Remy, ed. Pp. 165 asas 1987 V Congreso de la FederaciÃ³n Nativa del RÃo Madre de Dios y Afluentes (FENAMAD). Sur: BoletÃn Informativo Agrario 10(108):25 26. 1996 La SituaciÃ³n de los Pueblos IndÃgenas de la Selva Peruana frente a la Pro specciÃ³n/ExplotaciÃ³n de Hidrocarburo s y Recursos Minerales en sus Territorios. Unpublished MS. 2003 La EtnografÃa Tradicional Arakmbut y la MinerÃa AurÃfera. In Los Pueblos IndÃgenas de Madre de Dios: Historia, EtnografÃa y Coyuntura. Beatriz Huertas and Afredo GarcÃa, eds. Pp. 58 90. Lima: I nternational W ork G roup for I nternational A ffairs . 2006 IntroducciÃ³n. In Exploraciones de los RÃos del Sur. Pp. 15 73. Iquitos: CETA.
261 Morin, FranÃ§ oise 1992 Les Premiers CongrÃ¨ s Shipibo Conibo dans le Contexte Politique et Religieux des AnnÃ©es 60 70. Journal de la SociÃ©tÃ© des AmÃ©ricanistes 78(2):59 77. Mueller, Carol McClurg 1992 Building Social Movement Theory. In Frontiers in Social Movement Theory. Aldon D. Morris and Carol McClurg Mueller, eds. Pp. 3 25. New Haven: Yale University Press. Murra, John 1986 Notes on Pre Columbian Cultivation of Coca Leaf. In Coca and Cocaine: Effects on People and Policy in Latin America. D eborah Pacini and C hristine Franquemont, eds. Pp. 49 52. Cambridge: Cultural Survival. Narayan, Deepa 1999 Bonds and Bridges: Social Capital and Poverty. Policy Research Working Paper 2167. Washington, DC: The World Bank. Nash, June 2001 Cultural Resistance and Class Consciousness in Bolivian Tin Mining Communities. In Power and Popular Protest: Latin American Social Movements . Susan Eckstein, ed. Pp. 182 202. Berkeley: University of California Press. Nugkuag, Evaristo 1985 Las Experiencias del Consejo Aguaruna y Huambisa y el Desarrollo de la Provincia de Condorcanqui. AmÃ©rica IndÃgena 5(9):23 31. Osborn, Ann 1 968 Compadra zgo and Patronage: A Colombian Case. Man 3(4):593 608. Overing, Joanna 1988 Personal Autonomy and the Domestication of the Self in Piaroa Society. In Acquiring Culture: Cross Cultural Studies in Child Development. Gustav Jahoda and I. M . Lewis, eds. Pp. 169 192. London: Routledge. 1996 Who Is the Mightiest of Them All? Jaguar and Conquistador in Piaroa Images of Alterity and Identity. In Monsters, Tricksters, and Sacred Cows: Animal Tales and American Identities. A. James Arnold, ed. Pp. 50 79. Charlottes ville: The University Press of Virgina. 2003 In Praise of the Everyday: Trust and the Art of Social Living in an Amazonian Community. Ethnos 68(3):293 316. Overing, Joanna, and Alan Passes 2000 Introduction: Conviviality and the Opening Up of Amazonian A nthropology. In The Anthropology of Love and Anger: The Aesthetics of Conviviality in Native Amazonia. Joanna Overing and Alan Passes, eds. Pp. 1 30. London: Routledge.
262 Overing Kaplan, Joanna 1975 The Piaroa: A People of the Orinoco Basin. Oxford: Clarendon Press. Pacuri, Felipe, and Thomas Moore 1992 Los Conflictos entre Mineros AurÃferos y el Pueblo Arakmbut en Madre de Dios, PerÃº. Unpublished MS, Centro Eori. Passy, Florence 2003 Social Networks Matter. But How? In Social Movements and Networks: Relational Approaches to Collective Action. Mario Diani and Doug McAdam, eds. Pp. 21 48. Oxford: Oxford University Press. Passy, Florence, and Marco Giugni 2001 Social Networks and Individual Perceptions: Explaining Differ ential Participation in Social Movements. Sociological Forum 16(1):123 153. Paul, Benjamin David 1942 Ritual Kinship: With Special Reference to Godparenthood in Middle America. PhD Disseration, University of Chicago. Perreault, Thomas 2001 Developing I dentities: Indigenous Mobilization, Rural Livelihoods, and Resource Access in Ecuadorian Amazonia. Ecumene 8(4):381 413. 2003 Social Capital, Development, and Indigenous Politics in Ecuadorian Amazonia. The Geographical Review 93(3):328 349. Perreault, T homas, Anthony Bebbington, and Thomas F. Carroll 1998 Indigenous Irrigation Organizations and the Formation of Social Capital in Northern Highland Ecuador. Conference of Latin Americanist Geographers Yearbook 24:1 15. Peterson, Nicolas 1993 Demand Shari ng: Reciprocity and the Pressure for Generosity among Foragers. American Anthropologist 95(4):860 874. Plattner, Stuart 1985 Equilibrating Market Relationships. In Markets and Marketing. Stuart Plattner, ed. Pp. 133 152. Lanham: University Press of Ameri ca. Price, David 1981 Nambiquara Leadership. American Ethnologist 8(4):686 708. Putnam, Robert D., Robert Leonardi, and Rafaella Y. Nanetti 1993 Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press.
263 Renard Casevitz, F. M., Th. Saignes, and A. C. Taylor 1988 Al Este de los Andes: Relaciones entre las Sociedades AmazÃ³nicas y Andinas entre los Siglos XV y XVII. Lima: Instituto FrancÃ©s de Estudios Andinos. Renard Casevitz, France Marie 1981 Las Fronter as de las Conquistas en el Siglo XVI en la MontaÃ±a Meridional del PerÃº. Bullet in de l'Institut FranÃ§ ais d'Etudes Andines 10(3 4):113 140. RÃ©nique, Gerardo 2009 Law of the Jungle in Peru: Indigenous Amazonian Uprising against Neoliberalism. Socialism and Democracy 23(3):117 135. Remy, MarÃa Isabel 1994 The Indigenous Population and the Construction of Democracy in Peru. In Indigenous Peoples and Democracy in Latin America. Donna Lee Van Cott, e d. Pp. 107 130. New York: St. Ma rtin's Press. Reyna, Ernesto 1942 Fitzcarrald, Rey del Caucho. Lima: Taller GrÃ¡fico de P. Barrantes C. Ribeiro, Darcy 1970 Os Ãndios e a C ivilizaÃ§ Ã£o: A IntegraÃ§Ã£o das PopulaÃ§Ãµes IndÃ genas no Brasil Mo derno. Rio de Janeiro: CivilizaÃ§Ã£ o Brasileira. Rojas Zolezzi, Enrique 2008 ParticipaciÃ³n PolÃtica de los Pueblos IndÃgenas de la AmazonÃa Peruana. Scientia 10(10):113 138. Rosengren, Dan 1987 Matsigenka Social Organization as Expressed in their Settlement Patterns. In Natives and Neighbors in South America: Anthropological Essa ys. Harald O. Skar and Frank Salomon, eds. Pp. 329 354. Goteborg: Goteborgs Etnografiska Museum. 2003 The Collective Self and the Ethnopolitical Movement: "Rhizomes" and "T aproots" in the Amazon. Identities: Global Studies in Culture and Power 10(2):221 2 40. Rummenhoeller, Klaus 1984 Los Santarrosinos en e l Departamento de Madre de Dios (PerÃº): Apuntes sobre su Desarrollo HistÃ³rico y SituaciÃ³n Actual. MiscelÃ¡nea AntropolÃ³gica Ecuatoriana 4:97 105. 1988 Shipibos en Madre de Dios: La Historia no Escrita. PerÃº IndÃgena 12(27):13 33. Rummenhoeller, Klaus, Clara CÃ¡rdenas, and MarÃa Lazarte 1991 DiagnÃ³stico Situacional de Comunidades Nativas de Madre de Dios: Propuestas para un Autodesarrollo. Lima: Instituto Indigenista Peruano.
264 Saignes, Thierry 1981 El P iedemonte AmazÃ³nico de los Andes Meridionales: Estado de la CuestiÃ³n y Problemas Relativos a su OcupaciÃ³n en los Siglos XVI y XV II. Bulletin de l'Institut FranÃ§ ais d'Etudes Andines 10(3 4):141 176. 1985 Los Andes Orientales: Historia de un Olvido. La Paz: CERES . Sahlins, Marshall D. 1965 On the Sociology of Primitive Exchange. In The Relevance of Models for Social Anthropology. Michael Banton, ed. Pp. 139 236. London: Tavistock. Salazar, Ernesto 1977 An Indian Federation in Lowland Ecuador. Copenhagen: International Work G roup for Indigenous Affairs. 1981 The Federation Shuar and the Colonization Frontier. In Cultural Transformations and E thnicity in Modern Ecuador. Norman E. Whitten, Jr., ed. Pp. 589 613. Urbana: University of Illinois Press. Santos Granero, Fernando 1991 The Power of Love: The Moral Use of Knowledge among the Amuesha of Central Peru. London: Athlone Press. 2000 The Sisyphus Syndrome, or the Struggle for Conviviality in Native Amazonia. In The Anthropology of Love and Anger: The Aest hetics of Conviviality in Native Amazonia. Joanna Overing and Alan Passes, eds. Pp. 268 287. London: Routledge. 2007 Of Fear and Friendship: Amazonian Sociality beyond Kinship and Affinity. Journal of the Royal Anthropological Institute 13:1 18. S antos Gr anero, Fernando, and Frederica Barclay 2000 Tamed Frontiers: Economy, Society, and Civil Rights in Upper Amazonia. Boulder: Westview Press. Sawyer, Suzana 2004 Crude Chronicles: Indigenous Politics, Multinational Oil, and Neoliberalism in Ecuador. Durha m: Duke University Press. Schwartz, Michael, and Shuva Paul 1992 Resource Mobilization versus Mobilization of People. In Frontiers in Social Movements. Aldon D. Morris and Carol McClurg Mueller, eds. Pp. 205 223. New Haven: Yale University Press. Schwar tzman, Stephan, and Barbara Zimmerman 2005 Conservation Alliances with Indigenous Peoples of the Amazon. Conservation Biology 19(3):721 727.
265 Scott, James C. 1976 The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia. New Haven : Yale University Press. 1985 Weapons of the Weak: Everyday Forms of Peasant Resistance. New Haven: Yale University Press. 1986 Everyday Forms of Peasant Resistance. The Journal of Peasant Studies 13(2):5 35. 1990 Domination and the Art s of Resistance: Hidden Transcripts. New Haven: Yale University Press. 1998 Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press. Shepard, Glenn H., Klaus Rummenhoeller, Julia Ohl Schacherer, and Douglas W. Yu 2010 Trouble in Paradise: Indigenous Populations, Anthropological Policies, and Biodiversity Conservation in Manu National Park, Peru. Journal of Sustainable Forestry 29(2 4):252 301. Smith, Gavin 1989 Livelihood and Resistance: Peasa nts and the Politics of Land in Peru. Berkeley: University of California Press. Smith, Richard Chase 1974 The Amuesha People of Central Peru: Their Struggle to Survive. Copenhagen: I nternational W ork G roup for I nternational A ffairs . 1982 The Dialectics of Domination in Peru: Native Communities and the Myth of the Vast Amazonian Emptiness. Cambridge: Cultural Survival. 1985 A Search for Unity Within Diversity: Peasant Unions, Ethnic Federations, and Indianist Movements in the Andean Republics. In Native Peoples and Economic Development: Six Cases from Latin Americ a. Theodore Mcdonald, Jr., ed. Pp. 5 38. Cambridge: Cultural Survival. 1996 Las P olÃticas de la Diversidad: COICA y las Federaciones Ã‰tnicas de la AmazonÃa. In Pueblos Indios, SoberanÃa y Global ismo. Stefano Varese, ed. Pp. 81 125. Quito: Abya Yala. 2002 El Don que Hiere: Reciprocidad y GestiÃ³n de Proyectos en la AmazonÃa IndÃgena. In El Cuidado de los Bienes Comunes: Gobierno y Manejo de los Lagos y Bosques en la AmazonÃa. Richard Chase Smith a nd Danny Pinedo, eds. Pp. 155 179. Lima: Instituto del Bien ComÃºn. Snow, David A., E. Burke Rochford, Jr., Ste ven K. Worden, and Robert D. Be nford 1986 Frame Alignment Processes, Micromobilization, and Movement Participation. American Sociological Review 51(4):464 481. Sno w, David A., Louis A. Zureher, Jr., and Sheldon Ekland Olson 1980 Social Networks and Social Movements: A Microstructural Approach to Differential Recruitment. American Sociological Review 45(5):787 801.
266 Stokes, Susan C. 1995 Cultures of Protest: Social Movements and the State in Peru. Berkeley: University of California Press. Sulmont, Denis 1977 Historia del Movimiento Obrero en el PerÃº (De 1890 a 1977). Lima: Tarea. Taylor, Anne Christine 1996 The Soul's Body and Its Stat es: An Amazonian Perspective on the Nature of Being Human. The Journal of the Royal Anthropological Institute 2:201 215. 2001 Wives, Pets, and Affines: Marriage among the Jivaro. In Beyond the Visible and the Material: The Amerindianization of Society in t he Work of Peter RiviÃ¨ re. Laura Rival and Neil Whitehead, eds. Pp. 45 56. Oxford: Oxford University Press. Tilly, Charles 1978 From Mobilization to Revolution. Reading: Addison Wesley. Torralba, Adolfo 1979 Los Ha rakmbut: Nueva SituaciÃ³n Misionera. Antisuyo 3:83 141. Touraine, Alain 1988 Return of the Actor: Social Theory in Postindustrial Society. Minneapolis: University of Minnesota Press. Tripp, Robert 1995 Diccionario Amarakaeri Castellano. Pucallpa: Instituto LingÃ¼Ãstico de Verano. van den Berghe , Gwendoline, and Pierre L. van den Berghe 1966 Compadrazgo and Class in Southeastern Mexico. American Anthropologist 68(5):1236 1244. Varese, Stefano 2002 Salt of the Mountain: Campa AshÃ¡ninka History and Resistance in the Peruvian Jungle. Norman : University of Oklahoma Press. Veber, Hanne 1998 The Salt of the MontaÃ±a: Interpreting Indigenous Activism in the Rain Forest. Cultural Anthropology 13(3):382 413. Viveiros de Castro, Eduardo 1992 From the Enemy's Point of View: Humanity and Divinity in an Amazonian Society. Chicago: University of Chicago Press. 1993 Alguns Aspectos de Afinidade no Dravidianato AmazÃ³nico. In AmazÃ³nia: Etnologia e HistÃ³ria. Eduardo Viveiros de Castro and Manuela Carneiro da Cunha, eds. Pp. 149 210. Sao Paulo: Universidade de Sao Paulo.
267 1995 Pensando o Parentesco AmerÃndio. In An tropologia do Parentesco: Estudos AmerÃndios. Eduardo Viveiros de Castro, ed. Pp. 7 24. Rio de Janeiro: Editora UFRJ. 1996 Images of Nature and Society in Amazonian Ethnology. Annual Rev iew of Anthropology 25:179 200. 2001 GUT Feelings about Amazonia: Potential Affinity and the Construction of Sociality. In Beyond the Visible and the Material: The Amerindianization of So ciety in the Work of Peter RiviÃ¨re. Laura Rival and Neil Whitehead, e ds. Pp. 19 43. Oxford: Oxford University Press. Von Hassel, Jorge 1904 Ferrocarriles en Algunos Istmos del Oriente. BoletÃn de la Sociedad GeogrÃ¡fica de Lima 14:247 256. Wahl, Lissie 1985 La FederaciÃ³n Nativa del Madre de Dios: Informe de un Congreso. AmazonÃa IndÃgena 5(9):15 23. 1987 Pagans into Christians: The Political Economy of Religious Conversion among the Harakmbut of Lowland Southeastern Peru, 1902 1982. PhD Dissertation, The City University of New York. Werlich, David 1968 The Conquest and Settlement of the Peruvian MontaÃ±a. PhD Dissertation, University of Minnesota. Wilshusen, Peter R. 2009 Shades of Social Capital: Elite Persistence and the Everyday Politics of Community Forestry in Southeastern Mexico. Environmental and Planning A 41:3 89 406. Wolf, Eric 1966a Peasants. Englewood Cliffs: Prentice Hall. 1966b Kinship, Friendship, and Patro n Client Relations in Complex Societies. In The Social Anthropology of Complex Societies. Michael Banton, ed. Pp. 1 22. London: Tavistock. Yashar, Deborah J. 1999 Democracy, Indigenous Movements, and the Postliberal Challenge in Latin America. World Politics 52(1):76 104. 2005 Contesting Citizenship: The Rise of Indigenous Movements and the Postliberal Challenge. Cambridge: Cambridge University Press.
268 BIOGRAPHICAL SKETCH Da nny Pinedo was born in Tarapoto, Peru, in 1969. At the age of three, his family moved to Lima, where he was raised and attended primary and high school. In 1994, he obtained a BA in anth ropology from the Universidad Nacional Mayor de San Marcos. After graduating, he worked for several years on research projects aimed at determining the social and environmental factors that shape the success of initiatives for common pool resource manageme nt in the Amazon region. In 2006 he obtained the Licentiate in anthropology with a thesis on the role of community in the conservation of pastoral commons in a small village in the northern Andes of Peru. In 2008, he entered the graduate program in anthrop ology at the University of Florida , with financial support from a Gordon and Betty Moore Graduate Scholarship for Tropical Forest Conservation. He came back to Peru in 2011 to carry out di ssertation research among the A rakmbut of Madre de Dios. His fieldwo rk was supported by an Inter American Foundation Grassroots Development Fellowship. In 2012, he returned to Florida to write his dissertation with the generous support of the Department of Anthropology and a Bernard Lelong Fellowship. He received his PhD f rom the University of Florida in summer of 2014 .